This paper is devoted to the development of nominal declension in two villages in Transcarpathian Ukraine. The Transcarpathian dialects are an example of the preservation of archaic features under the condition of separation from the Ukrainian-speaking mainland (the Transcarpathian lands belonged to Hungary for about a thousand years). However, though the dialects are archaic, they nevertheless continued to develop. This development was sometimes shared with the neighbouring dialects, but sometimes was unique.
The morphological development in the Transcarpathian dialects is especially interesting. It encompasses not only change in sets of endings within old inflectional types, but also redistribution of lexemes between inflectional types, which often leads to merger of the old inflectional types and appearance of new ones, conditioned by new factors. Relative ability of certain factors to underpin inflectional types varies across time. Researchers find that the general development in some Slavic languages has been from inherited, unmotivated, lexically conditioned inflectional types (where the inflectional type of a given lexeme had to be specified in the lexicon because it could hardly be inferred from anywhere else) to grammatically, semantically and pragmatically conditioned categories (Zaliznjak). It is interesting to test this hypothesis against the data of specific dialects, especially Transcarpathian, because their archaic character indicates a slower rate of development, which might mean that they display the stages already passed through by more innovative dialects and languages.
The paper, based on my field recordings in the two villages (Novoselycja near Perečyn, further Nov, and Lypec′ka Poljana near Xust, further LypP) discusses changes in DLIpl of non-feminine nouns. There is a significant difference between the morphologic processes in the two villages, and not only because one of them, LypP, preserves a more archaic state of affairs in terms of the number of archaic desinences, but also because the redistribution of desinences among lexemes seem to have different directions and to be determined by different factors:
1. In LypP, an important factor for the change is the type of stem-final consonant (hard vs. soft). In Nov, this doesn’t play an important role.
2. In Nov, the gender of the noun (neut. vs. masc.) seems to be an important factor in the change, which is not the case in LypP.
3. In Nov, in the majority of cases there seems to be a generalization of one ending for all the declension types, which is not so in LypP.
On the whole, in the more innovative dialect of Nov the factors determining the new distribution of endings are not morphophonemic (the type of stem-final consonant), as in LypP, but grammatical (gender). This is an unusual development for East Slavic languages which have generally preferred not to morphologize the distinction between masc. and neut. in the oblique cases of the pl. On the other hand, this distinction is also morphologized in neighboring Slovak.