글로버메뉴 바로가기 본문 바로가기 하단메뉴 바로가기

논문검색은 역시 페이퍼서치

언어와 언어학검색

Language and Linguistics


  • - 주제 : 어문학분야 > 언어학
  • - 성격 : 학술지
  • - 간기: 계간
  • - 국내 등재 : KCI 등재
  • - 해외 등재 : -
  • - ISSN : 1225-4967
  • - 간행물명 변경 사항 :
논문제목
수록 범위 : 54권 0호 (2012)

영어 학습자 오류 코퍼스 구축과 오류 탐지 규칙 생성

김동성 ( Dong Sung Kim )
7,600
초록보기
This paper introduces the error corpus of Korean learner English and mal rules to detect the errors. Based on the corpus, we classified 42 error types. Our criteria for error classification are more general in order to enhance agreement rate and decrease errors. For generating mal rule, we testified two different grammars. One is Context Free Grammar (CFG) from Penn Treebank. The other is the typed feature structure grammars based on the Head-Driven Phrase Structure Grammar (HPSG), using Natural Language ToolKit (Bird et al. 2009). We advanced grammatical formalism from CFG to HPSG since CFG needs abundant phrasal markers causing over-generation, structural ambiguity and complexity.
6,400
초록보기
This paper observed two German grammatical phenomena “future tense” and “imperfect subjunctive”, which are related to each other semantically. German grammarians have also attempted to explain and to describe both phenomina. But they have treated the grammatical categories in isolated sentences and their interests were limited to the formal structure and sentential semantics. Therefore, not many studies have been performed on the future tense and imperfect subjunctive in the contextual level. This paper aims to fill this gap: Each has its own semantic characteristics and the features are revealed in specific contextual structures. A sentence in the future tense means a guessed event, namely an unreal event which can be realized in most cases. High probability of realization is given in the context. The same applies to the imperfect subjunctive. A sentence in the imperfect subjunctive means either a potential or contrafactual event and each event demands a specific contextual condition.

The Interpretation of Scalar Implicature: The Acquisition of even

( So Young Kim )
6,300
초록보기
Despite the active discussion on the pragmatic feature of even from theoretical perspectives, there has been no experimental investigation into children’s sensitivity to scalar implicature in sentences with even. This study aims to investigate how English-speaking children aged from 4 to 5 derive scalar implicature of even. To examine children’s calculation of implicature, a guessing task was created. The results suggest that, unlike adults, preschoolers are unaware of the scalar implicature that the focus particle even induces. a general picture emerged of two child groups with different capabilities for computing scalar implicature. The first group consistently failed for both sentence types with even; the second group consistently succeeded for both sentence types with even.

부정초점과 단순초점으로 설명한 한국어 NPI 자료의 문제

김정란 ( Joung Ran Kim )
5,200
초록보기
Many researchers have argued that some kind of intervention constraint applies for the Korean NPI and the licensing negation at LF. Some even argued that this intervention constraint actually applies at S-structure. I find the intervention constraint does work at LF, but it alone cannot adequately explain some data. This paper shows that some data, not previously taken care of, can be explained away if we introduce the concepts of two different foci: negation target focus and simple focus, and hypothesize the location of NPI to be on Spec/NegP.

한국어 다중주어 구문 연구

김준한 ( Jun Han Kim )
6,700
초록보기
The purpose of this paper is to analyze the properties of Korean Multiple Subject Construction(MSC) in the light of current minimalist theories. It has been generally assumed that non-agentive predicates can appear in MSCs. I dicuss why this kind of non-agentivity condition should hold for MSCs. The second part of the paper focuses on Quantifier NP construction, in which agentive predicates such as unergative and transitive verbs can license multiple subjects. I also examine the relationship between the possibility of having multiple subjects and the so-called “Multiple Spell-Out” proposed by Uriagereka(1999). Lastly, I claim that there are in fact two different types of MSC and that the predication analysis can account for how agentive predicates can license the multiple subjects in Quantifier NP construction.

국어 말뭉치에서의 ‘넓다’의 의미와 용법 분석

김해연 ( Hae Yeon Kim ) , 유은정 ( Eun Jeong Yu )
6,900
초록보기
In Korean, many words are used to represent our perception of space, and nelpta ``wide/broad`` is one of such space- denoting words. This research explores meanings and uses of nelpta by analyzing a Korean written corpus from the perspectives of corpus linguistics and cognitive linguistics. This study examines instances of nelpta in the Sejong Project Corpus, characterizing grammatical properties of nelpta. It also classifies meanings of nelpta in terms of the following categories: (i) physical width in space, (ii) width of the bounded space, (iii) degree and scope of human perception and judgment, and (iv) scope of and width in abstract entities. Examination shows that the basic meaning of nelpta, the physical width/breadth of concrete objects, extends to indicate the conceptual width or breadth in human perception or cognitive processing, and even in abstract entities. Such metaphorical extension of the basic meaning is explained in terms of Metaphor Theory in cognitive linguistics. Finally, this research suggests that a corpus-based cognitive approach to language can serve as a foundation for research which seeks to characterize metaphorically extended meanings of space-denoting words.

"-주다" 구문의 양태적 고찰 -의지의 양태의미를 중심으로-

백정민 ( Jeong Min Baek )
6,000
초록보기
The purpose of this study is to research the structure of auxiliary verb ``-juda`` and it``s modal meaning. When the main verb in a sentence has a negative purpose, and it is followed by ``-juda``, this structure shows the volition to realize that action the main verb. I argue that in the sentence of ``-juda`` with other modal final-ending, volition is performed. It is already stablished that ``-gess-`` and ``-lguya`` express volition. I stand that ``-ji`` also has the same function. In conclusion, ``-juda`` has the volition meaning by itself, and when it gets combined with other ``volition modal final-ending``, this composition transmits a double volition will, which has a stronger meaning.

일본어 가고시마 방언의 말음약화: 조화직렬적 최적성이론에 기반한 분석

손범기 ( Bum Ki Son )
6,600
초록보기
Kagoshima Japanese exhibits deletion of word final high vowels. And word final consonants caused by high vowel deletion are reduced in systematic fashion. (i) Stops and affricates are reduced to glottal stop. (ii) Fricatives become voiceless. (iii) Nasals change to moraic nasal [N]. (iv) Flap becomes nonsyllabic high front vowel. The Kagoshima dialect shows overapplication of /s/-palatalization which occurs only before high front vowel: the motivation for palatalization in word final position, however, is not apparent on the surface because high vowel does no more exist in that position. Classic OT can never solve this type of opacity. I will solve this problem within the framework of Harmonic Serialism by positing gradual step: operation order is affected by the ranking of markedness constraints.

한국어 용언 위계구조 자동구축

송상헌 ( Sang Houn Song ) , 최재웅 ( Jae Woong Choe )
7,800
초록보기
This paper discusses an automatic way to derive a type hierarchy for verbal items in Korean based on their subcategorization. There are three steps: First, all the dependent categories of the each verb are extracted from the Sejong Treebank. Second, based on the frequency of the dependent categories of each verb, the most stable subcategorization frames are selected, and two statistical measures are tested with some variations in their cutoff values. The resulting subcategorization frames are then compared with those from the Sejong Electronic dictionary for evaluation. The final step is to form a type hierarchy for Korean verbal items, based on the chosen subcategorization information.

Avoidance Strategies in the Processing of Relative Clauses: The Why and The How

( Il Jae Lee )
6,400
초록보기
This paper argues that what previous studies of relative clause (RC) processing have called avoidance strategy can actually be a strategy to simplify the rather complex Agree relations (Schulz, 2011) between a multiple number of probes in CP phase and goals in 폐 phase. Object RC involves two occurrences of probe-goal relations (Chomsky, 2008), which must evidently be difficult to process at some instance because such a relation is subject to processing cost (Pesetsky & Torrego, 2006). The learner then generates a derivation in which two probes - C and T - share the same goal, and the spell-out turns out to be a passivized object RC in SVO languages like English (Belletti, 2009). In SOV languages like Korean, learner gets by intended object RC via introducing a resumptive NP and a base-generated operator - both are free of processing cost (Chomsky, 2008). This spell-out becomes an object head-internal relative clause (HIRC). The concluding remark states that avoidance strategy is therefore a way to reduce the processing cost via simplifying the intended derivation while sacrificing grammaticality or acceptability despite of marginality.
1 2 >