Suppose we wanted to tell a general story about human beings, and suppose we arrive at the idea that the Human Creature is rational. And suppose we want more: we want to explain what we mean by “rational.” So we look around, and we finally discover the idea of “autonomy.” We use the idea of autonomy to explain the idea of rationality.
Time goes by. A history of philosophy passes, and a river of uses of both the word “rationality” and the word “autonomy” go under the bridge. And one morning the analytic philosopher stumbles upon a computer. He is now looking for something general to say about computers. He vaguely remembers (perhaps it’s in his genetic memory?) how useful the ideas of “rationality” and “autonomy” once were. He thinks a bit, and finds the answer: the computer is a non-autonomous rational creature.