1,720,982 research outputs found

    Particles in Korean: Focusing on verbal particles

    Get PDF
    The purpose of this paper is to investigate the morpho-syntactic nature of Korean particles/little words." We begin with providing a survey of some works on nominal and cross-categorial particles. We agree with the view that nominative/accusative case markers are phrasal affixes" and that postpositions and delimiters are clitics. Then, assuming that -si-, -ess-, keyss- and -ta are inflectional affixes, we focus on analyzing those verbal particles which occur after the mood marker: the quotative marker -ko and other post-verbal elements like -nta in khu-ta-nta. We argue that these elements should be analyzed as clitics, at least for one reading. According to our analysis, many closing" verbal particles are inflectional affixes. However, some closing ones are ambiguous between inflectional affixes and clitics. Some connecting" verbal particles are also ambiguous, but some others are analyzed as clitics-only elements

    An Analysis of the Korean [manyak ... V-telato] Construction : An Indexed Phrase Structure Grammar Approach

    No full text
    Concord adverbial constructions in Korean show unbounded dependency relationships between two non-empty entities. There are two different types of unboundedness involved: one between a concord adverbial and a verbal ending and the other between the adverbial as a modifier and a predicate. In addition, these unboundedness relationships exhibit properties of “downward movement” phenomena. In this paper, we examine the Indexed Phrase Structure Grammar analysis of the constructions presented in Chae (2003, 2004), and propose to introduce a new feature to solve its conceptual problem. Then, we provide an analysis of conditional-concessive constructions, which is a subtype of concord adverbial constructions. These constructions are special in the sense that they contain a seemingly incompatible combination of a conditional adverbial and a concessive verbal ending. We argue that they are basically conditional constructions despite their concessive meaning

    Right Node Raising in an Indexed PSG

    Get PDF
    This paper is concerned with the analysis of Right Node Raising (RNR) constructions in English. These constructions have special properties which cannot be handled effectively with traditional mechanisms, regardless of whether some kind of displacement is assumed to be involved or not. Here I propose a GPSG-style analysis of RNR under the Indexed Phrase Structure Grammar (IPSG) framework of Chae (1992). The basic idea behind this analysis of RNR is that the factor remains in the second conjunct and the gap in the first conjunct is licensed by this factor

    A Categorial Analysis of Bracketing Paradoxes in English

    Get PDF
    In this paper I present a formal account of "bracketing paradoxes" in morphology under a categorial framework. In this kind of system, Category Raising (CR) plays an important role in removing the "mismatches" between phonology and semantics. However, no previous approach was successful in properly restricting the application of CR. So, I propose two constraints for this purpose: a formal restriction and a lexical restriction. The former says that CR applies only when the arrangement of the morphemes in a word fits with a specified format. The latter, which is imposed by the lexicon, is an auxiliary constraint to the first restriction. We will see that this system can also provide a good account of bracketing paradoxes in compounds

    An Analysis of the Korean [manyak ... V-telato] Construction : An Indexed Phrase Structure Grammar Approach

    No full text
    textConcord adverbial constructions in Korean show unbounded dependency relationships between two non-empty entities. There are two different types of unboundedness involved: one between a concord adverbial and a verbal ending and the other between the adverbial as a modifier and a predicate. In addition, these unboundedness relationships exhibit properties of “downward movement” phenomena. In this paper, we examine the Indexed Phrase Structure Grammar analysis of the constructions presented in Chae (2003, 2004), and propose to introduce a new feature to solve its conceptual problem. Then, we provide an analysis of conditional-concessive constructions, which is a subtype of concord adverbial constructions. These constructions are special in the sense that they contain a seemingly incompatible combination of a conditional adverbial and a concessive verbal ending. We argue that they are basically conditional constructions despite their concessive meaning

    Light Verb Constructions and Structural Ambiguity

    No full text

    Clitic Analyses of Korean Little Words

    No full text

    Verbal Nouns and Light Verbs in Korean

    Get PDF
    One of the main purposes of this paper is to establish the identity of verbal nouns (VNs): they constitute a special sub-class of nouns which are responsible for both subcategorization and case-marking. They are neither verbs (even though they have some verbal properties) nor do they have the same properties as regular nouns. As for their lexical category, they are nouns because they show the morphological and distributional characteristics of typical nouns. They also exhibit some syntactic properties which are not shared by regular nouns. Another purpose here is to provide syntactic criteria to factor out VNs from the set of potential VNs. Some verbs can function both as a heavy verb and a light verb (LV). LVs need to be combined with VNs to become regular predicates. We can factor out VNs from other nouns by considering whether the noun in Question has its own complements which can be verbally case-marked or not. This decision can be made on the basis of the subcategorization frames and/or the thematic roles of the verb under consideration when it functions as a heavy verb. If there are some complements in the clause which are not attributable to this heavy verb, they can be judged to come from some other source with verbal properties. which proves that the noun is a VN
    corecore