This article summarizes the first part of a research on syllable pattern in Kosovo dialects. It attempts to answer the question, whether Kosovo dialects have onepeak syllable pattern, twopeak syllable pattern or maybe some other type. Facing many theories about the syllabic unit, that have been created until now, for our study we choose the sonority theory, which is a very comfortable model for formulating distributional rules of speech sounds. Standard Albanian language is untypical against other European languages. Dialect of Morava e Eperme neither has the typical onepeak syllable pattern (it tolerates twopeak initial clusters), nor it has a twopeak syllable model (it does not tolerate twopeak final clusters). This local dialect can be classified, the same as standard Albanian, as a one with onepeak syllable pattern, where, however, nasal sonorants should be distributionally classified as obstruents (and not as sonorants).
