The research presented in this thesis investigates the feasibility of a corpus-influenced syllabus for primary literacy education in Japan. It achieves this with reference to two aspects of the context within which such an initiative might be developed. One is the cultural context; that is,the demands of primary ELT in Japan. Therefore this research explores policy makers’ and teachers’views, the texts frequently used in primary ELT classrooms, and some aspects of teacher training. The other focus is from a linguistic viewpoint, concerned with the identification of linguistic features which pupils need to learn for the development of their English literacy. This thesis describes an innovative method for identifying such features. The cultural context was investigated by means of three surveys, the first of which was used to inform the choice of texts to include in the corpus. The surveys reveal a lack of attention to literacy teaching and teacher education in primary ELT in Japan, but also point to some potential for syllabus development. The research offers support for a corpus-influenced syllabus for teaching English literacy, while concluding that there is a need for incorporating it into teacher education and developing teaching methodologies which suit the pedagogic context of the Japanese primary school classroom.

The interface of corpus linguistics and second language writing has led to extensive corpus-based research focusing on a description of academic writing. The overwhelming majority of this research, however, has focused on scholarly writing, which may not be a valid model for novice writing. This thesis proposes the teaching of second language writing should be informed by a staircase model of writing progression which aims instruction at the level of student writing. For English for academic purposes writers in intensive English programs, this target is first year undergraduate writing, specifically freshman composition as it is taught in North American higher education contexts. This study specifically compares the frequency of the noun phrase in freshman composition writing and scholarly writing with two main aims: to provide empirical evidence of the differences between the two levels of writing and to contribute to a description of freshman composition writing. The findings from this comparison clearly demonstrate that noun phrases in both levels of writing employ a discernible pattern, and there are distinct differences between those patterns. A critical need form pedagogical materials to focus more on phrasal structures in general, but especially nounphrases, is evident.

Researchers in corpus linguistics and applied linguistics have recommended the use of corpus data by language learners to promote independent learning (Bernardini, 2004; Yoon & Hirvela, 2004; O’Keeffe et al, 2007). However, it is not clear to what extent learners are able to use corpus resources independently, and how they can be trained to use a corpus more effectively. This thesis reports a study of learners using a corpus for error correction. The learners recorded their processes using a think-aloud protocol. The thesis records three main findings. Firstly, the learners found it easiest to spot and correct errors of clause structure, noun class, adjective pattern, and collocation; they found verb pattern the most difficult errors to correct. Secondly, the learners most frequently searched for information about colligation, collocation, acceptability/occurrence of strings in a corpus, and determiner-noun agreement; they searched for information about lexical pattern relatively infrequently. Finally, the learners worked most effectively with the corpus when they entered single words as the search terms and scrutinized the concordance lines for collocates and patterns; they worked least effectively with the corpus when they entered whole strings of words. The thesis also makes recommendations for facilitating corpus use in classrooms and specifies the training that learners need to use corpora effectively.

This study is a classroom-based, mixed methods study into the influence of transcribing, reporting, and task repetition on in class student oral task performances. The study investigates two questions. First do students in an intact classroom improve task performances when they repeat the same task in subsequent performances? In a previous exploratory study by Moser (2008) students did not take advantage of task repetition opportunities to improve a repeat task performance. It was concluded that the reason for this was that amongst many students there was a lack of perceived pedagogical rationale for task repetition. On this point and more specifically the study investigates does a more transparent pedagogical focus realized through a transcribing phase or a reporting phase prior to a repeat task performance result in improved subsequent task performances. Related to this, and the second question of this study, is does the more intensive transcription work result in improved task performances than the reporting work? The results of the study reveal no significant difference between transcribing or reporting on subsequent task performances; however, there was significant results for a task repetition effect on task performances. The classroom implications of these findings will be discussed.

The study of metaphor has enjoyed a great deal of interest in recent years. Far from being considered a mere rhetorical or poetic device, metaphor has now been shown to play a fundamental role in human language and cognition. However, despite its prominence and utility in discourse, learners of English have been shown to struggle with both the production and comprehension of metaphor. The concept of ‘metaphoric competence’ should thus be considered an important aspect of language teaching and learning, but its definition and measurement remain problematic. This thesis uses an investigation of metaphor use in the written examinations of French and Japanese learners of English to address four main areas. The first investigates the development of metaphor use across different levels of written language. The second seeks to exemplify the way in which metaphor use is related to other aspects of language in this context, particularly lexis and phraseology, while the third explores the functions learners use metaphor to perform. The fourth area draws together these insights to explore what ‘metaphoric competence’ might mean in the context of learners’ exam-based written language, and how it might be measured.

Often when teaching oral communication, great emphasis is placed on the use of target English language only in the classroom. Reasons often given to defend this policy include the use of L1 in learning English causes unwanted language interference and extended “thinking-time” slowing down a conversation. However this may not be the best policy when producing L2 writing, particularly in the early planning stage where the use of L1 might in fact reduce cognitive loads on L2 writers especially if the topic of the writing is linked to a writer’s L1 and may be best recalled in L1. This PhD study explores the questions and reservations regarding the optimum methods of planning an English essay by Japanese writers of L2 English, both in the UK and in Japan, at intermediate and advanced proficiency levels, with particular focus on the variables of language of planning and topic choice The overarching aims of this PhD study are * To investigate whether planning in L1 about an L1 related topic or planning in L2 about an L2 related topic (language and topic match conditions) enhances L1 Japanese writers’ final essay texts in L2 English. * To investigate whether topic choice independent of planning language, or planning language independent of topic choice (language and topic mismatch conditions) have any impact on plans or resulting L2 English final essay texts. This investigation takes place in three common contexts in which L1 Japanese writers of L2 English operate. The design of the study and methods used to collect, analyse, discuss and compare data are done both quantitatively and qualitatively, that is empirically and also hermeneutically.

This thesis explores the phraseologies of the two first person pronouns, ‘I’ and ‘We’, and their verb collocates in Chinese novice writers’ academic discourse. Quantitatively, the study compares the use and the function of the FPPs in Chinese EFL learners’ academic texts across two disciplines, Business and Management and English Literature, and at two academic levels, undergraduate and postgraduate. It is found that the phrases serving these functions are highly formulaic. There is correspondence between the identified frequent phraseologies and the proposed textual functions in the novice writers’ academic texts. Specifically, a textual function is often realised by one or two phrasal frames including the two FPPs. When the two disciplines are compared, more similarities than differences in relation to the phrases and the textual functions of ‘I’ and ‘We’ are observed. When the undergraduates and the postgraduates are compared, the postgraduates use more impersonal expressions and more retrospective textual organisation expressions to organise and develop their academic texts. The qualitative case study illustrates the importance of quality of using ‘I’ and ‘We’ to interact with readership and claim authority. This study concludes by discussing the insights offered into the teaching and learning of academic writing in EFL contexts.

This thesis provides an original contribution to ongoing research on so-called Islamic State (‘IS’) by using a multiple case-study approach to offer an in-depth analysis of Arabic and English language narratives related to four atrocities committed by the group: (1) the mass killing of hundreds of Iraqi soldiers, known as the Speicher massacre, (2) the captivity and sexual enslavement of Ezidi girls, known in Arabic as sabi, (3) the executions of a number of western, Arab, and Kurd victims, and (4) the destruction of cultural artefacts in Nineveh province. The analysis engages with the discourses of ‘IS’, western, Arabic, Iranian, and Kurdish media, survivors, ‘IS’s’ religious opponents, and other actors. The dissertation uses a social narrative theory as its conceptual framework that I seek to develop by focusing on the fragmentation in narratives, on one hand, and on the multimodal resources through which narratives circulate, on the other. To this end, I combine the theory with Boje’s (2001) notion of antenarrative and Kress’(2009) understanding of the three resources of discourse, genre, and mode, to investigate ways in which narratives first unfold and how they later change as they are translated. Translation is understood in the thesis as a multi-directional movement that simultaneously takes place across multiple resources without necessarily crossing language boundaries. The findings of this study reveal that the aforementioned resources contribute to transforming narratives. In translation, ‘IS’s’ narratives can be delegitimized and confronted, or the opposite. Examining the changes in these narratives as they are translated in multiple directions is a novel contribution to the field of translation studies in relation to the digital media environment.

The topic of my three-part thesis is pragmatic information in EFL dictionaries. The thesis started with literature review and theoretical explorations of pragmatic information for EFL dictionaries (Module 1). Based on the findings from this first Module, I approached pragmatics focusing on lexical items and their pragmatic behaviours in context, seeking to collect empirical data for describing pragmatics in EFL and lexicographical contexts (Module 2). However, it is important to raise the question of whether pragmatics and lexicography can ever be made compatible at all, since they have different goals, approaches and methods in dealing with different types of meaning. Their units of descriptions are also different; while dictionaries are in principle concerned with words and phrases, pragmatics deals with utterances and discourses. More importantly, dictionaries are basically concerned with decontextualised meanings, and are expected to set out relatively fixed meanings, perhaps prescriptively, in the form of a dictionary definition or explanation. In contrast, descriptive pragmatics treats meaning in context. In this third module, I will be working towards my conclusion that they are indeed compatible, with particular emphasis on politeness.

This thesis reports a study using a corpus of lecture introductions to explore the generic features of the language used. The main research questions: What are the main communicative functions of this sub-genre? How are the functions realised? Are there any disciplinary differences in university lecture introductions? are explored using two complementary methods of analysis; genre analysis and corpus analysis. Analysis of eighty-nine lecture introductions from the BASE corpus1 resulted in a Lecture Introduction Framework which posits a two-level classification which first differentiates lecture content orientation (Thompson, 1994) versus listener orientation (Dubois, 1980). This yields three main functions, with additional sub-functions original to this study. The study reveals differences in the frequency of different sub-functions in the four disciplinary domains, reflecting different disciplinary cultures and knowledge. The corpus analysis generated a word frequency list for lecture introductions, giving a rich linguistic description of the lexis used. Further analysis on one of the three main functions; the Set Up Lecture Framework Function uses WordSmith Tools 5 to analyse the linguistic realisations of the three subfunctions, which are also the top three most used sub-functions in the corpus. Analysis of pronouns and keywords further supports the finding that there are disciplinary variations in lecture introductions.