July 11, 2006: Whose Intelligence is it Anyway?

My primary reaction upon completing Jeff Hawkins' On Intelligence seems to be, "I should have stopped after Chapter 6!" My "corollary" reaction, so to speak, is one of my favorite clichés: "To a small boy with a hammer, everything looks like a nail." If Hawkins were an admirer of Keats (which I suspect he is not), he would probably say, "Intelligence is neocortex; neocortex is intelligence. That is all you know on earth and all you need to know."

The problem is that this does not necessarily wash. As I had previously observed, when you take such a narrow view of the brain, that brain may as well be in a vat; and, when Hawkins turns his attention to "The Future of Intelligence" (Chapter 8), it seems as if he does not think that such a vat is that bad an idea. Now I do not want these final remarks on Hawkins' book to be a matter of choosing sides. I think that there is a lot of subtle philosophy that still needs to be worked dealing with the relationship between mind (wherever it may reside, if it "resides" at all) and body and the implications of that relationship upon being itself. This is neither the time nor the place to deep-end on that philosophy (although Hawkins has certainly provided me with the incentive to study it in greater depth)! Instead, I shall bring my own rant to closure through a single observation:

The Hawkins model of intelligence is totally autonomous: Intelligence emerges through sensorimotor interactions with the rest of the world, but I doubt that anyone (including Hawkins) would call such interactions "social." So, for the sake of argument, let us assume that we all live long enough to see Hawkins' model implemented in some realistically operational fashion; and let us further assume that his implementation has at least some of the scale-up properties (such as capacity or speed) that interest him so much. Would we (the all-too-human) be able to engage with these artifacts for our own benefit? My guess is that we would not. We probably would not even be able to communicate with them in natural language because of our radical differences in context. (This would be a living example of the Wittgenstein quip: "If a lion could talk, we would not understand him.") So it could very well be that Hawkins knows how to build devices that would make the world a better place, but who's world would it be? This is not meant to be a defeatist position that celebrates weaknesses that we have that we may never overcome; it is just a pragmatic suggestion that all of us can do no better that play with the cards dealt to us. At the end of the day, I suppose I am just a Voltairian meliorist!