Abstrakt
This pilot study aims to identify differences in native and non-native phraseologies, focussing on prepositional patterns. Previous research suggests L2 users' limited phraseological choices may hinder the accuracy of their language production, and prepositions can pose a particular challenge to Czech learners of English, given the lack of correspondence between translation equivalents.
Further, prepositional patterns contribute to text structuring, making them an important part of learners competence. Using representative corpora of English and Czech, 3-to 5-grams containing the equivalent preposition pair in/vare extracted.
The identified patterns are classified by their semantics and textual functions. While in/v patterns mostly fulfil corresponding functions in the languages compared, the distribution of these functions differs.
Specifically, some pattern types are only found in English, highlighting its analytic nature as opposed to inflectional Czech.Keywords:n-grams, prepositions, native and non-native phraseology, typologically distant language pair, Czech/English1.IntroductionThis study is based in cross-linguistic distributional (Granger and Paquot, 2008)or data-driven (Grangerand Meunier(eds), 2008)phraseology, i.e. examining recurrent word combinations through corpora. It was prompted by earlier findings provided by research into non-native phraseology (Ebeling and Hasselgård, 2015; Granger, 2017; Granger and Bestgen, 2014; Hasselgård, 2019; Vašků, Brůhová, and Šebestová, 2019), as well as by the interest in -and need for -teaching materials reflecting those findings (Reppen, 2011).
It is conceived as a pilot study, aiming to contrast a selected pattern group -prepositional patterns -between the typologically distant language pair of Czech and English. The results of this contrastive analysis can then be used as a springboard towards suggesting how n-gram based studies of phraseology can inform foreign language instruction.