The Effects of Ethnicity, Contact, Sense of Identity, and Social Attitudes on Dialect Contact in Jeddah, Saudi Arabia

Abstract

Dialect contact may lead to one of several linguistic outcomes, including (among others) levelling, accommodation, divergence, or convergence. Jeddah presents a unique linguistic situation in Saudi Arabia due to it being one of the three cities, along with Makkah and Madinah, which have a large Saudi non-Bedouin demographic. Given that Bedouin and non-Bedouin social groups are technically in daily contact in Jeddah, this urgent research question arises: do the marked linguistic features of each group weaken or perhaps even level out? Thus, this thesis investigates the dialect contact situation in Jeddah, Saudi Arabia, specifically how it affects the distinct varieties of Hijazi Arabic spoken by Bedouins and non-Bedouins in the city. This study investigates the influence of ethnicity, dialect contact, sense of identity, and social attitudes on variation in Hijazi Arabic in a sample of 32 young adult speakers (aged 19-22) in Jeddah. The sample was stratified by Bedouin versus non-Bedouin ethnic background and grouped by neighbourhood type: predominantly Bedouin, predominantly non-Bedouin, and mixed. Data were collected through sociolinguistic interviews and written questionnaires measuring participants’ sense of identity and social attitudes. The analysis focuses on three linguistic variables: (θ), (ð), and the third-person masculine pronoun (-ah). Distributional and regression analyses show that Bedouin and non-Bedouin Jeddawis belong to different speech communities because those in mixed and unmixed neighbourhoods (both Bedouins and non-Bedouins) do not share linguistic norms. Moreover, for a speech community to exist, there needs to be constant contact and interactions among its members, which cannot be said about the Bedouin and Non-Bedouin neighbourhoods. At the same time, the analyses show that the four social predictors (ethnicity, dialect contact, sense of identity, and social attitudes) affect two of the three linguistic variables: (θ) and the third-person masculine pronoun (-ah). Since the type of neighbourhood acts as a proxy for the degree of contact between Bedouins and non-Bedouins, the findings suggest that the degree of contact between the two groups affects the variation of the (θ) and (-ah). The variants of both (θ) and (-ah) in the two unmixed neighbourhoods (Bedouin and non-Bedouin) remain distinct from one another due to lack of contact. On the other hand, those in mixed neighbourhoods behaved linguistically as a one- speech community and used competing variants at similar rates regardless of individuals' ethnic backgrounds. i The results additionally show that there is a correlation between both sense of identity and social attitudes and the variables studied. Participants with a strong sense of identity favoured variants associated with their ethnic group. In contrast, those with a weak sense of identity used the variant associated with the other ethnic group. The same was found regarding attitudes: those with positive attitudes towards the other group exhibited higher rates of use of the variants associated with the ethnic group they do not belong to, and those who had negative attitudes exhibited little to no use of the variants associated with the other group. My results highlight the emergence of a supra-local Arabic variety in Jeddah wherein the urban marked [t] and [d] lose their distinctiveness in favour of the mainstream features [θ] and [ð] in the mixed neighbourhood. The results also suggest that they do so due to their markedness and association with the non-Bedouin demographic in Jeddah specifically and Hijaz as a whole. The results also show the levelling of the Bedouin 3rd person masculine suffix in the mixed neighbourhood in favour of an urban variant, unlike the phonological ones due to its disassociation with non-Bedouins.