فاطمه رجبی فخرآبادی

The Qur’ān is not only rich in linguistic construction but also deeply embedded with cultural and religious meanings, making its translation a challenging task. Among the many intricate linguistic features of the Qur’ān, proverbs, especially hidden proverbs, present one of the most significant challenges for translators. These proverbs carry culture-specific meanings that are often difficult to render accurately in other languages and are hidden due to their indirect nature, which can be challenging for readers to understand (Esmaeili, 1986). This study aims to identify and analyze hidden proverbs in the Qur’ān through translations generated by artificial intelligence (GPT-3.5) and by a human translator, Qarai’s English translation. The research adopts a descriptive-qualitative method, analyzing twenty-one Qur’ānic verses containing hidden proverbs. These 21 hidden proverbs were extracted from a technical source on Qur’ānic proverbs (Esmaeili, 1986). The study evaluated the translation strategies used by both GPT-3.5 and Qarai, examining how effectively each approach conveys the figurative and cultural meanings of the proverbs. Through comparative analysis, this research investigated the strengths and limitations of AI-generated translations versus human translations in handling cultural-specific linguistic elements. The findings reveal that GPT-3.5 demonstrated strong performance in applying Beekman and Callow’s (1974) translation strategies, particularly in cases where the source language proverbs were untranslatable and required a non-figurative explanation. Compared to the human translator, GPT-3.5 provided more consistent and contextually appropriate solutions to the challenge of translating hidden proverbs. These results highlight the growing potential of AI-assisted translation tools in addressing complex linguistic and cultural challenges, suggesting promising advancements in the field of Qur’ānic translation.

Linguistic issues are important in the textual analysis of translated texts. Among the most sensitive and significant texts translated into different languages, the Holy Qur’ān stands out. The text and texture of the Qur’ān are so unique that one cannot easily understand it without prior knowledge of its linguistic and extralinguistic aspects. One of the most challenging linguistic issues in the Qur’ān is proverbs, especially hidden proverbs that carry culture-specific meanings. The translator’s role in explicating the meanings of these culture-specific items is crucial. This research aims to identify and analyze Qur’ānic hidden proverbs using a technical reference (Esmaeeli, 1986) and to assess translation quality with Na Pham’s (2005) translation quality assessment model. In this study, two translation forms, AI and human (Qarai), were compared for their treatment of Qur’ānic hidden proverbs. Data collection and analysis followed a descriptive-qualitative design. Twenty-one verses containing hidden proverbs and their translations by GPT 3.5 and Qarai were analyzed. The study results indicated that, in terms of translation quality, GPT 3.5 performed better than Qarai.