Thursday, October 28, 2004

Science says we must be robots

The Ancient Greeks were the first to make a serious attempt to kill off our Free Will when they conceived of the existence of ‘atoms’ - the indivisible particles from which everything in the world, including the human brain, is constructed.

Our knowledge of science tells us that the behaviour of any atom or molecule is entirely automatic, dependent only on its physical properties and the laws of physics. So how can it be that when you put a billion entirely automatic atoms together to make a human… you create something that can do what it wants with its life?

It’s a question that has always baffled philosophers and scientists. How do 'we' get control of these automatic atoms of ours, if 'we' are nothing but atoms ourselves? For us to have free will, wouldn’t there have to be a part of us that isn't made of atoms; a part of us that’s free to tell all the atoms how to behave? But if so, then where is it - this non-atomic corner of our brains? And what kind of ‘stuff’ is it made of… if not atoms?

Of course, despite the brutal logic of this argument, free will springs back to vigorous life the moment the philosopher takes a break from his deliberations and has to make the difficult free choice of whether or not to have a Danish pastry with his coffee.

Or as science writer Matt Ridley puts it:

“I am quite capable of jumping in my car and driving to Edinburgh
right now and for no other reason than that I want to....
I am a free agent, equipped with free will.”

- Matt Ridley, Genome

Free will prevails the moment we leave the classroom and step back into the real world.

But then along comes Charles Darwin and his theory of evolution, and suddenly free will’s in trouble again: forget the atomic level – it seems we can’t be free at the ‘whole creature’ level either.

"We are survival machines - robot vehicles blindly programmed
to preserve the selfish molecules known as genes."
- Richard Dawkins, The Selfish Gene

The idea that evolution could create an animal that has the free will to do what it wants is like suggesting that water could run uphill. Even if a ‘free will gene’ could have been created by the chance mutations that created our eyes and ears, this ‘free will gene’ would simply never have been ‘naturally selected’ for. Natural selection, by definition, can only select genes that improve their own survival chances. ‘Doing what you want with your life’ is the evolutionary equivalent of a blind lion triumphing over a sighted one: it’s like being born with an insatiable desire to drink poison and jump in front of freight trains.

“The initial configuration of the universe may have been chosen by God, or it may itself have been determined by the laws of science. In either case, it would seem that everything in the universe would then be determined by evolution according to the laws of science, so it is difficult to see how we can be masters of our fate”
- Stephen Hawking

So what’s going on?

It makes no sense: how can there be such a gap between our scientific understanding and our personal experience of what our lives are all about? On the one hand, we know that we’ve got free will, while on the other hand we know that every time we exercise this free will of ours, we must be overcoming not only the laws of physics but everything science has taught us about the way that human beings came into existence.

What’s the answer?

Can we reassure ourselves that one day we’ll uncover the gaps in the laws of physics? Or is it time to start thinking the unthinkable about ourselves?

What if we don’t actually have the ability to jump in our cars and go to Edinburgh whenever we want to? What if we don’t really have the freedom to choose whether or not to eat that Danish pastry?

Could it all be just a delusion…?

“Free will is a delusion caused by
our inability to appreciate our true motivations.”
- attributed to Charles Darwin

Could it be that we’ve been fooled into thinking that we’re making our own choices, when in reality we’re just being told what to do by billions of years of natural selection?

Next: How we make a decision

Or, for a summary: The Story of the Conscious Robot

Or go back to the home page to see all the pages


stephenlawrence said...

I find most philosophical questions to be very difficult to answer but not the one about free will. surely it's simple. There are plenty of reasons to think we don't have free will and none to think we do. The reasons to think we don't have free will are already well put, I'll give you one more. Take something like the holocaust. 6 million Jews were killed. If you believe in free will you have to think how odd it was that so many bad people were born at around the same time to carry out those terrible deeds. I think it's much more reasonable to think that each person was just a bunch of genes acted upon by their enviroment. Put you or I in the same situation it's likely we would have acted in the same way!

Now I'll deal with your much less convincing arguments for free will. The Danish pastry example. The philosopher only thinks he has the choice between having a Danish pastry or not in fact he always makes the choice he is bound to make at the time. Imagine he is on a diet. Once he's eaten the pastry he could think to himself,"I shouldn't have done that" or "I could have chosen not to eat it" but of course, at that moment in time he could do nothing else but eat it or else he would not have done so!

I wonder if Matt Ridley really thought he was quite capable of jumping in his car and driving to Edinburgh. Well so could I and so could anybody but Matt didn't drive to Edinburgh at that moment I'm not driving to Edinburgh now and I doubt if anybody else who reads this or Matts book is going to. So we are not free to do so are we? If anybody reads this and jumps in their car and drives to Edinburgh they will do it either because they were going to anyway or as a result of reading this, which would just be the interaction of their genes and the enviroment NOT FREE WILL!

I'm well aware I could be wrong. If you think I am and have a well reasoned logical argument opposing mine, I'd like to hear it. Or maybe you agree with me, either way please feel free to email me. If you've got a strong conviction that I'm wrong but you can't give any more reason than feelings based on your conditioning or your belief system or your religion then please don't bother me.I hope to read some of your comments.

Stephen Lawrence

griffis said...

Say we accept a thoroughly materialistic perspective whereby nothing exists except the physical matter and energy that make up the universe, then there would seem to be no free will in the sense of some force that is independent of the law-goverened behaviour of matter and energy. So, at first glance, if we accept the picture of the world that physics gives us, we must reject free will.

However, the physical universe has provided a substrate wherein the process of evolution can take place. Evolution has, ultimately, produced entities with autonomous control systems - most notably nervous systems. These entities are able to take in information from their surroundings, process that information, and then make decisions to act based on their own interests combined with the information from the environment.

This might still sound very deterministic - that's because it is. But it seems to be the best sort of free will you can get if you accept physics and reject the supernatural. We are free in the sense that we have autonomous control systems (although these systems are based on law-governed physical processes). We are more free than rocks, for example, because we have these control systems. Furthermore, there are gradations of freedom based on the sophistication of the control systems involved. Humans seem to have the most sophisticated control systems, i.e. our brains, and are therefore the most free organisms that we know of.

It's largely a matter of opinion whether you want to call this freedom or not. I probably wouldn't actually, but it's the best we're going to get.

Scott said...

The rules of physics are not as harsh on "free will" as it would seem. At the most basic theoretic level (quantum mechanics) and therefore at every level, you can always determine what was from what is, but you can't generally deduce what will be from what is. This is because certain states are indeterminate until "measured" - that is, until the state has a bearing on the macroscopic world. In this case, "indeterminate" needs to be taken in the most extreme philosophical sense - even an omnipotent god could not know the state because the information simply doesn't exist.

Perhaps the easiest way of contrasting this situation with determinism is to start by allowing that everything is deterministic. That is, there was an initial state of the universe and that everything that followed is of a direct predictable effect of previous causes. Then we add one this one wrinkle, the initial state is only approximate. As the universe ages, the initial conditions become more resolved. Making a "free will" decision is thus a strictly deterministic event which further resolves those initial conditions.

Another point I will dispute is the notion that conscious "free will" would make no Darwinian sense. In my estimation, early mammals developed a "consciousness" gene to employ the special computational powers of quantum mechanics to assist in survival decisions. The fundamental brain problem is, "Given what is sensed and remembered, what action would be most beneficial to my genetic survival". One approach to this problem is to produce a mental model of what is sensed, generate a candidate response, predict what the outcome of that response would be, and then rate the survival value of that predicted outcome. This algorithm might have been used as is to audit decisions in some early animals. However, it has a major problem, there are almost endless possible combinations of candidate responses to evaluate - nothing close to a thorough evaluation could ever be completed as the algorithm stands.

However, this is exactly the sort of problem where quantum mechanical data processing excels. In essense, the encoding of the candidate responses is left indeterminate and a response is chosen based on the simultaneous evaluation of all possible outcomes. This sort of computation is known to science, but is, for the most part, a bit beyond modern technology. The reason for thinking that the human brain can somehow employ quantum mechanics is simply that there is nothing in classical mechanics that allows more than a tiny amount of information to exist at the same time and place - not enough to create the sort of routine awareness we enjoy.

Anonymous said...

Just finished reading your site completely....LOVED it. I know you heard this before but....You are not saying anything the ancient mystics have not already told us centuries ago. Even the apostle Paul stated "Why doe's God still find fault with us when He created us thus." Want to buy your book but I only trust Amazon.

dan from said...

The number one reason why there is confusion over the existence of free will is that everyone means something different by the term!

The way I understand it, there is no paradox. Complete determinism anf freedom of will are best friends.

conscious robot said...

In Reply to Stephen:

I would say I broadly agree with everything you posted. The notion of free will as the beginning of a causal chain, or as some kind of indeterminate decision maker which gives meaning to our otherwise determined actions is rubbish as far as I am concerned.

The problem with this I always find (and people constantly point out) is we can only give evidence after the event. When I have my eaten my Danish I can reason back and forth but I can never go back and repeat the experience to test whether I could have stayed faithful to my diet a second time round.

This unfortunately makes it rather difficult to reason meaningfully from this standpoint.

conscious robot said...

In response to griffis:

I think you are referring to what I have always described as the ‘argument from complexity’. I have always been slightly tempted to accept this argument. I do get the distinct feeling that it is just obfuscating the issue, however, and not really changing anything.

The argument runs that the level of complexity in our various systems has a chaotic effect and because you cannot analyse it effectively (or so chaos theory would tell us) out of this chaos comes undetermined free will.

This always seems to ignore that fact that chaos theory is only telling us that we cannot create effective model’s not that the data does not exist at all. Behind or this chaotic atoms bumping about in my brain there could still be a determined path of ‘random’ collisions.

This aside, I’m confused as to what you mean by “there are gradations of freedom based on the sophistication of the control systems involved”, could you provide an example of this?

To my mind if I have even the slightest decision to make then I am already free of hard determinism. As mentioned above even the tiniest unpredictability in a system creates chaos on a vast scale.