Presented by Owen Edwards as part of the ANU Centre for Research on Language Change seminar series

An examination of the historical phonology of the languages of Rote and Uab Meto reveals at least two strata of vocabulary, each attesting regular sound correspondences. A top-down approach to the data reveals one set of regular correspondences in inheritances from proto-Malayo-Polynesian. These correspondences include sound changes which link Uab Meto with the languages of western Rote. A bottom-up approach to the data reveals another set of regular sound correspondences found in vocabulary not known to be inherited from Proto-Malayo-Polynesian. These correspondences include sound changes which link Uab Meto with the languages of eastern Rote. I present this apparently contradictory data and discuss some ways of resolving it.