Genitive of Negation in Russian: Blocking of Case-licensing by Negation

Genitive of negation (GN) in Russian is a phenomenon in which an argument that bears structural Case in an affirmative sentence can surface bearing Genitive Case (GEN) in the corresponding negative sentence, as shown in (1). GN occurs only on Theme arguments, as shown in (2-4), and GN marked DPs refer to non-presupposed entities, whereas their structural Case-marked counterparts refer to presupposed things, as shown in (5-6). GN has received numerous accounts from various perspectives among others, and the references therein). In this paper, I offer a syntactic account of the Case-alternation facts about GN in Russian, with insights from the previous research that there is a correlation between Case on DPs and their presupposed or non-presupposed interpretations. I show that an elegant and proper treatment of GN is possible by utilizing the ingredients already available in recent linguistic theory, such as Relativized Minimality (Rizzi 1990) and the Mapping Hypothesis (Diesing 1992). I propose that GN instantiates a case of like blocks like: that is, NEG, being a Case-licensing head, blocks the Case-checking by other Case-licensing heads when it intervenes between these heads and their target DPs. I argue that what lies at the heart of GN in Russian is the blocking by NEG of structural Case-licensing on DPs which stay inside VP, which refer to non-presupposed things. In order to formalize this idea, I adopt Bailyn's (1997) argument that, in Russian, NEG licenses GEN and that Negation Phrase is located above VP and below Aspect Phrase, which is assumed to license propose that a Case-licensing head X can license Case feature F on Y if and only if there is no intervening Case-licensing head Z which c-commands Y but does not c-command X. Under the new proposal, the occurrence of GEN on DPs referring to non-presupposed things is explained as follows: Since NEG blocks structural Case-licensing on DPs inside VP by higher heads such as Aspect and Tense, the only way in which these DPs can satisfy the Case Filter is by having their Case features checked by NEG. Since these DPs stay inside VP, they get mapped into the nuclear scope and receive non-presupposed interpretations, as predicted by the Mapping Hypothesis. In order to account for the correlation between structural Case-marking on DPs and their presupposed interpretations, I posit that these DPs raise out of VP in overt syntax, and, by moving above NegP, they escape the blocking …