VARIATION IN VERBAL PREFIXATION: RUSSIAN DO- AND PERE-

The paper deals with variation in stem choice attested among the Russian prefixes do- and pere-. According to the classification by Tatevosov (2008; 2009; 2013), the completive do- ‘to finish V’ and the repetitive pere- ‘to do V again, better than before’ form a separate type of prefix with a unique cluster of formal features. One of these features is the ability to combine with both imperfective (cf. шить > дошить, перешить) and perfective stems (cf. проверить > допроверить, перепроверить). A further feature is the restriction on combining with secondary imperfectives (i.e., stems containing the imperfective suffix -ыва~-ва~-а).

The aim of the study is to describe in detail and to compare the formal behavior of these two prefixes based on informal Internet communications data. The expectation is that the rules of stem choice would be identical or similar for these two prefixes.

In the paper, I formulate usage-based rules of stem choice for different verbs classes. If the data of spontaneous informal speech are taken into account, the rules appear to be quite complicated and non-strict for both prefixes. For some verbs, parallel variants derived from different stems are used in natural texts (cf. перепаковать ~ переупаковать подарок). For such verbs, I give frequency distribution and assumptions on factors regulating choice.

Although do- and pere- belong to one and the same formal type according to their general features, they appear to have different rules of stem choice. The difference is partly motivated by the fact that in contrast to pere- and contrary to theoretical assumptions about the structure of Russian verbs, do- frequently combines with secondary imperfectives, at least in the type of texts under consideration. For instance, the variantsдопаковать and доупаковать, which are parallel to the above mentioned pere-derivates, compete with the verb доупаковывать, too (derived from the secondary imperfective упаковывать).

In the literature, Russian prefixes are divided into two positional types: inner (lexical) and outer (superlexical), both associated with a set of particular formal features. More detailed analysis shows that this two-part classification is not enough: in a more elaborated classification by Tatevosov (2009; 2013) superlexical prefixes are subdivided into three more types. If one turns to a larger mass of spontaneous texts and to a more finegrained analysis, more and more specific features typical of minor sub-types and of each individual prefix can be found.