Özet:
This thesis investigates the second language (L2) acquisition of the English article system by Turkish learners in order to examine first language (L1) transfer effects. More specifically, the aim is to see whether or not L1 Turkish learners would show any variability in the use of English articles due to their L1. Two groups of learners at different proficiency levels (20 beginner and 20 advanced learners) were tested through three different tasks. The tasks consist of a picture description, a writing, and a fill-in-the-article task. Data were analyzed using the Mann-Whitney tests. Results of the analyses indicated that Turkish learners of English were able to make semantic distinctions among different article environments. They supplied the required articles in obligatory contexts at a higher rate than the other articles. In general, it appeared that the absence of definite articles in Turkish did not pose much problem for L2 learners. Omission errors (i.e., omission of both definite and indefinite articles) appeared mostly in on-line speech and written production tasks, and learners also demonstrated some problems in zero article contexts; they tend to overuse the and a/an in these contexts. Results did not reveal clear evidence for persistent L1 influence in the use of L2 articles. The rate of variable use of L2 articles was lower in the advanced group. This suggests that variable article use in the L2 is not a permanent problem.
