Tuesday, October 03, 2006

I rant inexplicably about free will...

In response to Scott Adams, of Dilbert fame, here (or more accurately, his post here). I also saw many other lucid refutations, or at least, bones of contention, many of them more cogent than mine, or at least shorter...

As some have commented, this argument is somewhat futile, not necessarily because one cannot determine whether free will exists or not, but rather because if you accept personal responsibility as real (as Adams has it), then whether or not there is free will changes *nothing.* Nada. Zip. I've just been told that my sequence of actions, including writing this post instead of working on my dissertation was, I suppose, an inevitable result of the chain of events before it. But, for the purposes of living my life in a satisfying and orderly way, I cannot therefore use that as an excuse to never work again because I "can't help it." I suppose by Adams' reasoning, my previous experiences, my biology, etc. will drive me to spend (slightly) more time working than posting to achieve various things important to me.

BUT: despite this futility, I am compelled nevertheless to rebut a little bit of Scott's reasoning.

The more interesting (I think) rebuttal is that free will as Adams' describes it doesn't seem to be possible in a universe with rational laws (not rational actors, just laws of reality). He describes it as doing something breaking the chain of causal events before it. As I'm going to explain below, there's good reason to believe that many, many things in human society actually ARE indeterminate to some extent (based on the same finding in ecological systems). Meaning: if you wound history up somehow and replayed it, events might happen differently even with the exact same starting conditions. You could call this the vagaries of probability -- because behavior at least is, I think, probablistic -- a given person may choose differently in the same situation with, again, all the same starting conditions. That is, when faced with a big mac, a given person as a dieter may have a 95% chance of eating it, and a 5% of not. Or a newly-minted vegan a 50/50 chance. Mathematically, even given a complex model of decision making, the most you could get would likely be a probablistic answer. So is falling into probablity free will? I don't know. It doesn't break the chain of causal events...

...but then, really, what does? Adams' definition of free will requires people to act at random, it would seem -- that is, at odds with their own experience or circumstance. I guess it would be something like me getting up right now and killing myself -- when I have no reason to do so. Or does it have to be more crazy? A glass of water combusting? A person becoming a bird out of choice? If it is simply that I won't jump out a window because I'm a product of all the causal factors before now, well then, non-free will is a call for an irrational universe. I mean, given free will and all other things being equal, you'd think people would act according to their will in order to satisfy themselves the most in their lives -- monetarily, sexually, etc. So this system would seem to exactly emulate our system. The only way to have a system PROVING Scott's idea of free will would seem to be people running around setting themselves on fire, deciding to become a homeless person for the rest of your life, eating poisonous food for not reason, slapping your loved ones unprovoked -- I mean, only these things seem out of line of what a person with free will would choose, if you still assume people would want to make themselves happy. The only alternative I see is if you relax that assumption, and people don't want to make themselves happy nor successful. This, in turn, is very unlikely to happen in a universe with rational laws simply because such an population of entities without any sense of self-preservation would quickly become extinct. So, as Adams' defines it, free will is impossible unless the laws of the universe themselves are variable -- similar to my ex-girlfriend's believing that altruism isn't altruism if you enjoy, or at least feel emotional rewards for, helping people . One may define true altruism as only cases where you help people, at a cost to yourself, AND not liking it/thinking it's a stupid or horrible idea, but that's such a ridiculous definition as to not be useful. Same with this.

(now, the boring part... or at leat, MORE boring part)

For, though quantum mechanics do not, as a rule, trickle up into indetermination in the macro-world, chaos mathematics creates similar indetermination to quantum mechanics. There are two forms of chaos: determinate and stochastic. The second is easier to explain: it's "noise," it's all the bunches of little chances and changes that make, say, weather, hard to predict. But famous scientist (ecologist/theoretical physicist/mathematician) Robert May showed decades ago you could also have chaos from a simple equation and a time lag. To sum up, let's compare "conventional math" to "chaos math" -- in conventional math, the underlying assumption of ALL of it is that an incredibly small change in the "inputs" of an equation will lead to proportionally small change in the output -- or more precisely, that as the change in input goes to 0, the change in output also goes to 0. Chaos math turns that around, saying that in some systems, as the input change goes infinitely close to zero, the output can change unpredictably. It turns out that this can happen even in a simple mathematical system; imagine if you have bunches of these overlayed on each other? There simply IS no determinate result. There can be probabilistic results (as in quantum physics), and people may obey these as other systems do (i.e in a large enough sample, people will come out as the stats -- 70% doing A, 30% doing B). Is that free will? I don't know.

No comments: