Research Institute for Translation Studies (RITS)

Proper-Name Allusions in Audiovisual Translation: A Comparative Analysis of Human and ChatGPT Translation Strategies

Document Type : Translation Studies

Authors

1 MA in Translation Studies, Department of English Language and Literature, Faculty of Foreign Languages, University of Isfahan, Isfahan, Iran

2 Associate Professor, Department of English Language and Literature, Faculty of Foreign Languages, University of Isfahan, University of Isfahan, Isfahan, Iran

Abstract
Allusions are culture-specific references that are able to create important challenges in audiovisual translation, particularly in subtitling. This study investigates the translation of proper-name allusions in Persian subtitles produced by a human translator and by ChatGPT using both simple and detailed prompts. Leppihalme’s (1997) model of allusion translation served as the analytical framework. The corpus comprised nine English-language films and one episode of Fargo (2014), with proper-name allusions identified, classified, and analyzed to determine the strategies used by the human translator and ChatGPT. The findings showed that retention-based strategies predominated across all three datasets. The researchers proposed a specific retention type, which was the most frequently used strategy. Human translations employed it more often than ChatGPT, indicating greater attention to audience comprehension. Detailed prompts produced only minor variations, suggesting limited influence of prompt engineering on ChatGPT’s translation strategies. These findings contribute to providing insights into the translation of culture-bound allusions by both human translators and AI systems.

Keywords