The Acquisition of Spanish Differential Object Marking in Chinese-Speaking L2 Learners

In Spanish, animate and specific direct objects are marked with the preposition a, whereas other direct objects are not usually marked. This phenomenon can be found in many languages and has been studied by previous research under the term of differential object marking (DOM) (e.g., Bossong, 1995; Aisson, 2003). Spanish DOM is acquired by the age of 3;0 in monolingual speakers (Rodríguez-Mondoñedo, 2008) but it is challenging for Spanish heritage speakers and L2 learners (Cuza, Miller, Pérez-Tattam, & Ortíz-Vergara, submitted; Guijarro-Fuentes & Marinis, 2007, 2009; Montrul 2004; Montrul & Sánchez-Walker, 2013). Spanish DOM exists in both simple sentences and in clitic left-dislocated sentences (CLLD) where the direct object is moved to the topic position of a sentence and a clitic is used to double the direct object and check the accusative case. Previous research has focused on influential factors affecting this process, including cross-linguistic influence, structure complexity, and proficiency. Most of these studies, however, have examined the acquisition of DOM among learners with Indo-European languages as L1. L2 learners of Mandarin, a typologically distanced language, have not been investigated. The present study covers this gap in the literature by examining the acquisition of Spanish DOM among twenty-one (n=21) Chinese-speaking L2 learners of Spanish in North China. The learners were divided according to their proficiency level into low, intermediate and advanced speakers. They completed an Acceptability Judgment Task (AJT), testing DOM intuition in simple and CLLD contexts, and an Elicited Production Task, testing the target use of DOM in simple and CLLD contexts. Results showed that the learners showed some knowledge of the [animate] constraint of Spanish DOM but their performance was still different from the monolingual controls. In the AJT, they accepted grammatical sentences as the controls but they hesitated in judging the ungrammatical sentences. In production, they produced less DOM in simple sentences and more overextension in CLLD sentences than the controls. Both L2 learners and controls omitted the marking in CLLD sentences. The advanced learners patterned as the intermediate learners but performed more target-like then the low proficiency learners. The intermediate learners showed more variability than the advanced learners did. These results provide evidence on the effects of the three factors mentioned, however, the effects of the influential factors are weakened to different degrees in different acquisition stage. I argue that the learners were at different stages of acquiring Spanish DOM, and that Feature Reassembly Hypothesis (Ladiere, 2008, 2009) and Multiple Grammars Theory (Amaral & Roeper, 2014) can better account for these findings