“Translationese” (and “post-editese”?) no more: on importing fuzzy conceptual tools from Translation Studies in MT research

© 2023 Miguel A. Jiménez-Crespo. This article is licensed under a Creative Commons 4.0 license, no derivative works, attribution, CC-BY-ND. “Translationese” (and “post-editese”?) no more: on importing fuzzy conceptual tools from Translation Studies in MT research Miguel A. Jiménez-Crespo Rutgers University Department of Spanish and Portuguese 15 Seminary Place 5th Floor New Brunswick, NJ 08901 jimenez.miguel@rutgers.edu Abstract During recent years, MT research has imported a number of conceptual tools from Translation Studies such as “translationese” or “translation universals”. These notions were the object of intense conceptual debates in Corpus-Based Translation Studies (CBTS). A number of seminal publications recommended substituting them by less problematic terms, such as “the language of translation” or “typical” or “general features of translated language”. This paper critically analyses the arguments put forward in the early 2000’s against the use of these terms, and whether the same issues apply to current MT research using them. The paper discusses, (1) the impact of the negative or pejorative nature of the term “translationese” on the status of professional translators and translation products (2) the danger of “overgeneralizations” or overextending claims found in specific and very limited textual subsets, as well as (3) the need to reframe the search of tendencies in translated language away from “universals” towards probabilistic, situational or conditional tendencies. It will be argued that MT research would benefit from clearly defined terms and constructs for notions related to specific new variants of translated language. New terms will be proposed, such as “MT translated language” or “the language of MT”, or “general features/ tendencies in MT or PEMT”. 1 A Google Scholar search shows that since the emergence of NMT to date, (2016-2023), 35 papers in MT research use 1 Introduction During recent years, conceptual constructs that emerged in Translation Studies (TS), or more precisely in its sub-branch Corpus-Based Translation Studies (CBTS), have made their way into Machine Translation (MT) research. This paper deals with the recent adoption of the conceptual apparatus of CBTS related to “translationese” and “translation universals” in MT publications.1 Both theoretical constructs received extensive scrutiny in TS in the early 2000’s, primarily in terms of (1) their pejorative or negative connotations that could potentially impact the status of professional translators in academia and society at large, (2) the tendency to overgeneralize results obtained using limited textual subsets given the wide range of text populations, production conditions, language directions, etc., and (3) the need to reframe these “universal” terms towards probabilistic, situational or conditional tendencies. These tendencies could then be framed as more or less likely in certain textual subsets (genres, registers, domains, etc.), translation conditions (professional, non-professional, language combinations, use of technology, modality, etc.). The paper is structured as follows. It first critically analyzes how epistemological discussions evolved in the early 2000’s in TS, as well as the reasons why scholars proposed to abandon these conceptual tools in TS. It will be argued that MT research could avoid unnecessary debates over conceptual, theoretical and methodological issues if different proposed terms in CBTS were …

Content truncated to 500 words