| Summary: | 碩士 === 國立清華大學 === 語言學研究所 === 105 === This thesis aims to investigate the construction that has been called Small Clause, whose structure has been subject to much debate for a long time since this term firstly appeared in Williams (1975) and the hierarchical structure of nominal predicates in Mandarin Chinese has received scanty attention in the past. The main assumptions in this thesis is structured around three associated issues.
First, the analysis of the internal structure of small clauses and the dependent relationship holding between the subject and the predicate is provided. I have shown that copulas (covert or overt) play a crucial role in mediating the subject-predicate relationship between the subject and its predicate syntactically and semantically introducing a state or dynamic eventuality into the clause. In addition, it is found out that there are three copulas that can be inserted into the small clause, the patterns of syntactic distributions and semantic features setting these three copulas apart; in particular, the semi-copula zuo implies a change and a coming of states, whereas copulas BE and shi describe a state.
Second, the selectional restriction holding between the matrix verb and the entire small clause construction has been investigated. It is suggested that verbs taking the small clause as a complement can be divided into three types: verbs of calling including jiao ‘call’, chenghu ’call’, epistemic verbs, typically dang ‘consider’ and change of state verbs such as xuan ‘choose’, qu ‘marry’ and ren ‘accept’. The calling-verbs take the stative eventuality as a complement, headed by the null copula BE. Syntactically, the null copula is defective in that it cannot support a complex nominal structure except for bare nouns. The epistemic verb also prefers a state of being event, which can be headed by the null copula BE or the overt copula shi. If the clause is headed by shi, the nominal predicate structure can be not-bare, indicating a quantity meaning. Change of state verbs take a dynamic copula zuo only, accompanied by the additional meaning become.
Third, I propose that nominal expressions in Mandarin Chinese is a layered structure, composed of three distinct functional projections, in which SDP is linked to Referential, PDP is associated with Numerability, and KIP is related to the Delimitability. Nominal predicates do not project the SDP layer for it cannot be referential. As for PDP and KIP, I suggest that count nouns and proper names are base-generated in KIP layer. Notably, classifiers in Mandarin Chinese should be further divided into count-classifiers and massifiers. The [Num-Count classifier] sequence marked as [+Numerability] is mapped onto the PDP layer, associated with the quantity meaning. By contrast, to project the KIP, mass nouns must be coupled with massifiers to denote a property meaning.
|
|---|
