This study deals with the locative systems of seventeen northern dialects of the Dargwa language (the Dargwa group of the Northeast Caucasian language family). In the first part of the paper I discuss the relations between spatial and non-spatial uses of the localization morphemes. Here I prove that locatives are not equal in their ability to be used in non-spatial contexts and most of such uses concentrate around two morphemes: *cːi INTER and *ki SUPER, while other morphemes either develop a very limited set of non-locative uses or do not develop them at all. The second point of this part is that the semantic source of the non- spatial uses of the locatives is their spatial meaning at the moment of their grammaticalization and not their synchronic spatial meaning. In the second part of the paper I apply statistical methods to the distribution of the non- locative contexts among the morphemes of localization and orientation. Here I show that even though the non-spatial semantics of inter and super are not always connected to their locative semantics, the choice between the two localization is not random and the contexts form two clear clusters. In this section, I also analyze the non-locative uses of orientations and show that the vagueness of the difference between lative and essive that exists in spatial contexts is reflected in non-spatial contexts as well.
Four electronic corpora created in 2011 within the framework of the “Corpus Linguistics: the Albanian, Kalmyk, Lezgian, and Ossetic Languages” Program of Fundamental Research of the RAS are presented. The interface and functionalities of these corpora are described, engineering problems to be solved in their creation are elucidated, and the promises of their development are discussed. A particular emphasis is made on the compilation of dictionaries and automatic grammatical markup of the corpora.