Grammar patterns as an exploratory tool for studying formulaicity in English-to-Polish translation: a corpus-based study

In this chapter, we explore the use – and argue for the usefulness – of the concept of grammar patterns (Francis et al. 1996; 1998; Hunston & Francis 2000) in descriptive research on the English-to-Polish translation of formulaic language. Specifically, we use the Paralela English-Polish parallel corpus (Pęzik 2016) to explore – largely in terms of frequency distributions – the use of the Polish equivalents of selected English multi-word items, which are textual manifestations of grammar patterns. As a proof-of-concept, we will focus on a pre-selected grammar pattern (‘it v-link adj to-inf’), where a given word (e.g. the adjective possible) may convey different senses depending on the pattern in which it occurs (Groom 2005). We aim to investigate whether and to what extent somewhat similar lexico-grammatical patterns emerge from the Polish translations under scrutiny. Our analysis finds that the English pattern ‘it v-link adj to-inf’, when filled by adjectives conveying the sense of ‘difficulty’, corresponds to a set of Polish syntagmatic patterns, such as ‘adj v-link, aby’, ‘adv v-link’ or ‘adv’ (the last two followed by verbs in the infinitive form). It is argued that grammar patterns may be a useful starting point for the exploration of formulaicity in translation, but also that they help us explain some more general differences in terms of semantics, pragmatics and usage in source texts and their translations.