'East Formosan' and the PAN palatals
Résumé
This paper shows that the putative 'East Formosan' subgroup proposed in Blust (1999) on the basis of a sound change allegedly turning a voiced palatalized velar stop into an alveolar nasal, pays no attention to the geographical principle of conservation by the periphery, and to the need for phonetically natural sound changes in historical interpretations. It argues that the phoneme known as *j was a palatal nasal rather than a voiced palatalized velar stop, and proposes for it the new label *n y. With support from the history of the Chinese palatal nasal, it argues that *n y evolved to a sound combining nasality and friction, and that denasalisation of that sound occurred for lack of a prenasalized series in which it could be integrated. The paper also shows, based on an earlier proposal by Dahl, that the putative phoneme known as PAN *ñ, which competes with *n y for the palatal nasal slot in the PAN consonant system, arose no earlier than PMP, with an independent parallel development in the Formosan language Kanakanabu, when the outcome of the merger of PAN *niV and *NiV became palatalized. It concludes that reconstructing PAN by validating Dempwolff's PMP is a counter-productive strategy and advocates reconstructing PAN directly on Formosan evidence.
Domaines
Linguistique
Origine :
Fichiers produits par l'(les) auteur(s)
Loading...