Pronouns, DPs, and non-local A-movement

Some of us hope that one day we’ll be able to reduce all instances of Movement to Agree. About Agree, we know that “omnivorous” patterns exist, where a probe will skip a more local goal to target one further away, if the less local goal has some special feature(s) that the probe is interested in. If Movement is Agree, we should therefore expect similar movement patterns, where out of two potential goals, a more featurally special but less local one is favored for movement over a closer one. This is pretty much exactly what goes on in typical Ā-context: in object questions, C moves the wh-object and skips the closer (non-wh) subject. And indeed, we’re used to thinking that Ā-movement can be non-local, but A-movement is strictly local. Given enough featural distinctions, however, there’s no principled reason why we shouldn’t expect this kind of movement patterns to hold in the realm of pure A-movement as well (not even a little bit mixed A/Ā). Based on own fieldwork on Äiwoo (Solomon Islands; Oceanic < Austronesian), I argue that this is precisely the kind of pattern we see in a particular corner of the grammar of this language. Movement to spec,TP in this language favors pronouns over lexical DPs: T will skip a closer lexical DP subject to rather move an object pronoun instead. This also teaches us something about why and how syntax treats pronouns and full DPs differently, and I suggest that at least in this language, any explanations based on “greed” (“this pronoun needs to move because it’s deficient”) or any kind of “licensing” don’t work.