Abstract :
[en] Mandarin Chinese chengyu, fixed non-compositional four-syllable phrases genetically intertwined with their nation’s culture (Chen 1999, p. 122) which have been considered as the most representative type of Chinese idioms (Conti, 2019; 2020, p. 412), have long constituted a major hindrance for translators, causing linguistic issues and cultural tensions. They represent an even bigger worrying matter for translation educators (Lu, 2017) who have until now been deprived from suitable tools to tackle this challenging task. Indeed, chengyu, although commonly used in all types of Chinese discourse, have so far been almost disregarded by translation studies and translation education experts (Conti, 2020, p. 412 & Guo, 2017, p. 101).
Therefore, one might wonder if technological innovations such as AI and CAT tools may help extricate from this dilemma. Therefore, this paper will focus on whether and how the convergence of CAT tools (Google Translate, DeepL, Bing Microsoft Translator, 百度翻譯 Baidu Fanyi, 雲譯 Cloud Translation), AI softwares (ChatGPT) and parallel corpora (OPUS2 corpus available on SketchEngine) can help translators and translation trainees render better translations of these idioms, consequently contributing to improving translation techniques and thus transcending linguistic and cultural borders.
In a case study, we will use a parallel corpus (ZH-FR) composed of 魯迅 Lu Xun’s short stories contained in the collection 《吶喊》 Nahan, and of Sebastian Veg’s 2010 controversial French translation, Cris. Previously extracted chengyu will then be randomly selected, and their translations by the aforementioned tools will be analysed and compared to the French version. This multi-modal approach will contribute to a better understanding of AI and CAT tools applications to both lexicography and translation training.
