From Error Annotation to Quantitative Analysis: Patterns in Russian Language Learning
Résumé
This study relies on error annotation of Russian texts written by French students and collected in the Russian Learner Corpus (RLC, http://www.web-corpora.net/RLC/), and observes error progression in two groups of learners, foreign learners (FLs) and heritage learners (HLs), and across all CEFR language proficiency levels (A1–C2). The errors are annotated based on five linguistic categories: spelling, morphology, syntax, lexis and discourse. Linguistic questions discussed in each category are supported with authentic examples of learner errors. Additionally, possible causes of errors (intralinguistic, interlinguistic, and/or extralinguistic) and the particularities of learner errors depending on the group are briefly discussed. General results reveal that the most problematic category for the group of FLs is syntax, while HLs are most challenged by spelling.
This study demonstrates the value of quantitative analysis of a manually annotated corpus which reveals general trends in the acquisition of Russian and describes learners’ difficulties with specific linguistic categories. Therefore, it might be of interest not only for language instructors but also for linguists and specialists of foreign language acquisition.