Earticle

The deontic complementizers -key and -tolok in Korean encode the domain de re as opposed to the domain de dicto complementizer -ko. A deontic complementation denotes a indirect causation. Case markers of the subject of deontic complement clause alternate. Case markers encode semantic functions in Korean Case alternation in a deontic complement indicates a change of a semantic function and is conditioned by the classification of predicates into action and state. The deontic complementizer -tolok is sensitive to this categorization of predicates. The classification of predicates into action and state and the representation of semantic functions are necessary in describing the syntax in Koreas such as the morphological passive and the psych verb.

The aim of this paper is to investigate syntactic features which are carried by the finite complementizer þæt>þat>that and the non-finite complementizer for from diachronic perspective. Given that þæt/þat wasn't followed by a focalized or topicalized element, but permitted wh-word＋þæt/þat sequence from an embedded clause, þæt/þat had the finiteness feature of φ-features and a tense feature that was agreed with the finite feature of a finite verb in OE and EME. On the other hand, considering that that has been followed by a focalized or topicalized element in the embedded clause and φ-features and a tense feature in C has been inherited to T from LME onward PDE, that came to carry a declarative force feature rather than a finiteness feature from that time on. Meanwhile, the prepositional complementizer for began to be added to the to＋N/V＋enne to strengthen the grammatical function of to from EME, and this phrase has developed into an infinitival verb phrase for＋to＋V＋(O). As inflections were lost, V-2 was reanalyzed, and OV was reordered as VO, the for＋to＋V＋(O) sequence was changed into an infinitival verb clause and ultimately transformed into for＋Osubj＋to＋V in LME. With such restructuring of an infinitival verb clause, a non-finite future feature which was inherently carried by for came to be inherited to the infinitive marker to that started denoting an unrealized future tense ever since.

This paper tries to explain why preposition stranding is allowed in some languages like English but not in others, by extending den Dikken's (2006) phase extension analysis to DP to PP relation. Going through a number of examples with respect to preposition stranding and its relationship with the directionality of movements involved, this paper deals with the question why preposition stranding is different from language to language as well as within a language. By applying den Dikken's (2006) phase extension analysis to preposition stranding examples, this paper supports his theory of phase extension. This paper also deals with the consequences of the proposed analysis with respect to complementizer stranding. By equating C with P (including prepositional complementizer for), a unified account becomes possible to both PP-DP relation and that of CP-TP. The phase extension analysis is also proven to be superior to Fox and Pesetsky's (2005) linearization with respect to preposition stranding.

This paper is primarily concerned with the licensing of that-less complement clauses of nominals. In particular, the present study is purposed to answer the question why a null complementizer(NC) is not allowed with nominals while it is with verbs. In pursuit of an answer to the question the relationship between predicate raising and the presence/absence of NC is examined. This paper proposes a solution to the problem in terms of n to N (and to D) raising. The present study also deals with another interesting relationship between predicate raising and the presence/absence of NC with respect to the that-t effect and suggests that the alleged complementizer in Korean may not be a complementizer in a regular sense.

This paper revisits the distribution of null complementizers in English and develops a minimalist account building on Pesetsky’s (1992) affixal view of C. I argue that the absence of null C in certain contexts reflects constraints on movement and labeling. Such derivations yield headless configurations, violating interface conditions. The analysis unifies long-standing contrasts among verb classes under a structural condition on complementizer licensing. Drawing on phase theory, external pair-Merge, and the Labeling Algorithm, the paper refines the affixal view of C and highlights how syntactic structure constrains embedded clause formation.

This paper aims to investigate the syntactic structure of how come interrogatives. In particular, I focus on two interesting aspects where how come differs from other wh-interrogatives; 1) how come interrogatives do not show T-to-C movement, and 2) while C should remain null, it must be overtly realized when modifiers intervene between how come and the rest of the clause. To resolve the puzzles, I assume that how come externally merges to a relatively higher position than other wh-phrases. Thus, unlike other root wh-interrogatives where C values an uninterpretable clause-type feature on T, how come values the [uclause-type: ] feature on T instead, which prohibits T from undergoing movement to C. I also assume that C is split into Foc and Fin (Rizzi 1997), and a modifier head may intervene between the two heads (Rizzi and Shlonsky 2007). I argue that [uwh] and [Q] in root wh-interrogatives originate from separate heads ? Foc and Fin, respectively ? and morphological affixation takes place via Fin-to-Foc movement. If the head movement is prohibited by an intervening modifier head, Fin[Q] must be overtly realized with overt complementizer that.

Most studies on English complementizer that assume that overt and null that clauses have the identical underlying structure, predicting they show (nearly) the same syntactic distribution, contrary to fact. This paper explores overt and null complementizers in clausal complements of both non-factive and factive predicates in English, building on the assumptions that (i) feature specification on C differs from language to language and (ii) languages also differ as to lexicalizing a subset of these features on C. Adopting Rizzi’s (1997) split CP structure with two C heads, Force and Finiteness, we argue that null that clauses are FinPs, whereas overt that clauses have an extra functional layer above FinP, lexicalizing either the Force head under non-factive predicates or the light demonstrative head d under factive predicates. These three different underlying structures also successfully account for different syntactic patterns found between overt and null that clauses in various contexts.

This paper reviews recent works on the distribution of the null complementizers in English with a view to clarify the licensing conditions on the null Cs. Pointing out the problematic aspects of the analyses proposed in Pesetsky (1992), Bo?kovi? and Lasnik (2003), and Lee (2005), this paper argues that the account for the null C in terms of cliticization on the basis of phonological phrase has advantages over the others, in that it need not specify the host to which the null C will attach. Some puzzling cases are left, however, for further research.

This paper aims to submit a revised proposal on complementizers in Korean complex sentences. A complementizer as a terminal ending grammatical marker signifies the derivation of a finite or non?finite complement clause is completed. It is successively merged with a relevant morphological case?marking particle to make the case?marked complement clause to be selected by a matrix propositional predicate. In principle there appears to be five different types of complementizers; {?da, ?ni/?nya, ?la, ?ja, ?ma} heading a finite complement clause, {?eum} heading a non?finite nominal predicate clause similar to English gerunds, {?ki} heading a non?finite nominal verb clause similar to English infinitives, {?keot} heading a predicative nominal clause, and {?ji} heading an alternative interrogative clause. When a propositional predicate in the matrix clause is merged with a finite or non?finite complement clause as its complement, complex uninterpretable locutionary force features denoting sentence type and high/low status marker in C have just agreed with relevant interpretable force features which are carried by a terminal ending grammatical marker in the complement clause.

This paper is a study of the notion of complementizer and the complementizers of Korean. The notion of complemetizer has changed through the development of the Generative Grammar. GB and PP's notion of that is quite different form ST's. This paper showed two notions of complementizer don’t match with some of endings of Korean which had identified as complementizers in their own way. Especially quotative ‘-go’ makes big problem. This paper suggested tentative solutions for some problems, and argued the necessity of modification of the notion for other problems.

한국어와 일본어 그리고 영어의 보문소 의미론에 대한 비교 연구박기성본 논문은 한국어 일본어 영어의 일련의 보문소의 의미론(화용론)을 살펴보는 것이다. 보문소의 의미론에 관한 연구는 1960년대와 1970년대에 주로 영어를 대상으로 이루어졌고, 1980년대 이후 주로 기능유형론적 연구들과 인지언어학적 연구들이 주류를 이루고 있었다. 본 연구에서도 이러한 전통 속에서 이루어지고 있으며, 비교 연구를 통한 각 언어 간의 차이점과 평행성(parallelism)을 연구하고자 한다. 본 연구에서 비교 대상이 된 주요한 보문소들은 한국어의 '고, 도록, 게'이고, 일본어의 'to, koto, no'이고, 영어의 'that, to, -ing'이다. 각 언어의 이러한 보문소들 간에는 평행성이 있는 것으로 밝혀졌으며, 이러한 평행성은 de dicto/de re구분과 연관된 것으로 지적하였다. 또한 한 주절 동사가 여러 종류의 보문소를 선택할 수 있는 경우에, 그 선택은 의미적 요인에 의해 결정된다는 점을 지적하고 있으며, 그 통사-화용간의 상호 작용은 Van Valin & LaPolla(1997)의 IRH로 설명될 수 있음이 지적되었다. 이와 더불어, 각 언어에서의 구체적 보문소 선택원리로서 다양한 의미-화용적 개념들이 도입되었다. 한국어에서는 함축성, 동시간성, 직접/간접 지각, 일본어에서는 화자의 확신, 직접/간접 지각, 영어에서는 함축성, 직접/간접 지각, 사실성 등의 의미-화용적 요인들이 구체적 예를 통해서 논의되었다.

This study examines how Korean interrogative complementizers ?nci, -lci, -nka,-lkka interact with different types of attitude predicates, such as rogative, responsive (veridical), and responsive (nonveridical), to realize subjunctive interpretations in embedded questions. Unlike Indo-European languages, where mood distinctions are typically marked by inflectional morphemes in the embedded predicates, Korean employs a complementizer-based mood system in which choice of complementizer encodes fine-grained differences in epistemic commitment, nonveridicality, and emotive stance. A corpus-based frequency analysis and collostructional analysis of the Sejong corpus reveals systematic selectional preferences: ?nci and ?lci combine with veridical and nonveridical responsive predicates, with ?lci strongly linked to the latter. Combining with nonveridical predicates-nka prefers to express weak conjecture meaning, while ?lkka is favored in context of emotional concern and apprehension. The contribution of current study is that we provide the empirical corpus evidence that Korean interrogative complementizers behave as a mood-sensitive markers, paralleling systems such as Modern Greek, and show how language withouth employing overt mood inflection morpheme encode fine-grained mood contrasts through emplementizer choice.