This study focuses on limited English students’ cognitive engagement generated through learning interactions with others and tasks under a content-driven and discovery-oriented CLIL learning arrangement in a primary school setting. The main aims are: to explore students’ capability to access and inhabit learning spaces within the L2 mediated discourse; to investigate the potential the CLIL learning experience has to sustain dialogic learning and thinking of a higher order; and finally, to make a contribution to the debate over the potential of the CLIL learning experience to promote deep learning and foster life-long learning competencies. At the heart of its theoretical underpinning lies the idea that cognition develops through the merging of the social with the individual which points towards the need to corroborate a socio-cognitive theoretical framework. Thus, the investigation is conceptualised within the ‘inter-mental development zone’ (Mercer and Littleton, 2007), under the auspices of dialogic learning (Wells, 2001b), and in cognisance of the individual’s active construction of knowledge (Bruner, 1985; van Dijk’s, 2006a). This theoretical line originates in the, now, classic theoretical constructs of Zone of Proximal Development and the More Knowledgeable Other from Vygotsky (1978), and the notion of scaffolding from Wood, Bruner and Ross (1976). With regard to the methodological design, this explorative study falls under the qualitative paradigm within the boundaries of a case study, and corroborates close observations of the learning behaviours with introspective methods. A multilayered analysis is employed which allows elements from various contextual layers and dimensions to enrich the analytical insight. In addition, a fine grained-analysis is pursued regarding the dynamics as well as the substance of the learning events, which conveys a systematic and holistic investigation of the learning phenomenon. The findings of this study suggest that the CLIL approach be regarded as a learning interaction of three foci whereby alongside content-grounded and language-oriented strands run equally well-represented management-of-the-learning strands. Further, it is proposed that a reasonable level of ambiguity stemming from presenting content in the medium of a foreign language and from exposing students to new intriguing facts, prompts cognitive conflict thus giving rise to explorative conversational digressions which bring added cognitive value to the peer-sustained learning interaction. Moreover, this investigation also highlights the complementarity between conversational and instructional units on three levels of cognitive engagement (propositional, linguistic and managerial). Finally, an ability to activate and manipulate different manifestations of knowledge is documented. If this ability would be nurtured long-term, then a disposition for inquiry and criticality as well as an enhanced metacognitive awareness can develop which may translate into skills transferable across the curriculum. Overall, the study recommends the CLIL approach as a rich cognitive medium for learning, and an asset for promoting quality learning with the specification that its implementation needs careful context-bound consideration.

This research is carried out to explore teachers’ perspectives upon their teaching practices with respect to the in-service training program, C.O.T.E. they had been through. Teaching and in-service training have always been my interest area. Besides, after having informal conversations with the administrators from the Ministry of Education and owing to my observations, in North Cyprus the current education system needs to reformed considering main themes such as curriculum, in-service, pre-service training, material and curriculum design, lifelong learning, duration of school hours, financial issues and so forth. In this qualitative research, two major research questions guided this study. As a researcher the constructivist philosophical stance has been employed due to the fact that the purpose is to investigate teachers’ perspectives and in order to achieve this teachers themselves constructed their own perspectives on their teaching practices. Questionnaire, semi-structured interviews, video recording methods were used to gather the data. Questionnaire is used at the beginning of this research to grasp the general picture and at the same time to determine the number of participants. Then, semi structured interviews as pre and post were used to grasp a deeper understanding of teachers’ perspectives and video recordings were used to reflect teachers’ perspectives on their actual teaching practices. I finalized the research study by analysing and triangulating the participants’responses in relation to two major points as follows: o The effectiveness of the programme on teachers’ practice as evidenced through video recordings and discussions of their teaching. o Teachers’ perceptions of C.O.T.E with respect to the effectiveness of the training programme in course design, materials and practical experiences. I ended the research with the voice of the participants by underlying the significance and effectiveness of in-service training programs in general. The limitations of this study with the recommendations for further research are suggested at the end.

In this thesis the concept of Interactive Approach (IA) to the teaching and learning of imaginative texts and language is addressed in the English as a Second Language (ESL) context. As the title suggests the focus of the study was the Integrated English Curriculum (IEC) in Kenya. Although this curriculum was introduced twenty years ago, little has been done to bring about the envisaged integration between the English language and literature at the classroom level. The first chapter describes the background to and the rationale for the study. The second chapter addresses the research setting and explains the current English curriculum in Kenyan secondary schools. The related literature is reviewed in chapter 3 with a special focus on the pedagogical relationship between language and literature and the case for or against integrating the two components. In chapter 4, the methods used during data collection and analysis are presented. The findings of the study as described in chapter 5 affirm that there is the continued use of Traditional Approach (TA) to teaching imaginative texts despite the IEC being in place. This was found to be attributable to the lack of relevant training and the overwhelming focus on examinations by the education system as results from teachers show. The use of the proposed Reader - Text Interactive Approach (RTIA) was positively received by learners as shown by their responses to the interactive exercises. On the basis of these findings, the thesis concludes that there is an urgent need for educators to re-think the way teachers of English are trained in the country and redesign the IEC curriculum materials so that these take into account the integrated curriculum. It is further recommended that RTIA be adopted because the approach was found to have the potential of ensuring that the IEC becomes a reality at the classroom level in Kenya secondary schools.

This thesis explores the implications of high frequency collocation for adult second language learners. It addresses three main questions. First, it asks to what extent high frequency of occurrence in a corpus indicates that collocations are independently represented in the minds of native speakers. A word association study indicates that high frequency of occurrence is a fairly reliable predictor of mental representation, though this methodology does not allow us to determine the precise strength of the relationship. A series of lexical decision studies also show a relationship between frequency and representation, but effects are limited to those collocations which are sufficiently salient to also register as associates. This suggests that psycholinguistic 'priming' models may not be the best way of understanding collocation. Second, the thesis examines the idea that adult second language learners usually fail to retain the collocations to which they are exposed. This is tested through a lab-based training study and a learner-corpus study. Results suggest that adult learners are capable of learning collocations from input, but that 1) the relatively low levels of input to which most learners are exposed mean that they nevertheless tend not to attain native-like profiles of collocation use, and 2) input which provides repeated exposure to collocations can dramatically improve learning. Third, the thesis asks whether a useful pedagogical listing of frequent 'academic collocations' can be compiled. Results suggest that an academic collocation list is viable, but that important caveats need to be made concerning the nature of the collocations included and the range of disciplines for which such a listing will be useful. Moreover, listings of two-word collocations should be seen only as a starting point for more comprehensive phraseological listings. Suggestions will be made for ways in which we might go beyond such two-word listings.

The present thesis is devoted to the analysis of how cognates are processed and represented in the minds of Japanese-English bilinguals. Cognates are an interesting and important category of words in languages as they are distinguished by their similarity across languages, which includes both formal and semantic features. This thesis presents the most comprehensive description and analysis of Japanese-English cognates and how they are processed and represented in the minds of bilinguals. A large number of rating and norming data are presented, which will be of use to researchers in the field of bilingualism who are interested in languages that differ in script, such as Japanese and English. Utilising measures of formal (phonological) and semantic cross-linguistic similarity derived from bilinguals’ ratings, the present thesis presents evidence that cross-linguistic similarity impacts bilingual processing and representation in a variety of tasks, but is modulated by task type and language dominance. The findings of the present study complement previous research, which has often focused on languages that share script (e.g., Dutch-English), while advancing the use of continuous measures of formal and semantic similarity. Such measures are argued to be more appropriate in terms of current cognitive models of bilingual processing and representation. Following a review of previously documented cognitive models, the results are interpreted in terms of the most relevant models that address the issues of cross-linguistic similarity and language proficiency/dominance. The results are important for cognitive science, psycholinguistics and bilingual studies and may also feed into applied linguistics in terms of the potential implications for language learning and teaching.

Much has been claimed recently for the role of idiomaticity in L1 acquisition and fluency and many of these insights have been applied, in my view, uncritically in many cases, to the context of L2 use. Until recently, very little attempt was made to test out the applicability of these insights to English as a Lingua Franca by examining naturally-occurring L2 discourse. This thesis sets out to explore the reasons why even successful L2-users may find the phenomenon of idiomaticity difficult. It investigates the apparent paradox between idiomaticity in L1 use and L2 use, whereby for the L1-user, idiomaticity, in all its guises, makes for ease of processing and the promotion of fluency while in L2 use it seems, in some of its manifestations at least, to be error- prone and elusive. Drawing on an original corpus of spoken English as a Lingua Franca, I apply a combination of corpus techniques and techniques of discourse analysis within a sociocultural framework in order to identify the underlying factors that differentiate L1 and L2 idiomaticity. I illustrate the argument by looking at two different manifestations of idiomaticity: ‘minimal’ units of idiomaticity (two word phrases) and more traditional ‘colourful’ idioms. The results suggest that L2-users avoid or have difficulty with ‘native-like’ idiomaticity because L1 idiomaticity involves more than formulaic sequences of greater or lesser semantic opacity; it is a more extended and diffuse phenomenon that generates subtle webs of semantic, pragmatic and discourse prosodies. It is through these situated webs of signification that L1-users achieve fluency and the promotion of self rather than in the manipulation of isolated idiomatic units in vacuo. Note: When I use the terms ‘native’ and ‘non-native’ I put them in inverted commas to indicate to the reader that I do not subscribe to the deficit view of L2 use that these terms are often associated with. My preferred terms are ‘L1-user’ and ‘L2-user’ (Cook, 2002).

This thesis attempts to do three things. Firstly, it attempts to bring a new contribution to knowledge about fixed idiomatic expressions in English, by demonstrating that the overall meaning of such expressions need not always be conveyed by the presence of lexical words. Linguistic observation of natural and authentic language use has shown the existence of fixed idiomatic expressions consisting solely of grammatical words and possessing a particular overall meaning. Some of these expressions commonly found in everyday informal communication (written and spoken) are "this and/or that", "either or ", "round and round ", "ups and downs ", "on and off', etc. Secondly, the thesis will seek to illustrate through descriptive analysis that fixed expressions consisting solely of grammatical words can be called "units of meaning", using Sinclair's (1991a) position regarding form and meaning. Thus, a part of the thesis will be devoted to investigating the lexico-grammatical behaviour of such expressions. The analysis focuses solely on prepositional clusters, whose frequent usage in informal spoken and written communication makes them suitable for investigation. These prepositional clusters are composed of prepositions or words that can function as prepositions, and formed as a result of the common syntactic patterns in which they occur. Besides analysing cluster patterns that are composed solely of prepositions or words that can function as prepositions, other clusters which are composed of prepositions with adjectives/adverbs and nouns are included in the investigation, for purposes of comparison. Hence, the prepositional cluster patterns analysed in this study are: a) Prep+and+Prep (egs. ins and outs, up and down), b) Prep+Prep (egs. roundabout, upside down, inside out), c) Prep+Adv/Adj (egs. at most, at least), d) Adj/Adv+Prep (egs. excited about, worried about, angry about), e) Noun+ Prep (egs. Reason for, request for, excuse for) and f) Prep+Noun (egs. by mistake, by chance, by coincidence). In examining the lexico-grammatical behaviour of prepositional cluster patterns, I have applied linguistic principles from both Corpus Analysis and Cognitive Semantics. This approach, which combines two fields of linguistics, lends more depth to the analysis. While principles of Corpus Analysis are useful in determining common meaning usages and grammatical functions of prepositional clusters, principles of Cognitive Semantics are able to extend the interpretation of the meaning usages, with regard to metaphoricity. Consequently, I will utilise the principles in both fields to suggest a semantic representation of all the prepositional clusters analysed in the study, based on a superordinate classification rather than on a network one. The third and final part of the thesis seeks to apply the lexicogrammatical findings and the linguistic principles used in the study to pedagogy. More specifically, these findings, together with the linguistic principles of Corpus Analysis and Cognitive Semantics, have been utilised to construct activities which demonstrate a particular ELT methodology, which I have termed Investigative-Oriented Learning (IOL). IOL is meant to address the limitation of Communicative Teaching in developing investigative questioning in language learners. The aim of IOL thus is to empower learners with skills of Conscious Investigation which may enable them to be sensitive to patterns of language, and to their idiomatic and metaphorical meanings and grammatical functions. Prepositional clusters, which illustrate idiomaticity and metaphoricity in authentic language use, have been used as an example of language patterns to illustrate the methodology behind IOL.

This mixed-methods investigation reports on the under-researched issue of silence within Japanese university second language (L2) classrooms. An extensive, multi-site study using a structured observation methodology was employed to investigate the classroom behaviour of over 900 language learners across nine universities in Japan. To effectively measure the extent of macro-level silence within their classrooms, an original observation instrument called the Classroom Oral Participation Scheme (COPS) was specially developed for the task. A total of 48 hours of data was collected using a minute-by-minute sampling strategy which resulted in some startling results. Learners were found to be responsible for less than one percent of initiated talk within their classes, while over a fifth of total class time observed was characterised by no oral participation by any participants. Complementing the COPS' quantitative evidence of a robust national trend of silence in Japan's universities, a parallel qualitative phase of the investigation gave students a voice about their silences by drawing on over seventy-thousand words of transcribed data collected during a series of semi-structured interviews. This phase of the research provided a valuable individual-level analysis of learners' fundamental beliefs about and personal experiences of not speaking in L2 educational contexts. The final phase of the project adopted an event-specific focus on classroom silence by utilising a stimulated recall methodology to uncover what students were actually thinking and feeling whilst silent episodes were in progress during lessons. Using Dynamic Systems Theory (DST) as its conceptual background, the investigation moves away from reductionist, single-cause explanations for learner reticence to suggest that silence actually emerges through multiple, concurrent routes. These routes (termed attractors in DST) are so abundant, and appear to be so well supported both educationally and culturally in the Japanese context, that silence has fossilised into a semi-permanent attractor state within university language classrooms.

This thesis is a corpus-based investigation from a pragmatic perspective or the phenomenon of idiomatic creativity. The ultimate aim of the thesis is to provide comprehensive empirical evidence of the intertwining relationships between different factors of idiomatic creativity and the effects of such relationships on the reassessment of the determinants of idiomatic creativity in context-specific communicative events. It has often been suggested in the psycholinguistics tradition that the semantic transparency of some idioms facilitates the creative manipulations of such idioms in discourse, while the lack thereof hinders creativity in some others. In other words, the internal characteristics of idioms, including compositionality, analysability and motivation, which contribute towards transparency, arc postulated to determine idiomatic creativity. While theoretical arguments and illustrative examples appear to support such a hypothesis, empirical validation of the hypothesis using large bodies of authentic data is still required for it to gain acceptance in the linguistic community. In the first half of the thesis, therefore, quantitative analyses of corpus data and statistical tests arc carried out to empirically validate the psycho-linguistic hypothesis concerning the semantic determinants of idiomatic creativity. It is found in the analyses and tests that there are significant trends in the data that point towards the prevalence or transparent idioms over opaque ones in terms of both the range and the frequency of their creative manifestations in authentic English discourse. In the second half of the thesis, with a view to flesh out the otherwise decontextualised and superficial results from such quantitative analyses, qualitative analyses of individual instances of idiom variants in the data are carried out, thus elaborating on the varying degrees or importance of the underlying factors of idiomatic creativity and, notably, their interaction with one another in specific contexts. The results suggest that, together with idioms' semantic characteristics, external factors, including context, cognitive constraints and phraseological constraints, also contribute to idiomatic creativity. It is further revealed that the roles of the facilitating and constraining factors of idiomatic creativity vary according to the particular context-specific types of variant, in which context has the overarching power to overrule certain constraints as well as to disambiguate unusual creative uses. As such, the true functional profile of idiomatic creativity is argued to be best described and predicted in relation to context. A context-based model of idiomatic creativity in authentic English discourse is therefore proposed in this thesis in the hope that it will contribute to the existing literature on idiomatic creativity and offer a better understanding of the possibilities and constraints on the phenomenon in real life discourse.

Speechreading is a complex skill affected by both the observer's method of extracting visual speech information and talker-specific variation in speech production. This thesis focuses upon accent, a factor that can influence both an observer's viewing strategy and talker speechreadability. Auditory research demonstrates that an unfamiliar accent reduces speech intelligibility. The primary aim here was to determine whether accent type, familiarity or variation would alter visual speech intelligibility with consequential effects upon speechreading performance. Experiments 1 and 2 considered visual discrimination of native and non-native accented speech and the influence of non-native accent upon speechreading performance. Results indicated that observers were able to utilise visual cues for discrimination and were significantly poorer at speechreading a non-native accent. Experiments 3, 4 and 5 examined the influence of regional accent on speechreading performance. Results indicated that visual speech performance was significantly worse for Glaswegian-accented talkers than for talkers with a Nottingham accent. However, no clear advantage for accent familiarity was found. Experiment 6 examined the influence of accent type and talker variability upon speechreading performance. Accent type was consistently the dominant influence upon speechreading performance, above familiarity and variation. Experiments 7, 8, 9 and 10 examined the influence of exposure, context and repetition upon the effects of a Glaswegian accent. Here, the effect of the Glaswegian accent on talker speechreadability was reduced by context and repetition, but not removed entirely. In conclusion, while visual accent type mostly determines visual speech intelligibility, accent familiarity mostly determines auditory speech perception. Although spoken accent effects can be quickly reduced through exposure, no such effect was found here in the visual modality. Both context and repetition were necessary to improve the intelligibility of accented speech. This indicates a potential difference in the processing of accented speech across the two modalities and has implications for speechreading training.