L2 acquisition of Mandarin zai and -le

Abstract Although the Aspect Hypothesis has been tested in many European languages, it has not been investigated extensively in Chinese. The present study tested the Aspect Hypothesis in relation to two predictions: the Association Prediction, which predicts that perfective aspect (in Chinese, –le) will be associated with telic verbs and progressive aspect (zai) with activity verbs, and the Developmental Prediction, which predicts that such associations will be stronger at early stages of development. The study employed a controlled experiment, which elicited learners’ judgments on perfective –le and progressive zai in obligatory, incorrect, and optional contexts. The results show that the Association Prediction is only partially supported and that the Developmental Prediction is not supported, in that higher-level learners associate lexical aspect more strongly with the grammatical aspect marker. The results are more consistent with the Default Past Tense Hypothesis (Salaberry 1999. The development of past tense verbal morphology in classroom L2 Spanish. Applied Linguistics 20. 151–178), which we propose to be extended to the Lexical Insensitivity Hypothesis.