This paper argues that the so-called antiagreement effect (AAE: Ouhalla 1993), which refers to suppression of subject-predicate agreement that is induced by subject-extraction, should be accounted for as agreement intervention, in which an element with a certain feature intervenes between the extracted subject and T. The analysis is based on Chomsky's (2000 et seq.) probe-goal theory with some non-trivial revisions: phrases can probe (Fujii 2001; Bošković 2007; Kato et al. 2014) and lower copies are invisible for probing (cf. Chomsky 2013, 2015a). Focusing on the data from Berber, Breton and Kinande, it will be argued that phi-feature-bearing C, which triggers so-called wh-agreement (Chung 1998; Watanabe 1996) with the extracted subject, is an AAE-inducing intervener in those languages. The proposed analysis can correctly predict the following facts: (i) there are AAE-languages and non-AAE-languages, (ii) the AAE is only induced by subject-extraction, (iii) the AAE-languages differ as to whether long subject extraction induces the effect, and (iv) the AAE can be undone in negative clauses in Berber and Breton. Finally, it will be shown that the intervention approach can extend to hitherto unaccounted AAE data in Berber: the AAE which is induced by a negative concord item (Ouali 2005).
Proceedings of the 34th West Coast Conference on Formal Linguistics
edited by Aaron Kaplan, Abby Kaplan, Miranda K. McCarvel, and Edward J. Rubin
Table of contents