The present study investigates how English learners of Spanish compute gender agreement in Spanish–English asymmetrical switches (i.e. el key vs. la key ‘the.masc/fem key’). Twenty-six English learners of Spanish at intermediate-to-advanced second language (L2) proficiency completed a forced-choice elicitation task involving two codeswitching environments: Spanish determiner–English noun switches (Task 1) and English–Spanish switched copula constructions (Task 2). English nouns occurring in these codeswitching environments are controlled for semantic gender in the case of human nouns and for the grammatical gender of the Spanish translation equivalent in the case of inanimate nouns. The study also explores whether L2 proficiency and codeswitching experience influence L2 learners’ gender decisions in codeswitching. Gender agreement was almost categorical with human nouns whereby female nouns are feminine and male nouns are masculine. A mixed-effects logistic regression analysis further revealed that participants were more likely to apply masculine gender to inanimate nouns in both tasks, while the majority of the participants performed below chance with feminine gender. We stress the importance of separating animate from inanimate nouns in experimental L2 research. We argue that L2 learners’ preference for masculine gender in codeswitching is a reflex of a probabilistic default gender driven by the L2 grammar, rather than evidence for the retrieval of the grammatical gender of the translation equivalent.
Read full abstract