The L2 Acquisition Of English Articles By Chinese Speakers

This dissertation examines the L2 acquisition of English articles, exponents of functional categories DP and NumP, by Chinese-speaking learners. The L2 acquisition of functional categories and their features is receiving considerable attention in the generative approach to second language acquisition (SLA). In terms of the minimalism, functional categories are the locus where language variation resides. Whether functional categories and their features are acquirable determines whether UG is accessible to SLA and how. Specifically, the study investigates the role of the L1 at the initial state, and the acquisition of the semantic and syntactic features of English articles beyond the initial state.Through linguistic analyses of English and Chinese nominal domains, a working analysis is proposed. English has a DP, the maximal projection of NP, with D as the grammaticalization of definiteness. The definite article the is the exponent of DP while the indefinite articles a and ? are the exponents of NumP which encode indefiniteness. Chinese, an article-less language, has a DemP which has the definiteness feature and a ClP with the head Cl encoding the count-mass distinction. When the numeral, the spec of the ClP, is yi'one', it has the [-definite], [±specific] and [singular] features which are the near equivalents of the features of the indefinite article a. Thus, it is hypothesized that the initial state of L2 acquisition is the final state of L1 acquisition. The learners will transfer the features of yi to the acquisition of a and the definiteness feature of the demonstratives to the acquisition of the. It is further hypothesized that due to L1 transfer Chinese speakers will not fluctuate between definiteness and specificity as the Fluctuation Hypothesis predicted (Ionin et al., 2004). For the syntactic features of the articles, it is hypothesized that the learners acquire them incrementally. For the indefinite articles, they first acquire the singular and then the plural features. For the definite article, they first acquire the uninterpretable singular feature and then the uninterpretable plural feature. When there is no one-to-one correspondence between the form and the features, learning challenge will appear. Likewise, when there is no one-to-one correspondence between form and meaning, difficulty will occur. Thus, it is hypothesized that for Chinese speakers the difficulty in the acquisition of the generic feature of English articles lies in the syntax-semantics interface, since the articles have both the semantic features of definiteness and genericity.In order to test the hypotheses, four experiments are conducted. Experiment 1 tests the hypothesis on the initial state. A group of 10 junior high students did a forced choice elicitation task based on Ionin et al. (2004). Experiment 2 tests the FH using the same task as in Experiment 1 on 40 participants, 20 intermediate and 20 advanced. Experiment 3 tests the acquisition of the syntactic features of English articles by another 40 participants, 20 intermediate and 20 advanced, using another forced choice elicitation task. The last experiment tests the acquisition of the generic feature of English articles, using two tasks, one forced choice elicitation task and the other translation task. The following are the major findings.1) L1 transfer does occur at the initial state. The beginners'accuracy on the use of a is as high as about 80% while the accuracy on the use of the is about 60%, just above chance level. Besides, the learners can differentiate between definiteness and specificity, suggesting L1 transfer.2) The intermediate and advanced learners do not fluctuate between definiteness and specificity as predicted by the FH (Ionin et al., 2004), thus confirming the results of Ting (2005). The results suggest that L1 transfer takes place in the acquisition of English articles by the Chinese speakers.3) The learners acquire the syntactic features of English articles incrementally. They first acquire the combination of [±definite, +singular] features and later on the combinations of [±definite, -singular] and [±definite, +uncountable]. The acquisition is made difficult when there is no one-to-one correspondence between the form and the features.4) Another difficulty lies in the syntax-pragmatics interface, since even the advanced learners make errors in the larger situation use of the definite article.5) The syntax-semantics interface is also where acquisition difficulty resides, as the results of the experiment on the acquisition of generic articles suggest. The advanced learners have acquired the zero article with bare plurals to refer to kinds while the intermediate learners are still in the process of acquiring it. The generic use of the definite the and indefinite a is the last to acquire due to the asymmetry in the form-meaning correspondence.The L2 acquisition of English articles by Chinese speakers is a difficult process. The difficulty lies in various components such as the syntax, the syntax-pragmatics interface and the syntax-semantics interface, rather than just one component. The study provides evidence against the current trend in SLA research to attribute the difficulty in L2 acquisition of functional categories to just one factor. Besides, there is not a uniform pattern for L2 learners whose L1s do not have articles in acquiring English articles. Still, the L1 may play a role in the acquisition. The present study has shown that UG is accessible in the L2 acquisition of English articles by Chinese speakers via the role of L1, thus providing evidence to the Full Transfer/Full Access Hypothesis (Schwartz & Sprouse, 1994, 1996).