المدونات اللغوية لمتعلمي اللغة العربية: نظامٌ لتصنيف وترميز الأخطاء اللغوية "Arabic Learner Corpora (ALC): A Taxonomy of Coding Errors"

The present paper aims to introduce learner corpora and the two only-existing Arabic learner corpora. This paper highlights the enormous potential of this type of corpora for studies of language learning and teaching, such as contrastive analysis, error analysis and the role learner corpora play in learners' dictionaries and instructional materials design. Given the fact that designing a learner corpus requires the compiler to follow particular criteria depending on its purpose, this study summarises such criteria specifically for Arabic learner corpora. These criteria include determining the participants (Arabic language learners), the corpus size, the nature of the materials to be included, the methodology of collecting the texts, the approach employed to mark up the errors, and finally the methods for searching and analysing the corpus. The paper concludes by suggesting new error taxonomy for Arabic learner corpora. This taxonomy has been compared with that Abuhakema et al. (2008, 2009) used. The result reveals that the suggested error taxonomy is more appropriate for errors mark up in Arabic corpora because of its accuracy and comprehensiveness.