Abstract:

Cultural phenomenon, which is handled in today’s language teaching approaches to present a new perspective in terms of social, cultural and actional aspects, has become one of the topics frequently discussed in translation studies. As a requirement of translation profession, producing acceptable translations in the target language has made it necessary to acquire both linguistic and cultural skills. Due to the development of technological tools, intercultural communication and individual interaction have begun to be established through translation, and at this point, the use of translation technologies has gained more importance. In this context, the study aims to evaluate the functionality of neural machine translation in transferring cultural elements in Nasreddin Hodja jokes. This study, prepared as a qualitative approach, was carried out by a survey model and its data were collected through purposive sampling technique. As a result of the search with keyword of “Nasreddin Hodja jokes” on the platform Google Search Engine, anecdotes were compiled from the various sources and among them were selected examples in accordance with the classification, prepared by Şanlı (2009) within the scope of cultural elements in Nasreddin Hodja jokes. These examples were translated into French via Google and DeepL Translator programmes using a neural machine translation system, and then the translations were comparatively analysed. The research findings show that material cultural elements need to be properly explained in order to avoid loss of meaning in the target culture readers, while the way of expression of non-material cultural elements needs to be restructured according to the target language rules. It is also seen that some words and expressions were translated out of context. This study has revealed that Google and DeepL systems produced translations in accordance with the syntactic structure of the target language, but did not make semantically and culturally acceptable translations.

Keywords: