Abstract
The study explores the potential of quantitative methods to shed light on how texts originally written in English (EO) and texts translated into English (ET) from Norwegian cluster in terms of functional classes. The object of study are sequences of three words (3-grams), classified into 15 functional categories. The investigation establishes that EO and ET do not differ significantly in half of the categories. As for the categories that do differ, two (Comparison and Spatial) are investigated in more detail, uncovering that the more frequent use of Comparison and Spatial 3-grams in ET is most likely a result of source language shining through. The findings are important in the context of both descriptive translation studies and translation-based contrastive studies. With regard to the former, the current study shows that, in many cases, ET does not seem to constitute a ‘third code’ at the level of 3-gram functions, since the same functions are equally attested in EO. As far as contrastive studies are concerned, the investigation reveals few, if any, lexico-grammatical differences between EO and ET that overturn the belief that translations are a good tertium comparationis when comparing and contrasting language systems.