Please use this identifier to cite or link to this item:
|Title:||Direct objects in Persian|
|Publisher:||University of Southern California|
|Abstract:||This work challenges some prevalent claims and assumptions about Persian syntax and offers an alternative account. The standard assumption regarding the direct objects in Persian is that they form a natural class syntactically and differ from one another semantically, hence, the widespread specific/definite versus nonspecific/indefinite categorization of these elements (where only the former are marked with the morpheme –râ). This dissertation argues that the perceived differences between different object types are, in fact, a result of differences in their syntactic structure, since only some of the direct objects in Persian have a full DP projection (hence, also challenging the idea that all argument nominals are DPs (e.g., Longobardi 1994)). It is shown, furthermore, that the DP versus non-DP dichotomy is not sufficient for accounting for data pertaining to direct objects. While, superficially the non-DP direct objects appear to behave similarly, a closer inspection reveals that they do not form a natural class. Some only have an NP, while others have a NumP projection. The former are property-denoting expressions subject to Pseudo Noun Incorporation (Massam 2001) and the latter are weak determiners in the sense of Milsark 1974 and Diesing 1992. Building on the assumption that case (and referentiality) are only associated with the DP projection, it is argued that the morpheme –râ is a default case marker on DPs, which could not have their case features valued either as ‘nominative’ or ‘dative’. Therefore, this morpheme not only marks direct objects that are DPs but also the DPs in the (sentential) topic position. This would account for the differences in the interpretation of the –râ-marked DPs as well as their relative order. In order to have a full picture of the direct objects in Persian, this dissertation also examines cases in which the direct object appears as a clitic. It is argued that in those instances, the nominal expression that merges with the verb is the non-overt element pro, which is identified by the clitics’ -features. This analysis can also account for the Clitic Doubling facts as well as cases in which the clitic appears as the indirect object.|
|Appears in Collections:||Dissertations (restricted access)|
Files in This Item:
|Ganjavi_diss.pdf||1.38 MB||Adobe PDF||View/Open Request a copy|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.