Please use this identifier to cite or link to this item:
Title: COMP and the Left Periphery. Comparative Evidence from Romance
Authors: Paoli, Sandra
Keywords: Indo-European Languages
Romance Languages
Issue Date: 2003
Publisher: University of Manchester
Abstract: The combination of the recent interest in functional projections and their breaking down into syntactically and semantically different heads on the one hand, and of the recognition that the head of the Complementiser Phrase (CP) does more than simply demarcate clause boundaries on the other, has motivated the formulation of a more complex CP structure (cf. Rizzi, 1997; Benincà, 2001; Benincà and Poletto, 2002). In this system, what was traditionally known as the CP is split into a series of projections, semantically and syntactically distinct. Force and Finiteness delimit the system upwards and downwards respectively, acting as interfaces with the superordinate structure and the propositional content of the clause. Force° encodes information on the illocutionary force of the clause and Finº expresses information on relating to finiteness and modality. Through a comparative investigation this thesis addresses the issue of the content and function of an element’s belonging to the category traditionally labelled as ‘COMP’, the finite complementiser che. The aim of this work is two-fold. Empirically, it aims to present a detailed description of new or little studied data, focusing on the position that the elements hold with respect to each other. Theoretically, it aims to make a contribution to the understanding of the left periphery of the clause and of the status of the elements hosted therein. Furthermore, through the investigation of mood and agreement features, it addresses the relation and modality of interaction between the inflectional and the complementiser domains. The empirical evidence is derived from two dialects – i.e. non-standard varieties – spoken in North-western Italy, Turinese and Ligurian, from Romanian, from Early Romance and child French: these languages all share a construction in which what appear to be two finite complementisers are allowed to co-occur in the same sentence. In spite of the obvious similarities the constructions are the instantiation of different strategies: mood expression in the dialects and Romanian and topicality in Early Romance and child French. After having ascertained that both instances of che fill a position within the left periphery in Turinese and Ligurian, and that the lower che is a mood marker, Rizzi’s (1997) claim that the complementiser system reduplicates syntactic information expressed at the inflectional level is reinforced. The clear separation of the IP and CP systems is called into question when analysing the Romanian particle s ă and the Romanian left periphery. The conclusion is that in Romanian the CP and the IP can be syncretic categories: the IP thus is specified for features typically associated with the CP, such as [+wh] and [+focus]. A final comparison is made with another construction in which it seems that the complementiser is realised twice. The source of the data are Early Romance texts and child French: the ‘double complementiser’ is used as a strategy to give discourse prominence to a thematised phrase. As well as reinforcing the close connection between the CP and the IP domains, my findings support the idea of a differentiation within the complementiser system, and suggest that the use of the term ‘complementiser’ is not adequate and should be instead replaced by ‘subordinating particle’, ‘modal particle’ and ‘topic-marker particle’.
Appears in Collections:Dissertations (restricted access)

Files in This Item:
File Description SizeFormat 
RomanceCOMP(Paoli).pdf1.36 MBAdobe PDFView/Open    Request a copy

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.