The Acquisition of Relative Clauses: How Do Second Language Learners of Arabic Do It?

Abstract

The new developments in syntactic theory under Minimalism reconsiders the relation between the language faculty and general cognitive systems whereby language acquisition is accomplished by the interaction of Chomsky (2005)'s three factors: (F1) a minimally specified UG (Genetic endowment); (F2) Primary Linguistic Data (PLD), i.e., input; and (F3) non-language faculty-specific considerations, including principles of efficient computation and principles of data analysis employed in acquisition. Based on this assumption, this study examines the role of economy conditions of (F3) on syntactic derivation and feature interpretability in accounting for the process of second language acquisition by investigating the nature of interlanguage grammars of Arabic-English relative clauses. Arabic and English differ with respect to the operations involved in the derivation of relative clauses, such that, while English uses Move of an operator to generate relative clauses, in Arabic they are base-generated with a resumptive pronoun in the extraction site and are hence derived through (external) Merge. The two languages also differ with respect to the use of overt versus covert relative complementizers and the use of agreement features. Focusing on these three aspects of relativization, I carried out a study with 16 adult English-speaking learners of Modern Standard Arabic (MSA). The participants completed 3 tasks: (1) a grammaticality judgment task, (2) a sentence combination task, and (3) a picture description task. Based on syntactic differences and the cost of syntactic derivations I hypothesized that: (1) individual interlanguage grammars obey the economy conditions of Merge-over-Move and the Shortest Derivation Requirement in the distribution of resumptive pronouns within different types of relative clauses, (2) individual interlanguage grammars would show a systematic pattern in the use of relative complementizer within definite and indefinite relative clauses, and that (3) individual interlanguage grammars would show a systematic pattern in the distribution of agreement features on relative complementizers and resumptive pronouns as the complexity of the relative clause structure increases. The results of the study show that: (1) the distribution of resumptive pronouns in the interlanguage relative clauses systematically varies depending on the extraction site which still obeys to economy conditions of syntactic derivation, (2) individual interlanguage grammars show a systematic use of overt versus covert relative complementizers within both definite and indefinite relative clauses, and (3) individual interlanguage grammars show a systematic use of matching/mismatching agreement features on relative complementizers and resumptive pronouns but the pattern does not appear to be related to the increasing complexity of the relative clause structure. I argue that these results suggest that a minimalist account can be implemented to specify what language features and operations are least accessible to the learner. Distribution of resumptive pronouns is constrained by economy conditions of Merge-over-Move and the Shortest Derivation Requirement. Errors with complementizer's use and agreement are related to feature interpretability and Agree. We conclude that, the general principles of computational efficiency of syntactic derivations are operative on the process of second language acquisition.