Earticle

The purpose of this paper is to study constraints on middle verbs and implicit argument focusing on its way of presence and genericity. We examine affectedness constraint, aspectual constraint, and causativity constraint, and among them, aspectual constraint is proved to be the most convincing constraint of middle verb formation. We also prove that presence of an implicit argument is a defining characteristic of middles. The implicit argument of middle is syntactically unexpressed but semantically present. However, there are different opinions of the way of presence of an implicit argument. In addition, although the external argument is understood to be people in general, even middles lacking genericity involve an implicit argument. A specific person can be engaged in the interpretation of an implicit argument. Therefore, genericity is not an essential property of middles.

영어 전치사 ‘of’는 [X of Y]의 형식을 가지고 한 개체가 다른 개체의 내재적이며 제한된 하위 부분에 해당하는 종속관계로 두 개체 사이에는 특별한 관계에 놓여있는 것으로 보고 있다. 또한 X는 탄도체(trajector)로서 부분을 나타내고 Y는 지표(landmark)로서 전체를 나타내는 동시에 탄도체에 해당하는 요소의 자리는 지표의 요소에 비해 현저성이 부여된다고 본다. 그러므로 전체인 지표를 강조코자 할 때는 [Y of X]로의 전도현상이 발생하여, X와 Y의 자리가 전환될 수 있다. X에 해당하는 요소는 다양한 품사로 나타날 수 있을 뿐만 아니라, 경우에 따라서는 전혀 나타나지 않는 경우도 있다. 본 연구는 영어 전치사 ‘of’를 재고찰 하면서, 그 동안의 선행연구에서 보여준 ‘of’-구의 구조를 X의 출현 여부에 관계없이 2항 술어로 간주하고, 전치사 ‘of’가 가지고 있는 다양한 기능이 외부논항의 적절한 정의에 따라 일관성 있게 설명될 수 있음을 보이고자 한다. 이를 위하여 의미역(theta-roles)에 대한 포화 이론 (Saturation Theory)을 근거로. 논항이 술어로부터 의미역을 부여받는 것이 아니라, 술어가 취하는 논항 자리에 의미역이 제공됨으로서 해당하는 항(place)의 자리에 논항이 위치하여 의미역 조건이 해소되는 것으로 본다. 본 연구는 [탄도체(X:명사) ‘of’ 지표(Y:명사)]를 ‘of’의 기본 형식을 보며, [동사 of 명사] 또는 [형용사 of 명사]의 형식에서처럼 탄도체가 표시되지 않는 경우는 암시적 논항(implicit argument)으로 간주되거나 탄도체 자리에 기저 생성된 주어가 상승하여 흔적이 남겨진 경우라는 대안을 제시한다. 또한 소절(small clause)이 공(empty) INFL을 머리어로 하는 IP임을 근거로 소절 형식을 취하는 유형과 그렇지 않는 유형의 의미적 차이점을 비교하여 선행연구에서 제시되어온 전치사 ‘of’의 [X of Y]의 구조가 [X/(IA) of Y]의 구조임을 주장하고, 이러한 구조로 전치사 ‘of’의 다양한 활용이 일관성 있게 설명될 수 있다는 사실을 밝힌다.

It has been assumed that the preposition ‘of’ takes the form ‘X of Y,’ where one entity falls under a restricted inherent sub-part of the other entity. Typically, ‘X’ plays as the trajector, being the part of ‘X of Y,’ while Y, as the landmark, being the whole of it. The position of the trajector is considered to be always more salient than that of the landmark. Therefore, the shift from ‘X of Y’ to ‘Y of X’ can occur to emphasize the landmark. Based on the facts above, this paper proposed that the preposition of in English uniformally be treated as a two place predicate. Given its nature of transitivity, the preposition ‘of’ takes two arguments of ‘X’ and ‘Y.’ This proposal is based on Higginbotham’s (1985) revised version of Theta Theory and Burzio’s (1986) the analysis of implicit argument in the causative construction. Under the premise that the preposition ‘of’ adopts a predicate construction like a verb, this paper further suggested that the prepositional phrase where the trajector (noun) does not appear like ‘X(verb) of Y(noun)’ or ‘X(adjective) of Y(noun) carries an implicit argument in the place of ’X’ in the structure of the preposition ‘of.’ This paper further contended that the subject generated from the place where the trajector is supposed to appear is raised for the Case reason, leaving the trace behind. The passive sentences were also discussed to make a distinction of the cases where the ‘X of Y’ is involved in the small clause and the cases where the ‘X of Y’ is not. Given that the small clause has an IP, which carries the empty INFL, the form of ‘product (IA) of material is made’ was defined as being constructed within the small clause, whereas the form of ‘product is made (IA) of material’ is not. Therefore, this paper has shown that the analysis of two place predicate for the preposition ‘of’ in English is well maintained, such that not only the cases where the preposition ‘of’ takes the external as well as internal arguments, but the cases where the external argument does not appear can be accounted for in a unified and systematically way.

This paper aims to study on a phenomenon where adverbials are strongly needed for Korean middle constructions in light of semantics and pragmatics. First, in light of pragmatics, it was found that middle construction is a universal phenomenon in typology, in which case adverbial is also strongly required in many languages. In the process, the prototypical features of middle constructions were discovered, which leads to the notion that they describe the general properties of generic subject ultimately. It was found in this study that the attributes of generic subject in the middle constructions can be applied to anyone, and these universal features originate from implicit argument. In addition, the attributes of subject cannot be explained enough only with semantic information of middle verbs so the middle construction itself lacks in information regardless of valency. This lack of informativeness is complemented by the role of “adverbial,” and this paper, with a focus on pragmatics, described that adverbial has a function of relational new information. Meanwhile, with regard to the informatively domain by adverbial, some cases were explored where several verbal elements were simultaneously involved to provide informativeness for the middle constructions, which resulted in the finding that the informative role of adverbial relatively decreases when the focus markers “i/ka” are added to the subject, but still remains larger than when combined with demonstrative and subject.

In the previous literature on sprouting, most attention has been paid to the elided structure. This paper, however, focuses on the remnant of the elided part. More specifically, this paper claims that, in sprouting, the wh-phrase, which is the remnant of ellipsis, echoes a cataphoric implicit argument that is locally accommodated to the argument structure of the antecedent clause. Under this view, sprouting is not very different from merger in the respect that the former comes when the antecedent of the wh-phrase is manifested as an implicit argument rather than being completely absent whereas, in the latter, the antecedent of the wh-phrase is realized as an explicit argument. The implication of this study is that backward formation should, in general, be allowed in linguistics.

This paper concerns some crucial properties of the middle construction. It is well-known that middle constructions do not behave in a uniform way so that it is quite hard to characterize the construction. First, it refuses to be parasitic on the independently established structures including ergatives, unaccusatives and passives. Hence, it claims its own structure. Second, this paper deals with the generic reading and the agentivity in the middle construction. To empirically attest its generic reading, typical tests for stative are adopted here. The middle construction behaves just like statives. This paper also seeks to find where an agentivity reading comes from in the construction. If the agentivity has something to do with the presence of an agent. an agent should be either semantically or syntactically active. None of the evidence in the literature in favor of the existence of the agent turns out tenable.

This paper is concerned with how to account for the backward binding phenomena in psych-predicates. The backward binding can be structurally explained if we make the following two assumptions: first, the bindee is generated in the c-domain of the binder and second, the bindee moves across the binder. Unfortunately, however, the two assumptions contradict each other, since the bindee cannot cross over the binder in accordance with the Minimal Link Condition (MLC). The upshot of the problem is that interleaving movement is required in order to structurally account for the backward binding, but it is not permitted by the MLC. In this paper I claim that the interleaving movement approach is not a viable option since it does not conform to the MLC, claiming that the backward binding takes place on account of the implicit argument via λ-conversion. That is, this paper provides a semantic account for the backward binding in psych-predicates, after pointing out that the interleaving movement approach is on the wrong tract.

It is well known that a phonetically unrealized argument is syntactically present in short passives where a by-phrase is omitted. The purpose of this paper is to investigate the syntactic properties of this implicit argument under the probe-and-goal theory formulated in the framework of the Minimalist Program. This paper adopts the notion of lethal ambiguity in order to identify the syntactic position occupied by the implicit argument. Then it argues for the necessity to postulate a new kind of covert element called IMP, which is assumed to have a proper subset of the phi-features found in PRO. This proposal is used as a basis for providing an account of how the internal argument PRO can move over the implicit argument to the subject position in infinitival passive constructions without resulting in a violation of the Feature-Based Relativized Minimality Condition.

본 연구는 그간 국어학계에서 유사 개념으로 처리되어 온 반논항과 잠재논항이 구분되는 개념임을 규명하는 데 목적이 있다. 이를 위해 우선 반논항과 잠재논항에 관한 기존 논의를 검토하였다. 반논항의 경우 문장에 실현되었을 경우의 처리 방식에 관한 쟁점을 살피고, ‘가변중간투사론’이라는 이론적 모형이 반논항 처리에 어떤 영향을 끼쳤는지를 비판적으로 논의하였다. 잠재논항의 경우 그간의 논의가 수의성을 기반으로 잠재논항을 규정하는 방식을 취해 왔음을 밝히고, 이러한 방식이 일반언어학에서 잠재논항을 규정하는 방식과 어떤 차이가 있는지를 논의하였다. 이러한 검토 결과를 바탕으로 반논항은 수의성에 기반한 개념이고, 잠재논항은 잠재성에 기반한 개념이라는 점을 밝히고, 논항의 수의성과 잠재성이 개념적으로 어떻게 구분되는지를 논의하였다. 아울러 잠재논항의 잠재성이 정도성의 문제라는 점을 밝힘으로써 반논항, 잠재논항의 처리 방식에 관한 그간의 논행을 해결할 수 있는 방안을 제시하였다.

The purpose of this study is to identify the distinction between ‘semi-argument’ and ‘implicit argument’ which has been previously recognized as being a similar concept. For this purpose, related studies were reviewed. In the case of semi-argument, we examined the issue of how to handle the case when it was realized as phonetic form in the sentence, and critically discussed how the theoretical model called ‘Variable Media Projection Hypothesis’ influenced the concept of semi-argument. In the case of implicit argument, it was revealed that the previous discussions have defined implicit argument on the basis of arbitrariness unlike the definition in general linguistics. Based on the results of this review, it is revealed that the semi-argument is a concept based on arbitrariness, and implicit argument is a concept based on implicitness. We discussed how the arbitrariness and implicitness of the argument are conceptually distinguished. In addition, by suggesting that the implicitness of implicit argument is a matter of degree, this paper suggests a way to resolve the debate about semi-dispute and implicit argument.