Analyzing AI vs. Human Translation in Translating Metaphors from Persian into English based on Lakoff and Johnson's Conceptual Metaphor Theory

This study aimed to compare the performance of AI and human translators in rendering Persian metaphors into English, focusing on structural, orientational, and ontological metaphors as categorized under Lakoff and Johnson’s Conceptual Metaphor Theory (۱۹۸۰). The research sought to evaluate translation accuracy, highlight the challenges faced in metaphor translation, and identify areas where AI systems could improve. To achieve these aims, the study employed a descriptive, corpus-based, and quantitative methodology. A corpus of ۹۰ metaphorical sentences was compiled from Persian fiction and non-fiction texts, with ۳۰ examples for each metaphor type. These sentences were translated into English using GPT-based AI systems and professional human translators. Accuracy was assessed by comparing translations to the original source text, and statistical analyses, including Chi-square tests, were conducted to identify significant differences in performance across metaphor types. The findings revealed that while AI systems performed relatively well with orientational and ontological metaphors, human translators significantly outperformed AI in translating structural metaphors, which are highly dependent on cultural and contextual nuances. Human translators demonstrated superior accuracy across all categories, underscoring their critical role in handling complex and culturally embedded figurative language. These results have implications for the translation industry, emphasizing the need for hybrid approaches that integrate AI tools with human expertise. They also highlight the importance of improving AI systems’ contextual understanding and cultural adaptation capabilities. In translator education, the findings advocate for prioritizing metaphor translation skills.

کلیدواژه ها:

Artificial intelligence ، Conceptual Metaphor Theory ، Human translation ، Ontological metaphor ، Orientational metaphor ، Structural metaphor