Free Will

John K Clark (
Wed, 25 Feb 1998 09:03:37 -0800 (PST)


On Wed, 25 Feb 1998 Damien Broderick <> Wrote:

>Freedom of choice for humans does not mean acting at random like
>the Dice Man; that would be psychosis, not freedom.

True, except that it wouldn't even be psychosis because even insanity has a
cause, it would just be convulsions. I agree with the previous poster who
said in effect that first we must define what free will is, then we can
debate if human beings have this interesting property or not.
I gave my definition:

A man, animal or machine has free will if it can not always predict what it
will do in the future even if the external environment is constant. A third
party might be able to make such predictions but that's irrelevant, the
important thing is that the person himself can not know what he will do next
until he actually does it.

I maintain that this definition is clear, internally consistent and produces
no contradictions when used with most everyday uses of the word "free".
I've never heard of another definition that could do as well.

>our consciousness or ego is a module (either executive or
>interpretative) with very restricted information about the full
>state of the self.

I agree.

>This means that when we opt, we do so from many more `unconscious'

I have no argument with Freud, there are factors in our mind that we aren't
conscious of, but there are other things that effect our behavior that aren't
even "unconscious", they are not in the brain at all because they have not
been calculated yet. I think this is intimately related to Alan Turing's
discovery that a computer program has free will, that is, it can't understand
itself, that is, it can't (we can't either) predict if it will ever stop.

Let me suggest a thought experiment; a man is walking down a road and spots a
fork in the road far ahead. He knows of advantages and disadvantages to both
paths so he isn't sure if he will go right or left, he hadn't decided. Now
imagine a powerful demon able to look into the man's head and quickly deduce
that he would eventually choose to go to the left. Meanwhile the man, whose
mind works much more slowly than the demon's, hasn't completed the thought
process yet. He might be saying to himself I haven't decided I'll have to
think about it, I'm free to go either way. From his point of view he is
correct, even a robot does not feel like a robot, but from the demon's
viewpoint it's a different matter, he simply deduced a purely mechanical
operation that can have only one outcome.

But is it really a purely mechanical operation, what about the uncertainty
principal? I don't see how it effects matters one way or another. It says
that some things can happen for no cause and thus are truly random, but
happenstance is the very opposite of intelligence and even emotion. Things
either happen because of cause and effect or they don't and if they don't
then they are by definition random and have nothing to due with volition.
Those who claim that this is the source of the will must also believe that a
nickel has free will when you flip it. This topic muddies the question but
does not change it.

In my example the demon did not tell the man of his prediction but now lets
pretend he did. Suppose also that the man, being of an argumentative nature,
was determined to do the exact opposite of what the demon predicted. Now our
poor demon would be in a familiar predicament. Because the demon's decision
influences the man's actions the demon must forecast his own behavior, but he
will have no better luck in this regard than the man did and for the same
reasons. What we would need in a situation like this is a mega-demon able to
look into the demon's head. Now the mega-demon would have the problem.

>I think that the more we make our values `our own', the more free we
>feel - that is, against your claim that we would feel like robots.

My claim was that self knowledge would make us feel like robots but we don't
have to worry about it because self knowledge is impossible, I don't see what
that has to do with deciding we like a value and adopting it.

Let's simplify things to their essentials. Imagine a world in which the
environment was so simple it could be predicted with complete accuracy.
Doubtless we would find such a place boring and unpleasant but I don't think
we would feel like robots. Thus the origin of the sensation of autonomy can
not be external. What does it mean to "feel like a robot"? If you could
always forecast your own behavior and thoughts with complete accuracy then
you would feel like a robot. Subjective (not objective) uncertainty is at
the root of freedom and choice.

>John Clark wrote at random, that is, freely (by his own account):

I like to think it wasn't random but it was certainly free by my definition.
When I read your post there were several ways I could have responded to it,
when I sat down at my computer I had all the information I needed but I didn't
know what I would do, I could have written many things, I had to think about
it, I was free, in fact, I wasn't absolutely sure how it would come out until
right NOW.

John K Clark

Version: 2.6.i