Morphological Passivization of Object-Verb Idioms

Lee, Doo-Won 미래영어영문학회 2017 영어영문학 Vol.22 No.2

In Korean, the causative and passive morphemes –i/-hi/-li/-ki on the phase head v may be part of the domain for the idiomatic interpretation in the sense that the crucial factor for passivizibility of object-verb idioms may be whether a phase head belongs to the idiomatic interpretation. Some object-verb idioms may undergo morphological passivization where the passive morphemes –i/-hi/-li/-ki appear, but they may not undergo (-a/e) ci-passivization without their having the causative or passive morpheme belonging to the idiomatic interpretation.. The passive morphemes –i/-hi/-li/-ki are on the head of vPASSP and (-a/e) ci may follow them on vPASS. The Encyclopedic knowledge domain for the idiomatic interpretation may be extended to the phase head vDO, vCAUS, or vPASS. Only when the root phrase forming part of an idiom is merged with the head vDO, vCAUS, or vPASS, the idiomatic expression may be realized, which is a phase-cyclically based operation (Chomsky 2001). This is involved in the phenomenon that only when the head vCAUS or vPASS appears, the object-verb idioms may undergo (a/e) ci-passivization, which provides the reason why the passive morpheme and (-a/e) ci can form double passives.

Passivization of Double Object Constructions in Uzbek

Esanova, Shohida,Song, Yangming,Kang, Seung-Man 미래영어영문학회 2022 영어영문학 Vol.27 No.2

This paper delves into some properties of double object constructions (DOCs) cross-linguistically in English and Uzbek, centering around the passivization of an indirect object (IO) and a direct object (DO). The two objects of DOCs exhibit a cross-linguistic asymmetry in terms of passivization. That is, DO is freely passivized, but IO is somewhat restricted or prohibited in passivization across languages. Uzbek likewise allows DO to be freely passivized but prohibits IO from being passivized. This intuitive L1 knowledge of Uzbek EFL learners has a great impact on their perception of English DOCs, as demonstrated in the survey among 50 Uzbek EFL learners, who perceive IO-passivization as ungrammatical (51.0%) and weird (11.0%). We attribute the illicitness of IO-passivization in Uzbek to our proposal that IO is a postpositional phrase (PP) per se, not an NP, headed by the suffix -ga attached to IO. We argue that the suffix -ga is actually a postposition (P), not incorporated into V in the sense of Baker (1988), which thereby hinders IO from being passivized as instantiated in pseudo-passives in English.

Passivization of Transitive Verbal Nouns in Korean : Factors that Facilitate Passivization in the VN-cwung Construction

Yutaka Sato,Sungdai Cho 서울대학교 언어교육원 2013 語學硏究 Vol.49 No.3

This paper investigates whether transitive verbal nouns (VNs) can appear passivized without –toy ‘become’ and, if so, what factors are pertinent. We have demonstrated, based on the results of internet search and questionnaire, that some transitive VNs, particularly agentive ones, can appear passivized when followed by the aspectual element cwung ‘during’ in Korean, as in Japanese. We have also argued that passivization of a transitive VN is possible only with a light verb lacking agentorientation, which is determined in terms of whether a light verb has an agentive heavy counterpart. Agent-oriented light verbs like ha- ‘do’ never allow such passivization.

Control or Raising?

Sun-woo Lee,Jun-suk Park 한국중원언어학회 2014 언어학연구 Vol.0 No.33

As Radford (2009) notes, two syntactic peculiarities which differentiate raising from control are 1) long-distance passivization and 2) no theta-role assignment for an external subject at Spec TP. In this paper, we examine whether or not threaten and fail constructions involving to-infinitival complements allow long-distance passivization. The current study shows that both threaten which requires a subject with agent theta-role and fail allow long-distance passivization. From this finding, we argue that the embedded infinitive clauses of threaten and fail are TPs. we also claim that these two verbs are neither typical control verbs nor typical raising verbs. For this claim, we provide two conflicting pieces of evidence, (i) the embedded infinitive clauses of threaten and fail cannot be cleft at all and (ii) still, constructions involving threaten and fail must have subjects with theta-roles. We conclude that fail shares more syntactic similarities with typical raising verbs than threaten does.

A Comparative Analysis of English and Chinese DOCs

Long, Yu,강승만 미래영어영문학회 2019 영어영문학 Vol.24 No.2

Chinese EFL learners often make errors in their perception of English DOCs. Accordingly, in order to examine their perceptual patterns, we have conducted a survey on English and Chinese DOCs among 50 EFL learners in China. The subjects were given three tasks. First, they were asked to turn English DOCs into Chinese ones. It is shown that three ditransitive verbs, gei 'give' (100%), tou 'steal' (6%), and mai 'buy' (48%), vary in their accuracy rate, which we attribute to their different argument structure from those corresponding to them in English. Second, they were asked to put Chinese passives into English ones. It is shown that some of them omit the indispensable preposition to in English DO passive constructions, which is due to a cross-linguistic difference between English and Chinese passives in terms of Case assignment. Third, DO passivization in the verbs, tou 'steal' and mai 'buy', shows a remarkably higher frequency rate (92%) than IO passivization. This is due to negative transfer from the Chinese language, which highly prohibits Beneficiary IOs from occurring in passive subject position. The findings of this research work suggest some teaching implications in terms of Case and argument structure of English DOCs in comparison with Chinese.

Control or Raising?-Peculiar syntactic patterns of threaten and fail-

이선우,박준석 한국중원언어학회 2014 언어학연구 Vol.0 No.33

As Radford (2009) notes, two syntactic peculiarities which differentiate raising from control are 1) long-distance passivization and 2) no theta-role assignment for an external subject at Spec TP. In this paper, we examine whether or not threaten and fail constructions involving to-infinitival complements allow long-distance passivization. The current study shows that both threaten which requires a subject with agent theta-role and fail allow long-distance passivization. From this finding, we argue that the embedded infinitive clauses of threaten and fail are TPs. we also claim that these two verbs are neither typical control verbs nor typical raising verbs. For this claim, we provide two conflicting pieces of evidence, (i) the embedded infinitive clauses of threaten and fail cannot be cleft at all and (ii) still, constructions involving threaten and fail must have subjects with theta-roles. We conclude that fail shares more syntactic similarities with typical raising verbs than threaten does.

On the Passivizability of Benefactive Ditransitives

문지순 한국언어학회 2014 언어 Vol.39 No.2

English ditransitive constructions have received their fair share of attention in linguistic literature. However, a curious omission in the otherwise exhaustive coverage of the phenomenon has been the apparent asymmetry in passivizability between dative ditransitives (constructions that correspond to to-datives) and benefactive ditransitives (constructions that correspond to for-datives). In other words, while dative ditransitives seem to passivize freely, benefactive ditransitives are much more constrained in their capacity to passivize. The present investigation attempts to provide a cogent explanation for this asymmetry by building on the widely-acknowledged assumption that passivizability is a function of transitivity. Specifically, the study adopts Rice's (1987) notion of "transitive prototype”and argues that the passivizability of benefactive ditransitives is governed by an interplay between the semantics of the ditransitive construction and the various conceptual dimensions of transitivity.

Hierarchical Inductive Bias in the L2 Textbook-T5 and Child-T5 Language Model: A Study of Data and Architecture

구건우,Jaemin Lee,박명관 한국응용언어학회 2023 응용 언어학 Vol.39 No.4

This study aims to investigate neural language models in alignment with Chomsky's (1965, 1980) proposition regarding the innate human tendency to acquire syntactic rules based on hierarchical structures rather than linear order. We have examined the architectural and dataset factors influencing the acquisition of a syntactic inductive bias during the pre-training of the T5 model. To achieve this objective, particularly concerning the pre-training dataset, we inquire whether there are differences when using datasets of varying complexity. We use two distinct pre-training datasets: Child-Directed Speech and L2-Textbook dataset. Upon examination, we observe that these datasets exhibit different levels of syntactic complexity. Then, with our models, we employ two syntactic transformation tasks: (i) question formation and (ii) passivization. Our results indicate that model depth (number of layers) exerts a more significant influence on hierarchical generalization compared to other model components. Furthermore, we observed that models learn the hierarchical aspects of language more efficiently when exposed to the simpler complexity of the dataset.

On the Passive of Double Object Constructions in English

Kang, Seung-Man 한국중앙영어영문학회 2010 영어영문학연구 Vol.52 No.2

Some English grammar books illustrate that the indirect object of some ditransitive verbs such as write and buy is not allowed to be passivized, and secondary school English teachers accordingly teach their students. To describe this linguistic aspect, I have conducted a survey on the passive of English DOCs among 20 high school students in the US. The survey exhibits a different result; only 2 out of 20 subjects respond that the passivization of the indirect object is ungrammatical. Some theoretical grounds such as Case and Ɵ-role assignment also lend support to the IO-passive movement. Nevertheless, its movement is not approved to be completely natural by the subjects. I attribute this to a diachronic observation that the passivization of IO has expanded from Old English up to the present. I also suggest that the trace of P, which is ‘active’, left by Preposition Incorporation somehow hinders its movement, and the degree of its being active varies across ditransitive verbs.

Passivization, Structural Case and A’-movement in Double Object Constructions of American and British English

Myung-Kwan Park 한국생성문법학회 2012 생성문법연구 Vol.22 No.3

This paper investigates immobility of A-moved/passivized DO and A’-moved IO in double object constructions of American English, in contrast to mobility of both IO and DO in those of British English. To account for the issue at hand, we propose that there is a strong connection between structural Case and A’-movement. In other words, if IO always receives structural Case as clearly manifested by passivization in a certain language, it cannot undergo A’-movement in that language. We suggest, following Rizzi (1996, 1997) and Rizzi and Shlonsky (2007), that structural Case brings about so-called freezing effects, blocking further movement for a relevant expression.