Sentential Object Complements in Modern Standard Arabic

This dissertation comprises a corpus-based study of the form and function of sentential object complements in Modern Standard Arabic (MSA). The study is motivated by the scarcity of investigations into MSA syntax as opposed to the syntax of Classical Arabic. Complements to three classes of complement-taking predicates (CTPs) ­ manipulation, cognition and utterance ­ are considered. The major part of the thesis is descriptive, covering morphological, syntactic and semantic features of MSA complementation. MSA complements are divided into eight structural patterns. The data are then analysed in a search for correlations between the semantics of CTPs on one hand, and the syntactic form of their object complements on the other. Theoretically, it would be possible for a language to make use of a very limited number of CTPs with rather unspecified basic meanings, where the meaning in a specific context would be decided by the complement pattern used. The CTPs and complements recorded in the corpus suggest that, to some extent, MSA does make use of the possibility of letting complement patterns modify the meaning of CTPs. The majority of CTPs in the corpus have a rather well-defined semantic content, and are also rather tightly associated with specific complement patterns. Nevertheless, there are some instances where the choice of complement pattern modifies the semantics of the CTP. One area in which typological studies have pointed to correlations between syntax and semantics in complementation is the use of indicative vs. subjunctive mood. This is manifested in MSA complementation mainly in the choice between the complementizers `anna (or `inna) and `an, the former being used with an indicative complement, and the latter with a subjunctive one. The principles found in the literature for choosing one or the other are, however, rather vague. More specific information about the use of these complementizers has been provided through the analysis of the MSA corpus performed in the last chapter of the thesis. The choice between the complementizers `an and `anna in the corpus is noted to be both systematic and consistent. Moreover, the analysis gives rather clear indications as to when complementizers are not used