Contact-induced Change and the Comparative Method: Cases from Papua New Guinea1

Abstract The overriding goal of the comparative method is to reconstruct linguistic prehistory and thereby to contribute to human culture history. Unfortunately, the method is blinkered in this endeavour by its prerequisites. It deals only with genetically related languages, and then only with regular correspondences among them. Yet the history of almost every language entails contact between its speakers and those of other languages, contact which often impairs the language’s correspondences with its kin. Whether the languages in contact are genetically related or not, or so closely related that one should more accurately speak of dialects, the comparative method usually leaves this noncorresponding material unaccounted for, either by ignoring it or by labelling it ‘borrowed’.