This study is a sociolinguistic investigation of the dialect contact between the Bedouins (Bs) and Sedentary groups (Ss) in central Najd in Saudi Arabia. Three linguistic variables have been examined quantitatively. The variables are: the merger /u/, /i/ > /i/ (B /u/ as in kull and S /i/ as in kill ‘all’), the 3SM suffixed pronoun (as in B kita:b-ih and S kita:b-ah ‘his book’) and the imperfect prefixes (as in B ya-ktib and S yi-ktib ‘he writes’). The data is based on sociolinguistic interviews of 23 Bs, grouped by age and tribe, and 13 Ss, grouped by age. Additionally, the relative sociolinguistic salience of the variables has been measured using the Social Category Association Test (SCAT), online identification task and multiple interviews.
The interview data has been statistically analysed using generalised linear mixed model in R. The data show that Bs are converging towards the S dialect, to varying degrees across the examined linguistic variables. The Bs’ convergence on the S variant of the merger (/i/), exhibiting the highest rate, is conditioned by the factors of tribe, education and level of contact. The Dawasir tribe were more likely to converge on /i/ than the other tribes, while speakers with low levels of education and contact with the Ss were less likely to converge on /i/. The Bs’ convergence on the S variant of the 3SM suffixed pronoun, which is quite low, is conditioned by age. Younger Bs were more likely to converge on -ah. As for imperfect prefixes, the Bs did not show any convergence. The interview data for the Ss, on the other hand, show consistent use of their variants, except for 4 speakers who code-switched to
Standard Arabic. The data from the SCAT, online identification task and multiple interviews reveal that the merger is salient (it is a marker), while the other variables are not (indicators).
The quantitative data and qualitative observations of other linguistic variables indicate that a Najdi koine has not formed; however, it is expected to be complete after two or three generations and will most likely exhibit S and Standard Arabic features, but not B.
