At the end of lecture 3 of Mind and World, McDowell does a nice job of articulating a division I’ve been thinking a lot about recently:
Now there is a temptation to think it must be possible to isolate what we have in common with [animals] by stripping off what is special about us, so as to arrive at a residue that we can recognize as what figures in the perceptual lives of mere animals. That is the role that is played Evans’s picture by informational states, with their non-conceptual content. But it is not compulsory to attempt to accommodate the combination of something in common and a striking difference in this factorizing way: to suppose our perceptual lives include a core that we can also recognize in the perceptual life of a mere animal, and an extra ingredient in addition.
This is representative of a way of thinking which extends all the way back to Frege’s context principle, which Wittgenstein quotes in the Tractatus, and seemingly continues to believe in his later philosophy. One difference is that the Frege/Wittgenstein thought concerns propositions and symbols, whereas McDowell is concerned with the relationship between experience and conceptual capacities. Dennett, and Stich (circa 1983) are interested in the attribution of propositional attitudes. What is common to all the cases is the idea that we can only understand the activity of the system as a whole–what is primary is the fully functional, primarily veridical, activities of a rational being.
In contrast, the method of cognitive science and neuroscience involves a process of decomposition. From the fact that an individual with a particular type of brain damage can give definitions for words, but not recognize the object that those word refer to (for instance) we conclude that the abilities are distinct. This approach is essential if your concern is to understand how the trick is performed–it shows that two distinct mechanisms are involved in performing the trick.
Someone like McDowell can come back with the response that while the mechanisms involved are separable, they don’t tell us the nature of mental abilities unless we consider their combined functions. Only by looking at normal case can we understand the role of these particular mechanisms. That is, the part of our cognitive capacities that we share with lower animals has an entirely different significance once it is combined with our distinctive rational faculties. This view may sound bizarre, at least in part because we (myself included) are inclined to defer to the expertise of cognitive science. It is made a little more compelling by consideration of cases like the one Stephen Stich poses in “From Folk Psychology to Cognitive Science,” where an elderly lady will reliably tell you that “McKinley was assassinated” but cannot answer the question “is McKinley dead?” or “who was McKinley?” or even “are you dead?” The overwhelming temptation is to say that even if the physical structure which leads her to say “McKinley was assassinated” is the same as the structure that lead her to say it earlier in her life, it is not a belief in the former case.
For the time being, I will call it a draw.
Side-note: if that doesn’t sound like McDowell outside of the quotation, it’s because it’s not McDowell’s vocabulary, and it’s not necessarily how he thinks about the issue, it’s just me trying to springboard off of what he said in that passage from Mind and World (I did run Stich, McDowell and Dennett together…). I do think I’m gesturing at a way of seeing the issue which (in some moods) I think McDowell could be right about.