Application of the Comparative Method to Morpheme-Final Nasals in Nivkh

Abstract

Following up on recent work, we consider morpheme-final nasals in the Nivkh language family of northeast Asia using the Standard Comparative Method, and attempt to reconstruct the inventory and morphophonemic behavior of morpheme-final nasal phonemes in Proto-Nivkh (PN). Previous work has pointed towards PN nasals at four loci, /*m/, /*n/, /*ɲ/, /*ŋ/, of which at least /*ŋ/ could be phonemically either “strong”, triggering fricatives to surface across morpheme juncture, or “weak”, triggering plosives to surface; with weak /*ŋ/ place-assimilating to following plosives across morpheme juncture, and weak /*n/ and weak /*ŋ/ elided in the Amur and West Sakhalin lects. However, with the benefit of more and better data than were available to previous authors, we find instead that elision must have been conditioned by a feature other than the strong-weak contrast (provisionally, length), but which interacted with the strong-weak contrast (“short” strong nasals were inextant), and that this “length” contrast also conditioned assimilation or non-assimilation of final /*ŋ/ (only “short” weak /*ŋ/ assimilated, not “long” weak /*ŋ/). We confirm that the strong-weak morphophonemic contrast existed for at least /*m/, /*ɲ/, and /*ŋ/ (rather than only for /*ŋ/), and the “length” contrast for /*n/ as well as /*ŋ/. Keywords: Nivkh, Gilyak, Proto-Nivkh, nasal, Comparative Method