The present study explores cross linguistic influence (CLI) on L3 Korean acquisition in Hong Kong higher education through analyzing learner’s particle errors in composition, particularly identifying the specific source of CLI –L1 (Cantonese/Chinese) or/and L2(English) in comparative linguistic perspective. A hybrid research approach is designed. Respondent’s particle errors in text assignment data and survey data are analyzed in descriptive and statistical approach correspondingly. In this thesis, majorities of findings regarding EA of particles are consistent with literature and there are also evidences of CLI on morphology stated in the literature such as inter-language grammars in word-order and semantic equivalence in different syntactic categories between source and target languages have been found in our data. It is argued positive transfer of morphology is possible if the semantic and syntactic function of morphemes between source languages and target language are identical or very similar and concluded that analysis of morpho-syntax would be crucial to identify CLI on Korean particle acquisition. Some of the important findings about the negative relationship between proficiency in source languages and CLI on Korean particle acquisition have been pinpointed. Significantly, the different roles of proficiency in L1 Chinese and L2 English are assumed as factors indicating learner’s error pattern and particle error rates respectively. The final part of the thesis is to apply theory into pedagogy. It is hoped that a tailor-made pedagogy, teaching L3 Korean in Hong Kong higher education where L2 English is teaching medium, will be achieved with the better understandings of learner’ particle errors and their causes. Certain issues of causes of particle errors except CLI are also outlined for further research in this area. / published_or_final_version / Linguistics / Master / Master of Arts

As a result of migratory movements in recent decades, an increasing number of children are educated in a language which is not their mother tongue. For these children, learning to read takes place in their second language. This learning context is made unique by two challenges. First, the learning of the written code occurs simultaneously with the learning of the oral language. Second, no reference to the written code of the maternal language is made available to these children. / This particular learning context can lead to inequity between the second language learner and his or her unilingual peers in terms of phonetic encoding and decoding. Furthermore, the limited vocabulary of a beginning language learner can impede the direct lexical access used when learning to read. / Fourteen students were evaluated for their metalinguistical abilities, lexical and phonic knowledge. Following these tests, an analysis was conducted of student reading errors made in a real reading context. / This descriptive study explores the interaction between reading strategies used by second language learners: bottom-up (word comprehension derived from the context of the text) and top-down (text comprehension derived from word recognition). In addition, this study seeks to describe the linguistic and metalinguistic abilities of these second language students in the process of learning to read.

Within the framework of generative grammar, it is assumed that children are endowed with an innate language faculty called Universal Grammar (UG). Children learn their native language on the basis of the interaction of positive evidence in the input with the principles and parameters of UG. In terms of parameter setting, positive evidence consistent with just one value of a parameter causes the preemption of any incorrect settings hypothesized by the child, in accordance with the Uniqueness Principle (Pinker 1984, Wexler and Culicover 1980) which ensures that only one parameter setting can be held at a time in the child's grammar (Berwick 1985). / This thesis investigates the operation of preemption in parameter setting in second language (L2) acquisition in cases where the learner initially adopts the L1 value of a parameter. Focussing on a parametric difference between French and English, namely, the verb movement parameter (Pollock 1989) which relates to (among other things) the placement of adverbs, 58 grade 5 francophone students learning English as a second language in intensive programmes in Quebec schools were exposed over a two week period to a flood of positive evidence on adverb placement in English--input which was consistent with only the English value of the parameter. The results indicate that the subjects did not reset the verb movement parameter to the English value, suggesting that preemption does not function in L2 as in L1 acquisition. Possible explanations for these results and their implications for theories of L2 learnability are developed.

This study primarily examined the use of second language production as a measure of second language text comprehension in 18 first-year nursing students enrolled in French for Nursing courses (nine low-intermediates, nine high-intermediates). / Although few differences were found between the two levels of proficiency with respect to recall and inference, the high-intermediates were more proficient in reading highly variable material and difficult vocabulary in their second language than the low-intermediates as measured by sentence reading times. The within-subjects results indicate a greater amount of propositional recall of the second language text when it was interpreted in the first language than when it was interpreted in the second language.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.