The present study explains that ideology can affect translators’ linguistic selections which can consequently shape the receivers’ worldviews. Owing to the fact that after the Arab Spring, new leaders with different ideologies and belonging to different political movements sprung forth, their political discourse has become a subject of increasing interest. The language these leaders use to promote their own political and ideological visions and the way to interpret them requires analysis to detect the possibility of translators’ intervention in the translation of these speeches. Adopting a mixed approach of corpus linguistics and critical discourse analysis, the present study focuses on investigating the manipulation of the source text ideology in the translation of presidential speeches after the Arab Spring. The source texts analysed in this study are 20 speeches by the former Egyptian president Morsi translated into English by five translators of various ideological backgrounds. The analysis of these source texts is based on the extraction of keywords and a selection of keywords with ideological content. The analysis of the target texts, on the other hand, focuses on the use of ideological keywords in lexical patterns and grammatical structures to detect ideological manipulation in translation. The thesis aims to describe systematically the means through which translations transfer, strengthen, or mitigate the ideology underlying the source texts. Using five parallel corpora of the source texts and their translations, the thesis also aims to ascertain whether the lexical choices and the syntactic structures employed in the target texts engender changes in the ideological content of the source texts and their underlying ideology. The results reveal that two out of the five translations project a manipulated ideology that is at variance with the ideology underlining the original texts. One translation strengthens the ideology of the source texts, whereas the other two translations aim to maintain the original ideology unchanged. This indicates that instances of ideological manipulation are probable even in the translation of presidential speeches due to the nature of the source texts, the ideology underlying them as well as the possibility of an ideological clash.

Current thinking about writing considers spelling skills to be a crucial factor in its development. However, most research evidence to date comes from studies in English, which has a highly inconsistent orthography. Since the rate of spelling development is affected by degrees of orthographic consistency, cross-linguistic investigations are thus needed to determine whether English-based assumptions may be extended to more consistent orthographies. This work reports the results of a longitudinal investigation of early writing development from a cross-linguistic viewpoint. The writing development of children learning to write in English, a very inconsistent orthography, and children learning to write in Spanish, a very consistent orthography, was contrasted. Specifically, the studies included in this thesis explored the extent to which orthographic complexity moderated gains in a number of microstructural writing features, as well as whether it shaped the relationships between different levels of text construction. Finally, preschool-Year 2 predictors of writing outcomes were explored. The results showed that orthographic complexity moderates gains in wordlevel features of text writing, but, beyond the word level, both language groups showed remarkably similar performances. Spelling was found to contribute to development in the amount of text produced (text length) over a relatively short period of time. Moreover, the underlying factors driving the development of writing were common until the middle or end of Year 1. In short, orthographic consistency seems to moderate word-level writing, but has a reduced effect in text-level performance. The discussion considers the implications for models of writing development and the divide between word- and text-level writing performance.

Cynigir yn y traethawd hir hwn ddadansoddiad o’r math o wallau a wneir yn gyffredin wrth gyfieithu o’r Saesneg i’r Gymraeg, a hynny er mwyn galluogi’r rhai sy’n darparu hyfforddiant yn y maes i gynnig hyfforddiant pwrpasol, sydd wedi’i deilwra’n benodol i anghenion y diwydiant cyfieithu yng Nghymru. Yn rhan gyntaf y traethawd (Penodau 1 & 2) rhoir gwybodaeth gefndir yr ymchwil. Trafodir, i ddechrau, y diwydiant cyfieithu yng Nghymru a’r heriau y mae’n eu hwynebu; cyflwynir wedyn yr ochr ddamcaniaethol, gan fanylu ar y cysyniad o ‘ansawdd’ a’r hyn a olygir gan ‘wall’ yng nghyd-destun cyfieithu. Yn ail ran y traethawd cyflwynir yr ymchwil ei hun. Amlinellir ym Mhennod 3 y fethodoleg a ddefnyddiwyd i gynnal yr ymchwil, sef methodoleg wedi’i seilio ar ieithyddiaeth gorpws, a disgrifir cynnwys y corpws cyfochrog a grëwyd at ddiben y prosiect hwn. Wedyn, yn y ddwy bennod nesaf cynigir dadansoddiad meintiol ac ansoddol o’r gwallau systematig a welwyd yn y corpws, gan ymdrin â gwallau ‘iaith’ ym Mhennod 4 a gwallau ‘cyfieithu’ ym Mhennod 5. Trafodir achoseg bosibl y gwallau a’r gwahaniaethau a welir rhwng gwaith cyw gyfieithwyr a gwaith cyfieithwyr profiadol, gan ystyried y goblygiadau o ran hyfforddi cyfieithwyr. Mae’r traethawd yn cloi, ym Mhennod 6, drwy grynhoi canfyddiadau’r ymchwil, cyn nodi sawl maes lle y gellid gwneud ymchwil pellach.

Arabic and English are acknowledged as two major natural languages used by many countries and regions. Reviews of previous literature conclude that machine translation (MT) between these languages is disappointing and unsatisfactory due to its poor quality. This research aims to improve the translation quality of MT between Arabic and English by developing higher quality parallel corpora. The thesis developed a higher quality parallel test corpus, based on corpora from Al Hayat articles and the OPUS open-source online corpora database. A new Prediction by Partial Matching (PPM)-based metric for sentence alignment has been applied to verify quality in translation between the sentence pairs in the test corpus. This metric combines two techniques; the traditional approach is based on sentence length and the other is based on compression code length. A higher quality parallel corpus has been constructed from the existing resources. Obtaining sentences and words from two online sources, Al Hayat and OPUS, the new corpus offers 27,775,663 words in Arabic and 30,808,480 in English. Experimental results on sample data indicate that the PPM-based and sentence length technique for sentence alignment on this corpus improves accuracy of alignment compared to sentence length alone.

Despite the intriguing nature of metaphor and its acknowledged importance in the discipline of Translation Studies (TS), a relatively small number of studies have explored the translation of metaphor from the perspective of Conceptual Metaphor Theory, and very few of them adopted an experiential approach to the object of analysis. This research aims at exploring the translatability of creative metaphor in six Arabic translations of Shakespeare’s Othello and Macbeth based on a combined methodology that adopts the Conceptual Theory of Metaphor and the descriptive approach to text analysis in TS. The empirical study argues that metaphor translatability is an experiential process that is highly influenced by the diversity and richness of our conceptual system and the background knowledge shared by the metaphor producer and metaphor translator. Discussing metaphor translatability from the perspective of these factors involves dealing with different levels of variation in our metaphoric thinking including the cultural, contextual and pragmatic levels. The analyses and discussions of the empirical study mark a departure from text-linguistic approaches to the topic in that they deal with the Source Text’s and Target Text’s metaphoric content as physically embedded conceptual models rather than linguistic patterns with grammatically delineated features and structures. The arguments of the study answer several questions with regard to researching the translation of metaphor from the perspective of Conceptual Theory, providing a detailed description of what exactly influences the process and product of translation, and underlining the functionality of the variation factor in appreciating the conceptual nature of metaphor. The results of the empirical research reveal that, although our metaphoric thinking has a universally shared metaphoric structure, not all our metaphors are translatable or translated in a single way, which refutes the supremacy of the notion of metaphor universality, putting emphasis on the factors of experientialism, exposure and intentionality.

This thesis is a descriptive, evaluative and comparative study in the field of translation studies. One of the objectives for this thesis is to explore a valid criterion by which a literary translation can be evaluated efficiently and to assess the translation of the selected novel for this research. The aim of this study is to measure the shifts which occurred between TTl and TT2 when compared to the ST. The thesis also aims at highlighting the significance of culture and the way cultures are introduced to the Tar et readership through translation. It is thought that the strategy of Foreignization enriches target texts and introduces cultural elements to the target reader. The corpus of the study is Ziqiiq Al Midaq the well-known novel by Naguib Mahfouz, the Nobel Prize Laureate in Literature in 1988. This novel has been first translated by Trevor Le Gassick in 1966 and a revision of this translation was published in 1975. The main objective of this study is to explore the translation shifts which were applied in TTl and TT2. The methodology of this thesis relies on the Nord Model (2005). It focuses on the translation problems introduced by Nord. The four aspects of translation problems Nord identifies are Pragmatic, linguistic, cultural and text-specific translation problems. A final assessment of the quality of both versions of translation is discussed at the end of the study.

This research investigates the effect of the interplay of paralanguage and language on the accessibility of written texts. It identifies textual factors that may hinder the reader's understanding of the message. Studies of reading comprehension have traditionally concentrated on the language of texts. Many written texts, however, also rely on paralinguistic features to convey their messages, and to complement, illustrate, clarify and organise their content. Hence, it is paramount that a study investigating obstacles for reading comprehension be able to account for the complex web of paralinguistic devices present on the page. This work investigates such paralinguistic devices, focusing particularly upon the presentation of letters and words, the layout of the page, and the illustration of the written text. Parallel to the analysis of these nonlinguistic devices, language is also examined. A particular text type has been selected to provide the material for the analysis: emergency procedures. This choice was motivated not only by the fact that this text type is a rich source of paralinguistic and linguistic features, but also because accessibility in emergency texts is vital. In addition, emergency procedures are only minimally influenced by political, religious and other ideological factors. A corpus of 126 emergency procedures is analysed and related to the literature on paralinguistic features of written texts. Several combinations of linguistic and paralinguistic features are examined and, subsequently, hypotheses about how they affect the reading process are formulated. Verbal protocols and interviews are used to gain access to readers' interpretative processes when dealing with texts from the corpus. They aim to provide an insight into the reader's perception of the interaction of elements from both paralinguistic and linguistic structures and to evince accessibility problems. The application of this procedure both verifies the hypotheses and reveals new facts which were not predicted, thus increasing understanding of the factors which contribute to the accessibility of written texts

The present study investigates issues in the subtitling of humour in the American sitcom, Seinfeld. The study also identifies the subtitling strategies used by Arab translators to solve the technical, linguistic and cultural problems of translating humour in Seinfeld, and it uncovers the factors that might have affected the subtitlers’ decisions. The study draws on the General Theory of Verbal Humour (GTVH; Attardo & Raskin 1991, Attardo 1994, Attardo 2001, Attardo 2002) and Pedersen’s (2005) model of subtitling cultural references. Applying the GTVH and Pedersen’s (2005) model to the subtitling of humour between English and Arabic will test their reliability (i.e., the analysis of the data using these theories will reveal their strengths and weaknesses). The findings show that different types of humour were used in the selected data (eight types). These types were rendered into Arabic using a variety of strategies which were governed by a number of parameters, including cultural and linguistc differences between the source and target languages and cultures, intersemiotic redundancy, intertextuality, media-specific constraints, viewers’ knowledge of the show, and the simplicity of the humour in the source text. The Arab subtitlers managed, in many instances, to translate humour successfully into Arabic. However, some instances of humour that contained cultural references, wordplay, and catchphrases proved to be more challenging. The study also discusses the subtitling of a wide range of types of humour, some of which have not been explored before in the Arabic context (e.g., retorts, register clash, spoonerisms, and catchphrases). This study should contribute to filling the gap in research into the subtitling of humour in the Arab world, as this field is relatively new and there is a huge shortage of research in the field of subtitling comedies. In addition, a new model of analysing and subtitling humour in sitcoms is proposed, based on the GTVH and Pedersen’s (2005) model and the findings of the current study. This new model consists of the external and internal parameters of the joke. The new proposed model of analysing and subtitling humour presented in this study contributes to the overall knowledge of the treatment of humour in AV productions.

Vocabulary knowledge has been conceptualized as consisting of four components: the size of the mental lexicon, the depth or quality of knowledge of the individual lexical items in the lexicon, the degree to which those items are connected to each other (organlzation], and the speed at which those items can be retrieved and employed (e.g. Chapelle, 1998; Quian, 2002; Read, 2004; Schmitt, 2010b). There is now a considerable literature on each of the first three aspects, but lexical processing speed has hardly been explored outside the laboratory. In vocabulary acquisition research, studies have mainly focused on the acquisition of new words' form and meaning but research into the learning conditions leading to automaticity of vocabulary knowledge is still scarce. Furthermore, in the field of vocabulary testing, measures of automaticity and fluency of lexical access have seldom been used as an indication of participants' actual vocabulary knowledge of the words being assessed. This thesis attempts to address this lack in research in vocabulary acquisition and assessment. The first two studies of this thesis (Study 1 and Study 2) examine the acquisition of automaticity in the language classroom, comparing the effectiveness of two different approaches, i.e. incidental learning and explicit teaching. Results of these studies show that, the incidental approach does not lead to a significant improvement in the speed and automaticity of learners' lexical decisions in any of the measures used, i.e. reaction time (RT) and coefficient of variation (CV). However, with the explicit treatment learners' lexical decisions are not only faster but also more automatic, as measured by the CV. Results of these two acquisition studies show that automaticity of vocabulary knowledge can benefit from classroom instruction, although it might need more engaged, explicit exposure. The last two studies of this thesis (Study 3 and Study 4) examine the use of measures of RT to assess the veracity of learners' vocabulary knowledge and the use of that information to score Yes-No vocabulary size tests. Results of these last two studies show that RT carries important information about learners' accuracy and veracity of responses in vocabulary tests. These studies also present a new approach that uses this RT information to score Yes- No vocabulary size tests and that can help to make the test more valid. Overall, results of the studies presented in this thesis point out the importance of researching the acquisition of automaticity in instructional settings and the use of the time element in vocabulary tests to assess learners' vocabulary knowledge.

The aim of this study is to isolate the individual stylistic traits of one translator, Humphrey Davies, within the framework of descriptive translation studies. Davies’ English translation of the Arabic novel Midaq Alley is compared, using a corpus-driven approach based on keyword lists, to another English translation of the same source text by another translator, Trevor Legassick. By making this initial corpus-driven comparison and subsequently generating a keyword list for Davies’ Midaq Alley, the stylistic features regarded as indicative of the translator style and meriting further investigation declared themselves and, accordingly, hypotheses regarding Davies’ translator style were constructed and then tested by carrying out a thorough corpus-based investigation. A consistent pattern of choices was identified in the translation of four types of words: culture-specific items, including culture-specific common expressions and proper nouns; terms of respect; reporting verbs and function words, including the contraction ‘’d’ and ‘that’ as complementizer, relativizer, demonstrative pronoun and demonstrative determiner. For lexical words, the results show that Davies’ tends to transliterate foreign words and supplement them with extratextual gloss, reproduces the structures of proper nouns, preserves the terms of respect by literal translation and translates literally the reporting verbs. Regarding function words, Davies tends to make heavy use of contractions and all types of ‘that’. Generally, the findings show that Davies stays close to the source text compared to Legassick who moves much further from the source text. The identified stylistic features are investigated in Davies’ English translation of another Arabic novel (The Yacoubian Building) to check whether these features are stable across one of his other translations. The findings show that most of the features revealed through the comparison of Davies to Legassick are stable across the Davies’ two translations. Despite limitations, it is anticipated that the approach developed in this study will be fruitfully adapted for further rigorous and replicable analysis of translator style.