Chapter Two Literature Review

2.1 Predicate NP Licensing:an Understated Field

Literature on non-argument NP licensing in small clauses in English shows that it is an understated topic, which has been relatively ignored in syntactic study with the background that argument NP licensing has been heated discussed in the field. The apparent contradiction partly derives from Chomsky's revision of Case Filter in GB theory which relates NP licensing with theta-role assignment, making Case Theory a part of Theta-criterion in form of Visibility Condition.Being casemarked/assigned/checked is regarded as a condition for arguments'theta-role being visible at LF.Only in this way can argument NPs be fully interpreted and the sentence acceptable（the derivation converges at LF in MP approach）.This argument-specific version of Case Theory naturally overlooks those nominals at non-argument positions.Constrained by this theoretical background and some other technical limitations, the licensing mechanism underlying non-argument nominals like predicate NPs and bare adverbial NPs has not received due attention.

Recall that the original purpose for Vergnaud's proposal is to identify NP distribution with regard of its syntactic relations with other elements in the structure; some linguists still try to identify the case-marking of non-argument NPs.But due to the technical limitation in GB theory, they fail to locate the case-assigner of non-argument NPs.

In the following parts of this chapter, we are going to introduce four possible approaches to the current issue.They are Caseless Hypothesis, Default Case Hypothesis, Case Agreement Hypothesis and Structural Case Hypothesis. These four approaches to the Case-licensing problem of predicate NPs diversify dramatically from the basic conception of the Predicate Case to the syntactic mechanisms conducting case-assignment.

Caseless Hypothesis denies the predicate NPs carry Case simply out of the non-argument position of the elements.This notion changes in Default Case Hypothesis, which assumes Cases on predicate NPs under the basic conception that all NPs should be subject to Case Filter.But grammarians have great difficulties in finding the case-assigner for the predicate NPs.So they attribute the Case on predicate NPs to a default of last-resort mechanism.Case Agreement Hypothesis makes progress in defining the co-indexation relations between the subjects and predicates as the passage for predicate NPs inherit Case from the subjects.Structural Hypothesis for the first time points out that predicate NPs should be treated like argument NPs as to their Case-licensing patterns.A relatively unified analysis is proposed of the Case-assignment of predicate NPs.That is predicate NPs, just like the argument ones, receive Cases through structural relations and mechanism.Although still having problems, Structural Case Hypothesis is the most theoretically valuable under the generative framework in terms of its interpretive capacity in dealing with the various Cases found in predicates cross linguistically.In the following part, we will present a thorough examination of these Hypotheses.