PROSODIC PHRASING AND WORDBREAKS’ StRENGTH IN THE SPOKEN TEXT (DATA OF PERCEPTION EXPERIMENT)

 The paper discusses the problem of oral speech prosodic phrasing, gives a brief description of the history of the issue, clarifies the basic terms, explains the concept of prosodic break and word boundary strength. The role of Russian linguists and textologists in the discovery of this phonetic phenomenon and in understanding its hierarchical and functional nature is shown. Examples of prosodic labeling of spoken Russian texts are given, which are available in the works by L. V. Shcherba and R. I. Avanesov. The experimental study is described in this paper has a twofold purpose: 1. To test the hypothesis about the hierarchical nature of prosodic phrasing and its mapping in the prosodic breaks’ depth; 2. To substantiate the adequacy of distinguishing 5 levels of the wordbreak’ strength as a means of quantifying the prosodic breaks depth in the spoken text. The experiment design was developed under the assumption that perception would be most consistent when the number of distinct levels of prosodic breaks in the evaluation scale would be most closely correspond to the reliable perception of the listeners. As a result, it was shown statistically reliable that listeners consistently distinguish no more than 5 levels of break’s strength. So it makes no sense to use more detailed evaluation scales in further studies. This conclusion is confirmed by the results of appraisal perception experiments on the material of other languages (English and Dutch). It can be assumed that the maximum 5-level scale reflects the universal properties of human perceptual sensations in the analyzed phonetic space. The final part of the paper presents the results of the obtained data analysis in the statistical package STATISTICA: numerical characteristics of dependencies between the prosodic breaks’ depth, pauses duration on the word boundaries, punctuation marks, elements of syntactic labeling.