Case as an Anaphor Agreement Effect: Evidence from Inuktitut

Abstract

The Anaphor Agreement Effect (AAE) is the cross-linguistic inability for anaphors to co-vary with Φ-agreement (Rizzi 1990; Woolford 1999), with languages making use of a variety of strategies that conspire to circumvent this effect. In this short paper, I identify and confirm a prediction arising from two previous observations by Woolford (1999) concerning the scope of the AAE, based on new evidence from Inuktitut (Eastern Canadian Inuit). I propose that anaphors in Inuktitut are lexically specified as projecting additional syntactic structure, spelled out as oblique case morphology; because Φ-Agree in Inuktitut may only target ERG and ABS arguments, encountering an anaphor inevitably leads to failed Agree in the sense of Preminger (2011, 2014). I moreover argue that this exact AAE pattern is previously unattested, yet is predicted to arise given the range of existing strategies. Finally, this paper provides evidence against previous detransitivization-based approaches to reflexivity in Inuktitut (e.g. Bok-Bennema 1991).

Deriving the anaphor–agreement effect and the violations of it

“…Considerations such as these led Rizzi to claim (p. 27) that the AAE generalization holds "quite systematically in natural languages." Since Rizzi's original observation, there has been both empirical and theoretical study of the AAE generalization, which has yielded interesting results (Woolford 1999, Shiraki 2004, Tucker 2010, Sundaresan 2016, Preminger 2019, Yuan 2021. The empirical side of this research has focused on the different repair strategies that languages employ to avoid a violation of the AAE, while the theoretical side has tried to develop an explanation for the inability of the anaphor to control agreement.…”