This study investigates Omani academic writing in English as an example of English for Special Purposes produced by authors in an Arab context. Specifically, it explores the differences in attitudes towards academic authorities/sources manifested in the construction and use of deontic verb phrases within the particular academic genre on the topic of English Language Teaching (ELT). In response to the perceived threats from the spreading of English as an international language in academia to the cultural values of the Sultanate of Oman, this conservative Islamic nation has stated aims to modernise itself but still preserve its inner cultural thrust by producing ‘global workers with local values'. However, the difficulties experienced by many Omani novice academic writers in engaging with the wider academic community are well documented and a difference in attitudes towards authority has been mooted as a potential cause for their pragmatic failures. In addressing this issue, this project uses a qualitative evaluation to compare the work of locally published Omani writers alongside established authors, i.e. (non)native speakers using English systematically in the Western world. The results identify a number of key differences between the two sets of authors in the construction and use of deontic verb phrases which reveal the culturally embedded values and attitudes towards academic authority of the Omani writers.

Lexical relations have been thoroughly investigated cross-linguistically (Lyons, 1977; Cruse, 1986; Murphy, 2003). Antonymy is particularly interesting because antonymous pairs share both syntagmatic as well as paradigmatic relations. Studies (such as Raybeck and Herrmann, 1996) agree on the universality of this lexical relation; however, different perspectives towards opposition have been noted among different cultures (Murphy et al., 2009; Jones et al., 2012; Hsu, 2015). The present corpus-driven study investigates antonym use in Modern Standard Arabic text using an on-line corpus (arTenTen12) and a newspaper corpus (arabiCorpus). This thesis shows that antonym functions in Arabic are to a certain degree similar to those found in other languages. A new classification of these functions is presented and compared to previously identified functions in English text (Jones, 2002; Davies, 2013). The main difference between this classification and previous ones is in the category Ancillary Antonymy. In this category, canonical antonyms trigger contrast in non-contrastive pairings. The ancillary use of antonyms is presented as an effect projected on other words regardless of the hosting construction. As a consequence of removing this category, other functions of antonym use were identified. The present study also shows that a Sign-Based Construction Grammar (SBCG) account of antonyms can capture their syntagmatic and paradigmatic properties. Antonyms lend themselves well as pairings of meaning and form and therefore can be treated as constructions (Jones et al., 2012). Therefore, a treatment of antonyms using SBCG is presented in this study. Based on this treatment, I present a SBCG account of Arabic coordination as a contrastive construction in which antonyms frequently occur. The coordination construction is then compared to one use of coordination that presents antonym pairs as units referring to one concept.

This thesis provides a synchronic morphosyntactic description of the Hadari dialect, a variety of Gulf Arabic spoken in Kuwait, and presents a current documentation of this rapidly changing, under documented spoken dialect of Arabic. The description covers the basic morphology and syntax of Hadari, focusing mainly on the syntax. The description refers to Modern Standard Arabic both as a point of comparison and a point of reference when describing the spoken dialect's morphology and syntax. The study also draws on discussion of existing descriptions of the dialect and reflects upon their current adequacy. This thesis adopts a typological approach to describing the Hadari dialect, making reference both to Greenbergian typology and to modern typological theory. Two of the main typological theories applied in this description include an application of Matthew Dryer's exceptionless properties of V-initial languages (1990) and of the Branching Direction Theory (Dryer1992), to the spoken dialect. Furthermore, the study sheds light on the similarities and differences between Modern Standard Arabic and Hadari, regarding the expression of various syntactic aspects. One of the more significant contributions in this section is the typological description of the relative clause in Hadari. Furthermore, the thesis provides descriptions of clause structure, word order, modality, valency, copular clauses, interrogatives, negation, and subordination, in Hadari. The analysis is based on empirical data from recordings of everyday interactions in uncontrolled environment, television shows, radio broadcasts, and personal interviews.

Unsupervised approaches to learning the morphology of a language play an important role in computer processing of language from a practical and theoretical perspective, due their minimal reliance on manually produced linguistic resources and human annotation. Such approaches have been widely researched for the problem of concatenative affixation, but less attention has been paid to the intercalated (non-concatenative) morphology exhibited by Arabic and other Semitic languages. The aim of this research is to learn the root and pattern morphology of Arabic, with accuracy comparable to manually built morphological analysis systems. The approach is kept free from human supervision or manual parameter settings, assuming only that roots and patterns intertwine to form a word. Promising results were obtained by applying a technique adapted from previous work in concatenative morphology learning, which uses machine learning to determine relatedness between words. The output, with probabilistic relatedness values between words, was then used to rank all possible roots and patterns to form a lexicon. Analysis using trilateral roots resulted in correct root identification accuracy of approximately 86% for inflected words. Although the machine learning-based approach is effective, it is conceptually complex. So an alternative, simpler and computationally efficient approach was then devised to obtain morpheme scores based on comparative counts of roots and patterns. In this approach, root and pattern scores are defined in terms of each other in a mutually recursive relationship, converging to an optimized morpheme ranking. This technique gives slightly better accuracy while being conceptually simpler and more efficient. The approach, after further enhancements, was evaluated on a version of the Quranic Arabic Corpus, attaining a final accuracy of approximately 93%. A comparative evaluation shows this to be superior to two existing, well used manually built Arabic stemmers, thus demonstrating the practical feasibility of unsupervised learning of non-concatenative morphology.

This thesis investigates the phonology of loanword adaptation focusing on English loanwords in Urban Hijazi Arabic (UHA). It investigates the segmental adaptations of English consonants that are absent in UHA as well as the various phonological adaptations of illicit syllabic structures. It is based on dataset of around 100 English loanwords that were integrated into UHA that contain several illicit consonants and syllable structures in the donor language. This dataset is compiled from different published sources along with a data collection exercise. The first significant source is Abdul-Rahim (2011) a dictionary of loanwords into Arabic, while the other one is Jarrah's (2013) study of English loanwords into Madinah Hijazi Arabic (MHA) adopting the on-line adaptation. The third source is original pronunciation data collected from current UHA speakers. Furthermore, the Oxford English Dictionary (OED) was consulted for the etymology and transcription of the English words. The goal is to provide a thorough analysis of these phonological patterns whether consonantal or syllabic ones found in the adaptation of English loanwords into UHA. To accomplish this, the adaptations have been analysed according to two theoretical frameworks: the Theory of Constraints and Repair Strategies Loanword Model (TCRSLM) proposed by Paradis and LaCharité (1997) and Optimality Theory (OT) introduced by Prince and Smolensky (1993). The different proposed analyses in this study facilitated an evaluation of the adequacy of each of these theories in accounting for the discussed phonological patterns found in UHA loan phonology. The thesis concludes that OT better explains the adaptations, but neither theory fully accounts for the variety of adaptations found in UHA.