Rethinking Hurtado?s observation WCO effects across clitic doubling patterns in Rioplatense Spanish

SAAB, ANDRÉS

Congreso; Romania Nova XII, XIX Congreso de la Asociación de Lingüística y Filología de América Latina; 2021

ALFAL

I argue that Hurtado?s observation (HO), according to which clitic doubling (CD) and clitic left dislocation (CLLD) structures always repair WCO, is only partially correct. I propose then that whereas CD triggers A-movement to the doubled nominal, CLLD corresponds to an A-bar chain. As for the empirical point, Di Tullio et al (2019) have shown that the position of the possessive is crucial for distinguishing CD from CLLD. Whereas CLLD always repairs WCO, regardless the position of the possessive, in CD WCO is repaired only when the possessive is in the same embedded clause as the clitic. (1)a. A Juani, sui profesor cree que loi criticó María.b. A Juani, María cree que loi criticó sui profesor.(2)a. *? A JUANi cree sui profesor que loi criticó María.b. A JUANi cree María que loi criticó suiprofesor.Therefore, HO cannot capture the ungrammaticality of (2a). I argue that the problem vanishes once we assume that CD of focus moved objects always involves a first step of A-movement, whose visible effect is the occurrence of a doubling clitic. As is well-known, A-movement repairs WCO, which directly explains why only (2b) is a licit configuration: only there the moved object A-crosses the subject containing the possessive. Following Sauerland (1998), Ruys (2000, 2004) and van Urk (2015), I adopt the idea that A-movement of the relevant constituents results in abstraction over individual variables, whereas focus movement, as other related movements abstract over choice functions. This results in a situation in which the possessive, an e-denoting element, can never be bound by a choice function abstractor. As for CLLD, I conjecture that either the dislocated topic abstract over individuals, or that the dislocated topic is always base-generated (Cinque 1990).