Font Size: a A A

Rethinking The Syntactic Structure Of Infinitives

Posted on:2011-08-08Degree:MasterType:Thesis
Country:ChinaCandidate:G C ChuFull Text:PDF
GTID:2155360305480032Subject:English Language and Literature
Abstract/Summary:PDF Full Text Request
Infinitives have been studied by many generative linguists, but interestingly, the categorization of infinitives is a puzzle and the majority of the studies take some infinitives as CP, while others as IP. There is no consensus on the status of infinitive constructions. Meanwhile the particle to which is a constituent to form an infinitive construction is generally categorized into INFL since Chomsky (1981).The following studies have been devoted to specifying the status of to, yielding various positions. For some analysts, to is taken as a Tense morpheme, while others take it as Modality since there is an"irrealis"meaning in control infinitive clauses. The question is that not all infinitives contain"irrealis"property; thus we have reasons to doubt the Tense or Modality proposal to the status of infinitive marker to.Devoted to the categorization of infinitives, on the basis of syntactic distributions of infinitives, this thesis posits that the infinitival marker to is a pure marker, not related to INFL, i.e. it is neither Tense nor Modality element. Actually, so-called"irrealis"is determined by the matrix verb, in other words, it is outside the infinitive structure.This thesis hypothesizes that the infinitive marker"to"is a Deverbalizer, which Deverbalizes its complement VP. Traditionally, infinitives, gerunds and participles are dubbed"verbal noun", which implies that those constructions show both Verbal and Nominal features, i.e. they have the features of [+V] and [+N]. Assume that the whole infinitive is DEVERBALIZED, the to-infinitive construction attains other non-verbal features, such as nominal and adjective features.The postulation that to is a deverbalizer can be applied to other nonfinite constructions like gerund and participles. The morphemes"-ing","ed"and"to"belong to the same category, that is, they are all Deverbalizers. Thus, we can assume that deverbalization is a process in which the VP is not inflected for Tense or Agreement but attains other features like [+N] and [+A]. As a Deverbalizer,"to"instantiates the marked non-finiteness and deverbalization of VP. The Deverbalizer is meaningless, merely implying Markedness of non-finiteness. While bare infinitive indicates the Unmarked non-finiteness and deverbalization of VP, the to-infinitive refers to the Marked non-finiteness and deverbalizaion of VP. Then, we suppose that the Deverbalizer takes a VP as its complement, projecting into a DevP, which accounts for the basic structure of infinitives. Since the infinitive has an overt NP or a covert PRO as its subject, it is generally considered as nonfinite clause. The syntactic structure of nonfinite clauses can not be captured by IP in that they lack INFL elements. Thus, this thesis attempts to postulate NonIP hypothesis to capture the structure of nonfinite clauses. NonIP is a maximal functional phrase, projected by its head Non-INFL Elements.In summary, based on their syntactic distributions, the thesis asserts that infinitives are clausal, but they are NonIP rather than IP. Therefore, the syntactic structure of the infinitive and the syntactic relations between its constituents should be reanalyzed as NonIP under the framework of X'theory.
Keywords/Search Tags:infinitive, category, syntactic structure, Devervalization
PDF Full Text Request
Related items