Predictive power of contrastive analysis : Syrians' learning of the English DP
I This thesis is an investigation of the Contrastive Analysis Hypothesis. This hypothesis is founded on the assumption that second language learners tend to transfer their native language structures when learnIng a, second language. In its strong version, this hypothesis claims that by contrasting two or more languages, it is possible to predict probable areas of difficulty and hence errors on part of the foreign language learner. Contrastive analysis yields two types of prediction: (i) second language learners will transfer their isomorphic Ll structures into the second language and thus produce correct target constructions, and (ii) they will transfer the anisomorphic structures of their mother tongue thus producing erroneous structures which reflect those of their mother tongue. The second hypothesis which I seek to verify in the present work claims that the more advanced the second language learner is, the more successfully he/she will perform in the second language. Long exposure to the new language will enable the learner to improve his/her linguistic competence in this language and as a result, he/she will utilise more positive transfer and less negative transfer than the less advanced learner. The validity of the two hypotheses will be investigated with reference to Syrian learners of English. The two languages under focus are English and Modern Standard Arabic. I focus exclusively on one syntactic structure viz., the noun phrase. My contrastive analysis of English and Modern Standard Arabic noun phrases is based on their description in terms of the general framework of the theory of Government and Binding. The predictions yielded by contrastive analysis were empirically tested by carrying out a small scale empirical investigation which consisted of three tests: a Completion Test, a Translation Test and a Judgment Test. The three tests were administered to two groups of Syrian students studying English at the University of Tishrin, Latakia, Syria. The first group comprised 25 first year students, and the second group included a similar number of fourth year students. I then carried out an error analysis of the data obtained in order to determine the source of each error and separate transfer from non-transfer errors. Quantificational measures were applied to the results in order to determine the relative frequency of each prediction in percentages. iii The degree of success of the predictions were taken as measures for the validity of the hypothesis on which they were based viz., the Contrastive Analysis Hypothesis. In order to verify the second hypothesis, I compared the mean percentages of transfer scored by both groups for each prediction in each test. Conclusions as to whether there were significant differences between the two groups in the degree of transfer were drawn by using the T-Test, which is statistical measure used to assess the significance of the differences between two given average scores.