This article proposes a new approach to restructuring which unites complex head approaches with a bare VP complementation approach. With the former, I argue that restructuring involves incorporation, however, in contrast to complex head (V-V) approaches, I argue that only the voice head of a restructuring complement undergoes incorporation. With the VP complementation approach, I assume that a restructuring complement contains a syntactically and semantically independent VP. In contrast to the bare VP complementation approach, however, I propose that restructuring complements also involve a voice head (but no embedded subject). Motivation for a voice head in restructuring comes from the subject interpretation of the embedded predicate, German stem allomorphy, and voice marking in several Austronesian languages. This hybrid account has the advantage that argument structure ‘sharing’ only applies to the subject of a restructuring configuration, whereas the remaining argument and event structure properties of the matrix and embedded VPs remain separate, which is supported by cross-linguistic properties pointing to the morphological, syntactic, and semantic independence of the two VPs. The account proposed achieves a larger empirical coverage than previous accounts and also improves in several respects on the theoretical details of previous analyses.
In this paper it is proposed that a basic difference in structural dependency, the difference between selected complements of a head on the one hand and unselected adjuncts to a projection of a head on the other, correlates with a basic interpretive difference which is central to understanding the range of complex predicate constructions. Adjuncts, such as depictive predicates, are interpreted as conjuncts, as is commonly assumed. The standard assumption about complements is that their interpretation is by function application, where the specifics are then determined by the semantics of the selecting head. Here it is proposed that there is a common interpretive core to complementation, one which is monotonic and in some abstract sense mereological. This applies both to extended projections and to complements which are embedded in another extended projection, and is exemplified by resultatives as well as other secondary predicate types. The interpretation is monotonic in the sense that the selecting head preserves the interpretation of the complement, elaborating on it. It is mereological in the related sense that the complement is interpreted as an integral part of the interpretation of the whole.
In this paper I describe and analyze location verbs like auzi-pe-ra-tu (trial-under-all-tu) ‘prosecute’ in Basque. Basque location verbs consist of the allative adposition (-ra), an optional morpheme lexicalizing Axial Part (like pe ‘under’), and its Ground complement (like auzi ‘trial’). I show that Basque location verbs can only lexicalize a specific type of PP: the morpheme -tu (the participle or infinitive suffix) can only attach to a GoalP complement, and not to a PlaceP (*etxe-an-tu [house-ine-tu] ‘be at home’) SourceP (*etxe-tik-tu [house-abl-tu] ‘come from home’), nor a ScaleP (*etxe-rantz-tu [house-appr-tu] ‘go towards home’). This fact shows that there is not a silent V to which its P complement has been incorporated (cf. Hale & Keyser 1993). If that were the case, we would expect location verbs to be built on any type of P which can be the complement of a verb like go. This way, this paper shows that there is a contrast between Goals and Sources/Scales in derived location verbs, a fact which can be considered another example of the well-known Goal/Source asymmetry (Lakusta 2005, Lakusta & Landau 2005, Gehrke 2008, Pantcheva 2011 among others). In order to account for that, I will suggest that Sources and Scales have a more complex structure than Goals, following particularly Pantcheva’s (2011) analysis on the decomposition of Path. In order to explain the restriction on the formation of location verbs, I will consider the topological and structural isomorphism of the events (Ramchand 2004 2008a et seq.) and adpositions (Svenonius 2006 2008, Pantcheva 2011).
Léa Nash and Pollet Samvelian
Léa Nash and Pollet Samvelian
Léa Nash and Pollet Samvelian
This paper argues for a faithful mapping between event/argument structure and morphosyntactic structure in Malayalam. The empirical focus of this article is the distribution of the verb ceyy- ‘do’ in a subset of coordinate constructions in the language, specifically VP coordinations involving action verbs. Under our account, this ceyy- serves to resolve a conflict between a morphological constraint (Stray Affix Filter, Lasnik 1981, 1997) on the one hand, and a syntactic constraint on movement out of coordinate structures (Coordinate Structure Constraint Ross 1967) on the other, making its overt appearance “last resort”. However, in spite of similarities with do-support in English, we find that formal accounts for the latter (Chomsky 1957, 1991, et. seq; Halle & Marantz 1993; a.o.) fall short of explaining the Malayalam facts. In our modified account, we claim that this ceyy- is the phonological realization of an Agent-introducing functional head in the sub-lexical structure of Malayalam action verbs (e.g. kick, eat, push). We motivate the selective presence of such a functional head in action-sentences and not statives by appealing to differences in their respective underlying argument- and event-structures, a distinction strongly argued for in the neo-Davidsonian literature, and now well-established within mainstream linguistic theory. Our evidence for the crucial role of syntactic and semantic factors in accounting for Malayalam ceyy-support challenges the widely-accepted formalization of do-support as a necessarily post-syntactic operation.
Verbo-nominal complex predicates, the bulk of the Hindi predicate lexicon, a highly heterogeneous category regarding compositionality and idiomaticity, can however be sub-classified in three types regarding agreement, internal and external syntax, separability, omissibility, extraction and acceptable modifiers. However, in spite of the highly heterogeneous nature of the category, they behave similarly in widening the scope of Aktionsart, aspect and voice specifications in comparison to the grammatical markers of TAM in simplex predicates. The most important fact deals with the diachronic evolution of the language. All sub-types of complex predicates, massively introduced during the renewal of the verbal lexicon by means of borrowing, have contributed to a general shift towards semantic alignments by licensing non-canonical subjects, mainly dative and genitive subject constructions.
Adele E. Goldberg
The present work investigates English verb particle combinations (e.g., put on) and argues that item-specific and general information are needed and should be related within a default inheritance hierarchy. When verb particle combinations appear within verb phrases, a tripartite phrasal syntax is defended, whether or not the V and P are adjacent (e.g., She
the wrong shoes; she
the wrong shoes
on). The < V NP P > order is motivated as the default word order by explicitly relating a verb-particle construction to the caused-motion construction (e.g., she put the shoes on her feet). Well-known and independently needed processing considerations related to complement length, information status, and semantics motivate system-wide generalizations that can serve to override the default word order. Lexical verb-particle combinations (e.g., a
showdown) and an idiomatic case, V-off are also briefly discussed as providing further evidence for the need for both item-specific and more general constructions.
The Obligatory Coding Principle accounts for the inventories of possible coding frames in languages that, according to the current terminology, can be characterized as consistently accusative or consistently ergative in their system of argument coding. In coding frame inventories fully consistent with the Obligtory Coding principle, every coding frame includes a given type of coding, either A (in obligatory A coding languages) or P (in obligatory P coding languages). However, languages with coding frame inventories violating this principle are not exceptional. This paper examines the questions raised by light verb constructions with respect to the Obligatory Coding Principle, in particular the possible impact of the univerbation of light verb constructions on argument coding systems initially consistent with the principle or obligatory P coding. The discussion is based on an analysis of the role of the univerbation of light verb compounds in the changes that have affected the situation of Basque with respect to the Obligatory Coding Principle, and a comparison of the situation of Basque with that of Andic languages (East Caucasian).