09 January 2007

Free Will (Part III)

(This entry is the third part of my train of thoughts about free will. The previous two entries can be accessed here: Free Will (Part I) and Free Will (Part II).)

This is an absurdly overdue entry. But there is a reason for the delay - the concept of free will has burrowed into the depths of obscure philosophy, and each logical step has to be taken carefully and examined thoroughly, and often ending up in a terrible blur. Since the last post, I've read numerous articles and had several discussions, and I think it's time to pull some meat slices out from the murky soup.

In summary, the previous two entries talked about how free will cannot be possible if our physical laws are deterministic (classical mechanics) or random (quantum mechanics). Since then I explored the various underlying assumptions of this view and how they can be made to be otherwise. One possibility Yao has highlighted to me was that there is a concept that is not subjected to physical laws, which is, for a lack of a better word, "reason", or the connection between cause and effect.

By this I mean something like, for example, the presence of a mass will result in a force - (Newtonian) gravitation. The presence is the cause and the force is the effect. The connection is the ability to see that the presence of mass causes the force. If I apply a force on an object, the object will accelerate. In this case, force is the cause and acceleration is the effect. To be able to understand that this cause "causes" the effect, is what I mean here by "reason".

But what does "reason" has to do with free will? Well, nothing that I know of! Yes, "reason", as far as I can see, has nothing to do with free will. But all Yao is trying to raise is the argument that there may exist certain mental process that are not subjected to physical laws, and it is there that free will can exist.

The trouble with this concept is that, perhaps, it's just inventing something up to satisfy our opinion. That is, there is no meaning in objectifying the bridge between cause and effect, because there is simply none! It's somewhat like saying, there's an apple in a basket and I put in one more apple. So 1 + 1 = 2, and now I have two apples in the basket. And free will hides in that equal sign.

But if we assume that everything is subjected to physical laws, then how can free will exist? One notion I've heard/read more than once is that although the basic laws, the first principles, the underlying equations... they're all deterministic (or random), when you add them up, you'll get something extra. So the equation becomes 1 apple + 1 apple = 2 apples + 1 free will. Putting it another way, if we can make computers as complex as the brain, then it will have free will.

But does it? Does a sufficiently complex mind have free will? Or does it have the illusion of free will?

The sad thing is, we may not even find out, because mathematically that is impossible, thanks to mathematical genius Kurt Gödel. Quoting from this article which touches on his theories on logic,

Another implication is there is no algorithm, or recipe for computation, to determine when or if any given computer program will finish some calculation. The only way to find out is to set it computing and see what happens. Any way to find out would be tantamount to doing the calculation itself.

According to this, it means that when a certain complex mind, build up by basic blocks of deterministic calculations, is presented with two options, whether its choice is fixed or based on free will, there is no simple formula, or a shortcut, to find out. The only way is to run through the algorithm of making the choice, which is the mind making the choice itself.

Zooming out, we see more and more experiements "infringing" on free will. From the same article, an experiment on making random actions showed that

brain signals associated with these actions occurred half a second before the subject was conscious of deciding to make them.

Of course, this does not mean free will doesn't exist. As the article summarises,

the conscious brain was only playing catch-up to what the unconscious brain was already doing.

Which means we may not have control over the unconscious brain, there is still this glimmer of hope that we may have control over our conscious brain and hence our choices.

But that experiment was conducted decades ago. This more recent one has shown that, in an experiment involving fMRI,

the authors were able to successfully predict whether the study participants would decide to purchase each item.

Again, this doesn't mean that free will doesn't exist. What it has showed is that many other measureable factors play a part in decision-making. And these may even be undeterministic (i.e. free will).

But somehow, as science rumbles on, I'm beginning to feel that the arguments for free will wearing thin...


warhammer said...

Hey man, maybe I can suggest something 'extra' arising from the mind other then Yao's reason.
That is our mind's ability to conceive of the un-physical, that is, we can think of situations and things that are in complete contradiction to all known physical laws and mathematics. Like 1+1=3.

This shows that the output from our cognition need not be bound by physical laws. Unlike everything else which we observe to behave within the bounds of physical laws, our imagination need not. This cognitive ability cannot be claimed to be made up can it? So maybe that's one more place where free will can take refuge in.

Pandemonium said...

Then I think things become very... sketchy. So if a computer is programmed to be able to explore all possible mathematical or physical constructs like 1 + 1 = 3, does it mean it has free will? Or does that mean an intellectually inferior person may not have free will?

Nonetheless, I think it worth taking into good consideration. Thanks for the input!

Ponder Stibbons said...

Google compatibilism, which is a philosophical stance that free will is compatible with determinism. This sounds counter-intuitive at first, but I find it very reasonable on further thought.

Imagine that the world is not deterministic (which it isn't). Imagine that our decisions are somehow made in conjunction with some quantum indeterministic process in our neurons. This would seem to allow for indeterminism in our brains, and hence free will. But on further thought, this does not seem to give us free will as we imagine it, because the moment of quantum indeterminism in our neurons is not something that 'we' can control either. If, just before you make every decision, someone rolls a die to determine which course of action you should take, that doesn't give you any free will regarding the final decision. Quantum indeterminism in our brains would be something like that. So, the argument goes, free will cannot be dependent on there being physically indeterministic processes in our decision-making. Rather, we should look carefully at what it really means to have control over our decisions.

A good exposition on the compatibilist position is given by Daniel Dennett in his book Freedom Evolves.

Pandemonium said...

With regards to compatibilism, I think compatibilists are just trying to get round the problem by re-defining free will. In my opinion, their version of "free will" is not free will in the loose sense at all.

And a treatment of free will derived from quantum mechanical indeterminism has been discussed in the first part of this discourse, and I've came to the same conclusion as you that if free will is something that we cannot control, then it's not free will at all.

Thanks for your comments by the way!