Is the Overt Pronoun Constraint Universal?

It has been observed that overt and null subjects do not have the same distributional properties within the same pro-drop language. That is, there are certain grammatical and discourse principles that determine the occurrence of overt versus null pronominal subjects in a particular context (Enç, 1986; Erguvanlı-Taylan, 1986; Pérez-Leroux & Glass, 1997, 1999). For example, it is known that overt and null pronouns demonstrate different interpretative properties as bound variable or referential pronouns (Montalbetti, 1984; Saito & Hoji, 1983). The binding contrast between overt and null pronouns in the context of quantified antecedents is formulated by Montalbetti (1984) under the Overt Pronoun Constraint (OPC) and postulated as a property of Universal Grammar (UG). Following Montalbetti’s claim on the universality of the OPC in pro-drop languages, L2 researchers have looked at the acquisition of this phenomenon in Spanish (Pérez-Leroux & Glass, 1997) and in Japanese (Kanno, 1997; 1998) and found native-like performance in adult L2 acquisition of the OPC. They have attributed this to the presence of UG-constrained L2 grammar in adult learners (see also White, in press). In this paper, we will look at the OPC effects in Turkish, a null-subject language, which, despite arguments for the universality of the OPC, presents evidence for the similar distribution of null and overt pronouns. L2 Turkish data from native English speakers will be discussed within this context in relation to Binding Principle B.