In lieu of an abstract, here is a brief excerpt of the content:

The Two "Possessor Raising" Constructions of Bulgarian* Guglielmo Cinque and lIiyana Krapova 1. Introduction. In this article w e present an analysis of a specific phenomenon of Bulgarian syntax, which can be better understood, we argue, through a comparison with Romance. As is often the case when one compares different lan guages, certain constructions appear not to correspond neatly. How ever, before surrendering to the conclusion that no neat correspondence exists across languages, one should try to see if one can find it by decomposing the complexity of the data. This is what we attempt to do here. Bulgarian clausal dative clitics can, as in other languages, be interpreted as external possessors of a DP, provided they are contained in the same minimal clause containing the DP, and that they c-command the DP (or its trace); exam ples of each of these, respectively, are shown in (1-4) (d. Gueron 1985: 48; 2003: 193f.):1 (1) a. Kuceto mu otxapa prasta. dogoEF himoAT bit-off fingeroEF 'The dog bit off his finger.' * This paper is dedicated to Wayles Browne as a token of our appreciation and respect. A version of this article was presented in Paris, in December 2008, at the Ecole Normale Superieure. We thank the audience for their comments , and in particular Jacqueline Gueron, Richard Kayne, Hilda Koopman, and Dominique Sportiche. We also thank Richard Kayne and an anonymous reviewer for their comments on a previous version of the article. 1The literature on so-called "possessor raising" in various languages is extensive, and we are able to review it here only partially. References to specific studies are made where they directly bear on points of our analysis. Steven Franks, Vrinda Chidambaram, and Brian Joseph, eds. A Linguist's Linguist: Studies in South Slavic Linguistics in Honor of E. Wa yles Browne. Bloomington, IN: Siavica, 12348 . 124 G UGLI ELMO CI NQUEAND ILlYANA KRAPOVA (1) b. Te mu namerixa cadiira. they himOAT found umbrellaoEF 'They found his umbrella.' c. Te ne mu saobstixa imeto. they not himoAT communicated nameoEF 'They didn't communicate his name.' (2) a. Kaza se [ee sa mu namerili cadiiraj . said3sG REFL that are3PL himoAT found umbrellaoEF 'It was said that they found his umbrella.' b. Kaza mu se [ee sa namerili cadiiraj . said35G himoAT REFL that are3PL found umbrellaoEF 'It was said to him that they found the umbrella.'! *'It was said that they found his umbrella.' (3) a. Kaza [ee ne mu se vartjala glavata ot vinotoj. said35G that not himoAT REFL spinEVID headoEF from wineoEF 'He said his head was not spinning because of the wine.' b. Kaza [ee glavata; ne mu se vartjala t; ot said35G that headoEF not himoAT REFL spinEVID from vinotoj. wineoEF 'He said his head was not spinning because of the wine.' (4) a. *Jumrukiit ne mu udari masata. fistoEF not himoAT hit tableoEF b. Jumrukiit mu ne udari masata. fistoEF himoAT not hit tableoEF 'His fist did not hit the table.' The examples in (1) have been taken in the literature on Bulgarian to constitute a homogeneous construction, and have been analyzed as involving either movement of the clitic from the DP expressing the possessee (Franks and King 2000: 276; Stateva 2002; Moskovsky 2004) or direct base generation of the clitic in the clausal dative clitic position [18.224.32.86] Project MUSE (2024-04-18 10:29 GMT) THE T wo " POSSESSOR RAISING" CONSTRUCTIONS OF BULGARIAN 125 (Schick 2000; Schtircks and Wunderlich 2003, section 4; Tomie forthcoming). Here we argue that in fact two distinct constructions should be recognized. The first, identical to what is sometimes referred to as "possessor raising" in Romance, imposes a benefactive/malefactive reading on the possessor, is limited to inalienably possessed body parts (with some extensions), and shows properties of a base-generated construction; the other, which does not have any benefactive/ malefactive connotation, nor limitation to inalienably possessed DPs, involves instead movement of the clitic from within the DP that expresses the possessee.2 To see this, it may be useful to start from a puzzling contrast between the Romance and the Bulgarian constructions3 2. A Comparative Puzzle The Romance construction corresponding to (1) is subject to a number of well-known restrictions, listed here in (Ia-c)A 2 With respect to these properties Romanian appears to pattern with Bulgarian rather than with the other Romance languages (see £n. 16 below). 3We ignore here certain differences among the Romance languages, which are orthogonal to our concerns. For...

Share