Abstract Syntactic Representation in a Second Language: Processing and Acquisition of Wh-dependencies and Definiteness in L2 English

Abstract

This thesis investigates the second language (L2) processing and acquisition of English wh-dependencies and definiteness. Two studies were conducted to test whether adult L2 learners can process and acquire L2 properties that are not present in their first language (L1). The first study replicates Canales (2012) to compare L2 real-time processing of filler-gap dependencies in English wh-sentences by speakers of two typologically different languages, Jordanian Arabic and Mandarin, which both lack wh-movement. The results show that the L2 participants can process filler-gap dependencies incrementally in real-time and provide evidence that L2 processing exploits the same syntactic knowledge (wh-constraints) as L1 processing. These results challenge the predictions of the Shallow Structure Hypothesis (Clahsen & Felser, 2006, 2018) which states that adult L2 learners are ‘shallow processors’ who rely less heavily on morpho-syntactic knowledge during real-time processing than on lexical semantic knowledge.
The second study investigates the acquisition of definiteness in English relative complementizers by L1 Jordanian speakers. The appearance of relative complementizers in Jordanian Arabic is the phonological reflex of the [+definite] feature of the head noun, unlike English relative complementizers which are not specified for definiteness. This study examines whether adult L1 Jordanian Arabic L2-English speakers will transfer the [+definite] feature to English relative complementizers, by investigating their acceptance of null and overt relative complementizers in definite and indefinite English relative clauses. These combinations are all grammatical in English, but in the participants’ L1, null complementizers are incompatible with a definite marker and overt complementizers are incompatible with indefiniteness. The results show evidence of L1 transfer since the L2 participants had significantly lower ratings for definite relative clauses with a null complementizer and indefinite relative clauses with an overt relative complementizer. Further, the size of this apparent L1-transfer effect was bigger in participants with lower L2 English proficiency but attenuated with increased proficiency, suggesting that the higher proficiency participants were able to acquire the target representation of definiteness with relative complementizers. These results are compatible with the Feature Reassembly Hypothesis (Lardiere, 2008, 2009) which argues that learners can acquire L2 features that would be incompatible with their L1 features.
Overall, the results of both investigations do not support models of L2 processing and acquisition that propose shallower syntactic representation (the Shallow Structure Hypothesis). Instead, they support models which argue that adult L2 learners can acquire full syntactic processing and representation of L2 properties that are absent in their L1 (e.g., the Feature Reassembly Hypothesis).