Diglossic code-switching in Kuwaiti newspapers

Abstract

The present study investigates the phenomenon of diglossic code-switching between Standard Arabic, as a High variety, and Kuwaiti Arabic, as a Low variety, in Kuwaiti newspaper articles. The study was precipitated by the paucity of research on the linguistic characteristics of newspaper discourse generated within this region as well as Kuwaiti perceptions towards this medium of communication. The frameworks adopted in this research were extended to novel contexts and were also utilised to gain new insights into several dimensions of diglossia, most of which have never been explored before. The findings of this study indeed revealed important insights into how diglosia is changing and how participants both use and perceive diglossic code-switching. The investigation was carried out to explore three main dimensions of code-switching in newspaper articles in Kuwait: changes in attitudes and frequency of code-switching use, the social motivations for it and the morphosyntactic constraints associated with it in this context. The first involves the study of changes in both language attitudes and in the frequency of code-switching in newspaper articles over the last 29-30 years. The second part of the study offers a social motivations’ analysis of code-switching in newspaper articles by appealing to the ideas captured in the Markedness Model (MM), proposed by Myers-Scotton 1993a. The primary goal of this element of the research was to seek explanations for the diglossic code-switching strategies identified in a sample of twelve newspaper articles. The third aspect explored in the research relates to the testing of the Matrix Language Frame (MLF) model which applies specifically to the morphosyntactic constraints thought to operate in spoken code-switching contexts (Myers-Scotton 1993b, 2002). A key objective of the research overall was to evaluate the models themselves which have not, to my knowledge, been appraised heretofore using written data of this kind. In general terms, my findings regarding attitudinal change and code-switching frequency suggest that, despite the differences exhibited by a range of social variables, the nature of the attitudes expressed by the readers and columnists alike still reflect the traditional diglossic situation in Kuwait. Moreover, an analysis of language attitudes, ix employing the ‘apparent time’ hypothesis, shows that there is indeed a change in language attitudes in Kuwait between one generation and the next. This change, however, is contrary to predictions as it actually shows a favouring effect within the community at large for the H variety, i.e. SA. Non-parametric statistical analyses (specifically the Wilcoxon Signed Ranks, Mann-Whitney and Kruskal-Wallis tests) were selected as most appropriate for discriminating quantitative distinctions in the analysis of attitudes. Furthermore, an investigation of how common code-switching has become over the last three decades reveals that there is, in fact, static code-switching frequency, indicating that the practice of code-switching has remained relatively stable between 1985 and 2014-15. As for the second and third dimensions of the research, it was shown that the MM offers a very useful explanation of the linguistic behaviour of columnists and reveals the intricacies of their code-switching strategies which can be related to their understanding of community perceptions towards diglossic codeswitching in Kuwait as captured in other aspects of the research. A key finding with respect to the testing of the MLF model itself was how difficult it actually was to diglossic code-switching in a written context. My research clearly shows that the MLF approach does not, in fact, provide as much insight into the dynamics of the phenomenon as it clearly does when applied to conversational exchanges and this is partially due to the problems identified in my thesis regarding the analysis of Arabic code-switching in writing.