| Home Page | Web Logs at CCIL |
|
Oct_08_2006 So I'm ~still~ wondering about evolution. I'm thinking about "survival of the fittest". What does that mean? Fittest? I suppose at the level of the organism, it means best able to survive and reproduce. What does it mean at the level of the machine though? If we look at nature as one big natural-object production machine, what does "fittest" mean? Nature produces pebbles and people. Implicit in the idea of survival of the fittest is the requirement for a goal. You can't be fittest if there's no definition of better or worse... closer to or further a goal. If we suppose that the same evolutionary process created life from non-life, was that step goal driven or random? I don't suppose the pebble outside in the driveway has much interest in survival or reproduction. So nature's cruising along for millions or billions of years and everything in the machine is blindly following physical or chemical or quantum laws. Changes are driven exclusively by random collisions of different forms of matter and energy. Suddenly, that unimaginably unlikely random event, life from non-life. Even more puzzling to me, than life from non-life is that suddenly out of randomness comes goal driven behavior. It's the million monkeys typing Shakespeare. Whether life has free will is a question that's subject to debate, but I don't think there's much room for debate about whether life is goal driven. So there's an heuristic built into "survival of the fittest" for life which doesn't exist for non-life. That heuristic is... preserve the organisms that are best able to survive and reproduce. For pebbles, the heuristic doesn't exist, for people it does. Where could that heuristic have possibly come from? I suppose the answer I'd expect to hear would be that it's an "emergent property". Combine the right chemicals in the right conditions with just the right amount of energy and boom... life emerges with this goal-driven behavior property. Luger points out in "Artificial Intelligence", though, that that's really not an answer. When my 4 year-old asks me questions about complicated subjects, an answer I often fall back on is "that's just the way it works". Well, I think "an emergent property" is a scientist's fancy way of saying "that's just the way it works" or fundamentally, "I can't explain it". As I wrote once before on this topic, life to life, I think evolution is rock-solid. I'm having a real hard time with a million monkeys typing Shakespeare to get from non-life to life though. It just seems thin. In those computer evolution simulators, one thing I think all the virtual organisms have in common is a goal. I wonder how long they'd have to run to produce something interesting if the v-organisms weren't programmed to accomplish something. A pretty long time, I bet. Something like forever. In the face of forever, 14 billion years is insignificant. It's not enough time for a million monkeys to type Hamlet and I don't think it's enough time to produce a goal where one didn't previously exist. Link for this entry: http://home.ccil.org/~remlaps/weblog_2006/metaphysics_index.html#SurvivalOfFittest100706 |
|
Jun_15_2006 http://www.edepot.com/liftext.html Which is linked from... http://users.aristotle.net/~diogenes/meaning1.htm Link for this entry: http://home.ccil.org/~remlaps/weblog_2006/metaphysics_index.html#Aristotle061506 |