SOAS Research Online

Abstract

Research on the Chinese language from the perspective of Government-Binding Theory requires the satisfactory execution of the GB modules in Chinese syntax. This thesis investigates a number of aspects of Mandarin Chinese syntax from this point of view. Chapter 1 outlines the GB framework and raises the issue of constituency in Chinese. Chapter 2 discusses a number of GB theorems with reference to Chinese. They include X-bar syntax and head ordering, the governing function of INFL, distribution of PRO and pro, 0- and Case-marking, Case-adjacency, the notion of accessible SUBJECT, empty-topic binding, the ECP and its rival approach, generalised binding. Chapter 3 examines the structural process of BA-insertion (object-marking) and BEI-insertion (passive-marking). It is argued that Move a should apply to [vp...NP...] where NP = [NP, VP/PP] and VP = a quanti-ficational phrase (QP), but restructuring to [NP, PP] which is at the same time a member of a higher argument structure. Chapter 4 investigates wh-quantifier constructions. A new COMP-marking system, Ett4H, +/-QUANO, is introduced to license syntactic wh-quantifier movement, whose trace needs to be locally bound if the moved wh-quantifier is an adjunct, but need not be locally bound if the moved wh-quantifier is a non-adjunct. Chapter 5 considers descriptive clause constructions. It is argued that there is an empty operator movement in descriptive clauses which display undounded dependency, and that descriptive clauses are not within the NPs they modify, as they differ from relative clauses in distribution and in escaping the CNPC. Chapter 6 analyses serial verb constructions. The long-standing issue of whether a V-V series is of coordinate or subordinate structure is re-examined. A subject-control analysis is proposed for these constructions, as they are more likely to be of a subordinate structure. Evidence for empty operator movement in some purposive constructions is also studied. Chapter 7 deals with the so-called pivotal constructions. It shows that determining the constituent structure of these constructions is the key to treating them as object-control, as there is no distinction between finite verbs and infinitives in Chinese. Chapter 9 attempts to treat subordinate clause constructions. It is shown that raising is not possible in these constructions containing the item BA, whose occurrence (in these constructions) is to Case-mark the object of the matrix verb, which becomes defective after suffixed with the particle 'De'. Other alternatives to BA-insertion are verb-reduplication, NP-movement (of the object of the matrix verb into the subject position of the subordinate clause), or the attachment of an additional verb to the matrix verb (hence the so-called "resultative verb-complement"). Evidence for not treating 'De' as a complementiser is independently produced. Chapter 9 summarises the findings of this thesis.