Deletion vs. epenthesis: Intra- vs. inter-paradigmatic arguments in linguistics

Abstract

In this paper, an attempt is made to strengthen the arguments for an epenthesis rule of French liaison given in Klausenburger 1974, and challenged in Churma 1977, by (1) integrating liaison into the whole complex of linking processes, showing how the question of h-aspiré can be solved concretely as a result of the epenthesis formulation, and (2) applying the theory of morphologization proposed in Klausenburger 1976a to the history of liaison, thus explaining the diachronic side of the process. In addition, the deletion vs. epenthesis debate is examined within the wider context of the apparent paradigmatic change-over in phonology today.