Russian Nominal Semantics and Morphology

10. Concluding Remarks

CHAPTER 10 Concluding Remarks The idea running through this book is that lexis and grammar make up a single coherent structure. The results of the investigations provide evidence for the claim that there is something that can be labeled language structure. Every language has its own overall, typologically specific structure that permeates the numerous lexical and grammatical subsystems. Linguistic research must proceed from the detail, taking one or the other subsystem of form distinctions as its point of departure . For this reason we need a number of empirical studies of different sections of lexis and grammar in order to understand how the subsystems making up the structure of the given language are dependent on each other and ultimately reflect a system of basic semantic distinctions. In a way one can say that nothing-or at least very little -is random in the subsystems. Nevertheless, the present investigations of modern Russian have only been concerned with nominals, broadly defined as covering nouns, adjectives, pronouns, and numerals, and certain fields and categories , notably case, have even been left out of consideration. On the other hand, I have included topics that are usually not considered in grammar studies, in particular derivation, and emphasized the interdependence between patterns of suffixation and grammatical distinctions by relating them to one and the same underlying semantic features . The basic claim of the present study is that both the various patterns of lexical classification and the grammatical distinctions of the Russian nominals reflect a hierarchy of semantic features (section 2.2.) that represent basic characteristics of the items denoted by the nouns. It has been shown how this system can be interpreted as the nominal reflection of basic verbal semantics, in particular the distinction between activities and states, as presented in Durst-Andersen's (1992, 1995, 2011) studies of aspect and typology. Pursuing this line of investigation, it has proved possible to draw lines of dependency from verbs to nouns and further to adjectives, pronouns, and numerals, thus demonstrating how lexical and gram- 342 RUSSIAN NOMINAL SEMANTICS AND MORPHOLOGY matical subsystems are interwoven to reflect the language structure. The argument allowed us, among other things, to set up content-form correspondences, signalled in predicate position, for the basic parts of speech (section 7.6). While the function of the noun is to denote a concept , verbs are primarily linked to activities and adjectives to states. This is just one of several reflections of the systematic semantic approach pursued throughout the presentation. The classical concept of sign is taken seriously in that I have systematically looked for content distinctions behind all paradigmatically opposed forms. This also accounts for oppositions that are often, directly or indirectly, treated as fully or nearly semantically empty, such as gender (e.g., Hubenschmid 1993: 47). Further, it accounts for all levels of description. Thus, if a part of speech represents a true word classification, relevant to the language structure, it should also be possible to isolate a generalized semantic feature that is common to all members of the class, and which, ideally, delimits the class unequivocally in relation to other parts of speech. I have approached the nominal parts of speech from this point of view and, I believe, successfully described their basic semantics. On the other hand, it should be acknowledged that a fully unequivocal demarcation of each part of speech on semantic grounds is hardly possible. In order to delimit the parts of speech from each other, one must also refer to certain purely formal features as a supplement to the semantic specification. The same is true of certain grammatical distinctions, in particular declensional class, where semantics only plays a minor role, and of gender, which is assigned to the nouns on a mixed formal and semantic basis. It is, however, interesting to note-as has been demonstrated in my investigation of gender-that semantic distinctions influencing the assignment of gender tend to extend their field of influence to subclasses of nouns for which the assignment of gender is, in principle, purely formal. Thus, it was shown through an investigation of constructions used for comparing persons with things (section 5.4.4) that even these nouns are associated with gender and reflect the female/non-female distinction. Further, even in the cases when gender does not seem to have anything to do with meaning, a combined consideration of gender and declensional class proved to be revealing, uncovering a reflection of semantic distinctions through specific sets of declensional class and gender and, ultimately...