SubCom: A parallel corpus of English-Arabic comedy shows subtitles

Abstract

Translating humor presents numerous challenges due to its inherent connection to specific linguistic, interactional, and psychological contexts. This paper introduces the SubCom, a new Parallel Corpus of English-Arabic Comedy Shows Subtitles, designed as an educational resource for translating authentic instances of humor. The researchers experimented with the corpus to develop customized translator training tasks to assess the accuracy of translating verbal irony in a humor translation workshop. These tasks are constructed based on a combination of bottom-up and top-down scaffolding, aligning with a revised model of Bloom’s taxonomy (2001) for effective learning outcomes. Furthermore, the study proposes Mateo’s model of translating irony (1995) as a viable approach for translating verbal irony within the context of comedy shows. Finally, the researchers advocate for using parallel corpora as pedagogic tools in humor translation training workshops and classrooms, as they facilitate trainees’ creativity and enable them to tackle contentious translation issues confidently.