A Phase Extension Approach to Double Object Constructions – Evidence from Modern Greek*

Abstract

This article examines A-movement (a)symmetries in the double object construction in Modern Greek. Two types of double object constructions are attested in this language, the ''dative construction'' and the ''double accusative construction''. These constructions, which are analysed as low applicatives, differ with regard to (i) the Case marking of the Goal, (ii) A-movement properties of Themes and (iii) formation of adjectival passives with Goal externalisation. The article shows that (i)-(iii) are interrelated phenomena in Greek and that they can be accounted for by exploiting recent findings on phases, namely phase extension whereby a phase expands as a result of movement of the phase head. The phase extension approach to double object constructions illustrates that head movement interacts with locality as a narrow syntactic operation and that locality-violating passives can be licensed without phase EPP-features or multiple Specifiers. The article also dwells on the controversial issue concerning the quirky Subject status of fronted dative Goals in Greek and shows that the phase extension approach, in which dative clitics are analysed as phonetic realizations of applicative heads, supports the quirky Subject status of dative Goals.

External Arguments and Dative Cliticization: Evidence of Selective Spell‐out of Functional Heads

“…The paper also revisits my previous criticism (Kupula , ) of Anagnostopoulou's () proposal regarding the interaction of dative cliticization with locality‐violating movement. I refine the assertion that Anagnostopoulou's generalization—according to which dative cliticization obviates locality effects—as such is too narrow and therefore misrepresents the empirical facts.…”

“…As I pointed out in previous work (Kupula ), the grammaticality judgments associated with the construction in (10), specifically the degree to which the presence of the dative clitic tis is mandatory, exhibit considerable variation. Crucially, the presence of the clitic is not mandatory for all speakers; all of my three informants regarded (11) as well formed (see also Anagnostopoulou's remark):…”

“…The following section refines the above discussion by highlighting the interaction of dative cliticization with voice/transitivity of the base predicate. I then revisit my earlier (Kupula ) criticism of Anagnostopoulou's () MLC‐based approach, which I have shown to involve a number of empirical inconsistencies. All these inconsistencies can be resolved by deemphasizing the analytic relevance of MLC and by instead treating dative cliticization as a general morphosyntactic reflex of head movement, rendered visible by the interaction of the generalized DFCF and the principles of Distributed Morphology.…”

“…If constructions like (13) are addressed transformationally, whereby the preverbal dative argument is moved to (rather than base‐generated into) the sentence‐initial position, the data in (10) and (13) indicate an indiscriminate presence of a clitic item (here, tis ) whenever movement takes place (see also Kupula ). I return to discuss this issue in greater detail in section .…”

LetCausatives and (A)symmetricdat‐nomConstructions

“…This analysis of the symmetric–asymmetric distinction has been proposed independently by Kupula () on different grounds.…”