No description available.

This thesis is a comparative study that investigates English and Kurdish connectives which signal conjunctive relations in online newspaper opinion articles. This study utilises the Hallidayan framework of connectives in light of the principles of Relevance Theory established by Sperber and Wilson (1995). That is, connectives are considered in terms of their procedural meanings; i.e. the different interpretations they signal within different contexts, rather than their conceptual meanings. It finds that Halliday and Hasan’s (1976) classification of conjunctive relations and connectives needs to be modified, in order to lay out a clearer classification of English connectives that could account for their essential characteristics and properties. This modified classification would also help classify Kurdish connectives with greater accuracy. The comparison between connectives from both languages is examined through the use of translation techniques such as creating paradigms of correspondence between the equivalent connectives from both languages (Aijmer et al, 2006). Relevance Theoretic framework shows that any given text consists of two segments (S1 and S2), and these segments are constrained by different elements according to the four sub-categories of conjunctive relations. Different characteristics of connectives are considered in relation to the different subcategories of the Hallidayan framework of the conjunctive relations as follows: additive: the semantic content of the segments S1 and S2; adversative: the polysemy of the connectives; causal-conditional: iconicity in the order of the segments and temporal: the time scenes in the segments S1 and S2. The thesis comprises eight chapters. Chapter One introduces Kurds and Kurdish language, provides the rationale for conducting this project, and outlines the research aims and questions. Chapter Two reviews the existing research on connectives in particular and discourse markers in general. Chapter Three outlines the data and the combined methodology used in the following chapters. Chapters Four, Five, Six and Seven are dedicated to the four subcategories of conjunctive relations and connectives: additive, adversative, causal-conditional and temporal relations respectively. Finally, Chapter Eight reflects on the contribution of the research to the field in terms of findings and methodology and gives suggestions for future research.

No description available.

No description available.

No description available.

No description available.

While it has been recognized that the use of collocations is significant for L2 learners, much research has not been carried out on the knowledge and use of learner's collocations. The present study investigated differences on the knowledge and use of collocations between French and Japanese learners with regard to: 1) L1 influence; and 2) combinability and transparency influence. The test materials included four categories of the lexical collocations: 1) verb + noun; 2) delexicalised verb + noun; 3) adjective + noun; and 4) adverb + adjective. The two types of tasks, Multiple Choice Question Tasks and Translation Tasks, are performed, and the learner corpora are also investigated in order to examine whether the learners from different L1 backgrounds demonstrate different results. Since both French and English belong to IndoEuropean background languages, they share a number of cognate words. Thus, originally it was expected that L1 influence of the French learners would be higher in all of the four lexical collocations than that of Japanese learners, who have non-IndoEuropean backgrounds. Though L1 influence by both French and Japanese learners was demonstrated, the Japanese learners showed a greater L1 influence in the [adjective + noun] category than the French learners. The investigation also found that L1 influence does not necessarily result in accuracy of the collocations. With regard to the combinability and transparency influence, the results of the two types of tasks followed the previous remark made by Kellerman (1978) who argues that L2 learners are unable to transfer words with figurative meaning. However, some contrasted results were also identified in learner corpus investigation. Thus the combinability and transparency influence were not necessarily identified. The results of the present study have a potential to improve teaching/learning of collocations through recognizing the learners' tendencies of learning collocations.

The purpose of this research was to investigate the effects of Processing Instruction (VanPatten, 1996, 2007), as an input-based model for teaching second language grammar, on Syrian learners’ processing abilities. The present research investigated the effects of Processing Instruction on the acquisition of English relative clauses by Syrian learners in the form of a quasi-experimental design. Three separate groups were involved in the research (Processing Instruction, Traditional Instruction and a Control Group). For assessment, a pre-test, a direct post-test and a delayed post-test were used as main tools for eliciting data. A questionnaire was also distributed to participants in the Processing Instruction group to give them the opportunity to give feedback in relation to the treatment they received in comparison with the Traditional Instruction they are used to. Four hypotheses were formulated on the possible effectivity of Processing Instruction on Syrian learners’ linguistic system. It was hypothesised that Processing Instruction would improve learners’ processing abilities leading to an improvement in learners’ linguistic system. This was expected to lead to a better performance when it comes to the comprehension and production of English relative clauses. The main source of data was analysed statistically using the ANOVA test. Cohen’s d calculations were also used to support the ANOVA test. Cohen’s d showed the magnitude of effects of the three treatments. Results of the analysis showed that both Processing Instruction and Traditional Instruction groups had improved after treatment. However, the Processing Instruction Group significantly outperformed the other two groups in the comprehension of relative clauses. The analysis concluded that Processing Instruction is a useful tool for instructing relative clauses to Syrian learners. This was enhanced by participants’ responses to the questionnaire as they were in favour of Processing Instruction, rather than Traditional Instruction. This research has theoretical and pedagogical implications. Theoretically, the study showed support for the Input hypothesis. That is, it was shown that Processing Instruction had a positive effect on input processing as it affected learners’ linguistic system. This was reflected in learners’ performance where learners were able to produce a structure which they had not been asked to produce. Pedagogically, the present research showed that Processing Instruction is a useful tool for teaching English grammar in the context where the experiment was carried out, as it had a large effect on learners’ performance.

No description available.

Although negation is a linguistic universal (Dahl 1979; Horn 2001: xiii), the ways in which it is expressed are highly variable within and across languages (Miestamo 2005; de Swart 2010: 245). This thesis focuses on this variation in English, using corpora of informal conversations recorded in Glasgow (Scotland), Tyneside (North East England) and Salford (North West England) to study three variables: 1. Not-/no-negation and negative concord e.g. I don’t have any money / I have no money / I don’t have no money 2. Non-quantificational never and didn’t e.g. I never saw / I didn’t see that programme last night 3. Negative tags e.g. It’s a nice day, isn’t it / int it / innit? This research aims to bridge the gap between two typically distinct sub-fields of linguistics: variationist sociolinguistics and formal linguistic theory. The investigation draws upon formal theory in (i) defining the linguistic variables and their contexts; (ii) generating hypotheses to test using the spoken data; and (iii) interpreting the results of the quantitative variationist analysis in a theoretically-informed manner. The analysis takes a comparative approach (Poplack & Tagliamonte 2001) to examine whether the factors conditioning negation are subject to regional differentiation. The results demonstrate that all three variables are most significantly constrained by internal factors such as verb type and lexical aspect. Although the relative frequency of variants always differs across geographical space, the underlying system is the same. Discourse-pragmatic factors apply consistently for the not/no/concord and never/didn’t variables, whereas the negative tags are more variable in this regard and are sensitive to social and situational factors. These different strands of evidence together provide support for particular theoretical accounts of how variants are derived from the grammar (not/no/concord) and how they have grammaticalised over time (never/didn’t and negative tags).