Theories of features, continued

As we’re still digesting two in-depth presentations by Omer Preminger against abstract Agree as a way to analyze anaphora and other predicate-argument relations, the discussion from the Thirty Million Theories of Features workshop continues unabated in various venues — Gillian has a new blog entry entitled Define semantics, and I have one on the question of whether plural properly contains singular. And now Thomas Graf has posted a lengthy response to my blog post on his own blog, Outdex, entitled Omnivorous number and Kiowa inverse marking: Monotonicity trumps features?