惯青扁包 郴 埃青拱

  • 埃青拱 郴 八祸 八祸

积己巩过楷备

八祸搬苞 :
6
傈眉急琶 Endnote Refworks
In this study, we propose an analysis of a novel subtype of modalized questions in Korean. Modalized questions express the speaker’s non-commitment to the propositional content by emphasizing the equal possibility of p and ¬p worlds given what the speaker’s doxastic world is (Kang and Yoon, to appear). Thus far, the inquisitive disjunction marker nka has been analyzed as the only modalized question marker, but we identify (u)l-kka as another subtype of modalized question marker. Our proposal comprises two main parts: First, we show that (u)l-kka can be morphologically decomposed into the irrealis marker (u)l and the question marker kka; specifically, (u)l can be further analyzed as an epistemic subjunctive whose semantic role is equivalent to the English epistemic might. Second, given three core properties of modalized questions, inquisitiveness, epistemic modality, and subjunctive mood, the behavior of nka and (u)l-kka can be properly incorporated under the framework of nonveridical equilibrium. The current analysis thus suggests an extended spectrum of (u)l from the previously known irrealis adnominalizer to the epistemic subjunctive suffix.
This paper investigates the structure of the postnominal numeral classifier construction, where a numeral classifier occurs after its associated Caseless nominal. I give some pieces of evidence that the classifier is not a functional head, but a noun, which has its own functional heads. So there are two head nouns, i.e., the classifier and its associated noun, in the postnominal numeral classifier construction. Regarding the structure of this construction, I assume, following Park S.-Y. (2009), that the classifier phrase and the associated noun phrase merge to make a small clause. But, unlike Park S.-Y., I propose that the former is merged as a subject and the latter as a predicate, and then, the latter (a predicate) moves over the former (a subject), which gives the correct order of the associated noun followed by the numeral classifier. In brief, the structure that I propose for the construction is an inverted predication structure.
This paper aims to present an analysis of the causative make. The English causative verb make is categorized as a syntactic causative construction. The causative make conveys different meanings with respect to the properties of the complement construction. make in direct causation constructions takes a bare infinitive complement, while make in indirect causation constructions takes a passive bare infinitive complement. The functional category ‘Voice(P)’ is used for a structural analysis of the causative make. In addition the two semantics features, [【direct] in matrix constructions and [【volition] in complement constructions, are adopted for a feature analysis of the two different constructions. First, [-direct] and [+volition] are suggested for indirect causation. The matrix construction cannot be passivized because of one of these features and its value, such as [-direct]. Second, [+direct] and [-volition] are posited for direct causation. There is a semantic condition on passivization such that make involving direct causative meaning can be passivized.
This paper investigates the strategy of extracting the so-called remnant constituent out of the portion to be pseudogapped/ elided. There have been three proposals on this phenomenon: (i) overt A-movement (Lasnik 1995); (ii) covert QR (Thoms 2016); (iii) overt QR (Tanaka 2017). We show that despite their effectiveness, these proposals stop short of accounting for all the empirical aspects of remnant extraction in deriving Pseudogapping. We go on to show that another conceivable option of covert A-movement as an instance of remnant extraction is superior in its empirical coverage. To boot, this proposed strategy provides a handle for accounting for the difference between coordinate and subordinate structure in the formation of Pseudogapping.
This paper attempts to provide an account for the Scope Economy effect in Right Node Raising (RNR) in English. In general, the pivot in RNR can take scope above both subjects in TP-coordination. However, it is not allowed when the subject in one TP-conjunct is not a scope-taking element, exhibiting the Scope Economy effect in the sense of Fox (2000). Assuming Scope Economy, I suggest that this effect can be best captured by the multidominance analysis: Being multidominated/shared between both conjuncts, QR of the pivot is permitted only if it induces a semantic effect for both conjuncts. This effect, however, constitutes a non-trivial problem for the rightward ATB movement analysis (cf. Sabbagh 2007) since being overt and unbound, the rightward movement would be blind to Scope Economy. As a consequence, RNR is taken as another context where Scope Economy can be detected, due to the fact that the pivot is shared.
Ahn & Cho (2017a) suggests that multiple fragments consist of two clauses plus additional ellipsis. Two important claims are as follows. First, prior to ellipsis, the sentential source can be a syntactically and semantically complete answer to the question. Second, island violation is not repaired by ellipsis. Based on these, Ahn & Cho (2017a-c) account for ill-formed examples related to multiple fragments. Recently, Chung (2019) discusses problems that the bi-clausal analysis of multiple fragments may have. This squib critically reviews the defects pointed out by Chung (2019) and attempts to show that they are not real. This squib also provides further evidence for the no-repair-by-ellipsis analysis of multiple fragments in Korean.
1