A Study On The Blocking Effect Of Ziji In Mandarin Pivotal Sentences

This thesis is conducted under the framework of the Minimalist Program to explore the blocking effect of ziji in Mandarin pivotal sentences(henceforth PS).The simple reflexive ziji exhibits some distinctive referential properties,which seem to be exempt from the Principle A proposed by Chomsky(1983),which stipulates that an anaphor must be bound in its governing category.A number of cases can be found in which ziji is bound by a long-distance antecedent outside its local domain.Previous studies addressing the referential properties of ziji focus primarily on complex sentences with object clauses in Chinese,in which either the embedded subject or the matrix subject may serve as the antecedent of ziji.However,it has been found that long-distance binding of ziji may not be possible if the two subjects differ in person features,a phenomenon called “the blocking effect”,which was first observed by Huang(1984).The blocking effect of ziji is noteworthy because it may involve the interaction of different factors such as(in)consistency of person features of possible antecedents and particular sentence patterns.A comprehensive account of ziji’s referential properties cannot be achieved without adequately addressing the blocking effect of ziji.Extant analyses on the binding relations,especially the blocking effect of ziji are mainly conducted under the framework of Chomsky’s Generative Grammar(G&G).Studies under the Government and Binding theory have come up with several analyses to account for the general properties of ziji including the blocking effect,such as Anaphoric Pronominal Analysis,Parameterization Analysis,LF-movement Analysis,and Relativized Subject Analysis.Representative of the studies under the Minimalist Program are Hu’s(1998)and Cheng’s(2004)analyses.Although a number of important findings have been made in previous studies,little attention has been paid to the referential properties of ziji in Mandarin PS,especially the blocking effect of long-distance reference,which presents a different picture from what has been observed in literature with regard to complex sentences with object clauses.The present study focuses on the blocking effect of ziji in PS.Under the framework of MP,we intend to address the following three questions.Firstly,what are the general differences in the blocking effect of ziji between PS and complex sentences with object clauses? Secondly,how does the blocking effect of ziji vary in different types of PS? Thirdly,how can we account for the blocking effect of ziji in different types of PS under an integrated framework of MP?Given the above research questions,the objectives of this study are twofold.On the descriptive level,we aim to present an adequate depiction of the blocking effect of ziji in Mandarin PS by exploring the semantic and syntactic properties of different types of PS and how these properties exert influence on the reference of ziji.On the explanatory level,we aim to give a principled account of the blocking effect of ziji in different types of PS within MP by analyzing how the syntactic structures of PS are generated and explaining why the blocking effect of ziji in different types of PS can be induced(or not)with resort to the Theory of Feature Checking.This thesis has made the following major findings.Firstly,the blocking effect of ziji displays a symmetrical relation in PS whereas it is non-symmetrical in complex sentences with object clauses.Besides,the blocking effect is not merely restricted to person features but also to the properties of the predicates in PS.In the type of PS with strong causativity,the blocking effect is mainly activated by mismatch in person features of the matrix subject and the pivotal element.But in the type of PS with weak causativity,the properties of the predicates play a decisive role in the blocking effect,whereas the person features may not take effect at all.Secondly,different types of PS may involve different types of empty categories in their underlying structures,which is another factor contributing to the activation of the blocking effect.In the type of PS with strong causativity,the empty category is PRO and ziji that seems to be bound by DP2 on the surface actually takes PRO as its antecedent in Deep Structure.It is the unspecified feature of zi valued by PRO that can give rise to the blocking effect.In the type of PS with weak causativity,it is the feature checking of ji that can determine whether the blocking effect is induced or not.Thirdly,when the predicate verbs have strong [+cause] feature or belong to the two subtypes of mingling ‘order’ and qingqiu ‘ask’ groups with weak causativity,the light verb v is checked before Spell-out and ziji checks its features in the full derivation of PS.The difference between the two is that in the former,zi checks its features against PRO and ji moves up to check against DP1.If the valued person features of zi and ji do not match,it is the feature of zi that determines the activation of blocking effect in line with the Economy Principle.In the latter,zi cannot check against the trace but ji can move to check its features against the higher DP1,with no blocking effect induced at all.When the predicate verbs belong to the subtypes of haozhao call for’ and quanling ‘persuade’ with weak causativity,the light verb v is checked after Spell-out and ziji checks its features during the derivation of v P.But zi still cannot check its features with t left by DP2 and ji can only move to check against DP2,which will give rise to the blocking effect.