Sandhi phenomena in Polish
Irena Sawicka
Nicolas Copernicus University, Toruń
The term "sandhi" in Polish, as in all Slavic phonetics, refers mainly to the voicing and de-voicing of word-final obstruents. The rules are always formulated in purely phonetic terms, i.e. as depending on phonetic context. That is true, and this is the main difficulty when we try to determine the place of the phenomenon in the language, because most of these phenomena, in terms of natural phonology, should be qualified as rules, not processes. Only part of these assimilations may be qualified as processes.
In the paper, I shall present arguments for the thesis that most of the sandhi phenomena in Polish are not alive processes and should be discussed as morphophonemic alternations (or equivalent rules, depending on the theory applied). Certain part of contexts, however, do undergo alive phonetic assimilation, but, again, this is not unequivocal because the dominating rules, formulated on other contexts, interfere.
References:
Andersen H., 1986, Sandhi and prosody: reconstruction and typology, [in:] Sandhi phenomena in the languages of Europe, ed. H.Andersen, Berlin, Mouton de Gruyter, 231-246.
Steffen-Batóg M., 1996, On the pronunciation of some Polish consonant clisters containing sonorants, Studia Phonetica Posnaniensia 5, 61-85.
Sawicka I. An outline of the phonetic typology of the Slavic languages, Wydawnictwo UMK, Toruń, 2001.