Facilitator effects in middles and more

Abstract

In this paper, I will argue that obligatory adjuncts in many languages' middle constructions are a byproduct of an independent constraint on movement, called Anti-locality (Abels 2003; Grohmann 2003; Boskovic 2007; Schneider-Zioga 2007; Erlewine 2016; Brillman & Hirsch to appear). Anti-locality rules out movement that is too local; in other words, movement cannot produce a structure that is too similar to the pre-movement configuration. Though Anti-locality has typically been proposed for A-movement, one of its signatures is the presence of obligatory adjuncts (Erlewine 2016; Brillman & Hirsch to appear), which are also found in many languages' middle constructions. I argue that the profile of obligatory adjuncts in middles is best understood if we extend Anti-locality to A-movement (Deal 2019), thus providing support for the unity of A- and A-movement. The proposal will include a tightly restricted set of assumptions about the structure and derivation of a middle, which combined with a formalized Anti-locality constraint, predicts the obligatory adjunct effects we find in different languages' middle constructions. In particular, middles will be argued to lack a VoiceP and have a two step derivation, in which the object moves through Spec vP en route to subject position. Each movement step will be subject to Anti-locality, which can be ameliorated by adjuncts, reflexives, or lexical structure in the predicates that can form middles.