This research analyses the representation of Turkey in four British broadsheets (the Guardian, the Daily Telegraph, the Independent and the Times) as well as their Sunday sister papers between 2007-2013. Using the concepts of self and other as a theoretical basis, this research seeks to determine whether Turkey, a predominantly Muslim, secular and partially Westernised EU candidate, was represented as a part of the European Self or as an Oriental Other in British broadsheets during the time period specified. As well as defining modern Turkey’s unique position on the Self-other axis of Europe in the context of British quality media, the research examines how applicable Edward Said’s theory of Orientalism is to the Turkish example. Taking Turkey’s multifaceted national identity and Britain’s exceptionalist attitude towards Europe (as well as the British media’s prejudices about Islam) into consideration, the research goes on to demonstrate that the straight-forward, binary understanding of Self-Other relationships in Said’s Orientalism is not applicable to the Turkish example. Instead, the research utilises the concept of Nesting Orientalisms (created by Bakić-Hayden to explain the self-other relationships within Europe) in the analysis and concludes that Turkey was perceived and represented as an agreeable, useful yet still inferior Model Other in British media texts during the time period analysed in this study. The research, which consists of a quantitative content analysis conducted on 731 news items and a qualitative textual analysis conducted on 150 representative news articles, 60 editorials and 10 front-page stories, creates the most detailed map of Turkey’s coverage in the British print media to date as well as providing continuity to the existing relevant literature.

This study investigates the beliefs and practices of writing tutors regarding written corrective feedback (WCF) in a Saudi Arabian university. The central focus of this work is threefold: (a) the relationship between teachers’ beliefs and practices regarding the provision of WCF on students’ L2 writing, (b) the factors that prevent teachers from enacting their beliefs into practices, and (c) the relationship between students’ preferences and teachers’ practices regarding WCF. Ten writing tutors and their thirty students at the English language and Literature department at the University of Dammam participated in this study. Semi-structured interviews were utilized to understand teachers’ beliefs and students’ preferences. In order to understand teachers’ actual WCF practices, teachers were observed while teaching writing over almost a full semester. The think-aloud protocols of teachers while providing feedback on their students’ writing were examined and teachers’ feedback on student writing was analysed. Finally, stimulated-recall interviews were conducted to understand the factors that influence teachers’ WCF practices. The study showed both congruence and tensions between teachers’ beliefs and practices. Teachers’ beliefs were greatly congruent with their practices regarding the amount and focus of WCF. Conversely, teachers’ beliefs were incongruent concerning the explicitness of WCF, the use of positive feedback, and the source of WCF. Several contextual factors related to the university overall context (e.g. time allocated to cover the syllabus), teachers (e.g. teaching experience), and students (e.g. proficiency levels) were found to affect teachers’ practices. As for the relationship between students’ preferences and teachers’ practices, complete congruence was found in terms of teachers being the key providers of feedback. Conversely, students’ preferences were incongruent with teachers’ practices regarding the explicitness of WCF, the focus of WCF, and the provision of positive feedback. Finally, although students valued teachers’ WCF and placed a great importance to it, they faced difficulties understanding some of their teachers’ comments. This study concludes by providing some implications which could serve more than one purpose by creating knowledge which will be useful for researchers in the field of language teacher cognition and WCF.

As global social networks expand, couples are increasingly comprised of partners from divergent sociocultural backgrounds (e.g. Piller, 2007; Dervin, 2013). This unfolding trend inspires research into complex identification processes in such transnational relationships. To explore these processes, I conduct a qualitative discourse analysis of interactions in five UK-based Polish-British families. The data include the families’ interactions during celebratory meals, which they video-recorded, and my semi-structured interviews with the participants, which were audio-recorded. The study focuses on how the participants’ food-related interactions project ‘stance’ (Du Bois, 2007), that is, how talk about food and food practices can discursively and semiotically index the speakers’ positioning towards their own and others’ sociocultural fields. The analysis reveals that as the speakers negotiate their foodscapes, they constantly engage with various sociocultural repertoires and appeal to multiple ‘centres’ of normativity (Blommaert et al., 2005). This negotiation at times occasions contrasting positioning acts, highlighting the dynamism of the speakers’ stancetaking, and thus of their identities. On one hand, the participants reproduce and exoticise what they imagine as their ‘traditions’, ‘cultures’ and ‘nations’, on the other, they echo postmodern discourses of ‘choice’ (Giddens, 1991), individualism and post-national cosmopolitanism. Following the theories of ‘reflexivity’ (Giddens, 1991; Urban, 2001), I demonstrate how in postmodernity even food interactions surface as reflexive spaces. Through culinary performances and meta-talk, the speakers reinterpret cultural signs, creating ‘third spaces’ (Bhabha, 2004 [1994]) – discursive zones with ever-evolving cultural meanings. These reflexively co-constructed ‘third spaces’ display the participants’ identity as hybrid and cosmopolitan families. The family members successfully negotiate the perceived differences between them, which challenges the ideologies of problematic intermarriage (see also Piller, 2002; Gonçalves, 2013). Their complex sociocultural repertoires do not ignite a ‘cultural clash’. They rather offer the speakers versatile vistas for identification and constitute ‘symbolic capital’ (Bourdieu, 1977), thus reflecting the increasing commodification of hybrid forms and pursuit of transcultural identities.

Three central topics that have been at the heart of research into second language (L2) acquisition over the past 30 years are the extent to which properties of a speaker’s first language (L1) transfer into their L2 mental grammars, the extent to which L2 learners’ mental grammars are constrained by an innate language faculty (Universal Grammar (UG)), and the nature of the development of grammatical knowledge. Much of the evidence bearing on these topics has come from the investigation of the acquisition of syntactic properties. There have been comparatively fewer studies of these topics in other domains of the grammar. This thesis investigates the role of L1 transfer and UG in the acquisition of two pre-syntactic properties in English by L1 speakers of Libyan Arabic: noun compounding (a lexical operation) and argument structure realization (a property at the semantics-syntax interface). The participants were selected at different stages of learning English in the classroom to provide a measure of possible development. Using elicited production and a grammaticality judgement task, results suggest some possible evidence of L1 influence on plural marking in noun compounds and knowledge of the morphological marking of constructions realizing argument structure. But in the latter case L1 influence appears to lead to a general problem with the realization of intransitive verbs, rather than direct transfer of L1 properties into the L2. There is also some evidence of the influence of UG on the representation of unaccusative versus unergative verbs, but no evidence of UG influence in other areas investigated (constraints on number marking in noun compounds and on the the linking of thematic arguments to syntactic positions). Little development was observed across the two groups investigated. Broadly, the results are consistent with iii an L1 transfer/access to UG view of the L2 acquisition of pre-syntactic properties, without providing strong support for this position.

This thesis explores the mixed language and dialect setting of an international school in Spain across two age groups, young children and adolescents. It aims to investigate the impact of frequent and sustained interaction between Anglo and Non-Anglo informants, and if within such a turbulent, fluid and messy speech community, processes of dialect contact can be traced. Dialect contact research to date has usually focussed upon two distinct varieties coming together in a melting pot situation, with eventual stability. This speech community is fluid and turbulent. People are joining and leaving the community constantly. The study aims to reveal that despite previous research claims that the process of koinéisation takes two or three generations (see Kerswill and Williams 2000; Trudgill 1986; 2004), a speech community such as the one in the present study shows that processes of koinéisation can be demonstrated in a shorter time span. Through the analysis of the BATH vowel and the (t) variable across the two age groups and three broad ethnic groups, the study finds evidence of focussing over time in this highly diverse speech community. It also reveals that certain social factors which have been shown to influence new dialect formation in other studies, have an impact upon variation in this rather unique speech community.

This study investigates how the two pragmatic markers 'I mean' and 'you know' are used by Chinese EFL learners as compared to British speakers. To describe how these two markers by used by the Chinese learners, this study first investigates how they are used in the British speakers’ data. To obtain a finer picture of how they are used by the native speakers, the interpretation of their pragmatic meanings is open to all plausible explanations instead of being confined to one single theory or framework. As a result, this study sets up its own categories and comes up with much longer function lists than previous studies. In addition, a new framework is proposed. Following the completion of a detailed description of the pragmatic functions of I mean and you know, a detailed and systemic comparison between them in terms of the specific functions they play and their positioning are carried out on the grounds that these two markers are analyzed by following the same approach in the same data set. By highlighting the similarities and differences between them and explaining why, the comparison improves our understanding how they relate to each other in conversation. Compared to the British speakers, the Chinese learners show different patterns of I mean and you know in their L2 English. The main features of the learners’ uses of I mean and you know are: firstly, I mean is markedly under-represented and less pragmatized while you know is markedly over-represented and more pragmatized; secondly, both I mean and you know are used in more restricted contexts; finally, the pragmatic functions of I mean are more evenly distributed while you know heavily depends on a very small number of functions. Since I mean and you know are very unlikely to be taught in the classroom, the accounting for the patterns of them in the learners’ data is approached from the perspective of second language acquisition. This study follows the assumption that learners’ L2 production can be seen as the result of the interaction of all potential factors and the importance of a certain factor varies from one L2 phenomenon to another. The analysis seems to suggest that the learners’ uses of I mean are greatly influenced by the congruence between the pragmatic meaning and semantic meaning of I mean while the learners’ uses of you know are mainly affected by L1 influence. Other factors that seem to have impact on the production of both markers include the tasks performed by the learners and the learners’ proficiency level.

The main purpose of this study is to investigate language practices in the linguistic landscape (LL) of Jordanian cities. There have been few research studies that examine the LL of Jordanian cities, and none has investigated multilingualism. This study is intended to fill this gap in LL research. By means of qualitative and quantitative methods, it aims to discover the extent to which multilingualism is reflected in the LL. The main fieldwork was conducted in November and December 2012 in urban Jordan. Ten streets were selected in each of six major Jordanian cities, including Irbid, Salt, Zarqa, Amman, Karak and Aqaba, sixty streets in total. A LL item represents ‘any piece of text within a spatially definable frame’ (Backhaus, 2007). 4070 signs were recorded as multilingual (c. 51%), whereas 3967 signs were categorised as monolingual (c. 49%). To discover correlations between types of signs and existing languages and scripts, and to measure these against conflicting language policies, signs are categorised as ‘top-down’ or ‘bottom-up’. The notions of ‘code preference’ (Scollon & Scollon, 2003) and dynamics of language contact are employed to understand the semiotics of writing in the LL of Jordanian cities. The main data findings indicate that minority languages are almost absent, so a questionnaire was introduced as an additional supportive source to the analysis of the findings, providing a qualitative dimension to the study. The study was conducted in July 2013, during which period the researcher interviewed 32 participants. The primary objective of this secondary study is to reflect on plausible reasons explaining the limited presence of minority languages in the visual public space. The main data indicate a dominance of both Modern Standard Arabic (MSA) and English on signs, because they are closely related to Arab nationalism and globalization respectively. Jordanian Arabic is deleted from the top-down LL, because it is closely linked to informal domains. Classical Arabic (CA) is mainly used to convey religious functions in the LL. Mixed codes, Romanised Arabic (RA) and Arabacised English (AE), are commonly used in the LL to reflect ‘glocalisation’. French, German, Italian, Spanish, Turkish and Russian are found mainly to be used on brand name and business name signs for reasons of European linguistic fetishes and tourism. The data indicate that minority languages are significantly marginalised on both top-down and bottom-up signs. Several reasons lie behind the limited visibility of established minority languages in the LL. Spatial distribution of migrant communities, the small size of minority communities, lack of (sufficient) institutional and parental support, migration and absence of close ties with families and linguistic peers are behind different stages of language maintenance and shift among older migrant groups. Linguistic russification, hostility, instrumentality of both Arabic and English and top-down language policies enacted by the Jordanian government contribute to the limited visibility of minority languages in the LL. Although foreign workers’ minority languages tend to be maintained, the instrumental functions of both Arabic and English, Islam, and the small sizes of economic minority groups have each played a key role in the limited visibility or invisibility of minority languages in the LL.

The present research, drawing upon Martin and White’s (2005) Appraisal Theory, investigated the deployment of interpersonal meanings in discussion sections of doctoral theses produced by both First Language (hereafter L1) Chinese and L1 English speakers from the Centre of Applied Linguistics (hereafter CAL) at the University of Warwick. This study adopted a corpus-based approach to examining the choices of Appraisal options made by L1 Chinese and L1 English writers. It also explored the patterns of discussing the authors’ own research findings in relation to previous literature by means of Appraisal options or co-articulations of options, which is considered as a key aspect of the rhetorical purpose of discussion sections. The statistical tests of this study showed that no significant difference was found in the use of Appraisal options between the L1C and L1E sub-corpora. This finding indicates a similar command of these interpersonal resources by both sets of writers and suggests that L1 may not be a constraint for English as second language (hereafter L2) writers on using interpersonal resources at the doctoral level. The qualitative analysis identified different preferences for co-articulating with the three main Appraisal options that the authors adopted to engage with the literature while discussing their findings. It also identified the congruent and non-congruent linguistic realizations of the two main Appraisal options that the authors used to present their claims about findings. Part of the qualitative results was shared with Masters students at CAL for the purpose of raising their awareness of the use of interpersonal language through exploration of extracts from corpus data.

As shown by research conducted in the fields of corpus linguistics, psycholinguistics and second language acquisition (SLA), formulaic sequences (e.g. collocations, idioms, phrasal verbs, lexical bundles etc.) form a crucial part of second language learners’ language competence. One of the main characteristics of some of these multi-word combinations is their idiomatic nature and idioms have been widely researched as being the most compositional type. However, little is known whether collocations can be idiomatic too. Therefore, this thesis aims to fill that gap with a specific focus on statistically-derived collocations. Three studies have been carried out to this end. The first study examined the extent of collocations whose meanings were the sums of their component parts, those that were figurative (compositional or idiomatic) and finally, collocations that were polysemous, i.e. had both literal and figurative meanings. A corpus search was performed to find the most common V+N and Adj+N combinations and results showed that the majority of them had additive meanings (as reflected in traditional definitions of collocations), very few were completely opaque and a substantial percentage were those that allowed both literal and figurative interpretation. The second study was conducted in the Chilean classroom context and it focused on the knowledge of the figurative meanings of collocations. A meaning-recall test was used as a measurement instrument and variables such as corpus frequency, semantic transparency and language use factors were manipulated. The findings demonstrated rather weak figurative knowledge and a positive relationship between this knowledge and the amount of outside engagement. The last study investigated the same type of collocations, namely how non-native speakers acquired figurative meanings incidentally through reading. An authentic novel was used as a means to manipulate one single variable, frequency of exposure. Results showed that the figurative meanings could be acquired incidentally from reading and that the point at which a substantial amount of learning occurred varied greatly among individuals. Overall, results of the studies presented in this thesis stress the importance of researching the acquisition, use and knowledge of collocations with figurative meanings as they have important implications for instructed SLA.

Formulaic language constitutes an essential part of English vocabulary and is necessary for performing a wide range of communicative functions, but knowledge and acquisition of formulaic sequences is typically found to be lacking and problematic for L2 learners. Whilst much research has been carried out on formulaic sequences such as idioms and collocations, comparatively little has been done on phrasal verbs which are nonetheless commonly perceived as one of the most challenging aspects of English vocabulary. This thesis attempts to contribute to filling this gap by exploring the usage, knowledge and acquisition of phrasal verbs by native and non-native speakers of English. Study 1 explores the semantic frequencies of the 150 most frequently used phrasal verbs using the Corpus of Contemporary American English (COCA). Results show that, whilst the vast majority of these phrasal verbs are polysemous, only two meaning senses on average are enough to cover three-quarters of the occurrences of each of them. The most frequent meaning senses of all 150 phrasal verbs are listed in the PHrasal VErb pedagogical List (PHaVE List), in frequency ranking order along with frequency percentages. The list thus offers teachers and learners the possibility of prioritising these most frequent, and thus most important, meaning senses, thereby allowing for a more systematic approach to tackling phrasal verbs. Study 2 explores L2 learners’ knowledge of a sample of phrasal verbs and meaning senses on the PHaVE List at a form-recall level of mastery, and the effect of a number of factors on this knowledge. Results show that only about 40 % of meaning senses were known, with a 20 % chance that all the various meaning senses attached to a given phrasal verb would be known. A mixed-effect modelling analysis reveals a significant effect of two factors on scores: item frequency and learner engagement in leisure activities in the L2 such as reading and social networking. This is consistent with previous research showing the robust effect of frequency for L2 knowledge of individual words and formulaic sequences, and the benefits of reading for L2 language acquisition. Study 3 investigates L2 learners’ acquisition of novel phrasal verbs through three intentional, word-focused learning activities: rote memorisation, textbook exercises, and guessing from context. Knowledge of the items was measured both immediately and one week after the teaching treatment at meaning-recall and meaning-recognition levels of mastery. Results show encouraging learning gains, similar to those found by previous research for individual words and idioms, with higher L2 proficiency and general vocabulary knowledge leading to significantly higher scores. A Friedman test reveals no significant difference in learning gains between the three activities. Taken together, these studies provide empirical evidence for the gap in L2 learners’ knowledge of phrasal verbs, but suggest that a restricted number of phrasal verbs and meaning senses can go a long way and be effectively learned using the same explicit activities commonly used for learning single words. Overall, they offer useful insights for learning and teaching English phrasal verbs in a more systematic and efficient manner.