On Roman Jakobson’s Concept of Irregularity in Russian Conjugation

Ronald F. Feldstein, Indiana University

This paper examines several examples of Russian verbs which are on the borderline between being regular and irregular. It seeks to demonstrate that revised notions of “regularity” and “irregularity” are needed for the treatment of Russian conjugation, whether for scholarly or pedagogical purposes. Several problematic verbs are identified and arguments for and against their status as regular or irregular are discussed. Finally, a modified verb classification is proposed, which takes these considerations into account.

Treatments of Russian conjugation must confront the issue of what constitutes irregularity, which is not always  obvious. When it appears that verbs break rules which apply to others, they often possess a unique environment. We may then ask whether that environment is responsible for the irregularity. E.g. the verb tkat′ is said to be irregular, due to the absence of the velar mutation k>č in the present tense (e.g. tku, tkeš′, tkut). However, the presence of velar mutation would have induced a phonological problem: initial tč-, which cannot be initial, but only on morpheme boundaries (e.g. letčik). Two possibilities exist: either tket is an irregular verb, which should have mutated k>č, or the verb is regular, but k>č is blocked after a root-initial t- (perhaps because it would lead to a single long [čč] as the only root consonant). In the latter case, the verb is not irregular, but simply obeying a phonological rule.

The -avaj- type verbs have stems ending in - avaj- in the imperative, and are supposed to follow the rule of dropping the - av- in the present tense. Three stems and their compounds follow this pattern (davaj-, -stavaj-, and -znavaj-). But, other verbs (e.g. plavaj- and rare pomavaj-) are exceptions, since they do not drop the -va- in the present. Many treatments classify both davaj- and plavaj- as regular, but this seems contradictory. One solution is for the -avaj- class to take stress into consideration.

Jakobson’s 1948 Word article, “Russian Conjugation” (1948:163) suggested two basically different types of irregulars: “single deviating forms” and “discrepancy between the prevocalic and preconsonantal stem- shape.”  These were all presented as lists of irregulars, but there is a major difference between the two categories. Single deviating forms must be left as anomalies, but the class with a discrepancy between pre-V and pre-C endings really has many regular properties. The latter can be divided into two large groups, based on an unexpected mutation or lack thereof (e.g. stonat′, stlat′), or а mutation of the first stem-vowel (different before -V and -C endings: spat′, molot′, pet′).

Тhe latter group of verbs (spat′) is usually treated as irregular, yet has regular properties, while the verbs mentioned above (plavaj-, davaj-) are usually treated as regular, but have some irregular properties. This indicates that the entire situation is in need of revision and re-classification. This paper hopes to contribute to that process.