Re: Strong AI Thesis (No Chinese room, I promise)


"JGCASEY" <jgkjcasey@xxxxxxxxxxxx> wrote in message
news:1144979237.016430.152610@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Allan C Cybulskie wrote:
"JGCASEY" <jgkjcasey@xxxxxxxxxxxx> wrote in message


So if the "soul" caused the choice as long as the "soul"
wasn't determined we'd be okay.


Essentially we are at an impasse here because I cannot
imagine something as "uncaused". If it happens for a
reason (which is a cause) then it is determined. I don't
have examples of things that "just happen" that I can
then show really did "just happen" without anything from
which it could happen.
...

Ultimately, though, you are going to have to, since you cannot go through an
infinite regress of causes. So something has to be uncaused. The issue is
figuring out what is uncaused, why it is uncaused, and how it is uncaused.


Think about why you made that less than ideal choice
and then tell me you didn't do what you wanted.

In cases I forget things that I thought of when I originally
decided what I should do. How is this doing what I wanted?

Nothing nondeterministic about that.

Did I claim it was, necessarily? I was merely refuting your "we always do
what we want" argument.



When one system is coupled to another system we can call
certain kinds of interactions "intentions".

We can? News to me. Intentions are NEVER described in
that manner.

Well I just did, so you can't say NEVER :-)

If a block moving robot moves the blue block off the top
of the red block and you ask why I might say its "intention"
was to make room for placing the yellow block on the red
block. To say something "intends" something is too explain
why it does something, it is all deterministic. Why did
I go into the kitchen. Because my intention is to cook a
meal.

But none of these are really coupling systems to other systems, necessarily.
And I did miss the "certain kinds of interactions" above, but that doesn't
make things clearer, since I imagine that we'd still call the same things
intentions (you and I). But then we need to know what are the defining
qualities of those interactions that make them intentions.

If the intention was there and they are not constrained
by other wants they do exactly what they want. If the
intention was not there it was either an accident or
they wanted to do it.

How can you claim that they wanted to do it if the intention
was not there? This is just completely out there.

I realised after posting that was written incorrectly.
I meant to write,

"If the intention was not there it was an accident. If the
intention was there it means they wanted to do it."

The problem is I often am thinking of something else while
my fingers type my previous thoughts and they get muddled.

S'allright [grin].

In the case of habits or conditioned behaviour, we don't
want to do it -- the intention is not there -- but the
conditioning takes over because we don't consciously
interrupt it. And that proves that it is not the case that
we always do what we want to do, making your translation
false, as well as unhelpful.

Perhaps then I should say "intention" is a "conscious want".
Just don't ask me what consciousness is :-)

Ah, so then we'd agree. There is a problem with this, however, since some
things that we call intentions don't seem to be conscious [grin].

My point about the circularity is that it seems like you
can only make that stick if you claim that the action you
did was the one you wanted, regardless of what your conscious
thoughts and analysis of the matter was. And that's circular,
because you define your wants in terms of what you do, which
means that you cannot say that you do what you want when you
claim that what you want is what you do.

Well I think your "conscious thoughts and analysis of the matter"
are determined by your "wants" which in this case the "wants" are
the basic physiological factors that motivate us. In simple terms
we seek pleasure and avoid pain in the broadest meanings of those
words.

Well, the problem is that if this was actually true then we'd never have a
clash between what we immediately prior thought we should do and what we
actually do. But this occurs fairly frequently.



There is some use in describing "do what you want" in terms of
a more unconstrained choice, but this must be used delicately
since a) we don't have ANY totally unconstrained choices and
b) it can contradict your own idea. For example, if I say that
I want to go and walk through the park instead of going to work,
but that I'll go to work because of other concerns -- like other
people's opinions -- then you'd want to say by your ideas that
I did what I wanted to do, but my choice was constrained by other
people. You can avoid this by reducing your scope to only those
actions that are indeed TOTALLY controlled by other people ...
but that limits you to direct "brainwashing" cases and gets you
in trouble with behaviourism since if you accept their premise
(that you seem to accept later) ALL of our actions are indirectly
controlled by other people and other things ... i.e. the environment.

Actually I resist the idea that the environment controls
anything as such.

Excellent. So we are now getting closer to being on the same page. This
idea is the main argument that materialists and behaviourists will be forced
into in dealing with free will.

It is very difficult to argue that we don't do "what
we want". But there is much evidence to suggest that
we can choose our wants. For example, when we prioritize
wants, this seems to be done "on the fly", where the
relative weights of wants can change in a short period
of time with no indicative change in the environment.


There may not be any indicative change in the environment
but there is clearly plenty of changes inside the head.

But these changes are not necessarily meaningful. You cannot
simply translate that change into the reason why we prioritize
as opposed to it simply being a reflection of prioritization.
Moreover, you have no evidence to suggest that the brain is
being changed -- even in terms of the weights of the links in
the brain as per connectionism -- on the fly in these cases.
And it does seem odd to suggest that it would be.

I would say that any change in behavior has its counterpart
change in the brain. It would be odd to me if it wasn't the
case.

Oh, I agree it would be odd for you to argue otherwise ... but note that you
don't really have any evidence to suggest that it is the case. Looking at
connectionism, for example, in order to reprioritize wants you'd have to
reshuffle the weights of the connections in the brain, but we have no
evidence to suggest that this is the case.

Perhaps ... or perhaps not. The problem is that you have
no real reason to assert that there are other genes involved
other than the fact that you want to insist that it's all
due to genes and the environment. But we have no evidence
of any major environmental difference that can explain the
difference in behaviour, which leaves genetics. But asserting
that there is a genetic difference that explains this is
indeed an unevidenced claim. What we know is that children of
alcoholics seem to have a genetic and environmental
predisposition towards becoming alcoholics. The free will
hypothesis is in accord with this evidence because it agrees
that those conditions will make children of alcoholics more
likely to be alcoholics, and they are. But your idea has a
hard time explaining why a significant number of those children
DON'T become alcoholics. You end up simply asserting that
there must be some reason why the predisposition fails in
those cases, but you cannot provide any particular explanation
other than asserting that there must be one. That makes your
theory suspect, to say the least.

People's brains are as different as their faces. The idea that
a single "alcoholic gene" requires a "mystical free will" to
explain its effect or lack of effect seems a stretch when a
simple physical explanation is that what we are a product of
all our genes interacting with the environment. At least it
is potentially testable which is a requirement for a scientific
theory as opposed to religious "theory" about "free will".

Um, you are going off-base here. Don't bring religion into this. The idea
is this: environment and genetics seem to lead to a predisposition for some
people to be alcoholics. In a significant number of cases, this
predisposition is resisted. The original proposition was that it was their
"willpower" that leads to them overcoming their predisposition. Your reply
is basically that there are other environmental or genetic factors that lead
to that situation. But you don't know what those factors are. So why
should anyone accept that there are any? All you are doing is asserting
that there is a case, even though you don't know what it is yet.

This doesn't make you wrong ... but it doesn't make the theory especially
credible either.



You are simply trying to assign "free will" to those
things we can't find causes for.

Well, if we COULD find the causes for it, there'd be no
room for free will, wouldn't we? [grin].

Of course there is a use for the phrase "free will". It means
that if something wills something and it is free (unconstrained
from doing what is wills) it has done something as an act of
free will. This separates the act of doing something you don't
do as an act of "free will" such as jump off a cliff because
you were pushed. Free will means you alone determine what
you do at that point in time.

But this is clearly not true in any case. I make no unconstrained
conditions, even if we get away from physically determined neural reactions
to the environment. So this isn't useful after all.

I recently read an attempt to do just this, and it fails.
The reason is that the right way to look at it is to ask if
it is possible to do anything else. But if there was a cause
in the past that determines the chain of events that lead to
my actions now, it is not possible to do anything else
... no matter how we semantically slice the term "possible".

It only fails if you stick to your other worldly idea of free will.

Nope. It fails because it seems reasonable to only care about it being
possible for me to do something else at the time the choice is made ... and
if a cause or set of causes in the past make it certain what choice I will
make at the time I make it, then it is not possible to choose anything else
even though you can make a semantical argument about how it is "possible".


If I am not locked in a room it is possible for me to exit,
it does not mean I will exit.

Well, my example OF this case deals with this: I would still
have free will even if the door WAS locked if I had no desire
to leave it.

Yes well I don't see free will as something you "have". If I
open your brain I don't think I will see any of this "free will"
stuff in there. You see "free will" in the singular. I see it
as two words, "free" (unconstrained), "will" (how something will
act if it can).

Whatever. Putting aside the terminology, could you address the issue?


.

Relevant Pages

  • Re: Strong AI Thesis (No Chinese room, I promise)
    ... of the red block and you ask why I might say its "intention" ... Actually I resist the idea that the environment controls ... specifically, the children of alcoholics. ... no real reason to assert that there are other genes involved ...
    (comp.ai.philosophy)
  • Re: Parents murdered daughter by prayer
    ... there is good reason for this as life's experiment daily demonstrates ... But of course we influence the environment around us. ... If you are that way by intention, then that intention must have been formed by the way you were at a previous time. ...
    (uk.religion.christian)
  • Re: Mvbase or D3 on Microsoft Win 2003 SBS
    ... knew that was a reason I don't hang out with any business people! ... Conversion of an existing Pick ... environment to Pick flavour UV is about the smallest "conversion" step ... that's what I/we think about for any migration effort. ...
    (comp.databases.pick)
  • Re: Hulls sentence doesnt seem to refer to Darwinian
    ... other life forms, ... in the face of environmental pressures, some organisms will survive at ... For whatever reason, some organisms will survive at higher rates than ... of that environment, ...
    (talk.origins)
  • Re: McAfee Virusscan: Generic.dx!sux - false positives?
    ... environment and then the malware engages. ... Sandboxes are just more ... there is no reason to move most software out of a sandboxed ...
    (alt.comp.anti-virus)