This article starts really well:  the idea that one can have “competence without comprehension” really is a stunning insight of both the theories of evolution and of computation, and I’d never thought of putting the two side by side like the author, Daniel Dennett, does.  Then he ruins the fun with the mother of all non sequiturs:  Alan Turing’s great insight is that a machine can do a lot of things without understanding anything.  Therefore, if we make the mindless algorithms that the machines follow complicated enough, the machines will be able to understand!  It seems to me that the Turing’s insight would imply the opposite:  the more refined our computers, the more we will find that it is possible to do without understanding, while not moving a step closer to making a sentient machine.  Following his connection between natural selection and computing, Dennett sees artificial intelligence skeptics as the analogue of creationists, but it’s an funny sort of fundamentalism that would include Kurt Goedel, John Searle, and Roger Penrose.  (I was very impressed, though, with the respect Dennett shows for his opponents.)