FASL 24, ###-### Michigan Slavic Publications 2015 Prominence Redistribution in the Aŭciuki Dialect of Belarusian * Lena Borise Harvard University One of the South-Eastern dialects of Belarusian exhibits an unusual phonological property: in certain environments, the immediately pretonic syllable is pronounced with prominence which is equal to or greater than that of the stressed syllable. This phenomenon has been analysed, albeit tentatively, as stress retraction (Kurylo 1928; Kryvicki 1959; Belaja 1974), and also as pitch peak retraction (Bethin 2006a, 2006b). Instrumental data presented in this paper confirms that the pretonic vowel can be higher in intensity and longer in duration than the stressed one, as well as comparable to it in pitch, depending on the respective heights of the pretonic vowel and the stressed one. However, the acoustic data does not lend support to either the stress retraction or pitch peak retraction hypothesis. Instead, this paper argues that the phenomenon at hand results from redistribution of the acoustic prominence associated with stress over two syllables. The paper is structured the following way. Section 1 lays out the basic facts of the Aŭciuki dialect, spoken in the villages of Malyja Aŭciuki and Vialikija Aŭciuki (Kalinkavičy region, Homel province) in Belarus. Section 2 presents acoustic data illustrating the Aŭciuki phenomenon, collected during fieldwork done in 2014 and 2015. Section 3 summarises an earlier investigation by Belaja (1974). * I would like to thank Kevin Ryan, Christina Bethin, Draga Zec, Patrick Jones, Maria Polinsky, Juliette Blevins, and Donca Steriade for their guidance and advice on this project, and the audiences at Polinsky Lab and FASL 24 for the most helpful discussion. Special thanks are due to Veranika Kurcova for generously sharing with me some of the audio materials from Aŭciuki, as well as for the many valuable discussions on the topic. The comments by two anonymous reviewers significantly contributed to the improvement of the paper. All remaining errors and shortcomings are mine. 94