Natural languages go through two major cycles in their diachronic change. A high synthetic marking characterizes the first cycle, and a high analytic marking characterizes the second. This thesis investigates an emerging analytic passive in Modern Standard Arabic (MSA), representing the analytic cycle. This construction is designated periphrastic passive since two grammatical morphemes mark the passiveness. The older morphological passive construction in Classical Arabic (CA) and MSA, representing the synthetic cycle, is juxtaposed with the periphrastic passive. Given the inconsistent passive characterization in the literature, the comparison between the two passive forms is couched in the prototypical passive analysis. This thesis seeks to show that the periphrastic passive in MSA has grammaticalized to perform the passive function. It argues that the main verb in the periphrastic passive, i.e., tamma/yatimmu, has grammaticalized to a passive auxiliary. The corpus data of CA and MSA about tamma/yatimmu complementation, the subjectverb agreement, and the frequency of tamma/yatimmu show the grammaticalization of the periphrastic passive. The lexical source of the auxiliary tamma/yatimmu, i.e., ‘finish,’ is also attested to perform the passive function in Colloquial Icelandic (CI). The commonality between the lexical sources in the two passive constructions in MSA and CI suggests that the lexical source ‘finish’ could serve as a lexical source of passive constructions.

This sociolinguistic study examines the various functions of Arabic-English code switching in the Egyptian talk show ‘Shabab Beek (literally: Young by You; communicatively: The Young Speak)’. In addition, this study investigates the syntactic categories and types of switches to English. The data consist of approximately four hours and forty-five minutes of YouTube videos of the talk show in which code switching to English occurred. The videos are collected from six episodes of the show that were aired in October 2010. The show featured three categories of speakers, show hosts, guests, and callers. The findings show that most of the switches were produced by show hosts and guests while callers produced very few switches due perhaps to the limited number of phone calls received in the selected episodes. The speakers mostly used nouns when they switched to English. Nouns are followed by adjectives and noun phrases. The most prevalent type of switches in the data is tag switches followed by intrasentential and intersentential switches, which occurred rarely. Finally, analysis revealed eight functions of code switching in the data. These are difficulty retrieving an Arabic expression, quotation, euphemism, reiteration, message qualification, academic or technical terms, association with certain domains, and objectivization.