CRITERIA FOR THE CHOICE OF DISTINCTIVE FEATURES IN SLAVIC LANGUAGES

Phonological descriptions of Slavic languages in most cases use ’structuralistic’ distinctive features, based on the work of Jakobson, Fant and Halle (1952) and Jakobson and Halle (1956). A few use ’standard generative’ sound features, proposed by Chomsky and Halle (1968), which were designed not only to capture the phonological contrasts of languages, but to describe the phonetic content of segments derived by phonological rules, as well as underlying segments. A number of modifications to be found both in the specific set of distinctive features and in the conceptualization of these features which linguists introduced since 1968 are hardly used in Slavic phonology. If distinctive features are needed just for classification and general distribution of sounds in a particular Slavic language, there is a wide choice of feature inventories. In stating phonological rules, the features which seem to best explain the motivation of the prosesses in question should be chosen. However, if a phonological description of a Slavic language is aiming to address certain empirical or conceptual issues in the theory of Universal Phonology, the choice of features and the phonological model is more limited. A useful approach in the comparative phonology of Slavic lanugages would be to take the set of universal phonetic features which lanugages implement in various linguistic ways. A phonetic feature which is not used phonologically by one Slavic language or dialect but is by another, even features responsible for ’foreign’ or ’dialectal’ Slavic accents, should be included in the description of each language