ИСТИНА

Two case theories are currently competing in syntactic literature: Inherent Case Theory (ICT) and Dependent Case Theory (DCT). According to the former theory (cf. Woolford 1997, 2006, inter alia), case represents a relation between a noun phrase and some verbal head (e.g. v, T, or C); in contrast, DCT takes case to be a relationship between two noun phrases in some defined domain (cf. Marantz 1991, Baker 2015, Baker and Bobaljik 2015). Evidence for or against those theories has typically been drawn from ergative languages, such as Basque, Shipibo, or Chukchi, or from quirky subjects in languages like Icelandic. In this paper, we shed new light on this debate by bringing in experimental and corpus data from Russian, a language which, although not ergative in the clausal domain, exhibits an ergative-style alignment in eventive nominalizations: objects of transitives and subjects of intransitives are marked GEN, and the subjects of transitives are marked INSTR. We examine the predictions of both ICT and DCT and show that neither theory, as currently defined, can account for all of the Russian data.