Post-editese: an Exacerbated Translationese
This addresses translation quality assessment for linguists and practitioners, revealing nuanced differences beyond error counts, though it is incremental in scope.
The study investigated whether post-edited machine translations exhibit stronger translationese traits than human translations, finding that post-edited texts are simpler, more normalized, and show higher source language interference.
Post-editing (PE) machine translation (MT) is widely used for dissemination because it leads to higher productivity than human translation from scratch (HT). In addition, PE translations are found to be of equal or better quality than HTs. However, most such studies measure quality solely as the number of errors. We conduct a set of computational analyses in which we compare PE against HT on three different datasets that cover five translation directions with measures that address different translation universals and laws of translation: simplification, normalisation and interference. We find out that PEs are simpler and more normalised and have a higher degree of interference from the source language than HTs.