Oblique Arguments Research Articles

Mixed categories and argument transfer in Korean light verb constructions

The Korean Light Verb Construction (LVC) contains a Sino-Korean main predicate (tayhwa-lul), a Light Verb (ha-ta), and semantic arguments of the main predicate (John-i, Tom-kwa): John-i Tom-kwa tayhwa-lul ha-yess-ta. John-Nom Tom-with talk-Acc do-Pst-Dc 'John talked with Tom.' We defend a three-part analysis: (i) The subject of the main predicate is thematically controlled by the LV's subject. Evidence: Korean verbs assigning Accusative take an external argument (Wechsler/Lee 1996; Burzio's Generalization). Since the main predicate is Accusative, ha-ta must theta-mark its subject. Moreover ha-ta selects a non-stative Verbal Noun (VN) (cp. *kyumson-ul ha-ta 'humble-Acc do-Dc'); non-stative theta-structures typically take an external argument (Kang 1986). This control arises through complex predicate formation. (ii) Oblique arguments (PPs) are optionally transferred (cp. Grimshaw/Mester 1988) — but Accusative NPs are not. Evidence comes from relativization and pronoun replacement. (iii) Accusative is assigned by a mixed category Verbal Noun. This can be supported by adverbial clauses with VN's assigning Accusative without LV's. We review cross-linguistic evidence for both argument transfer (German; Hinrichs & Nakazawa; i.a.) and mixed categories (many languages, Malouf; i.a.) and show that Korean LVCs provide the right environment for both to occur.

Abaza Applicatives

Abaza Applicatives Brian O’herin The applicative is a construction in which a verb bears a particular morpheme which licenses an oblique, or non-core, argument that would not otherwise be considered a part of the verb’s argument structure. Abaza, a Northwest Caucasian language, has an applicative construction which differs systematically from applicative constructions reported in many other languages. The properties of the Abaza applicative construction include the ability of both transitive and intransitive verbs to host applicative prefixes, the stability of the underlying argument structure despite the presence of applicatives (i.e. the underlying direct object is not demoted to an oblique argument), the ability of a single verb complex to host multiple applicatives, and the use of additional agreement morphology in the verbal complex corresponding to applied object(s), beyond the normal agreement requirements of the verb. This paper attempts to present a thorough description of the applicative construction in Abaza, as well as a brief analysis along the lines of Baker (1988).* The applicative In the applicative construction the verb bears a particular affix (the applicative morpheme), and this morphological marking licenses an oblique, or noncore, argument that would not otherwise be considered a part of the verb’s argument structure. Compare the simple Chichewa sentence in 1 with the applicative example in 2. Typically, the applicative construction involves arguments in such roles as instrumental, benefactive and adversative (or malefactive). Some languages also allow comitative and locative applicatives.1 The applicative construction contrasts with a construction in which oblique arguments are licensed by means of case and/or an adposition. (1) (2) A wide variety of languages have been described as having constructions of this type, including several Bantu languages (Baker 1988), Mohawk (Iroquoian) (Baker 1996), Rama (Chibchan), Nadëb (Maku) and Winnebago (Siouan) (Craig and Hale 1988), Slave (Athapaskan) (Rice 1989), Bajau (Austronesian) (Donahue 1996), Bahasa Indonesia and Tagalog (Austronesian), and Mayan languages. All of these have the property that the addition of an applicative affix to the verb licenses particular oblique [End Page 477] arguments, without the use of an independent adposition or oblique case. A constellation of certain other properties frequently occurs with the applicative construction as well. Most notably, the applicative object tends to behave syntactically like a direct object, while the underlying direct object tends to lose the normal syntactic properties of a direct object. Marantz (1984) analyzes applicatives as the merger of two independent morphemes (V and P), each of which bears its own argument structure. Feature percolation conventions laid out in his work and the assumption that verbs, simple or derived, may each assign only one semantic role combine to account for the observed properties. Baker (1988) argues that that applicative construction crosslinguistically can be analyzed as the incorporation into the verb phrase of the adposition which normally licenses the applicative object, and that the properties frequently associated with applicatives result from this type of head movement. Den Dikken (1995) gives an alternative analysis in which the applicative morpheme is an incorporated particle heading a small clause. All three of these analyses have in common that the applicative construction is one in which the oblique argument is external to the verb at some level, but becomes part of the verbal word at a different level. General principles interact to produce the properties commonly found with applicatives. Abaza (Northwest Caucasian) also exhibits a construction in which a morphological mark on the verb licenses certain oblique arguments, including benefactive, adversative, instrumental, comitative, and locative. In this respect the Abaza construction looks much like the applicative construction documented for other languages. The Abaza applicative construction is unusual in that it does not exhibit the full range of properties usually associated with applicatives. This article is an attempt to document this construction in Abaza. Ultimately, I will argue that the Abaza construction is also an applicative construction of the same general type as the others, and that the different surface properties are the result of a single general difference in how arguments are licensed in Abaza. 1. Background The Abaza verb complex necessarily indicates tense and mood, and optionally aspect, all as suffixes. The verb complex indicates polarity, and the negative...

Core-Oblique Distinction and Nominalizer Choice in Japanese and Korean

It has been suggested that a core–oblique distinction is not manifested in the grammatical structures of languages like Japanese and Korean, wherein both core and oblique arguments are marked by so-called case-marking particles. This study argues, based on elicitation data from Japanese and Korean as well as Japanese questionnaire data, that a core–oblique distinction in these languages is indeed reflected in the differing selection of sentential nominalizers to mark core and oblique nominalized clauses. Furthermore, this study presents an explanation for the differing selectional patterns of core and oblique marking sentential nominalizers from the viewpoints of markedness and information processing.

Papeles semánticos, preposiciones y objetos preposicionales

In this work, I have examined some interpretations of the relational meaning of several oblique arguments. I have pointed out some discrepancies in the interpretations of these complements. Finally, I have commented on some problemational questions about semantic roles and I have remarked that these questions could be involved in the mentioned discrepancies.

On the role of conceptual structure in argument selection: A reply to Emonds

Emonds (1991) claims that many phenomena that I attribute to the effects of conceptual structure on syntax in Jackendoff (1987a) are more properly accounted for in terms of syntax alone. The present paper shows that Emonds has misconstrued the role intended for conceptual structure in the interface between language and cognition. It then specifies precisely where Emonds's approach to argument selection differs technically from mine, and shows that the technical differences favor my approach as further developed in Jackendoff (1990). Finally, an improved account of the semantic selection of oblique arguments is proposed, but it is shown that still not all syntactic argument selection can be eliminated from lexical entries of verbs.

The Notion of Source in Language Acquisition

When children describe an event, they do not always make use of the conventional means for marking oblique arguments. In early attempts at passive constructions, for example, children often mark oblique agents with from instead of conventional by, as in He isn't going to get hurt from those bad guys. They also make use of from instead of because, as in From I put it under there. Such uses of from typically appear prior to by and because for agency and cause. We argue that these uses of from stem from children's conceptualization of agents and causes as types of SOURCE. If so, children should extend a similar analysis to other types of source as well. There is evidence that they do, both in longitudinal records of single children and in experiments on imitations and repairs.*