Defining the three binding domains of Basque

Abstract

In this paper, I will argue that Chomsky's revised binding theory (Chomsky 1986) has one main flaw: it is type-oriented, in the sense that the narrow binding category (NBC) and the wide binding category (WBC) it defines each concern a specific type of pronouns: the pronominals on the one hand, and the anaphors on the other. My main contention is that these BCs are in fact type-independent, since (eastern) Basque possesses five classes of (non-emphatic) pronouns: if some pronominals must be free in their NBC, others must be free in their WBC, and if some anaphors must be bound in their WBC, others must even be bound in their NBC; moreover, the reflexive possessive here is a typical pronominal anaphor, because it must be free in its NBC, and bound in its WBC. I will next sketchily discuss the nature of the two BCs, to conclude that if subjects must be incorporated into the definition of the NBC, Comp is the decisive element in the case of the WBC of non-emphatic pronominals. Finally, I will show that two distinct WBCs are in fact necessary, since the binding properties of the would-be emphatic pronoun(s) are different from those of the other pronominals: they must be free both in their NBC and in the domain of a c-commanding SUBJECT.