On the Shapes of the Polish Word: Phonotactic Complexity and Diversity

Abstract The aim of this contribution is to identify the dominant shapes of the Polish word with reference to three criteria: cluster complexity (i.e., cluster size), saturation (the number of clusters in a word), and diversity (in terms of features of consonant description). The dominant word shape is understood as the most frequent or typical skeletal pattern, expressed by means of alternations or groupings of Cs (consonants) and Vs (vowels), e.g., CVCCV etc., or by means of specific features (of place, manner, voice, and the sonorant/obstruent distinction). Our work focuses on 2 aspects of Polish phonotactics: (1) the relation between cluster complexity and saturation of words with clusters, (2) the degrees of diversity in features of place, manner, and voice within clusters. Using corpus data, we have established that only 4.17% of word shapes have no clusters. The dominant word shape for a one-cluster word is CVCCVCV. The most frequent scenario for a word shape is to contain two clusters, of which 67% are a combination of a word initial and a word medial cluster. We have found that: (1) cluster length is inversely proportional to the number of clusters in a word; (2) nearly 73% of word types contain clusters of the same size, e.g., two CCs or two CCCs (Polish words prefer saturation over complexity); (3) MOA is more diversified than POA across clusters and words.