Comments on Jäger “Anything is nothing is something”: on the diachrony of polarity types of indefinites

Abstract

This response focuses on two synchronic and two diachronic issues raised by Agnes Jager’s contribution to this volume. On the synchronic front, we take a closer look at the limits of the featural underspecification approach adopted by Jager, arguing for the need to draw a featural distinction between “ordinary” indefinites and positive polarity items, and also highlighting that the [affective] feature appears to play a rather central role not only in the domain of indefinites, but also beyond. In the diachronic domain, we briefly take issue with the feasibility of constraint reranking as an explanation for change, and then turn to a consideration of how a reinterpretation of Roberts and Roussou’s (2003) analysis of “increasing negativity” may be captured in terms of an Agree-based system (cf. Biberauer and Roberts 2010b). Like Jager’s proposed system, this alternative does not predict that indefinite-related developments should always be unidirectional, as has sometimes been supposed (cf. i.a. also Hoeksema 2010).