Abstrakt
This study aims to design corpus-informed teaching materials for advanced Czech students of English, reflecting differences in native as opposed to non-native phraseologies. Specifically, I investigate phraseological patterning involving prepositions.
The data comes from SYN2015 (Czech) and BNC (English; 100 mw each). A list of the 10 most frequent prepositions was compiled for either corpus.
In this work-in-progress report I focus on the corresponding preposition pair in - v. I extract 3-5-grams containing the preposition in any slot.
The resulting n-grams are examined to identify prepositional patterns. These are classified by their semantics and textual functions, and compared between Czech and English.
Results suggest that in/v patterns mostly fulfil corresponding functions in the languages compared, although the distribution of these functions differs. Some pattern classes are only found in English, highlighting its analytic nature as opposed to Czech.
Further, while some functional-semantic pattern classes comprise a diverse set of expressions (e.g. adverbials of place), others are limited to few patterns (e.g. emphasisers), suggesting that the corresponding functions are associated with more conventionalised forms of realisation.