[EN] The Limits of Neural Machine Translation in Translating Qur’anic Arabic into English: A Case Study of Surah al-Kawthar

Abstract

The emergence of Artificial Intelligence (AI) has transformed a wide range of global activities including translation. Neural Machine Translation (NMT) is increasingly used to process large data within short period. However, translating Qur’anic Arabic into English poses some challenges because of the richness of Arabic vocabulary, the complexity of its syntactic structures, the significance of context, and the cultural and religious meanings embedded in the Qur’anic text. The divine nature of the Glorious Qur’an further increases the need for accurate interpretation. This paper examines the limitations of NMT in translating Qur’anic Arabic into English, using Surah al-Kawthar as a case study. The study aims to highlight the linguistic challenges that may affect NMT, and examine the extent to which NMT conveys the meanings of the Qur’anic expressions when compared with selected human translations. The paper is divided into four main parts. The first part introduces the study. The second part examines the limitations of NMT at the level of Qur’anic vocabulary. The third discusses the challenges of NMT at the level of Arabic syntactic structure. The fourth compares NMT for translating Surah al-Kawthar with the four selected human translations in order to identify similarities and differences in translating the three verses. This study adopts a qualitative descriptive design and select (4) human English translations of the Glorious Qur’an to compare between NMT and human translations. The research concludes that NMT cannot replace human translators, but collaboration of both is needed for effective translation, especially when translating divine scripture like the Glorious Qur’an.