Integrating Corpora and AI Tools in the Teaching, Learning, and Assessing of Arabic/English Legal Translation

Abstract The use of corpora and artificial intelligence (AI) tools into education marks a significant evolution in teaching and assessment methodologies. Historically, Arabic corpora have been pivotal in linguistic research, expanding notably through scholarly publications and symposia since the late twentieth century. However, the scarcity of specialized resources such as parallel corpora and the prevalence of web-generated Arabic corpora have restricted thorough empirical studies of authentic Arabic texts and their application in educational settings. Additionally, while the integration of AI into Arabic pedagogy has increased, its application in Arabic/English legal translation remains scarce and underdeveloped. To address this gap, this paper aims to showcase the use of corpus tools and AI to advance research-driven and blended-learning approaches in Arabic/English legal translation. It uses advanced corpus tools and diverse corpora (large datasets) available on Sketch Engine, including the Leeds Monolingual and Parallel Corpora of Arabic and English Countries’ Constitutions and the UN Parallel Corpora . These resources, integral to research and instruction, have supported the MA translation programmes at the University of Leeds for five years. Moreover, ChatGPT 4 is applied as an interactive tool in designing teaching activities around legal documents such as contracts and agreements. This method not only mitigates the lack of specialized corpora in this genre but also explores new possibilities for using AI in education. The paper utilises a descriptive methodology to showcase how monolingual and parallel corpora can enhance interactive legal text learning, highlighting the educational and scientific innovation. Theoretical frameworks like Vinay and Darbelnet’s (1995) contrastive linguistic analysis and Vermeer’s Skopos theory (1978) provide foundational insights for designing and interpreting pedagogical activities involving corpus-based tools. Although the study illustrates the tools’ potential to enrich the learning environment and meet diverse educational demands, it acknowledges AI's limitations in ensuring accuracy and stylistic precision in complex legal texts. The scope of this study does not include an exhaustive evaluation of pedagogical impacts or a comparison with traditional methods, suggesting the need for further research to validate these findings and expand upon them.