ابراهیم داوودی شریف آباد

English for Academic Purposes (EAP) is derived from the umbrella term ESP, which means to meet learners’ and employers’ needs. It has gained a noticeable role as a language of knowledge dissemination, which has developed into an ongoing need for the improvement of language proficiency among non-English major university students. This has highlighted the significance of EAP courses, which are dominant in the context of Iranian universities. Although there is a reasonable body of contemporary research in the realm of ESP/EAP instruction nationwide, much of the literature in the Iranian EAP setting has been concentrated mainly on quantitative survey-based or experimental studies examining the syllabus and teaching contents with little or no attention to the teaching methodology. Moreover, there is scant research on the body of current EAP literature demonstrating a holistic picture of the running courses in terms of syllabus and methodology in EAP. The present paper aims at shedding light on the existing status of EAP courses for Iranian Islamic students in the tertiary educational system of Islamic universities of Iran through a comprehensive review of some major studies conducted in the related realm. Furthermore, it examines some of the major challenges and criticisms levelled at EAP instruction and EAP courses for Islamic students in Iranian higher education, with a focus on two major dimensions: syllabus and teaching methodology. The paper concludes by giving some practical and pedagogical implications for EAP practitioners, emphasizing the role of teaching methodology, aiming to raise the practitioners’ awareness in promoting the efficiency of EAP courses.

The present study investigated the translation shifts and most frequent types of translation shifts according to Catford’s model in the translation of Sayings 1-30 in the Nahj al-balagha from Arabic into English. The researcher selected the first thirty sayings of the Nahj al-balagha for the analysis. Throughout the analysis, each shift strategy was traced and manually counted to determine the most frequent type of translation shift. The findings revealed that unit shifts were the most frequent type of shift (59.70%), followed by structure shifts (23.30%), then level shifts (10.70%), intra-system shift (4.50%), and finally class shifts (1.50%). Both translators used unit shift and structure shift with the highest frequencies in their translations. The results also indicated that, unlike Sayyid Ali Reza, Sayyid Muhammad Askari Ja’fari went beyond the surface structure and translated words into phrases and explained them in some cases, thus providing an understandable translation. Sayyid Muhammad Askari Ja’fari used unit shifts more frequently than Sayyid Ali Reza. Sayyid Muhammad Askari Ja’fari most of the time used unit shift to convey the message to the target language reader in a comprehensible way. The translator who used unit shift was successful in conveying the message of the source language text to the reader of the target language. Using unit shift made the target language's readers understand the material better.

The extensive training data of ChatGPT has facilitated theorizing in the field of Translation Studies, particularly during the cultural turn. Following this theorizing, a standardization framework was proposed. This study employed a qualitative and interdisciplinary approach, utilizing a descriptive method to interpret the data. The validated questions were based on the works of theorists associated with the cultural turn. Two scholars participated in structured and semi-structured interviews, employing a three-point Likert scale to capture their opinions on the proposed theory. Data analysis focused on the accuracy of ChatGPT-4’s responses in relation to the scholars’ opinions and references. The findings associated with the Likert scale were linked to task-oriented benchmarks, including factual/contextual understanding, coherence, and resolution of ambiguity. The results indicated a 29.3% weakness in ChatGPT-4’s data analysis. The resolution of ambiguity resulted in a total of neutral responses at 44.8%. Scholars unanimously endorsed the cultural interaction theory, demonstrating a 100% capacity for theorizing by ChatGPT-4. In terms of coherence and summarization, the data suggested a stronger correlation with prompt engineering. The potential for theorizing concerning existing theory was found to be 77.7%, while the standard of theorizing by ChatGPT-4 was assessed at 68.9%. Surah Al-Fatiha was selected to exemplify cultural translation according to CIT, illustrating the theory’s effectiveness in translating cultural-ideological texts. A comparison of ChatGPT-4’s translations with those of six Quranic translators underscored its synthesizing abilities, with the incorporation of cultural interaction theory into prompts significantly enhancing its translation skills.

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.

Discourse analysis, a linguistic approach within the social sciences, investigates the relationship between writers' works and their social environments, as well as the impact of these environments on their methods and styles. This research reviews the translation of Imam Ali’s discourse, specifically examining the sermons addressed to the Kufis, through the lens of Fairclough’s (1980) critical discourse analysis model. The study assesses the effectiveness of a selected English translation in conveying the original discourse. The importance of accurately translating the discourse in these sermons cannot be overstated; when translating significant texts, particularly holy texts, a translator’s inability to convey the discourse effectively can result in a loss of the original message, meaning, and spirit. The translation analyzed in this research is Seyyed Ali Reza’s (1980) version of Nahj al-Balagha, recognized as the most accessible and comprehensive translation. By applying Fairclough’s (1980) discourse analysis, the study found that Imam Ali’s discourse features the highest frequency in the description stage at 65%, while the explanation stage has the lowest frequency at 9%. Using Berman’s (2010) translation model, the researchers found that clarification was the most frequently employed strategy in the analyzed translations, occurring 33% of the time. In contrast, the rationalization and qualitative impoverishment strategies were the least used, each at 8%.

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.