Idioms, as multi-word units that contain literal and figurative meanings, are inherently complex and thus unsurprisingly difficult to acquire for second language learners. Though experimental studies on idioms have been carried out with pedagogically minded foci, none have examined the differential effects picture type has on correct interpretation of meaning or meaning recall. Because idioms have both literal and figurative senses, they can be pictorially expressed via either or both of their dual meanings. However, no one has yet tested whether figurative elements in pictures will aid or confuse second language learners when presented alongside idioms. Thus, the primary aim of this thesis is to experimentally test how different kinds of pictures affect the way in which second language learners interpret and recall the figurative meaning of metaphorical idioms. Furthermore, the role of semantic transparency and how it impacts the effectiveness of the picture type is examined. The overarching finding suggests that metaphorically imbued pictures overall facilitate the learning of idioms. However, highly contextualized pictures have the potential to mislead learners in specific and often unpredictable ways. In addition to the pedagogical implications uncovered, this thesis also addresses the nature of semantic transparency and teacher attitudes on idioms.

Research into volunteers' motivations to volunteer (MTV) usually relies on surveys and there is a need for more qualitative research into volunteering motivations, particularly in the United Kingdom. Furthermore, existing qualitative research into volunteering motivations usually relies on the thematic analysis of interviews with volunteers, but while this analysis is necessarily the analysis of discourse, there is rarely a systematic or replicable analysis of such discourse. This thesis aims to contribute to current understanding of MTV and the beliefs underpinning them by drawing upon various tools for discourse analysis in the analysis of community-based volunteers' accounts of their volunteering experiences. However, given the generally positive reception that community based volunteering receives in the public domain, this thesis aims not only to describe and explain informants' MTV but also to critically evaluate volunteers' representations of volunteering practice. A central argument in this thesis is that while a strong sense of concern for the suffering and flourishing of beneficiaries is expressed throughout volunteers' representations, volunteering tends to be represented as an optional act of goodwill, rather than a moral duty or obligation. This raises questions about the dominant ways we think about helping others in contemporary society and about the ideologies underpinning such beliefs. It is argued that bringing to the fore and problematizing such beliefs is critical in a time where traditional collectives are breaking down and the social contracts underpinning British and European society together with the nature of our obligations as citizens in an increasingly globalized society are subject to increasing scrutiny and debate.

A body of work exists on the classification, by prosodic analysis and other means, of utterance types and dialogue moves in spoken corpora. Much of this output, while often linguistically well motivated, tends to rely on hand-crafted rules. This thesis presents a data-driven approach to the classification of utterances, using a novel combination of existing algorithmic approaches. Previous work has generally classified utterances according to such categories as wh- question, yes/no question, acknowledgement, response and the like; in general, the audio data used has been specially commissioned and recorded for research purposes. The work presented here departs from this tradition, in that the recorded data consists of genuine interaction between the telephone operator and members of the public. Moreover, most of the calls recorded can be characterized as queries. The techniques presented in this thesis attempt to determine, automatically, the class of query, from a set of six possibilities including "statement of a problem" and "request for action". To achieve this, a scheme for automatically labelling utterance segments according to their prosodic features was devised, and this is presented. It is then shown how labelling patterns encountered in training data can be exploited to classify unseen utterances.

Cognitive approaches to language teaching recognize conceptual metaphor awareness-raising activities as techniques that aid the understanding of metaphors in second or foreign languages (L2) such as English. However, the benefits of these techniques do not extend to the retention, production or interpretation of metaphors. These are important skills that could help language learners in their use of metaphor in the real world. This thesis focuses on the embodied nature of L2 metaphors through awareness-raising activities that incorporate bodily actions and tactile contact. Through four experimental classroom studies with university-level female EFL Saudi learners, this thesis explores the impacts of employing awareness-raising activities of conceptual metaphors and embodied metaphors. The first and second studies compare conceptual metaphor awareness to traditional semantic clustering and highlight the role of conceptual metaphor awareness in understanding metaphoric expressions. The third study introduces embodied metaphor awareness-raising activities through bodily actions and compares them to conceptual metaphor awareness and semantic clustering. The fourth study explores the benefits of using embodied tactile metaphor awareness-raising activities in the learning of linguistic and pictorial metaphors in advertising. The thesis finds that awareness-raising activities based on embodied metaphors have the potential to foster a deeper learning of metaphors in the L2.

This research focuses on improving stock tweet sentiment classification accuracy with the addition of the linguistic features of stock tweets. Stock prediction based on social media data has been popular in recent years, but none of the previous studies have provided a comprehensive understanding of the linguistic features of stock tweets. Hence, applying a simple statistical model to classifying the sentiment of stock tweets has reached a bottleneck. Thus, after analysing the linguistic features of stock tweets, this research used these features to train four machine learning classifiers. Each of them showed an improvement, and the best one achieved a 9.7% improvement compared to the baseline model. The main contributions of this research are fivefold: (a) it provides an in-­depth linguistic analysis of stock tweets; (b) it gives a clear and comprehensive definition of stock tweets; (c) it provides a simple but effective way to automatically identify stock tweets; (d) it provides a simple but effective method of generating a localised sentiment keyword list; and (e) it demonstrates a significant improvement of stock tweet sentiment classification accuracy.

The main aim of this thesis is to investigate the influence of self-assessment on the writing of inexperienced EFL Libyan university students’ writing. Data was collected from one hundred fourth year students majoring in English. The essays were each 300 words in length. The students were divided into experimental and control groups and 40 essays (20 from each group) were selected for analysis. The study uses a pre-self-assessment/post-self-assessment comparison of two groups of writers. The key research instrument was a self-assessment sheet which asked students to assess the content, organisation and language of their essays. Students were asked to revise their writing after using the self-assessment sheet. In addition to the self-assessment sheet, a post- study feedback form was given to elicit the students’ attitudes towards the use of self-assessment to guide their revision of writing. Further, a questionnaire completed by Libyans in the UK was also used in this study to elicit key contextual information about the writing context of the study. The results of the study showed that the writers in the experimental group who used self-assessment to guide their revision made more revisions than those in the control group. The result also demonstrated that the majority of students in the experimental group had a positive attitude towards the self-assessment technique. While there are limitations mainly due to the political and social climate at the time the research was conducted, this thesis demonstrates to some extent the important role of using self-assessment in developing inexperienced students’ writing.

This study investigates the interaction between adults with learning disabilities and their care staff. Many people with severe learning disabilities have little or no speech; for these individuals, augmentative and alternative communication (AAC) may enhance their communication. However, AAC non-use is a widely reported phenomenon. The study explores power and communicative competence within such interaction, as possible factors in AAC non-use. An ethnographic approach was adopted; data collection was carried out in five community homes, focusing on four residents. Field notes were accompanied by video and audio recordings of natural interaction between participants. Aspects of Critical Discourse Analysis were applied to the data within the themes of turn taking, topic control, exclusion from conversation, activity exchanges, test questions and politeness; the theme of AAC was also critically scrutinized. Findings regarding interaction between residents and staff were set in the context of the institution and of wider society. Substantial asymmetries in both communicative competence and power were evident. Staff tended to constrain interaction such that immediate participation of residents was facilitated; however, in the longer term, AAC use is likely to be thereby inhibited. Further application of critical approaches to AAC research is warranted.

This study examines language style of Arabic newspapers particularly in the world affairs and sport sections, using the word frequency analysis. The study is divided into seven chapters. The first chapter mainly focuses on background and aims of the study, while review of previous studies is presented in the second chapter. Chapter Three discusses Modern Standard Arabic (MSA), the selected newspapers and the chosen sections. Chapter Four concerns on the methodology applied in the study. Analysis of the language style is presented in Chapter five and followed by findings which are discussed in Chapter Six. The final chapter provides summary, conclusion and suggestions for further research. The study employed a self-constructed method in corpus building. A total of 30 articles (world affairs and sports) from seven Arabic newspapers were collected from the official online websites mostly in November 2007. Five of the newspapers are published in Arab countries and one in both the United Kingdom and Australia. The Wordsmith version 5.0 was used in analyzing the corpus data. Both descriptive and inferential statistics were employed to analyse word frequency list in both categories, while the Likelihood ratio test was applied in the comparison analysis. Findings revealed that most high frequency words have close relationship to their respective categories especially in the use of nouns. Slight differences were identified in terms of word spelling, loan word, verb transitivity and phrase amongst the newspapers published in different countries s. It is also proven that there is a tendency in MSA to practice new features in news writing which is different from the well known Arabic grammar, i.e, verb and subject agreement in gender. Different newspapers have also demonstrated their own focus in news reporting, and sports section is found to use more specific words than world affairs section.

This study investigates the language use through which Hillary Clinton constructs her political identity by examining the extent to which the First-Person Plural Pronoun (FPPP) is important in the political discourse of this American woman politician. Drawing on Brown and Levinson’s (1987) notion of face and face-work, this study demonstrates how Hillary Clinton, as a woman in a position of power, actively exploits the referential flexibility of the FPPP in her campaign discourse to construct and negotiate her identities to strategically enhance her political power and/or protect her threatened negative face in different face-threatening situations. The study also demonstrates how Clinton actively exploits self-affiliation to acquire, neutralize or challenge power in her campaign discourse. The study also suggests a role of the power differential between the affiliator (candidate) and the affiliated group in determining the strategic function of the FPPP in the candidate’s discourse. The study also shows a key role of the macro-linguistic context in interpreting the meaning of the FPPP in the candidate’s discourse, and thus, a more profound understanding of the political identity of the candidate.

In this thesis, I explore the physicality of the pregnant body in the womb imagery in Psalm 139 and Job 3 and in the simile “like a woman in labor,” specifically, the simile of YHWH as a woman in labor in Isa 42:14. I show that the metaphorical pregnant body is not an idea of a body; rather, it is a detailed physical body with images of gestation in the womb, physical descriptions of a body writhing in labor (e.g. face, breath, hands, heart, legs), and descriptions of a baby delivered from the womb. Using Benjamin Harshav’s theory of Integrational Semantics, I mine the text for details of and allusions to the physicality of the pregnant body. I look at the text through the lens of the pregnant female body and ask how the physicality of this body contributes to meaning. I show that the full impact of the text is lost when the physical properties of the pregnant body are not integrated within an interpretation.