It depends a lot on where you put the limit between neats and scruffies. I personally would have put him straight in the center, as more of a pragmatist. One of the big criticisms he had (from my memory of the Society of Mind lectures) is that some researchers spent too much time trying to find an overarching, simplistic theory of mind (some sort of physics-envy), which would put him in the scruffy camp. I do also remember him saying that trying to replicate the brain was pointless and that people should focus on trying to replicate its function rather than its architecture, which I am assuming is a criticism of the connectionist approaches, which would put him in the neat camp.