Abstract
This study investigates the use of conjunctive markers in translated Arabic texts against the background of corpus-based research on the recurrent features of translated language. Quantitative corpus-linguistic methods were used to investigate a comparable corpus of translated and non-translated Arabic that includes two registers, namely creative fictional narratives and legal writing. The aim of the analysis was to determine whether the independent variables of corpus and register have any significant main or interaction effects on the frequency of a set of conjunctive markers. The findings are interpreted within the framework of four features identified as characteristic of translated language, namely explicitation, normalisation, levelling out, and cross-linguistic influence. The findings confirm that there are differences between conjunction use in translated and non-translated Arabic, providing some support for features such as cross-linguistic influence, explicitation and normalisation, but also indicate that these effects are strongly conditioned by register.