| Summary: | Interpretive differences between overt subject pronouns relative to null subject pronouns are commonly considered a defining property of Consistent Null Subject Languages (CNSLs), in contradistinction to Partial Null Subject Languages (PNSLs). In this article, we present the results of two experimental studies which challenge this view with respect to two environments: variable binding and pronominal coreference (binding) under double embedding. Having shown that given the two phenomena, all four logically possible patterns of binding of overt pronouns are attested in languages with null arguments, we propose that a theoretically meaningful typological pro-drop classification should move away from facts surrounding the binding or interpretation of overt pronouns, and should instead focus on what kind of derivational mechanism a language has to yield a null subject structure. Specifically, CNSLs are characterised by a 'rich' inflection-based mechanism, in contrast to PNSLs, for which alternative proposals have been made.
|
|---|