This paper shows that so-called circumstantial predicates have PRO as their subjects. More specifically, we point out that circumstantial predicates and PRO form a small clause (SC), arguing that three types of secondary predicates, namely circumstantial, depictive, and resultative predicates, are affiliated to different positions in VP. We claim, from a theoretical viewpoint, that the SC formed by a circumstantial predicate and PRO is adjoined to VP at D-Structure. Furthermore, we argue that the segment of VP which inherits barrierhood from the SC blocks wh-extraction of circumstantial predicates.
This paper explores a way of reducing Rizzi's Relativized Minimality to Chomsky's Minimality Condition. It is argued that, given the X′ Convention and the Minimality Condition, it is possible to produce a unified approach towards the three domains which supply the empirical foundation for the Relativized Minimality. Furthermore, to account for the possibility of the extraction of an element from the domain of a head governor, it is proposed that there are two kinds of minimality barrier: a weak minimality barrier which blocks a chain of antecedent-government relations formed by wh-movement, and a strong minimality barrier which blocks both a chain of antecedent-government relations formed by wh-movement and one formed by NP movement.
In this paper, we will explore the possibility of raising-to-object within the framework of Government and Binding. We will slightly modify the parameter proposed by Authier (1991). It will be argued that the infinitival complement clause of verbs like believe is realized as CP. Assuming Rizzi's (1990) Relativized Minimality, the conjunctive ECP, and the VP internal subject hypothesis, we will show that a number of cross-linguistic considerations of the ECM construction support our raising-to-object analysis.
This paper presents an analysis of English ellipses, assuming no empty categories for missing constituents. It has several advantages over the approach of postulating empty categories and explaining their distribution in terms of the Empty Category Principle. Especially, it provides a principled account of the fact that overt wh-words may not occur in certain ellipses but are perfectly allowed in the corresponding non-elliptical constructions, which is totally inexplicable to the principle of no vacuous quantification under the assumption that ellipses involve syntactic empty categories.
Predicate Nominals are non-referential but allow long-distance movement, which is a contradictory fact under Rizzi-Cinque's approach to movement in terms of referentiality. This paper argues that this discrepancy can be explained under the VP-internal subject hypothesis in conjunction with the Split INFL hypothesis. It is also argued that predicate nominals must receive Case, contrary to Chomsky's (1986) Visibility Condition. The existence of AGRoP is discussed in connection with predicate nominals.
This paper presents a coherent account of the sentence internal reading of the English adjectives same and different. These adjectives are argued to have an anaphoric property in the sense that they are assigned a referential value by some external source in the way that anaphoric pronouns and other anaphoric expressions are. The analysis is empirically motivated by the observed parallelism between anaphoric expressions and same/different with respect to such a grammatical phenomenon as the crossover effect. Although the present analysis is not a total alternative to the previous study in Carlson (1987), an empirical ground is provided for maintaining the lines pursued here.
This paper deals with the structural difference between cause and make in terms of semantic representation. It is argued that the semantic structure of cause with a to-infinitive is more complex than that of make with a bare infinitive, in that the former includes the function GO as the second argument of CAUSE, whereas the latter does not. It is shown through discussion that the to-infinitive occurring with some causative verbs like cause can be interpreted as GOAL, as in the case of the preposition to. Furthermore, it is also argued that the passive of make with the to-infinitive complement does not constitute a counterexample to the present analysis.
This paper presents a study of the meaning of the English definite article using the framework of Cognitive Grammar developed by Ronald Langacker. What we focus on in particular is two kinds of non-specific usage: the generic definite and what we call the role definite. Our analysis in terms of Cognitive Grammar constructs and the newly introduced Identity Model will show that those definites designate imaginary individuals set up for expressive purposes. Definiteness itself is defined as the speech-act participants' mental access to the identity of the designatum.
This paper is concerned with the presentation of constraints in terms of Jackendoff's (1990) conceptual structure to accommodate the acceptability of English and Japanese gapped sentences. It is argued that the semantic aspects of Gapping are well captured by referring to the substitution mechanism based on the conceptual structure of the antecedent clause to get the conceptual structure of the gapped clause. The constraints prescribe the substitution possibilities and reflect perceptual strategies contributing to the difficulty of processing gapped sentences. This in turn accommodates some exceptional cases to the constraints and the difference between the English and the Japanese sentences.
Instruments generally fall into two classes: those which can appear as the subject of sentence and those which cannot. The difference stems from the semantic dependency between causation and instrumentality involved in the meaning of verbs. This article presents a lexical semantic analysis of instruments in relation to the mapping of semantic structure onto syntactic structure. It argues that the underlying structure of the instrumental construction is a direct structural reflection of semantic relations represented in verbs' lexical conceptual structures. Cross-linguistic evidence in support of this claim is partly drawn from the applicative construction and the serial verb construction. This account offers a challenge to a widespread assumption that the mapping between semantic and syntactic structures is mediated by argument structure.
This paper attempts to account for different scope behavior of quantified subjects appearing in three types of complement sentences, SC (=small clause), IP, and CP. In particular, I present an analysis where the scope behavior depends crucially on whether the traces of the subject QPs at LF are head-governed by elements internal to the complement sentences in question or by those external to them. I also argue that as a consequence of this analysis, a principle governing absolute scope of QP must be formulated in terms of head government and g-projection in the sense of Kayne (1983).
This paper argues that negative polarity items like any and ever are words which constrain mental processes in the sense of Blakemore (1987), and this may be viewed as an NPI-licensing condition. These words require that a proposition corresponding to an utterance containing NPIs should be processed in what I call the cognitive structure of negation. Relevance theory, proposed by Sperber and Wilson (1986), provides an adequate way of describing this mental process. The infelicity of NPIs in certain environments which would pose problems for Ladusaw (1979) is explicable under my analysis. It leads us to an interesting claim about the human congnitive processing procedure.
This paper investigates the intrinsic semantic meaning of kedo (‘but’) in utterance-final use and what kind of pragmatic effects an utterance appended with kedo would derive. I explain how softening or hedging effects are derived by the utterance-final use of kedo. I compare kedo in utterance-final use with that in utterance-medial use and point out the difference in its pragmatic function. I identify the pragmatic function of kedo within the framework of Relevance Theory which is a theory of communication and cognition.
Proposals for foot structure in Old English have hitherto been restricted mainly to syllable-based accounts and formulated such a phonological process as high vowel deletion in terms of syllable-counting feet. I argue here, however, that syllable-based accounts are inadequate both descriptively and explanatorily even if they utilize two metrical planes, because there are some examples whose stress location is not expected and other processes than high vowel deletion (e.g. resolution, Sievers's Law, and West Germanic gemination) are not captured uniformly. I present an integrated and uniplanar theory of Old English prosody, making judicious use of moras instead of syllables. This amounts to saying that the language is mora-counting.