From Basic to Independent: Refining an Error Taxonomy for Dutch Word Order from Indonesian Learners at Three Proficiency Levels

Word order errors are one of the most prominent issues in learning Dutch as a foreign language. This is particularly the case for learners whose first language is Indonesian, which has a very different syntactic structure. However, Indonesian learners are rarely the subject of research, and existing error taxonomies classify word order into a single, undifferentiated general category. This study examines the acquisition of Dutch word order among 30 Indonesian learners at CEFR levels A1-B1 who were divided into four groups and took tests at the end of three program phases. This corpus consists of 13,810 words and 3,251 coded clauses. Data were then extracted from 1,300 predefined contexts, including declarative inversion, questions, and subordinate clauses. This study has two main objectives: (1) to describe differences of learners’ word order patterns in clauses with inversion and in subordinate clauses over time, and (2) to evaluate the adequacy of existing taxonomy in capturing the specific error types that occur among Indonesian L1 learners and to refine the taxonomy accordingly. One coder coded all contexts and the intra-coder agreement for a stratified sample of 274 recoded contexts yielded a weighted κ value of 0.85 (bootstrap 95% CI 0.71–0.97). Word order errors were rare (62 of 1,300 contexts, 4.8%). The existing word order category, which emphasizes verb clusters, accounted for only 4 of the 62 errors; 56 errors concerned the position of the finite-verb in main and subordinate clauses; and 2 errors concerned the order of other elements. The refined taxonomy included three subcodes (verb-second order in main clauses, the position of the conjugated verb in the subordinate clauses, and verb clusters with combined middle-position order). No significance tests were run because errors were few and clustered within learners. Teaching and feedback can therefore treat finite-verb position separately from verb clusters.