In this paper, we discuss the possible composition, functions and aims of a parallel Scandinavian-Polish and Polish-Scandinavian text corpus. We begin by showing how different fields such as translation studies and contrastive grammar have used translations as study material to gather and interpret data, and what one can learn from those studies. Afterwards, we focus on the linguistic topics that so far have been studied based on literary translations in the language pairs Polish-Swedish and Polish-Norwegian. Our main resource here are bachelor and master theses written at Adam Mickiewicz University in Poznan between 2010 and 2016. We discuss both the composition of the corpora used, noting the asymmetries in the genres and source languages (narrative vs. documentary prose; Swedish texts in Polish translations and Polish texts in Norwegian translations) and examples of the use of corpora. The case
studies chosen here include one oriented towards strictly grammatical phenomena (conjunctions in Swedish and their Polish counterparts), one oriented towards the semantics of the source language (discourse particles in Norwegian) and one focusing on stylistics of a given author (typical features of Czesław Miłosz’ prose). All case studies have been prepared by manual search of accessible texts, rendering them more modest than they could otherwise have been, had the students had access to digital corpora. In the final part of the paper, we analyse the important features of a parallel Scandinavian-Polish text corpus, including choice of texts, minimal size of the corpus
and types of functions necessary. We argue that such a corpus would prove valuable for linguists, translation researchers and interpreters.
