Consiousness vs. Free Will
Keeva Speyer
Can there be consciousness without (possibly illusory) free will - if
only the apparent ability to control our attention and thoughts?
Obviously there are many types/definitions of consciousness but I'm
talking pragmatically from an AI point of view.
Passive reception and automatic responses to information doesn't
intuitively seem conscious to me.
If a machine had enough apparent free will (whatever that is, and all
that entails like ability to control at least something) and
attention, could it be sensibly called unconscious??
I know this doesn't help that much as even if it has merit it just
moves the question to free-will but at least it's another angle. (I'm
not trying to imply some non-determanistic model of consciousness,
after all we have apparent free will and in my view are probably
deterministic creatures).
biffy
"Keeva Speyer" <ke...@zip.com.au> wrote in message
news:87drats816vnsobu4...@4ax.com...
Ludovico Zaraga
Probably everyone has their own definition of "free will", so I shall try
to make clear what I mean for "free will" and show that it is strictly
connected to consciousness.
"Free will" is a way of making choices which occurs in a
non-deterministic way. If you believe that we live in a completely
mechanicistic world (which I don't believe) then there can be logically
no space for free will, it couldn't make sense. However in a purely
mechanicistic context even consciousness wouldn't make sense either.
There are cases in which we make decisions with insufficent data to work
out their consequences. Free will is the ability to act in these
situations. In fact if we knew everything we would know exactly how to
act in any situation and there would be no such thing as making
decisions.
The state in which we are which actively involves the continuous making
decisions, thanks to our free will, is consciousness.
Therefore if we assume consciousness to exsist, which we can assert only
if we are in a non-mechanicistic world, then its existence is based on
free will.
Ludovico Zaraga
Raphael Jolly
I am in total agreement. I think consciousness would'nt be possible if the
nature of reality was algorithmic.
David Mitchell
<zar...@tiscalinet.it> writes
>I firmly believe that consciousness needs free will to exisit.
>Probably everyone has their own definition of "free will", so I shall try
>to make clear what I mean for "free will" and show that it is strictly
>connected to consciousness.
>"Free will" is a way of making choices which occurs in a
>non-deterministic way. If you believe that we live in a completely
>mechanicistic world (which I don't believe) then there can be logically
>no space for free will, it couldn't make sense. However in a purely
>mechanicistic context even consciousness wouldn't make sense either.
Why?
>There are cases in which we make decisions with insufficent data to work
>out their consequences. Free will is the ability to act in these
>situations. In fact if we knew everything we would know exactly how to
>act in any situation and there would be no such thing as making
>decisions.
I think you're confusing "free-will" with the process of making
decisions without full knowledge.
>The state in which we are which actively involves the continuous making
>decisions, thanks to our free will, is consciousness.
You don't need "free-will" to make "Decisions". Computers can make
decisions, would you say they have "free-will".
>
>Therefore if we assume consciousness to exsist, which we can assert only
>if we are in a non-mechanicistic world, then its existence is based on
>free will.
>
Untrue. See above.
BTW, the reason I have put "free-will" in quotes is because unlike you I
don't believe in it. Consider the following thought experiment...
You're about to make a decision, I magically take a complete copy of
your entire body and environment (it's a thought experiment, I can *do*
this ;-). You make the decision, so does your copy.
Two possibilities: either the copy makes the same choice or it doesn't.
If it doesn't, what's different between the two (remember they're
identical right down to atoms).
If its the same - what's free about it?
My point is this: every decision you make is either due to the pre-
existing conditions in your brain (i.e.. what neurones are doing what),
or it's due to random noise.
In neither case can the decision making process be considered free in
any meaningful sense - "free will" is just an illusion.
--
==========================================================================
David Mitchell ===== Visit: www.thehungersite.com
================================
da...@edenroad.demon.co.uk ===== Feed someone for nothing.
==========================================================================
Raphael Jolly
> In article <3AAE9F09...@tiscalinet.it>, Ludovico Zaraga
> <zar...@tiscalinet.it> writes
> >I firmly believe that consciousness needs free will to exisit.
> >Probably everyone has their own definition of "free will", so I shall try
> >to make clear what I mean for "free will" and show that it is strictly
> >connected to consciousness.
> >"Free will" is a way of making choices which occurs in a
> >non-deterministic way. If you believe that we live in a completely
> >mechanicistic world (which I don't believe) then there can be logically
> >no space for free will, it couldn't make sense. However in a purely
> >mechanicistic context even consciousness wouldn't make sense either.
>
> Why?
Let me try an answer...
> >There are cases in which we make decisions with insufficent data to work
> >out their consequences. Free will is the ability to act in these
> >situations. In fact if we knew everything we would know exactly how to
> >act in any situation and there would be no such thing as making
> >decisions.
>
> I think you're confusing "free-will" with the process of making
> decisions without full knowledge.
>
> >The state in which we are which actively involves the continuous making
> >decisions, thanks to our free will, is consciousness.
>
> You don't need "free-will" to make "Decisions". Computers can make
> decisions, would you say they have "free-will".
Computers, yes. The algorithmic process they are executing, no.
> >
> >Therefore if we assume consciousness to exsist, which we can assert only
> >if we are in a non-mechanicistic world, then its existence is based on
> >free will.
> >
>
> Untrue. See above.
>
> BTW, the reason I have put "free-will" in quotes is because unlike you I
> don't believe in it. Consider the following thought experiment...
>
> You're about to make a decision, I magically take a complete copy of
> your entire body and environment (it's a thought experiment, I can *do*
> this ;-).
Not according to Heisenberg's principle.
> You make the decision, so does your copy.
>
> Two possibilities: either the copy makes the same choice or it doesn't.
>
> If it doesn't, what's different between the two (remember they're
> identical right down to atoms).
>
> If its the same - what's free about it?
>
> My point is this: every decision you make is either due to the pre-
> existing conditions in your brain (i.e.. what neurones are doing what),
> or it's due to random noise.
What if random noise wasn't a mechanical process ? (e.g. wasn't predictable,
even in principle)
Neil W Rickert
>You don't need "free-will" to make "Decisions". Computers can make
>decisions, would you say they have "free-will".
Computers don't make decisions. They just follow the rules given to
them by the programmer.
>BTW, the reason I have put "free-will" in quotes is because unlike you I
>don't believe in it.
I take it that I should assume that forces beyond your control
manipulated your muscles, so as to produce the above line. Since you
evidently deny any responsibility for having produced that line, I
see no reason to take it seriously.
Daryl McCullough
>
>David Mitchell <da...@edenroad.demon.co.uk> writes:
>
>>You don't need "free-will" to make "Decisions". Computers can make
>>decisions, would you say they have "free-will".
>
>Computers don't make decisions. They just follow the rules given to
>them by the programmer.
You are just being silly. Computers don't "follow rules". They just
act in a way that is describable by rules.
>>BTW, the reason I have put "free-will" in quotes is because unlike you I
>>don't believe in it.
>
>I take it that I should assume that forces beyond your control
>manipulated your muscles, so as to produce the above line. Since you
>evidently deny any responsibility for having produced that line, I
>see no reason to take it seriously.
What a silly response. What does "free will" have to do with
whether you should take his arguments seriously? Perhaps those
"forces beyond his control" constrain him so that he only makes
sensible statements. Perhaps someone with "free will" uses that
free will to make nonsensical statements.
There is no necessary correlation that I can see between free
will and the reasonableness of their statements.
--
Daryl McCullough
CoGenTex, Inc.
Ithaca, NY
Ludovico Zaraga
> You're about to make a decision, I magically take a complete copy of
> your entire body and environment (it's a thought experiment, I can *do*
> this ;-). You make the decision, so does your copy.
> Two possibilities: either the copy makes the same choice or it doesn't.
> If it doesn't, what's different between the two (remember they're
> identical right down to atoms).
> If its the same - what's free about it?
Interesting, but not at all convincing.
First of all you must know that determinism is just a peculiarity of a
model of the world's functioning that humans have created. You can't prove a
model to be undeniably true, you can just observe that it works in certain
situations. The only thing you can really prove of a model is it to be false.
Consider the following example: Before Einstein's theory of relativity our
model of space was an Euclidean one. Everybody thought that reality was
exactly how Newton had descrived it and they were sure about that because it
was "scientificaly proven" at that time by numerous experiments. After that
the theory of relativity came out it was understood that classic physics was
just a model of the world.
A model is extremely useful in certain circumstances. The model of classic
physics is still used when dealing with speeds which are much smaller than the
speed of light and doesn't give us any problems in these circumstances. This
model describes just a part of the universe. When it comes to talk about speed
close o the speed of light this model can't be used anymore.
The same thing must be said about your deterministic model. If it is
correct to use in certain circumstances it doesn't mean that it explains all
the circumstances in the world. Furthermore you can't prove your model to be
correct with an experiment, you can just prove it to be correct for THAT
PARTICULAR experiment. You would need an infinite number of experiments to
prove the model to be undeniably true.
Let us come right now to your experiment.
>My point is this: every decision you make is either due to the pre-
>existing conditions in your brain (i.e.. what neurones are doing what),
>or it's due to random noise.
It is not very clear if these are the premises or the conclusions of your
experiment.
If your experiment must prove either determinsm or free will to exist then you
can't assert them to exist somewhere in your premises. Keeping his in mind
let's see what happens:
Case A) You make the same choice.
Great! Let us assume that this means determinsm exists.
Case B) You don't make the same choice.
This seems to be quite a problem.... From your premises you can't assume
that this case would prove determinism not to exist!
This experiment has no sense.
Any experiment wouldn't have any sense to prove determininsm to exist.
Determinism is just a model of the universe which has success is certain
feilds and is doomed to failure in others. If you are in need of an example
just take a look at how quantum mechanics works.
Ludovico Zaraga
Neil W Rickert
>>David Mitchell <da...@edenroad.demon.co.uk> writes:
>>>You don't need "free-will" to make "Decisions". Computers can make
>>>decisions, would you say they have "free-will".
>>Computers don't make decisions. They just follow the rules given to
>>them by the programmer.
>You are just being silly. Computers don't "follow rules". They just
>act in a way that is describable by rules.
I presume, then, that if we change the program (the rules), then the
computer keeps acting the same way as before, but new replacement
rules will no longer accurately describe how the computer acts.
I sure wonder what you mean by "computer".
>>>BTW, the reason I have put "free-will" in quotes is because unlike you I
>>>don't believe in it.
>>I take it that I should assume that forces beyond your control
>>manipulated your muscles, so as to produce the above line. Since you
>>evidently deny any responsibility for having produced that line, I
>>see no reason to take it seriously.
>What a silly response. What does "free will" have to do with
>whether you should take his arguments seriously?
Just about everything.
Almoste the entire tradition of philosophical discussion of free will
has connected it with responsibility. We usually do not take people
seriously when they disclaim responsibility for their speech.
Jim Balter
>
> Neil says...
> >
> >David Mitchell <da...@edenroad.demon.co.uk> writes:
> >
> >>You don't need "free-will" to make "Decisions". Computers can make
> >>decisions, would you say they have "free-will".
> >
> >Computers don't make decisions. They just follow the rules given to
> >them by the programmer.
This sort of ad hoc semantics is right out of the vitalist tradition
a la Searle. It doesn't take much examination to see that there
is a false dichotomy here, borne by that reason-destroying word "just"
-- something could both follow rules and make decisions. I myself
do it all the time, as do computers via decision procedures. Only someone
in the grips of an ideology would deny obvious application of the
natural language word "decision" to what computers do and are commonly
described as doing. This is usually Weinstein's sin -- it's bizarre,
in the context of years of stagnant debate in c.a.p, to see you
take his line. Perhaps it's not stagnant -- he seems to have won
a convert, sigh.
> You are just being silly. Computers don't "follow rules". They just
> act in a way that is describable by rules.
And not even that in practice, just as with human or other complex
behavior. The code in a computer is no more a full description
of the behavior of a computer than the genome map or even a complete
understanding of biochemistry is of living entities, a point that Neil
appreciated in regard to dinosaur DNA.
> >>BTW, the reason I have put "free-will" in quotes is because unlike you I
> >>don't believe in it.
> >
> >I take it that I should assume that forces beyond your control
> >manipulated your muscles, so as to produce the above line. Since you
> >evidently deny any responsibility for having produced that line, I
> >see no reason to take it seriously.
>
> What a silly response. What does "free will" have to do with
> whether you should take his arguments seriously? Perhaps those
> "forces beyond his control" constrain him so that he only makes
> sensible statements. Perhaps someone with "free will" uses that
> free will to make nonsensical statements.
>
> There is no necessary correlation that I can see between free
> will and the reasonableness of their statements.
Ah, but if there is no free will then Neil isn't
free as to what he does or doesn't take seriously. One can get
nowhere with this debate -- as Dennett has pointed out, as soon as
one takes the intentional stance, one has already implicitly assumed
free will -- the level of description that uses words such as
"I" and "believe" and "reason" (and "silly") has at its core the notion
of rational agency. One can theoretically talk at a free-will-devoid level,
but it must be done consistently, and it is practically impossible
and pragmatically pointless to try to do so when talking about those things
toward which it is in fact most effective to take the intentional stance,
such as humans -- and computers, when discussing their behavior in regard to
its consequences for us (rather than its implications about the state or
nature of the computer itself -- and the same applies to humans).
--
<J Q B>
Neil W Rickert
>> Neil says...
>> >Computers don't make decisions. They just follow the rules given to
>> >them by the programmer.
>This sort of ad hoc semantics is right out of the vitalist tradition
>a la Searle.
There is nothing vitalist about it.
> It doesn't take much examination to see that there
>is a false dichotomy here, borne by that reason-destroying word "just"
>-- something could both follow rules and make decisions.
The next time Windows crashes on your computer, I expect you to file
a law suit charging your computer with negligence, instead of griping
about Bill Gates.
There was no false dichotomy in what I said. The fact is, we blame
the programmer, not the computer, when there is a failure.
> The code in a computer is no more a full description
>of the behavior of a computer than the genome map or even a complete
>understanding of biochemistry is of living entities, a point that Neil
>appreciated in regard to dinosaur DNA.
If you really believe that the role of the DNA in determining
dinosaur behavior is the exact analog of the role of the program in
determining computer behavior, then you are either very ignorant of
biology or very ignorant about computing.
Jim Balter
>
> da...@cogentex.com (Daryl McCullough) writes:
>
> >>David Mitchell <da...@edenroad.demon.co.uk> writes:
>
> >>>You don't need "free-will" to make "Decisions". Computers can make
> >>>decisions, would you say they have "free-will".
>
> >>Computers don't make decisions. They just follow the rules given to
> >>them by the programmer.
>
> >You are just being silly. Computers don't "follow rules". They just
> >act in a way that is describable by rules.
You guy are both entertaining a false dichotomy. A computer can be
said to follow rules; it can also be said to have behavior describable
by rules. In either case, one must be careful about what is held
to follow from such characterizations.
> I presume, then, that if we change the program (the rules), then the
> computer keeps acting the same way as before,
Sure, if we never make the computer run the new program.
But otherwise, why would you presume such a thing?
This absurd strawman has nothing to do with what Daryl wrote.
> but new replacement
> rules will no longer accurately describe how the computer acts.
The new program will, if properly implemented by the computer,
describe the behavior of the computer. That's what implementation
*means*. Getting a computer to follow a program is not simply a
matter of sticking a listing inside the cabinet. These days,
most of us get computers to run programs by exercising an
elaborate series of finger movements. Just how it is that
that causes the computer to end up in such a state that it
behaves in a certain way is a very elaborate story, and cannot
be captured adequately by "they just follow rules".
> I sure wonder what you mean by "computer".
Computers are made out of silicon and such, not "rules", so the
question goes to *you*. An algorithm doesn't make decisions, or follow
rules -- it is itself a rule. Computers aren't algorithms, they are
*embodied* (recall arguments against Weinstein) -- they *implement*
programs, they aren't coincident with them.
We design computers in such a way that we can describe them as following
rules. We write a set of rules and, through considerable technological
effort, we arrange the state of the computer in such a way that its
behavior becomes consistent with those rules. But this seems too subtle
for Platonists such as Weinstein, Searle, and apparently, you.
> >>>BTW, the reason I have put "free-will" in quotes is because unlike you I
> >>>don't believe in it.
>
> >>I take it that I should assume that forces beyond your control
> >>manipulated your muscles, so as to produce the above line. Since you
> >>evidently deny any responsibility for having produced that line, I
> >>see no reason to take it seriously.
>
> >What a silly response. What does "free will" have to do with
> >whether you should take his arguments seriously?
>
> Just about everything.
>
> Almoste the entire tradition of philosophical discussion of free will
> has connected it with responsibility. We usually do not take people
> seriously when they disclaim responsibility for their speech.
Has Weinstein stolen your keyboard? No one in the real world ever
makes such disclaimers, and this "usually" claim is baseless
ad hoc silliness. I take what people *say* seriously if what they *say*
merits it, regardless of their philosophical positions toward free will.
If *you* think that the fellow is making an error in regard to
whether he has free will, then you can hardly take his *erroneous*
(in your view) claim to lack responsibility as a reason to
disregard what he says. As for the claim that he is denying
responsibility, that's absurd, and you've got the import
of the tradition all wrong. We want to attach responsibility,
and the concept of free will is a (confused, many claim) outgrowth
of that. The confusion becomes apparent when all the reasons
for wanting to attach responsibility remain but applicability
of free will becomes questionable -- there is a continuum
from people with "normal" capabilities to people with
diminished capacity to animals to inanimate objects. All we
should really care about is causality, but the notions of
free will and responsibility are ancient and precede the causality
worldview of the enlightenment, as one can see from attitudes about,
say, the purpose of "penal"/"correctional" institutions.
BTW, it is blatantly dishonest to snip Daryl's justification for his
question:
> Perhaps those
> "forces beyond his control" constrain him so that he only makes
> sensible statements. Perhaps someone with "free will" uses that
> free will to make nonsensical statements.
This points out that your argument is ad hominem -- specifically,
poisoning of the well. How convenient that someone who holds
a certain position you disagree with is automatically disbarred from
presenting arguments for that position. That's so much easier
than actually thinking through a rebuttal. I am struck by how often
I agree with your position but am dismayed by your methods of
justification.
--
<J Q B>
Anders N Weinstein
Daryl McCullough <da...@cogentex.com> wrote:
>Neil says...
>>
>>>BTW, the reason I have put "free-will" in quotes is because unlike you I
>>>don't believe in it.
>>
>>I take it that I should assume that forces beyond your control
>>manipulated your muscles, so as to produce the above line. Since you
>>evidently deny any responsibility for having produced that line, I
>>see no reason to take it seriously.
>
>What a silly response. What does "free will" have to do with
>whether you should take his arguments seriously? Perhaps those
>"forces beyond his control" constrain him so that he only makes
>sensible statements. Perhaps someone with "free will" uses that
>free will to make nonsensical statements.
>
>There is no necessary correlation that I can see between free
>will and the reasonableness of their statements.
You are supposing it is wholly unproblematic that meaningful statements
are being made, quite apart from their being made by a responsible
agent who stands accountable for making and supporting them. But this
is far from clear.
It is correct that we can interpret the output of artifacts like a
computer theorem prover as meaningful statements in our language. Such a
thing might turn up interesting proofs or conjectures when interpreted by us.
And to assess these conjectures for "reasonableness" does not depend on our
viewing the artifact having free will or understanding or performing any
speech acts of asserting anything at all.
But that is because we can interpret them as meaningful. And this
possbility would seem to be parasitic on the existence of a meaningful
language used by agents who understand it and use it responsibly.
Jim Balter
>
> Jim Balter <j...@digisle.net> writes:
> >> Neil says...
>
> >> >Computers don't make decisions. They just follow the rules given to
> >> >them by the programmer.
>
> >This sort of ad hoc semantics is right out of the vitalist tradition
> >a la Searle.
>
> There is nothing vitalist about it.
Neil, why do you *so transparently* argue against strawmen?
It's the *ad hoc semantics* I was addressing.
> > It doesn't take much examination to see that there
> >is a false dichotomy here, borne by that reason-destroying word "just"
> >-- something could both follow rules and make decisions.
>
> The next time Windows crashes on your computer, I expect you to file
> a law suit charging your computer with negligence, instead of griping
> about Bill Gates.
You, OTOH, are likely to pointlessly bitch about Bill Gates instead
of replacing your flaky memory card.
> There was no false dichotomy in what I said. The fact is, we blame
> the programmer, not the computer, when there is a failure.
Non sequitur.
Since you have switched the subject to the cause of failure, I will
point out that it has been well understood since at least Aristotle
that events have multiple causes -- arguments about whether
guns kill people or people kill people depend on a false dichotomy.
(From your style of argument, I half-expect you to protest that
you aren't a member of the NRA.)
The claim that there was no false dichotomy displays very shallow
thinking.
> > The code in a computer is no more a full description
> >of the behavior of a computer than the genome map or even a complete
> >understanding of biochemistry is of living entities, a point that Neil
> >appreciated in regard to dinosaur DNA.
>
> If you really believe that the role of the DNA in determining
> dinosaur behavior is the exact analog of the role of the program in
> determining computer behavior, then you are either very ignorant of
> biology or very ignorant about computing.
Strawman.
Once again, as I have learned over the years, any attempt at
debate with you is pointless. I think you are actually quicker
to exercise bad faith methods than you used to be.
--
<J Q B>
Anders N Weinstein
Jim Balter <j...@digisle.net> wrote:
>Daryl McCullough wrote:
>>
>> Neil says...
>> >
>> >David Mitchell <da...@edenroad.demon.co.uk> writes:
>> >
>> >>You don't need "free-will" to make "Decisions". Computers can make
>> >>decisions, would you say they have "free-will".
>> >
>> >Computers don't make decisions. They just follow the rules given to
>> >them by the programmer.
>
>This sort of ad hoc semantics is right out of the vitalist tradition
>a la Searle. It doesn't take much examination to see that there
In the Kantian tradition, freedom is obedience to a law one gives
oneself (autonomy = self-legislation). I think Rickert is only suggesting
something similar, that free systems are those that follow rules of
their own construction, rather than conforming to rules imposed on them
from without.
Anders N Weinstein
Neil W Rickert <ricke...@cs.niu.edu> wrote:
>Jim Balter <j...@digisle.net> writes:
>>> Neil says...
>
>>> >Computers don't make decisions. They just follow the rules given to
>>> >them by the programmer.
>
>>is a false dichotomy here, borne by that reason-destroying word "just"
>>-- something could both follow rules and make decisions.
>
>The next time Windows crashes on your computer, I expect you to file
>a law suit charging your computer with negligence, instead of griping
>about Bill Gates.
>
>There was no false dichotomy in what I said. The fact is, we blame
>the programmer, not the computer, when there is a failure.
Indeed. It seems to me the interesting question is: what if anything
would have to be the case before it would be appropriate to blame the
computer and not the programmer?
Jim Balter
Rickert is playing the role of linguistic fascist that you so love.
No wonder he would get rare support from you in this instance.
I say that computers make decisions, and I am fully in accord with
prevailing social practice in doing so.
--
<J Q B>
Jim Balter
>
> In article <98omtf$g...@euclid.cs.niu.edu>,
> Neil W Rickert <ricke...@cs.niu.edu> wrote:
> >Jim Balter <j...@digisle.net> writes:
> >>> Neil says...
> >
> >>> >Computers don't make decisions. They just follow the rules given to
> >>> >them by the programmer.
> >
> >> It doesn't take much examination to see that there
> >>is a false dichotomy here, borne by that reason-destroying word "just"
> >>-- something could both follow rules and make decisions.
> >
> >The next time Windows crashes on your computer, I expect you to file
> >a law suit charging your computer with negligence, instead of griping
> >about Bill Gates.
> >
> >There was no false dichotomy in what I said. The fact is, we blame
> >the programmer, not the computer, when there is a failure.
>
> Indeed.
Ah, now there's a compelling argument!
> It seems to me the interesting question is: what if anything
> would have to be the case before it would be appropriate to blame the
> computer and not the programmer?
First, I would note that the original statement was about
making decisions, which people are much more ready (as an
observed fact) to attribute to computers than assigning
blame. The shift here is a standard ill faith move.
Even so, as a programmer, I blame the software when it makes a mistake,
and I go try to find out what is wrong with it and fix it.
It would be quite ineffective for me to blame the programmer --
I often can't even identify the programmer, who may in fact
be a whole set of programmers. It might even be a matter
of someone installing the wrong patch but having good reason
to think it is the right patch -- credit imperfect communication,
or a timing race condition a la Einstein's overthrow of
simultaneity.
Of course, all the same issues come up in regard to assigning
blame where humans are the final agents. Was it Dan White, or
twinkies? Do guns kill people, or people kill people?
Is it cigarettes, or smokers, or addiction to nicotine,
or cigarette manufacturers, or Congressmen who legislate
tobacco subsidies, or misleading ballots, or voters misled
to vote for these Congressmen, who are to blame for the cost to
society of smoking? Was Oedipus responsible for screwing his
mother? Was Oedipus responsible for screwing Jocasta?
Was his fate a just punishment?
The amazing thing here that the issues of the indeterminacy
of causation are very well known, and have been very well known
for a very long time, yet you two bozos go on about "there was no
false dichotomy", and base years of your behavior on ideological
commitments that are patently absurd.
--
<J Q B>
Daryl McCullough
>
>Daryl McCullough wrote:
>> You are just being silly. Computers don't "follow rules". They just
>> act in a way that is describable by rules.
>
>And not even that in practice, just as with human or other complex
>behavior.
Yes. I really didn't mean to be objecting to the use of the
phrase "computers follow rules". I was only pointing out that
it is an anthropomorphism, just like "computers make decisions".
I don't have any objections to either one, but I thought that
Neil was being inconsistent in rejecting "decision making"
in favor of "rule following".
Okay, I can accept "free will" as a description of a stance that
we take towards a system. But that contradicts claims along the lines
"A computer has no free will, because everything it does is determined
by its program."
Rick Craik
"Anders N Weinstein" <ande...@pitt.edu> wrote in message
news:98osj5$jv6$1...@usenet01.srv.cis.pitt.edu...
Good question. But I don't think we would be calling them computers
and programmers anymore. Perhaps such a case would develop when
devices have principles of behaviour that have to be trained properly.
Regards,
Rick
Pierre-Normand Houle
"Jim Balter" <j...@digisle.net> wrote in message news:3AAFFAF8...@digisle.net...
This is a different sense of "blame". It is the sense in which you can sometimes
blame bad weather. I did a google search with "the computer decided" and
only about one in ten results had the form of an intentional attribution. The rest
were blames for failure, crashes or buggy behaviors.
> and I go try to find out what is wrong with it and fix it.
Whereas you would not attempt to "fix" somebody who made
a bad decision. You would try to reason or convince a person.
> It would be quite ineffective for me to blame the programmer --
> I often can't even identify the programmer, who may in fact
> be a whole set of programmers.
But the question is whether the intentional stance is appropriate
not whether it is useful. It can be a fact that a programmer is
responsible even though you are in no position to blame him.
And it can be a fact that nobody is responsible at all without
this being a reason for "blaming" the computer in the sense of
attributing it responsibility and autonomy.
> It might even be a matter
> of someone installing the wrong patch but having good reason
> to think it is the right patch
This would then be a case where the erratic result is not
a decision of anybody or anything although the decision
of a human being can be at the origin (in the sense of its
being one cause) of the result.
Jim Balter
>
> In article <3AAFD729...@digisle.net>, Jim says...
> >
> >Daryl McCullough wrote:
>
> >> You are just being silly. Computers don't "follow rules". They just
> >> act in a way that is describable by rules.
> >
> >And not even that in practice, just as with human or other complex
> >behavior.
>
> Yes. I really didn't mean to be objecting to the use of the
> phrase "computers follow rules". I was only pointing out that
> it is an anthropomorphism, just like "computers make decisions".
As I don't see decision making or rule following to be particularly
"human shaped". "anthropomorphism" is better used in cases where
motives and causes are misascribed. But describing a computer
as making a decision isn't of that sort. Or, at least, to claim that
it is must be argued for; it cannot be established simply by fiat.
> I don't have any objections to either one, but I thought that
> Neil was being inconsistent in rejecting "decision making"
> in favor of "rule following".
Yes, there are several reasons to object.
Of course; it's a bad inference, based on a false dichotomy.
--
<J Q B>
Anders N Weinstein
There is no doubt that this practice exists. The question is whether
this acknowledged use of "makes a decision" has the same significance
applied to a computer that it does to a person.
I suggest there are crucial consequences that attach to the latter usage that
are not found in the former. If I tell you John made such and such a decision
you will treat him very differently than if I tell you the elevator control
system made a decision to change directions. These mark the former usage as
what I think common intution recognizes it as: a metaphorical usage that is not
literally true.
It is not ultimately fascistic prescriptions of usage but pointed
descriptions of concepts I am after. This requires helping someone who
is confused for ideological reasons to recognize usages and
distinctions they already make in their other moments -- a kind of therapy.
Anders N Weinstein
Anders N Weinstein <ande...@pitt.edu> wrote:
> If I tell you John made such and such a decision
>you will treat him very differently than if I tell you the elevator control
>system made a decision to change directions.
I meant: you will treat John very differently than you will treat the elevator
with its control software.
Jim Balter
No, it's not, at least not for me -- for me, blame applies to causes,
rather than being the sort of free-floating emotional "I want to
smack the one to blame" medieval anti-intellectual retributional
primitive reaction that is the alternative.
> It is the sense in which you can sometimes
> blame bad weather. I did a google search with "the computer decided" and
> only about one in ten results had the form of an intentional attribution. The rest
> were blames for failure, crashes or buggy behaviors.
>
> > and I go try to find out what is wrong with it and fix it.
>
> Whereas you would not attempt to "fix" somebody who made
> a bad decision. You would try to reason or convince a person.
In fact we do try to fix people, through training, education,
medication, surgery, etc. Reasoning is another sort of fix --
demonstrate an inconsistency and let the person use their
anti-inconsistency module to correct it. It's a non-invasive
option. Certain sophisticated programs are also subject to that
sort of correction.
> > It would be quite ineffective for me to blame the programmer --
> > I often can't even identify the programmer, who may in fact
> > be a whole set of programmers.
>
> But the question is whether the intentional stance is appropriate
> not whether it is useful.
No, that's not the question. Have you read Dennett at all?
> It can be a fact that a programmer is
> responsible even though you are in no position to blame him.
> And it can be a fact that nobody is responsible at all without
> this being a reason for "blaming" the computer in the sense of
> attributing it responsibility and autonomy.
This just assumes the blatantly erroneous notion that there are
single causes or responsibilities or sources of blame etc.
As I have already point out, all we should care about is *cause*;
the conception of "autonomy" is medieval, and there are thousands
of threads of evidence and thought, in sociology, physics,
biochemistry, psychology, neuroscience, cognitive science,
the law, etc. that argue against the notion of autonomy as
an *attribute*. The question is the usefulness of the
intentional stance, *not* whether it is "appropriate".
It is always appropriate to apply concepts where they are
useful, philosophical arrogance notwithstanding.
> > It might even be a matter
> > of someone installing the wrong patch but having good reason
> > to think it is the right patch
>
> This would then be a case where the erratic result is not
> a decision of anybody or anything although the decision
> of a human being can be at the origin (in the sense of its
> being one cause) of the result.
There are all sorts of cases. The sum total of them strongly
argues against the sort of ad hoc application of "blame",
"responsibility", etc. that you and others champion
to support your arbitrary claims as to who or what can be responsible
or blamed.
--
<J Q B>
Jim Balter
>
> In article <3AAFF5DB...@digisle.net>,
> Jim Balter <j...@digisle.net> wrote:
> >
> >Rickert is playing the role of linguistic fascist that you so love.
> >No wonder he would get rare support from you in this instance.
> >
> >I say that computers make decisions, and I am fully in accord with
> >prevailing social practice in doing so.
>
> There is no doubt that this practice exists. The question is whether
> this acknowledged use of "makes a decision" has the same significance
> applied to a computer that it does to a person.
>
> I suggest there are crucial consequences that attach to the latter usage that
> are not found in the former. If I tell you John made such and such a decision
> you will treat him very differently than if I tell you the elevator control
> system made a decision to change directions.
If so, it is not for reasons relevant to this discussion.
You erroneously assume that all reasoning applies equally well
to everything -- as Aaron Sloman so often complained, some != all.
When my cat decides to wake me up at 3am, I treat her very differently
than when my girlfriend does -- this is not reason to think that my
cat is not "literally" deciding -- my cat *is* "literally" deciding
according to the *literal* meanings of the words! Rather, it indicates
that my cat and my girlfriend differ in ways relevant to my changing
their behavior -- arguably a quite different thing. But you don't
want to argue it, you want to establish your conclusion by 1984-style
fiat as to where words can be applied "literally".
> These mark the former usage as
> what I think common intution recognizes it as: a metaphorical usage that is not
> literally true.
I have argued before that this use of "metaphorical" is erroneous.
You have never rebutted those arguments. So it goes, year after year.
> It is not ultimately fascistic prescriptions of usage but pointed
> descriptions of concepts I am after. This requires helping someone who
> is confused for ideological reasons to recognize usages and
> distinctions they already make in their other moments -- a kind of therapy.
What sort of therapy would correct someone under the delusion
that certain uses of "makes a decision" are "literally" true while
others are "metaphorically" true? It doesn't seem sufficient to
point out this is blatant misuse of these words per their
*literal* meanings, or that *real* metaphors, such as "he died laughing"
or "she's drowning in sorrow" are quite unlike "the computer made
a decision" -- I've done so in the past to no avail.
In fact, I cannot remember any instance where any argument or
presentation has changed your view on anything, whereas my
views have changed over the years from exposure to you, Rickert,
McCullough, Sloman, Lupton (r.i.p.), and others. So it goes,
year after year.
--
<J Q B>
Anders N Weinstein
>ande...@pitt.edu says...
>>
>>This formulation is a little confused: If you can interpret them, you
>>can of course take seriously ideas you can find in the words that
>>emanate from Edgar Bergen's dummy.
>>
>>Still Edgar Bergen's dummy does not make any arguments. The idea is that
>>a being without free will cannot make any arguments at all.
>
>I think that that is a distinction without a difference.
>I don't *care* whether something is *really* making
>arguments, or whether it is merely doing something that
>*appears* to be making arguments. The difference doesn't
>matter.
I bet we could find contexts and purposes for which you would be
interested to find that the entity you thought you were discoursing
with as a person was really a mindless dummy controlled by an unseen
agent. If nothing else this discovery would undermine your acknowledged
confidence that your putative interlocutor can be relied on to issue
sensible utterances.
I think what you must mean is: if you are only interested in
considering the quality of ideas, then it doesn't matter if they come
from a person or not. Perhaps someone puts up an ArgueBot on the Web,
and you can practice discussing something with it even though you are
under no illusion at all that it is a person you are sparring with.
Perhaps this could be the first step in testing some new idea you are
planning to present.
But that would seem to involve a somewhat specialized restriction in
our interests in having a discussion. One can abstract the concept
"apparent source of sensible utterances with which one might engage to
some extent" from the full concept of a person who means what they say,
but an abstraction is being performed nonetheless.
da...@cogentex.com
>
>In article <98qta...@edrn.newsguy.com>,
>Daryl McCullough <da...@cogentex.com> wrote:
>>>
>>
>>You claimed that one cannot take seriously
>>arguments made by a being without free will.
>>It was a nonsensical claim, and my arguments
>>showed it to be nonsensical.
>
>This formulation is a little confused: If you can interpret them, you
>can of course take seriously ideas you can find in the words that
>that come out of a teletype, words that are produced by a random
>sentence generation program, words you hear in a dream or from a
>disembodied voice, words from a spirit medium, from the oracle at
>Delphi, etc.
>
>Still Edgar Bergen's dummy does not make any arguments. The idea is that
>a being without free will cannot make any arguments at all.
I think that that is a distinction without a difference.
I don't *care* whether something is *really* making
arguments, or whether it is merely doing something that
*appears* to be making arguments. The difference doesn't
matter.
The difference between a person talking and a random character
generator *is* a difference that is important to me, because
in the case of the random generator, there is no reason to
believe that it will continue to behave sensibly in the future,
and there is no reason to believe that anything I say has any
affect on its future behavior. In another post, I enumerated
what I expect to get out of a discussion, and I don't have
any reason to think that I could get that from a random
character generator.
Neil W Rickert
>In article <98ru1g$4...@euclid.cs.niu.edu>, Neil says...
>>da...@cogentex.com (Daryl McCullough) writes:
>>>Neil says...
>>>>The suggestion that there could be reasons as to why I did
>>>>post is not sensible in a world without free will.
>>>That's not true. To have "reasons" for why something
>>>happens or doesn't, you only need causality, not
>>>free will.
>>But what would causality do for us without free will.
>I guess I don't know what you mean by "free will".
And I guess I don't know what you mean by causality. If you mean
that the universe is exactly emulable on a Turing machine, then I
deny it.
Daryl McCullough
>
>>The issue was your claim that you can't have a serious discussion
>>with someone without free will. To sharpen that point, let's try
>>to enumerate the reasons why you might want to have a "serious
>>discussion" with anybody:
>
>> 1. You are hoping to learn something about the discussion topic.
>> 2. You are hoping to learn something about the other participants.
>> 3. You are hoping to influence the future behavior of the other
>> participants.
>> 4. [Insert other reasons here...]
>
>>Which reason depends on the participants in the discussion
>>having free will?
>
>There is a lot of disagreement about what we mean by any of the terms
>"want", "hoping", "learn", "free will". However, we would no more
>apply "want" or "hoping" or "learn" to an ordinary computer running a
>typical program, than we would apply the term "free will".
I'm not applying them to a computer---I'm applying them to *you*,
Neil Rickert.
[stuff deleted]
>In case you haven't worked it out, I think all three of your reasons
>presuppose free will.
Then I really don't know what you mean by "free will". Can a system
whose actions are completely determined by a program have free will?
Such a system can definitely meet criteria 1-3, though.
Daryl McCullough
>>>post is not sensible in a world without free will.
>
>>That's not true. To have "reasons" for why something
>>happens or doesn't, you only need causality, not
>>free will.
>
>But what would causality do for us without free will.
I guess I don't know what you mean by "free will".
>We might know abstractly that the world runs based on
>causal principles. And we might have some causal
>principles encoded in our brains. But the
>causal principles encoded in our brains could be
>entirely bogus. What we really have to depend on
>is our own freedom and autonomy to
>test these causal principles for ourselves.
I guess I don't know what you mean by "freedom and autonomy".
>>Free will to me seems a completely theological
>>concept. I'm not using the word "theological"
>>pejoratively, I'm using it in the sense of religious
>>studies. The concept of free will was invented to
>>explain how a benevolent God could allow evil things
>>to happen. According to some theologians, God gave
>>humans freedom to make choices, so that the evil
>>done by humans wasn't to be blamed on their Creator.
>
>Forget the religious aspects. Instead, look at the philosophical
>tradition that connects free will to responsibility.
Since when did you have such great respect for philosophical
traditions? The fact that philosophers have said lots of
words about free will doesn't mean that it exists, or
even that it is a coherent idea.
>>Granted, the concepts of being "responsible for
>>one's actions" is important in many legal systems.
>
>Its importance goes far beyond legal systems. We rely on it in every
>day life.
That doesn't mean that anybody has a coherent notion of what
"responsibility" is.
Jim Balter
>
> In article <98tjs...@edrn.newsguy.com>, <da...@cogentex.com> wrote:
> >ande...@pitt.edu says...
> >>
> >>This formulation is a little confused: If you can interpret them, you
> >>can of course take seriously ideas you can find in the words that
> >>emanate from Edgar Bergen's dummy.
> >>
> >>Still Edgar Bergen's dummy does not make any arguments. The idea is that
> >>a being without free will cannot make any arguments at all.
> >
> >I think that that is a distinction without a difference.
> >I don't *care* whether something is *really* making
> >arguments, or whether it is merely doing something that
> >*appears* to be making arguments. The difference doesn't
> >matter.
>
> I bet we could find contexts and purposes for which you would be
> interested to find that the entity you thought you were discoursing
> with as a person was really a mindless dummy controlled by an unseen
> agent. If nothing else this discovery would undermine your acknowledged
> confidence that your putative interlocutor can be relied on to issue
> sensible utterances.
Why in the world would it do that? If Garry Kasparov, half way
through a game with Deep Blue, learned that his opponent's moves
were really being generated by a human, that wouldn't undermine
his confidence that they would continue to be good moves.
In the same way, when I see that a post is from Anders Weinstein,
I have considerable confidence that it will display a certain
amount of illogic, regardless of whether the mindless dummy
that generated it is sitting right here on my desk or is off
somewhere in Philadelphia.
A much better example is Edgar Bergen's dummy shouting "bomb!" on
an airplane -- we wouldn't throw the dummy in jail and let Bergen
continue on his flight. But this goes to being *effective* by
considering *causal relationships*, not some airy fairy notion of
"full concept of a person who means what they say".
--
<J Q B>
Neil W Rickert
>Neil says...
>>da...@cogentex.com (Daryl McCullough) writes:
>>>Neil says...
>>>>Your *point* still seems entirely irrelevant to what was at issue.
>>>The issue was your claim that you can't have a serious discussion
>>>with someone without free will. To sharpen that point, let's try
>>>to enumerate the reasons why you might want to have a "serious
>>>discussion" with anybody:
>>> 1. You are hoping to learn something about the discussion topic.
>>> 2. You are hoping to learn something about the other participants.
>>> 3. You are hoping to influence the future behavior of the other
>>> participants.
>>> 4. [Insert other reasons here...]
>>>Which reason depends on the participants in the discussion
>>>having free will?
>>There is a lot of disagreement about what we mean by any of the terms
>>"want", "hoping", "learn", "free will". However, we would no more
>>apply "want" or "hoping" or "learn" to an ordinary computer running a
>>typical program, than we would apply the term "free will".
>I'm not applying them to a computer---I'm applying them to *you*,
>Neil Rickert.
>[stuff deleted]
I note that you have snipped most of what I wrote. Perhaps you found
it non-relevant. That's fine. I point it out since both you and Jim
were critical when I snipped parts of your earlier posting that I
found non-relevant.
>>In case you haven't worked it out, I think all three of your reasons
>>presuppose free will.
>Then I really don't know what you mean by "free will". Can a system
>whose actions are completely determined by a program have free will?
I guess it depends on what you mean by "completely determined by a
program." In my opinion, the type of system envisaged by GOFAI, for
example, cannot have free will.
>Such a system can definitely meet criteria 1-3, though.
Again, assuming that we are talking about the type system known as
GOFAI, there is zero evidence that "such a system can definitely meet
criteria 1-3."
Jim Balter
I wrote
BTW, it is blatantly dishonest to snip Daryl's justification for his
question:
It was dishonest, because Daryl offered a context in which free will
might not be an issue, but you snipped the context when answering
the question of why it is an issue. That has nothing to do with mere
snipping, which could simply be that Daryl has no dispute with what
you wrote or has already responded to it. And you know this full well,
jackass.
--
<J Q B>
cyberdiction
>reality, not reality itself. If so their meaning depends on the
>context and for example free will has different meaning if you would
>be in the concentation camp and different if you are wondering what a
>car to buy.
cyberdiction wrote:
>is it that perceives the universe unfolding deterministically one step
>after another inexorable step? This is also a conceptualization.
Dz wrote:
some concepts seems to be eternally multi-meaningful, they are more
sensitive to the actual context than others more harboured in the
material world
Neil W Rickert
>Neil W Rickert wrote:
>> >[stuff deleted]
>I wrote
This is much like the fact that I had no dispute with the part that I
had snipped. My response to Daryl in no way depended on that.
What we see is a double standard at work here.
Or perhaps it is the single standard, that Balter is always right by
definition.
JRStern
wrote:
>>Which reason depends on the participants in the discussion
>>having free will?
>
>There is a lot of disagreement about what we mean by any of the terms
>"want", "hoping", "learn", "free will". However, we would no more
>apply "want" or "hoping" or "learn" to an ordinary computer running a
>use these terms metaphorically with respect to computers, but we
>recognize the difference.
Well, "learn" may be distinguished from the others.
I have long been a proponent of avoiding use of intentional terms when
their reality is exactly what is in question, just as you say above.
However, this is not to preclude that they *may* in fact be justified
after all. In recent days, my pendulum has been swinging just that
way.
>Serious discussion is not at all like the data flow that we see
>between computers.
Again, it might just be, though that would mean discounting some of
the elan vital of humans in some eliminative or reductive way, or else
discovering some better equalities between computer software and human
minds than is yet accepted.
>It depends on a degree of mutual trust that we do
>not expect to find in computers.
I don't know about "trust", in either computers or people. Do I
"trust" the messages on this newsgroup? Yet, surely it is data flow,
even "information" flow -- not that I want to have the "data vs.
information" discussion here, but it may not be avoidable.
(I just read a few pages from Millikan on this -- p. 195-201 of On
Clear and Confused Ideas -- and marked the margins with all sorts of
agreement)
> I might trust the data coming from
>a particular computer system, but only because I know its causal
>structure well enough, or because I trust the people who provided
>that system.
Or, you may accept the data with some reduced degree of trust, and
these things may never be absolutely known or true or trusted at all.
>I can think of occasions when I might engage in discussion,
>but with the other party not taking responsibility. Examples
>that come to mind are talking to a clown, or talking to the Eliza
>program. But in such cases, I would not be taking the discussion
>seriously. It would be something done for amusement, and not for
>any of the reasons 1,2,3 that you list.
Exactly.
Joshua Stern
JRS...@gte.net
Pierre-Normand Houle
> > rather say that I was externally constrained to act in conformity with
> > the rule?
>
> This is so sloppy! *You* weren't "constrained" at all, you simply
> had a tie placed around your neck, and this put your action of entering
> the restaurant into conformity with the rule -- as I said above,
Yes, I was sloppy in my formulation. I much prefer yours.
But I only wanted to draw you attention on this distinction between
following a rule (as a responsible intentional agent) and acting in
conformity with a rule (possibily accidentally).
A similar distinction appears in the following example:
Suppose that as I return home late at night I am walking a few
feet behind a lady who accidentally happens to be employing
the same route as myself. At some point, (after a few coincident
bifurcations, say) the suspicious lady might turn and ask with
a slightly irate tone: "Are you following me?". And I might
reply: "I am not following you. I live right there." Upon seeing
me grabbing my keys, unlocking the door and entering the house
I just indicated to her, she might also judge that I indeed told
the truth: In a clear sense I was not following her.
> This is separate from what *you*, or you *girlfriend*, as agents,
> did. There is no *dichotomy* between talk about agents following
> rules and rules not being violated by the state of affairs
I only claim that they are different concepts. Obviously, both
concepts can simultaneously apply (in the case of responsible
agents). A distinction need not be a relation of exclusion,
of course.
But I am claiming that computers, specifically, do not follow
the rules, or instructions expressed in their native language.
They only act in conformity with them.
Pierre-Normand Houle
"Jim Balter" <j...@digisle.net> wrote in message news:3AB05CE9...@digisle.net...> Pierre-Normand Houle wrote:
> > > No, it's not, at least not for me -- for me, blame applies to causes,
> > > rather than being the sort of free-floating emotional "I want to
> > > smack the one to blame" medieval anti-intellectual retributional
> > > primitive reaction that is the alternative.
> >
> > This seems to be a false dichotomy to me --
>
> *You* are the one who used the word "different", claiming a dichotomy.
To charge someone of making a false dichotomy is to submit that
the two contemplated alternatives are not exhaustive.
>> It is quite different, I think.
>
> Of course it is different -- so bloody what? The point is that
> both kinds of fixes apply to both people and machines, a
> point that you dishonestly snip rather than respond to.
It is a claim that I deny. We need not try to fix those people that
do not act for the fulfillment of our goals. We often choose to
respect their autonomy. And this is not an attitude that we have
toward machines, generally.
> > Because leading somebody to alter
> > his conduct through reasoning him (rather than through mere
> > threat or manipulation, say) will lead him to act purposively
> > and still in accord with his *own* acknowledged standards
> > and practical commitments (or intentions).
>
> You can always push your argument into such words as *own*,
> but it is at root dishonestly circular.
Autonomy can be understood as a status that is socially constituted
rather than a mysterious intrinsic property. It might be constituted
by interpersonal attitudes of attributing and undertaking commitments
and entitlements to these commitments: attitudes that we don't have
toward machines and that they do not have toward us or each other.
> > > > It can be a fact that a programmer is
> > > > responsible even though you are in no position to blame him.
> > > > And it can be a fact that nobody is responsible at all without
> > > > this being a reason for "blaming" the computer in the sense of
> > > > attributing it responsibility and autonomy.
> > >
> > > This just assumes the blatantly erroneous notion that there are
> > > single causes or responsibilities or sources of blame etc.
> >
> > I said that "it can be a fact that nobody is responsible". Of
> > course, it can also be a fact that many are responsible. And
> > it certainly is the case that there always are many causes.
>
> But these are single causes -- "no one is responsible" or
> "A, B, and C are responsible".
No. Responsibilities are normative statuses. Causal antecedents
are other kinds of beasts altogether. An individual can fail to
fulfill its responsibilities without this failure constituting a
denial of them (he need not even have acknowledged them in the
first place). But if A fails to cause B, then A is not a cause
of B after all.
Of course, a successful fulfillment of responsibility entails
causal efficacy. But it does not entail uniqueness of cause.
> But in fact there is a sense
> in which John von Neumann is responsible for every bug.
This is the sense in which a butterfly is responsible
for a thunderstorm.
> If you exclude him, you can't do it on principle,
I can't exclude him on physical principles, for sure.
> you have to do it ad hoc.
I can do it on legal or moral grounds for instance, and
although these might involve principles that are relative
to culture they need not be ad hoc.
In the "butterfly" causal sense anybody who was born
some time before Abraham Lincoln is responsible for
the death death of this president. Yet, there is a non
ad hoc sense in which they are not murderers (except for
people like John Wilkes Booth, of course)
> > > As I have already point out, all we should care about is *cause*;
> >
> > Why is that?
>
> Because that's what it takes to be effective.
Again, I might not have as a goal to change or fix another person.
In a vacuous sense, I want to be causally efficacious, for instance
I want that my interlocutor hear what I say and this involve him
having his physical state modified as a causal result of my utterance,
but this is only a secondary goal.
My primary goal might be to "set him free".
> > I do care about what my responsibilities are, for instance.
>
> Non sequitur -- we were talking about attributing blame.
> But, responsibilities to whom or what?
To other persons, mainly. Maybe some abstract causes also.
> Why do you have these responsibilities?
Because I am a social rational animal?
> What happens if you simply change your mind about what
> your responsibilities are?
My responsibilities will not change thereby. If my change
of mind is *justified* then I might have discovered what
my responsibilities were.
> I would be much more concerned with what my *goals* are.
No. I do not believe that we have intrinsic fundamental
goals. I am more interested in setting my goals than discovering
which ones are genuine in the sense of being innate or naturals.
Goals are mine that I have made mine, through justifying them and
committing myself to their accomplishment. Otherwise, they might be
goals that publicists have set for me, or those of my genes,
but they are not mine.
> and in fact that's why
> cause matters, because understanding cause is how I achieve
> goals;
But I didn't say that causes do not matter. I am merely pointing
out that causal efficacy can not be an end in itself.
> > Shouldn't I? If I could only look at causes of my behavior, I would
> > be clueless as to what I *ought* to do next.
>
> You seem to have quite misunderstood. Rather than seeking to
> blame someone by taking the emotional stance that they are not
> entitled, an utterly ineffective strategy, we should seek what
> caused the condition that thwarts our goals/desires, so we
> can change the course. For this we need to understand what
> causes what, including what our own actions cause. And we
> may even want to understand what causes us to act as we do,
> if we want to change the way we act but find that simply
> wanting isn't sufficient -- more sleep, better diet,
> psychoactive drugs, nicotine patches, etc. may bring about
> such changes. All of this is directed at *goals*,
> not *responsibilities*. Of course, you may have as a goal
> to meet what you feel to be your responsibilities, but that's
> a level removed.
It is not my concern what theoretical physics or evolutionary
biology has to say as to what my most fundamental goals are.
But physical law is what lay at the close of your quest, as I
understand it.
I might comment on the rest another time if it is still
relevant.
<snip>
Pierre-Normand Houle
Sent: Wednesday, March 14, 2001 9:50 PM
Subject: Re: Consiousness vs. Free Will
> Pierre-Normand Houle wrote:
> >
> > "Jim Balter" <j...@digisle.net> wrote in message news:3AAFFAF8...@digisle.net...
> > > Anders N Weinstein wrote:
> > > > It seems to me the interesting question is: what if anything
> > > > would have to be the case before it would be appropriate to blame the
> > > > computer and not the programmer?
> > >
> > > First, I would note that the original statement was about
> > > making decisions, which people are much more ready (as an
> > > observed fact) to attribute to computers than assigning
> > > blame. The shift here is a standard ill faith move.
> > >
> > > Even so, as a programmer, I blame the software when it makes a mistake,
> >
> > This is a different sense of "blame".
>
> No, it's not, at least not for me -- for me, blame applies to causes,
> rather than being the sort of free-floating emotional "I want to
> smack the one to blame" medieval anti-intellectual retributional
> primitive reaction that is the alternative.
This seems to be a false dichotomy to me -- as if the physical stance
was the only sound possibility. My blaming somebody can express
my denying him entitlement to his conduct. And only through the
intentional stance might I be able to attribute entitlements in the
first place.
> > It is the sense in which you can sometimes
> > blame bad weather. I did a google search with "the computer decided" and
> > only about one in ten results had the form of an intentional attribution. The rest
> > were blames for failure, crashes or buggy behaviors.
> >
> > > and I go try to find out what is wrong with it and fix it.
> >
> > Whereas you would not attempt to "fix" somebody who made
> > a bad decision. You would try to reason or convince a person.
>
> In fact we do try to fix people, through training, education,
> medication, surgery, etc. Reasoning is another sort of fix --
It is quite different, I think. Because leading somebody to alter
his conduct through reasoning him (rather than through mere
threat or manipulation, say) will lead him to act purposively
and still in accord with his *own* acknowledged standards
and practical commitments (or intentions).
> > > It would be quite ineffective for me to blame the programmer --
> > > I often can't even identify the programmer, who may in fact
> > > be a whole set of programmers.
> >
> > But the question is whether the intentional stance is appropriate
> > not whether it is useful.
>
> No, that's not the question. Have you read Dennett at all?
Well, *I* am saying that the intentional stance is not appropriate
when describing low level computational processes. I thought this
was what you were disagreeing with. I do not deny that merely
metaphorical uses of intentional vocabulary can be *useful* in
some circumstances. (If Dennett disagrees, more power to him)
> > It can be a fact that a programmer is
> > responsible even though you are in no position to blame him.
> > And it can be a fact that nobody is responsible at all without
> > this being a reason for "blaming" the computer in the sense of
> > attributing it responsibility and autonomy.
>
> This just assumes the blatantly erroneous notion that there are
> single causes or responsibilities or sources of blame etc.
I said that "it can be a fact that nobody is responsible". Of
course, it can also be a fact that many are responsible. And
it certainly is the case that there always are many causes.
> As I have already point out, all we should care about is *cause*;
Why is that? I do care about what my responsibilities are, for instance.
Shouldn't I? If I could only look at causes of my behavior, I would
be clueless as to what I *ought* to do next. I would no longer be
a rational creature.
> the conception of "autonomy" is medieval, and there are thousands
> of threads of evidence and thought, in sociology, physics,
> biochemistry, psychology, neuroscience, cognitive science,
> the law, etc. that argue against the notion of autonomy as
> an *attribute*. The question is the usefulness of the
> intentional stance, *not* whether it is "appropriate".
> It is always appropriate to apply concepts where they are
> useful, philosophical arrogance notwithstanding.
But I don't see how you can reduce semantic questions of
appropriateness (to appropriately apply a concept in judgment
is to make a true claim) to questions of utility. How do you
decide of the utility of applying a concept without making
(implicit) judgments of truth involving these concepts?
Also, what it useful to me qua rational and socially responsible
agent might not coincide with what is useful to me qua biological
creature and passive carrier of a subset of the human genome.
Do you have some alternative in mind when you are talking
about utility?
Pierre-Normand Houle
From: "Jim Balter" <j...@digisle.net>
Newsgroups: comp.ai.philosophy
Sent: Wednesday, March 14, 2001 9:50 PM
Subject: Re: Consiousness vs. Free Will
> >
> > "Jim Balter" <j...@digisle.net> wrote in message news:3AAFFAF8...@digisle.net...
> > > Anders N Weinstein wrote:
> > > > It seems to me the interesting question is: what if anything
> > > > would have to be the case before it would be appropriate to blame the
> > > > computer and not the programmer?
> > >
> > > First, I would note that the original statement was about
> > > making decisions, which people are much more ready (as an
> > > observed fact) to attribute to computers than assigning
> > > blame. The shift here is a standard ill faith move.
> > >
> > > Even so, as a programmer, I blame the software when it makes a mistake,
> >
> > This is a different sense of "blame".
>
> No, it's not, at least not for me -- for me, blame applies to causes,
> rather than being the sort of free-floating emotional "I want to
> smack the one to blame" medieval anti-intellectual retributional
> primitive reaction that is the alternative.
This seems to be a false dichotomy to me -- as if the physical stance
was the only sound possibility. My blaming somebody can express
my denying him entitlement to his conduct. And only through the
intentional stance might I be able to attribute entitlements in the
first place.
> > It is the sense in which you can sometimes
> > blame bad weather. I did a google search with "the computer decided" and
> > only about one in ten results had the form of an intentional attribution. The rest
> > were blames for failure, crashes or buggy behaviors.
> >
> > > and I go try to find out what is wrong with it and fix it.
> >
> > Whereas you would not attempt to "fix" somebody who made
> > a bad decision. You would try to reason or convince a person.
>
> In fact we do try to fix people, through training, education,
> medication, surgery, etc. Reasoning is another sort of fix --
It is quite different, I think. Because leading somebody to alter
his conduct through reasoning him (rather than through mere
threat or manipulation, say) will lead him to act purposively
and still in accord with his *own* acknowledged standards
and practical commitments (or intentions).
> > > It would be quite ineffective for me to blame the programmer --
> > > I often can't even identify the programmer, who may in fact
> > > be a whole set of programmers.
> >
> > But the question is whether the intentional stance is appropriate
> > not whether it is useful.
>
> No, that's not the question. Have you read Dennett at all?
Well, *I* am saying that the intentional stance is not appropriate
when describing low level computational processes. I thought this
was what you were disagreeing with. I do not deny that merely
metaphorical uses of intentional vocabulary can be *useful* in
some circumstances. (If Dennett disagrees, more power to him)
> > It can be a fact that a programmer is
> > responsible even though you are in no position to blame him.
> > And it can be a fact that nobody is responsible at all without
> > this being a reason for "blaming" the computer in the sense of
> > attributing it responsibility and autonomy.
>
> This just assumes the blatantly erroneous notion that there are
> single causes or responsibilities or sources of blame etc.
I said that "it can be a fact that nobody is responsible". Of
course, it can also be a fact that many are responsible. And
it certainly is the case that there always are many causes.
> As I have already point out, all we should care about is *cause*;
Why is that? I do care about what my responsibilities are, for instance.
Shouldn't I? If I could only look at causes of my behavior, I would
be clueless as to what I *ought* to do next. I would no longer be
a rational creature.
> the conception of "autonomy" is medieval, and there are thousands
> of threads of evidence and thought, in sociology, physics,
> biochemistry, psychology, neuroscience, cognitive science,
> the law, etc. that argue against the notion of autonomy as
> an *attribute*. The question is the usefulness of the
> intentional stance, *not* whether it is "appropriate".
> It is always appropriate to apply concepts where they are
> useful, philosophical arrogance notwithstanding.
But I don't see how you can reduce semantic questions of
Neil W Rickert
(reacting to my post)
>BTW, it is blatantly dishonest to snip Daryl's justification for his
>question:
>> Perhaps those
>> "forces beyond his control" constrain him so that he only makes
>> sensible statements. Perhaps someone with "free will" uses that
>> free will to make nonsensical statements.
I snipped it because it was entirely irrelevant to the issues
that I had raised and that Daryl was disputing.
But I see that you had a bad sleep last night, and are only capable
of making false accusations instead of engaging in rational debate.
In the circumstances, there is no point in continuing this exchange.
cyberdiction
> You're about to make a decision, I magically take a complete copy of
> your entire body and environment (it's a thought experiment, I can *do*
> this ;-). You make the decision, so does your copy.
> Two possibilities: either the copy makes the same choice or it doesn't.
> If it doesn't, what's different between the two (remember they're
> identical right down to atoms).
> If its the same - what's free about it?
lodovico wrote:
Interesting, but not at all convincing.
Daryl McCullough
>
>
>"Jim Balter" <j...@digisle.net> wrote
>> Even so, as a programmer, I blame the software when it makes
>> a mistake,
>
>This is a different sense of "blame". It is the sense in which
>you can sometimes blame bad weather. I did a google search with
>"the computer decided" and only about one in ten results had
>the form of an intentional attribution. The rest
>were blames for failure, crashes or buggy behaviors.
>
>> and I go try to find out what is wrong with it and fix it.
>
>Whereas you would not attempt to "fix" somebody who made
>a bad decision. You would try to reason or convince a person.
Yes. This way of dealing with a human (as opposed to a machine)
is, I think, part of what Dennett would call "taking the
intentional stance". For humans, we understand their behavior
in terms of beliefs, motives, intentions, etc. For computers,
we understand their behavior in terms of state transitions,
data processing.
But the reason that we take different stances towards people
and (current) computers is because the *behavior* warrants
the difference. When a computer deletes important data files,
I don't start thinking along the lines "What did I do to make
the computer angry at me? What can I do to make things up with
it?" We don't start thinking along those lines because it
doesn't do any good. It doesn't help avoid the next computer
error. With humans, consideration of motives and feelings
*does* help to understand (and influence) their behavior.
In separating systems that warrant being treated as "people"
and those that don't, I don't think that we need to go beyond
behavior---does it *work* to treat the system as a person?
Jim Balter
>
> ----- Original Message -----
> From: "Jim Balter" <j...@digisle.net>
> Newsgroups: comp.ai.philosophy
> Sent: Wednesday, March 14, 2001 9:29 PM
> Subject: Re: Consiousness vs. Free Will
>
> > >
> > > In article <3AAFD729...@digisle.net>, Jim says...
> > > >
> > > >Daryl McCullough wrote:
> > >
> > > >> You are just being silly. Computers don't "follow rules". They just
> > > >> act in a way that is describable by rules.
> > > >
> > > >And not even that in practice, just as with human or other complex
> > > >behavior.
> > >
> > > Yes. I really didn't mean to be objecting to the use of the
> > > phrase "computers follow rules". I was only pointing out that
> > > it is an anthropomorphism, just like "computers make decisions".
> >
> > As I don't see decision making or rule following to be particularly
> > "human shaped". "anthropomorphism" is better used in cases where
> > motives and causes are misascribed. But describing a computer
> > as making a decision isn't of that sort. Or, at least, to claim that
> > it is must be argued for; it cannot be established simply by fiat.
>
> in a certain restaurant.
Shouldn't you dwell at least a bit on the various ways the word
"rule" can be used and it's significance in this case? Such
a rule is essentially a warning from the restaurant management that,
if you don't wear a tie, they will bar you from entering the restaurant.
If you believe the warning, want to enter the restaurant, and are
rational, you will don a tie so as not to have your desires thwarted.
This is presumably not the sort of rule following that Rickert
means when he says "Computers follow rules". Just what position
are you arguing for with this example, and how do you see it supporting
that position? It seems prima facie irrelevant to the text about
anthropomorphism and decision making that you are responding to.
> Suppose furthermore that I am a rebellious
> spirit and choose not to wear one.
Perhaps you believe that you can convince the restaurant to
let you in despite not wearing a tie, i.e., the rule has exceptions,
it doesn't govern, it isn't a law, etc. It doesn't appear to me
that you are thinking very deeply or imaginatively about these things.
> However, my girlfriend, who is
> accompanying me for dinner and disapproves of my anti-social
> attitude,
Ah, now we have a quite different issue -- an implicit rule
that one should conform to posted notices, especially if one expects
others to conform to them -- not doing so is "anti-social".
It seems to me that you are revealing your own psychology
without even stopping to question it. Among my peers,
a certain degree of nonconformity is quite "social",
and excessive conformity to "establishment" (think about
*that* word) "rules" is anti-social. Additionally, the
mere donning of a tie can be considered anti-social,
and one must balance that against a calculation of how
many others may conform to the demand of the sign,
and how you may be judged one way or the other for standing out,
or even for appearing to care about being judged.
There are many elements here separate from the notion of
"rule following" that Rickert seemed to appeal to, but you
blithely conflate them and yet somehow expect clarity to
flow from yes/no answers to your complex (in the legal sense)
questions.
> puts a tie around my neck, unbeknown to me, pretending
> to be hugging me.
This action has no bearing on your *attitude*. There is
no apparent way in which this reflects any *disapproval*
on her part -- if she disapproves, she would display *that* by
letting you know what she has done and why. Rather, she apparently
doesn't want you thrown out but doesn't want to express disapproval,
or just feels uncomfortable about you not wearing a tie, worrying about
how she, as your companion, will be judged if you don't.
Your thinking here is appallingly shallow and unanalytical, and yet
you expect to actually hold opinions about these matters that might
have some legitimacy, and might have some beneficial effect on
the totality of mankind's knowledge. I suggest that these efforts
on your part have the opposite effect, and that mankind would be
better off if you stuck to some field for which you are better
equipped. (And before you get too insulted, keep in mind that
95% of people polled think they have above-average IQ's. Obviously,
a large number of people have a judgment of their own intellectual
talents that is inflated relative to reality, and there is no
reason to think you are not among them -- in fact, I am providing
reasons to think you are).
> So I enter the restaurant wearing a tie unwittingly
> and unknowingly.
Yes.
> Did I *follow* the restaurant's rule?
The sorts of rules one can follow are commands, such as
"add 5 to x" or "don a tie before entering this restaurant",
but not "you must wear a tie to enter this restaurant".
Talk of "following" the latter sort of rule is a category
mistake. But "don a tie before entering this restaurant"
is a rule you didn't follow, since you donned nothing -- it
was your girlfriend who acted. Express the rule differently,
and we can attempt to determine whether in fact you followed
it; e.g., "have a tie around your neck when entering the restaurant"
is a rule you followed or, if you wish, a constraint that you
didn't violate. I tried hard to phrase that in a way that
makes what you are saying here make sense, but that I had to
expend that effort exposes the weakness of your "wouldn't we
rather".
> Wouldn't we
> rather say that I was externally constrained to act in conformity with
> the rule?
This is so sloppy! *You* weren't "constrained" at all, you simply
had a tie placed around your neck, and this put your action of entering
the restaurant into conformity with the rule -- as I said above,
the rule states a constraint that was not violated, as a result
of your girlfriend's action; to refer to this as you being
"externally constrained" is mightily confused. But it didn't simply put
*you* in conformity of the rule, it put the state of the world in conformity
with the rule, i.e., the rule was not violated as a matter of empirical fact.
This is separate from what *you*, or you *girlfriend*, as agents, did. There
is no *dichotomy* between talk about agents following rules and rules
not being violated by the state of affairs. What we would *rather*
do depends on the details and our objectives and what is most useful.
And upon at least a modicum of care in stating a rule before
debating whether it was followed and by what or whom.
--
<J Q B>
Neil W Rickert
>>>> Neil says...
>>>> >Computers don't make decisions. They just follow the rules given to
>>>> >them by the programmer.
>>> It doesn't take much examination to see that there
>>>is a false dichotomy here, borne by that reason-destroying word "just"
>>>-- something could both follow rules and make decisions.
>>The next time Windows crashes on your computer, I expect you to file
>>a law suit charging your computer with negligence, instead of griping
>>about Bill Gates.
>>There was no false dichotomy in what I said. The fact is, we blame
>>the programmer, not the computer, when there is a failure.
>Indeed. It seems to me the interesting question is: what if anything
>would have to be the case before it would be appropriate to blame the
>computer and not the programmer?
It is an interesting question. But it is a question that has to do
with the details of how behavior is produced. Experience has shown
that this is something you are unwilling to discuss.
Jim Balter
>
> ----- Original Message -----
> From: "Jim Balter" <j...@digisle.net>
> Newsgroups: comp.ai.philosophy
> Sent: Wednesday, March 14, 2001 9:50 PM
> Subject: Re: Consiousness vs. Free Will
> > Pierre-Normand Houle wrote:
> > >
> > > "Jim Balter" <j...@digisle.net> wrote in message news:3AAFFAF8...@digisle.net...
> > > > Anders N Weinstein wrote:
>
> > > > > It seems to me the interesting question is: what if anything
> > > > > would have to be the case before it would be appropriate to blame the
> > > > > computer and not the programmer?
> > > >
> > > > First, I would note that the original statement was about
> > > > making decisions, which people are much more ready (as an
> > > > observed fact) to attribute to computers than assigning
> > > > blame. The shift here is a standard ill faith move.
> > > >
> > > > Even so, as a programmer, I blame the software when it makes a mistake,
> > >
> > > This is a different sense of "blame".
> >
> > No, it's not, at least not for me -- for me, blame applies to causes,
> > rather than being the sort of free-floating emotional "I want to
> > smack the one to blame" medieval anti-intellectual retributional
> > primitive reaction that is the alternative.
>
> This seems to be a false dichotomy to me --
*You* are the one who used the word "different", claiming a dichotomy.
> as if the physical stance
> was the only sound possibility.
I said nothing about the physical stance. The intentional stance
is taken precisely to make causal analysis tractable.
> My blaming somebody can express
> my denying him entitlement to his conduct.
No, it can *justify* your denying entitlement.
> And only through the
> intentional stance might I be able to attribute entitlements in the
> first place.
Yes, so? I stated that I don't apply this sort of blame -- "ooh, ooh,
you aren't entitled to behave that way, wah wah wah". I don't
think in terms of "entitlement" -- a *very* medieval notion.
> > > It is the sense in which you can sometimes
> > > blame bad weather. I did a google search with "the computer decided" and
> > > only about one in ten results had the form of an intentional attribution. The rest
> > > were blames for failure, crashes or buggy behaviors.
> > >
> > > > and I go try to find out what is wrong with it and fix it.
> > >
> > > Whereas you would not attempt to "fix" somebody who made
> > > a bad decision. You would try to reason or convince a person.
> >
> > In fact we do try to fix people, through training, education,
> > medication, surgery, etc. Reasoning is another sort of fix --
>
> It is quite different, I think.
Of course it is different -- so bloody what? The point is that
both kinds of fixes apply to both people and machines, a
point that you dishonestly snip rather than respond to.
> Because leading somebody to alter
> his conduct through reasoning him (rather than through mere
> threat or manipulation, say) will lead him to act purposively
> and still in accord with his *own* acknowledged standards
> and practical commitments (or intentions).
You can always push your argument into such words as *own*,
but it is at root dishonestly circular.
> > > > It would be quite ineffective for me to blame the programmer --
> > > > I often can't even identify the programmer, who may in fact
> > > > be a whole set of programmers.
> > >
> > > But the question is whether the intentional stance is appropriate
> > > not whether it is useful.
> >
> > No, that's not the question. Have you read Dennett at all?
>
> Well, *I* am saying that the intentional stance is not appropriate
> when describing low level computational processes.
How did "low level" get in there? We were talking about computers
and humans, not ICs or neurons.
> I thought this
> was what you were disagreeing with.
> I do not deny that merely
> metaphorical uses of intentional vocabulary can be *useful* in
> some circumstances. (If Dennett disagrees, more power to him)
"merely metaphorical" *presupposes* what I dispute.
> > > It can be a fact that a programmer is
> > > responsible even though you are in no position to blame him.
> > > And it can be a fact that nobody is responsible at all without
> > > this being a reason for "blaming" the computer in the sense of
> > > attributing it responsibility and autonomy.
> >
> > This just assumes the blatantly erroneous notion that there are
> > single causes or responsibilities or sources of blame etc.
>
> I said that "it can be a fact that nobody is responsible". Of
> course, it can also be a fact that many are responsible. And
> it certainly is the case that there always are many causes.
But these are single causes -- "no one is responsible" or
"A, B, and C are responsible". But in fact there is a sense
in which John von Neumann is responsible for every bug.
If you exclude him, you can't do it on principle, you have
to do it ad hoc. That's what a *careful* analysis of the
concepts of cause and responsibility reveal.
> > As I have already point out, all we should care about is *cause*;
>
> Why is that?
Because that's what it takes to be effective.
> I do care about what my responsibilities are, for instance.
Non sequitur -- we were talking about attributing blame.
But, responsibilities to whom or what? Why do you have these
responsibilities? What happens if you simply change your mind
about what your responsibilities are? I would be much more
concerned with what my *goals* are. and in fact that's why
cause matters, because understanding cause is how I achieve
goals; mere "blame" or opinions about "entitlement" do nothing
for me.
> Shouldn't I? If I could only look at causes of my behavior, I would
> be clueless as to what I *ought* to do next.
You seem to have quite misunderstood. Rather than seeking to
blame someone by taking the emotional stance that they are not
entitled, an utterly ineffective strategy, we should seek what
caused the condition that thwarts our goals/desires, so we
can change the course. For this we need to understand what
causes what, including what our own actions cause. And we
may even want to understand what causes us to act as we do,
if we want to change the way we act but find that simply
wanting isn't sufficient -- more sleep, better diet,
psychoactive drugs, nicotine patches, etc. may bring about
such changes. All of this is directed at *goals*,
not *responsibilities*. Of course, you may have as a goal
to meet what you feel to be your responsibilities, but that's
a level removed.
> I would no longer be
> a rational creature.
If you ignore responsibilities you might not be a *moral*
creature, but rationality has to do with effectiveness
in achieving a goal. There is no rational basis for goals
except as a means to other goals; at some point, it bottoms
out in the facts of physiology, which is a matter of how
evolution (in interplay with environment) constructed you.
> > the conception of "autonomy" is medieval, and there are thousands
> > of threads of evidence and thought, in sociology, physics,
> > biochemistry, psychology, neuroscience, cognitive science,
> > the law, etc. that argue against the notion of autonomy as
> > an *attribute*. The question is the usefulness of the
> > intentional stance, *not* whether it is "appropriate".
> > It is always appropriate to apply concepts where they are
> > useful, philosophical arrogance notwithstanding.
>
> But I don't see how you can reduce semantic questions of
> appropriateness (to appropriately apply a concept in judgment
> is to make a true claim) to questions of utility.
I haven't done any such reduction. True claims must be made
in terms of concepts; it is useful to apply concepts in
many contexts, and different true claims follow. Ruling
a priori that some applications are appropriate and some
are not blocks you from making many true claims
(e.g., that computers literally make decisions).
> How do you
> decide of the utility of applying a concept without making
> (implicit) judgments of truth involving these concepts?
There is a broad literature on the science of philosophy
on this subject.
> Also, what it useful to me qua rational and socially responsible
> agent might not coincide with what is useful to me qua biological
> creature and passive carrier of a subset of the human genome.
To me you have a quite irrational notion of "rational".
I do not evaluate my rationality in terms of the rules that society
externally imposes upon me; rather, I evaluate it in terms of my
*internal* values and goals -- these were largely *implicitly* imposed
upon me by society through my upbringing and indoctrination,
but as they are a part of my psyche, they really *are* my internal
goals and values, I have no need to repudiate them. When people ask
how an atheist such as myself can be moral, without an external code,
my answer is that I'm *built* that way. Surely that is at least as
effective, from society's POV, as enforcing moral behavior through
fear of divine retribution. The result of that is so often that
rational people come to realize that there is nothing real to be
feared, or fail to integrate this fear into their judgment process
just as smokers fail to integrate the fear of cancer into their
judgment process, and these people with externally imposed morality,
not being built as moral creatures, act amorally.
> Do you have some alternative in mind when you are talking
> about utility?
I'm talking about effectiveness -- a foreign concept to many
philosophers, apparently.
I'm way over my c.a.p. time allotment, so I'm got going to
be able to answer further posts in depth. This should not be
taken as any indication that I agree with or disagree with
anything. While its hard to let go of these debates, in the
end they aren't very effective, and it isn't really rational
to engage in them, but then the rationality of modern Western man,
so divorced from the rigors of survival, is a quite questionable thing.
--
<J Q B>
Jim Balter
>
> Jim Balter <j...@digisle.net> writes:
>
> (reacting to my post)
>
> >BTW, it is blatantly dishonest to snip Daryl's justification for his
> >question:
>
> >> Perhaps those
> >> "forces beyond his control" constrain him so that he only makes
> >> sensible statements. Perhaps someone with "free will" uses that
> >> free will to make nonsensical statements.
>
> I snipped it because it was entirely irrelevant to the issues
> that I had raised and that Daryl was disputing.
Since it was, as I said, a justification for his question,
it isn't irrelevant.
> But I see that you had a bad sleep last night, and are only capable
> of making false accusations instead of engaging in rational debate.
Well, no, that's not all I'm capable of. It's that sort of
hyperbole that makes rational debate with you hopeless.
> In the circumstances, there is no point in continuing this exchange.
Yup. And the less exchange, the less wasted bandwidth and the
more time each of us will have to do something worthwhile.
In the end, debating in c.a.p is quite irrational, a form
of addictive behavior.
--
<J Q B>
Daryl McCullough
>
>Jim Balter <j...@digisle.net> writes:
>
>(reacting to my post)
>
>>BTW, it is blatantly dishonest to snip Daryl's justification for his
>>question:
>
>>> Perhaps those
>>> "forces beyond his control" constrain him so that he only makes
>>> sensible statements. Perhaps someone with "free will" uses that
>>> free will to make nonsensical statements.
>
>I snipped it because it was entirely irrelevant to the issues
>that I had raised and that Daryl was disputing.
You snipped the *point* that I was trying to make.
>But I see that you had a bad sleep last night, and are only capable
>of making false accusations instead of engaging in rational debate.
Boy, the pot calling the kettle black!
>In the circumstances,
The circumstances being that you were caught making
a silly statement and then making a dishonest defense
of that statement....
>there is no point in continuing this exchange.
That's probably true.
jbze...@ck-sg.p.lodz.pl
<zar...@tiscalinet.it> wrote:
>I firmly believe that consciousness needs free will to exisit.
>Probably everyone has their own definition of "free will", so I shall try
>to make clear what I mean for "free will" and show that it is strictly
>connected to consciousness.
>"Free will" is a way of making choices which occurs in a
>non-deterministic way. If you believe that we live in a completely
>mechanicistic world (which I don't believe) then there can be logically
>no space for free will, it couldn't make sense. However in a purely
>mechanicistic context even consciousness wouldn't make sense either.
>There are cases in which we make decisions with insufficent data to work
>out their consequences. Free will is the ability to act in these
>situations. In fact if we knew everything we would know exactly how to
>act in any situation and there would be no such thing as making
>decisions.
>The state in which we are which actively involves the continuous making
>decisions, thanks to our free will, is consciousness.
>
>Therefore if we assume consciousness to exsist, which we can assert only
>if we are in a non-mechanicistic world, then its existence is based on
>free will.
>
>Ludovico Zaraga
>
>
Don't forget that consciousness and free-will are only concepts about
reality, not reality itself. If so their meaning depends on the
context and for example free will has different meaning if you would
be in the concentation camp and different if you are wondering what a
car to buy.
DZ
cyberdiction
reality, not reality itself. If so their meaning depends on the
context and for example free will has different meaning if you would
be in the concentation camp and different if you are wondering what a
car to buy. DZ
Neil W Rickert
>I'm quite happy to discuss the details of how 'behavior is produced'
>that *I* think are relevant to answering this question. I don't myself
>happen to believe these details concern mechanisms in the brain. They
>would have to be given in the irreducible vocabulary of intention,
>belief and the related family of concepts.
The mistake is to assume that the "vocabulary of intention" can be
treated as disconnected from physical processes.
>But if you could enunciate an account at that level and motivate
>the idea that this does characterize what makes it appropriate to "blame
>the system and not the programmer", I could certainly entertain it for the
>purpose of discussion.
It is a question of the physical requirements for agent autonomy.
Anders N Weinstein
Anders N Weinstein <ande...@pitt.edu> wrote:
>I'm quite happy to discuss the details of how 'behavior is produced'
>that *I* think are relevant to answering this question. I don't myself
>happen to believe these details concern mechanisms in the brain. They
>would have to be given in the irreducible vocabulary of intention,
>belief and the related family of concepts.
>
I meant: at your own favored level
Jim Balter
>
> Neil says...
>
> >Your *point* still seems entirely irrelevant to what was at issue.
>
> The issue was your claim that you can't have a serious discussion
> with someone without free will.
Actually, you said it of someone who *claims* not to have free will.
Rickert himself thinks this person does in fact have free will but is
mistaken in claiming not to -- which seems to eliminate any logical
justification for not having a discussion (and in fact he *is*
having a discussion with that person).
To sharpen that point, let's try
> to enumerate the reasons why you might want to have a "serious
> discussion" with anybody:
>
> 1. You are hoping to learn something about the discussion topic.
> 2. You are hoping to learn something about the other participants.
> 3. You are hoping to influence the future behavior of the other
> participants.
> 4. [Insert other reasons here...]
>
> Which reason depends on the participants in the discussion
> having free will?
>
> My point was that, if you have reason to believe that the
> partipants' statements are sensible (and informative, etc.)
> then reasons 1 and 2 can hold, regardless of whether they
> have free will. If you have reason to believe that your
> statements can influence the participants' behavior, then
> reason 3 can hold. So, I can't understand why you say that
> free will is a pre-requisite for serious discussions.
>
> Yes, I know you say that you want to talk with people
> who will stand by and take responsibility for what they
> say. But *why* do you care about that, if there are other
> reasons to engage in a discussion?
Your "necessary precondition ... responsive to what you say"
is relevant, captured by notions of good faith
and intellectual honesty. Without these elements, reason 1,
usually the primary reason, is thwarted. But it doesn't take
free will to meet these conditions -- in fact, it is restraint,
not freedom, that is required. And that goes for responsibility
in general -- ponder "care free".
--
<J Q B>
Daryl McCullough
>The suggestion that there could be reasons as to why I did
>post is not sensible in a world without free will.
That's not true. To have "reasons" for why something
happens or doesn't, you only need causality, not
free will.
Free will to me seems a completely theological
concept. I'm not using the word "theological"
pejoratively, I'm using it in the sense of religious
studies. The concept of free will was invented to
explain how a benevolent God could allow evil things
to happen. According to some theologians, God gave
humans freedom to make choices, so that the evil
done by humans wasn't to be blamed on their Creator.
Granted, the concepts of being "responsible for
one's actions" is important in many legal systems.
But that by itself doesn't mean that it is a coherent
concept. It seems to me that the sorts of distinctions
criminal lawyers make about degree of guilt has no
real basis.
Daryl McCullough
>Your *point* still seems entirely irrelevant to what was at issue.
The issue was your claim that you can't have a serious discussion
with someone without free will. To sharpen that point, let's try
to enumerate the reasons why you might want to have a "serious
discussion" with anybody:
1. You are hoping to learn something about the discussion topic.
2. You are hoping to learn something about the other participants.
3. You are hoping to influence the future behavior of the other
participants.
4. [Insert other reasons here...]
Which reason depends on the participants in the discussion
having free will?
My point was that, if you have reason to believe that the
partipants' statements are sensible (and informative, etc.)
then reasons 1 and 2 can hold, regardless of whether they
have free will. If you have reason to believe that your
statements can influence the participants' behavior, then
reason 3 can hold. So, I can't understand why you say that
free will is a pre-requisite for serious discussions.
Yes, I know you say that you want to talk with people
who will stand by and take responsibility for what they
say. But *why* do you care about that, if there are other
reasons to engage in a discussion?
--
Neil W Rickert
>Neil says...
>>Jim Balter <j...@digisle.net> writes:
>>(reacting to my post)
>>>BTW, it is blatantly dishonest to snip Daryl's justification for his
>>>question:
>>>> Perhaps those
>>>> "forces beyond his control" constrain him so that he only makes
>>>> sensible statements. Perhaps someone with "free will" uses that
>>>> free will to make nonsensical statements.
>>I snipped it because it was entirely irrelevant to the issues
>>that I had raised and that Daryl was disputing.
>You snipped the *point* that I was trying to make.
I readily grant your point.
Presumably Edgar Bergen's dummy is constrained by forces beyond its
control so that that it only makes sensible statements. That does
not lead us to try having a serious dialog with it. On the other
hand, young children often make nonsensical statements, but we do
attempt to have a serious dialog with them.
Neil W Rickert
>Neil says...
>>The suggestion that there could be reasons as to why I did
>>post is not sensible in a world without free will.
>That's not true. To have "reasons" for why something
>happens or doesn't, you only need causality, not
>free will.
But what would causality do for us without free will. We might know
abstractly that the world runs based on causal principles. And we
might have some causal principles encoded in our brains. But the
causal principles encoded in our brains could be entirely bogus.
What we really have to depend on is our own freedom and autonomy to
test these causal principles for ourselves.
>Free will to me seems a completely theological
>concept. I'm not using the word "theological"
>pejoratively, I'm using it in the sense of religious
>studies. The concept of free will was invented to
>explain how a benevolent God could allow evil things
>to happen. According to some theologians, God gave
>humans freedom to make choices, so that the evil
>done by humans wasn't to be blamed on their Creator.
Forget the religious aspects. Instead, look at the philosophical
tradition that connects free will to responsibility.
>Granted, the concepts of being "responsible for
>one's actions" is important in many legal systems.
Its importance goes far beyond legal systems. We rely on it in every
day life.
dzark
("cyberdiction") wrote:
>Don't forget that consciousness and free-will are only concepts about
>reality, not reality itself. If so their meaning depends on the
>context and for example free will has different meaning if you would
>be in the concentation camp and different if you are wondering what a
>car to buy.
DZ
>The same can be said about causality/no free will. Who or what
>is it that perceives the universe unfolding deterministically one step
>after another inexorable step? This is also a conceptualization.
some concepts seems to be eternally multi-meaningful, they are more
sensitive to the actual context than others more harboured in the
material world
....
>In order to obtain free will there must be a sense of self. The "I"
>that uses a mind to think with. The strong AI view was that when
>a computer ran a program that perfectly emulated human behavior
>that a mind would be created in the process because a human mind
>was the result of effective procedures called algorithms.
>
>This is a causal deterministic assumption. It is necessary to deny
>free will in order to consistently hold this computationalist position.
>So this thread is the emperor wearing slightly different clothing
>while discussing is the human mind completely algorithmic.
Yes, some cognitive techniques require to assume such or other meaning
in order to go forward.
>And it is not that far removed from asking how reality came to exist.
>There is an assumption that the universe if deterministic...the next
>moment is pre-ordained and is and effect of the current moment/cause.
>This happens sequentially extending into the future. It is also assumed
>that the universe has a history of successive causally chained moments
>which we call the past. Everything has a cause which produces an effect.
>
>
>As long as it is know that humans think one thing and do another it
>shall be unsafe to conclude that two people who do the same thing
>or demonstrate equivalent behavior have the same reason for doing so.
>And if another situation arises which is superficially similar, the same
>two people may exhibit different behavior because their thinking or
>reasoning differs in weighing the importance of some element of the
>decision process which was less important in the first circumstance.
>
>You can't judge a book by its cover,
>Stephen
>
This unpredictability creates a concept of free will as a
generalization of observation that people can act in a different way
and so we feel because we even don't know everything about ourselves.
()
David Mitchell
<ricke...@cs.niu.edu> writes
>David Mitchell <da...@edenroad.demon.co.uk> writes:
>
>>You don't need "free-will" to make "Decisions". Computers can make
>>decisions, would you say they have "free-will".
>
>Computers don't make decisions. They just follow the rules given to
>them by the programmer.
>
>>don't believe in it.
>
>I take it that I should assume that forces beyond your control
>manipulated your muscles, so as to produce the above line. Since you
>evidently deny any responsibility for having produced that line, I
>see no reason to take it seriously.
>
By my reasoning you did not "freely" produce that text - should I
therefore ignore it? If so, why did you post it?
--
==========================================================================
David Mitchell ===== Visit: www.thehungersite.com
================================
da...@edenroad.demon.co.uk ===== Feed someone for nothing.
==========================================================================
Anders N Weinstein
I'm quite happy to discuss the details of how 'behavior is produced'
that *I* think are relevant to answering this question. I don't myself
happen to believe these details concern mechanisms in the brain. They
would have to be given in the irreducible vocabulary of intention,
belief and the related family of concepts.
But if you could enunciate an account at that level and motivate
the idea that this does characterize what makes it appropriate to "blame
the system and not the programmer", I could certainly entertain it for the
purpose of discussion.
Ultimately though it seems to me a philosophical question not an
engineering one: it asks us to reflect on our concepts, on what we take
to be conceptually necessary for an ascription of blame of the relevant
sense. It does not primarily ask what is technologically necessary to
achieve this.
Neil W Rickert
>Neil says...
>>Your *point* still seems entirely irrelevant to what was at issue.
>The issue was your claim that you can't have a serious discussion
>with someone without free will. To sharpen that point, let's try
>to enumerate the reasons why you might want to have a "serious
>discussion" with anybody:
> 1. You are hoping to learn something about the discussion topic.
> 2. You are hoping to learn something about the other participants.
> 3. You are hoping to influence the future behavior of the other
> participants.
> 4. [Insert other reasons here...]
>Which reason depends on the participants in the discussion
>having free will?
There is a lot of disagreement about what we mean by any of the terms
"want", "hoping", "learn", "free will". However, we would no more
apply "want" or "hoping" or "learn" to an ordinary computer running a
typical program, than we would apply the term "free will". We may
use these terms metaphorically with respect to computers, but we
recognize the difference.
Serious discussion is not at all like the data flow that we see
between computers. It depends on a degree of mutual trust that we do
not expect to find in computers. I might trust the data coming from
a particular computer system, but only because I know its causal
structure well enough, or because I trust the people who provided
that system.
Incidently, when Jim Balter accused me of arguing in bad faith
(wrongly, IMO), that accusation too had to do with trust. Where
there is no trust, there could not be bad faith.
>My point was that, if you have reason to believe that the
>partipants' statements are sensible (and informative, etc.)
>then reasons 1 and 2 can hold, regardless of whether they
>have free will.
But the reason to believe that a person's statements are sensible
(i.e., to place some trust in them), usually goes back to having a
reason to believe that the person is a responsible autonomous agent,
or to having knowledge of causal structure which connects what is
said back to a responsible autonomous agent. In the latter case we
would consider ourselves as having our discussion with the
responsible autonomous agent, rather than with the mouthpiece that
delivered the words.
>have free will. If you have reason to believe that your
>statements can influence the participants' behavior, then
>reason 3 can hold. So, I can't understand why you say that
>free will is a pre-requisite for serious discussions.
>Yes, I know you say that you want to talk with people
>who will stand by and take responsibility for what they
>say. But *why* do you care about that, if there are other
>reasons to engage in a discussion?
I can think of occasions when I might engage in discussion,
but with the other party not taking responsibility. Examples
that come to mind are talking to a clown, or talking to the Eliza
program. But in such cases, I would not be taking the discussion
seriously. It would be something done for amusement, and not for
any of the reasons 1,2,3 that you list.
In case you haven't worked it out, I think all three of your reasons
presuppose free will.
Pierre-Normand Houle
> >> and I go try to find out what is wrong with it and fix it.
> >
> >Whereas you would not attempt to "fix" somebody who made
> >a bad decision. You would try to reason or convince a person.
>
> Yes. This way of dealing with a human (as opposed to a machine)
> is, I think, part of what Dennett would call "taking the
> intentional stance". For humans, we understand their behavior
> in terms of beliefs, motives, intentions, etc. For computers,
> we understand their behavior in terms of state transitions,
> data processing.
>
> But the reason that we take different stances towards people
> and (current) computers is because the *behavior* warrants
> the difference. When a computer deletes important data files,
> I don't start thinking along the lines "What did I do to make
> the computer angry at me? What can I do to make things up with
> it?" We don't start thinking along those lines because it
> doesn't do any good. It doesn't help avoid the next computer
> error. With humans, consideration of motives and feelings
> *does* help to understand (and influence) their behavior.
>
> In separating systems that warrant being treated as "people"
> and those that don't, I don't think that we need to go beyond
> behavior---does it *work* to treat the system as a person?
I think that one could simultaneously locate intentional
attributions at the social-behavioral level and resist treating
them as utilitarian moves.
I could understand the motivation for applying similar pragmatist
methodological strategies when dealing with alternative scientific
theories that must be weighted against each other in a neutral
way. One question could be: does it work (is it useful for making
predictions) to apply the atomic theory of Dalton to account for
chemical reactions generally? If it does, then, using that theory, it
will becomes possible to infer new theoretical facts (or hypothesis,
maybe) about nature (such as the chemical composition of substances,
in this example).
Once the theory is adopted, however, the inferences it licenses
and the experimental and observational procedures its interpretation
determines will then be mastered in an implicit fashion by new
members that are being initiated in the new scientific practices.
It will not be necessary that these initiates be taught
the new facts *as* theoretical facts that depend on pretheoretical
concepts. For it is sufficient that the employment of the new terms,
in quinean observation statements, for instance, be theory laden
in the right way and this will not require that they be explicitly
defined in term of historically more primitive concepts.
Intentional vocabulary similarly does not need to be explained
in term of a more primitive vocabulary only because it is
theory laden and useful. Just as we can come to know as an
objective fact that the earth is round -- as opposed to knowing
this as a useful theoretical claim that helps us navigating and
has to be inferred from other facts that are not geometry-laden,
say -- intentional attributions can be attributions of objective
and primitive properties (such as mind-laden descriptions of
behavior) and not necessarily be inferred from more primitive
observations.
So, the reason for our employment of intentional concepts
in our dealings with each other would not be their utility
for predicting non-intentional facts. The reason would
rather be that we perceive and treat each other in intentional
terms in the first place.
David Mitchell
<zar...@tiscalinet.it> writes
>
> Let us come right now to your experiment.
>
>>My point is this: every decision you make is either due to the pre-
>>existing conditions in your brain (i.e.. what neurones are doing what),
>>or it's due to random noise.
>
>
>Case A) You make the same choice.
> Great! Let us assume that this means determinsm exists.
>Case B) You don't make the same choice.
> This seems to be quite a problem.... From your premises you can't assume
>that this case would prove determinism not to exist!
>
I guess I wasn't clear in my original post - my contention is that there
is no such thing as free-will. The thought-experiment is an informal
one I use to try to convince people - the point being that if each copy
always makes the same decision then there's no "free" about it.
Neil W Rickert
>In article <98o5m9$f...@euclid.cs.niu.edu>, Neil W Rickert
><ricke...@cs.niu.edu> writes
>>David Mitchell <da...@edenroad.demon.co.uk> writes:
>>>BTW, the reason I have put "free-will" in quotes is because unlike you I
>>>don't believe in it.
>>I take it that I should assume that forces beyond your control
>>manipulated your muscles, so as to produce the above line. Since you
>>evidently deny any responsibility for having produced that line, I
>>see no reason to take it seriously.
>By my reasoning you did not "freely" produce that text - should I
>therefore ignore it? If so, why did you post it?
Presumably, by your reasoning, you have no ability to ignore it (nor
to not ignore it). Presumably these things just happen, and there is
no such thing as ignoring.
Similarly, by your reasoning about me, I could not have had the
option to not post it. The suggestion that there could be reasons as
to why I did post is not sensible in a world without free will. That
I posted it is just one of those things that happened, and no reasons
could be involved.
Anders N Weinstein
Daryl McCullough <da...@cogentex.com> wrote:
>>
>
>You claimed that one cannot take seriously
>arguments made by a being without free will.
>It was a nonsensical claim, and my arguments
>showed it to be nonsensical.
This formulation is a little confused: If you can interpret them, you
can of course take seriously ideas you can find in the words that
emanate from Edgar Bergen's dummy. You can also take seriously words
that come out of a teletype, words that are produced by a random
sentence generation program, words you hear in a dream or from a
disembodied voice, words from a spirit medium, from the oracle at
Delphi, etc.
Still Edgar Bergen's dummy does not make any arguments. The idea is that
a being without free will cannot make any arguments at all. Of course
noises can issue from them
>The only necessary precondition to having a serious
>discussion with some being is that what it has to say
>is sensible, interesting and informative, and
>is responsive to what you say.
I grant you can carry on a simulated discussion with any interpretable
source of words. But the teletype or Edgar Bergen's dummy does not
have anything to say.
Gregor Rot
cited author.]
In article <UoVyrAAi...@edenroad.demon.co.uk>,
da...@edenroad.demon.co.uk says...
>
> BTW, the reason I have put "free-will" in quotes is because unlike you I
> don't believe in it. Consider the following thought experiment...
>
I think there is no absolute argument (pro/contra) about the existence of
free-will. The problem is not free-will, but more widely speaking it is the
question of determinism.
> You're about to make a decision, I magically take a complete copy of
> your entire body and environment (it's a thought experiment, I can *do*
> this ;-). You make the decision, so does your copy.
Yes, theoretically everything ok, but the time/space environment that the
copy would live in wouldn't be the same as the original's environment. (you
can't put two copyes into the same space, can you? ;) So, there is the
difference.
>
> If it doesn't, what's different between the two (remember they're
> identical right down to atoms).
Ah, here you get your own self into trouble. In order to say two "things"
are identical their structure have to be identical. Identical atoms you
say. And what are atoms made of? Neurons? Electrons? And what are
neurons/protons/electrons made of? Energy? And what is energy? And what is
energy made of? And what is the smallest particle in the universe? Or are
there questions silly, stupid at all - maybe...if any physic is reading
this newsgroup, a point of view could be very interesting perhaps?
regards,
Gregor
Daryl McCullough
>>>I snipped it because it was entirely irrelevant to the issues
>>>that I had raised and that Daryl was disputing.
>
>>You snipped the *point* that I was trying to make.
>
>I readily grant your point.
Then why didn't you say that?
>Presumably Edgar Bergen's dummy is constrained by forces beyond its
>control so that that it only makes sensible statements. That does
>not lead us to try having a serious dialog with it.
If Edgar Bergen were a psychologist who only "spoke" to his
patients through Charlie McCarthy, I can easily imagine people
having a serious dialog with the dummy. Yes, there is a sense
in which the patient would be having a conversation with
Edgar Bergen, but so what?
In the case we're talking about, a usenet poster claimed
not to believe in free will. I have no trouble having a
serious discussion with such a person. If you want to
say that I'm really only having a discussion with the
forces that "pull the strings" (namely, chemical and
electrical forces in his brain), so what? How does
that make the discussion less meaningful?
>On the other hand, young children often make nonsensical statements,
>but we do attempt to have a serious dialog with them.
>
>Your *point* still seems entirely irrelevant to what was at issue.
You claimed that one cannot take seriously
arguments made by a being without free will.
It was a nonsensical claim, and my arguments
showed it to be nonsensical.
The only necessary precondition to having a serious
discussion with some being is that what it has to say
is sensible, interesting and informative, and
is responsive to what you say.
--
Daryl McCullough
>Daryl McCullough <da...@cogentex.com> wrote:
>>There is no necessary correlation that I can see between free
>>will and the reasonableness of their statements.
>
>You are supposing it is wholly unproblematic that meaningful statements
>are being made, quite apart from their being made by a responsible
>agent who stands accountable for making and supporting them. But this
>is far from clear.
>It is correct that we can interpret the output of artifacts like a
>computer theorem prover as meaningful statements in our language. Such a
>thing might turn up interesting proofs or conjectures when interpreted by us.
>And to assess these conjectures for "reasonableness" does not depend on our
>viewing the artifact having free will or understanding or performing any
>speech acts of asserting anything at all.
>But that is because we can interpret them as meaningful.
That's the *only* sense of "meaningful" that is relevant.
>And this possbility would seem to be parasitic on the existence
>of a meaningful language used by agents who understand it and
>use it responsibly.
Yes, language is only meaningful because there is a community
of speakers who use it meaningfully. But that has nothing to
do with free will, as far as I can see. I don't see any problem
with a community of robots using language meaningfully.
Jim Balter
>
> >> da...@cogentex.com (Daryl McCullough) writes:
> >> >Neil says...
> >> >>da...@cogentex.com (Daryl McCullough) writes:
> >> >>>Neil says...
>
> >> >>>>Your *point* still seems entirely irrelevant to what was at issue.
>
> >> >>>The issue was your claim that you can't have a serious discussion
> >> >>>with someone without free will. To sharpen that point, let's try
> >> >>>to enumerate the reasons why you might want to have a "serious
> >> >>>discussion" with anybody:
>
> >> >>> 1. You are hoping to learn something about the discussion topic.
> >> >>> 2. You are hoping to learn something about the other participants.
> >> >>> 3. You are hoping to influence the future behavior of the other
> >> >>> participants.
> >> >>> 4. [Insert other reasons here...]
>
> >> >>>Which reason depends on the participants in the discussion
> >> >>>having free will?
>
> >> >>There is a lot of disagreement about what we mean by any of the terms
> >> >>"want", "hoping", "learn", "free will". However, we would no more
> >> >>apply "want" or "hoping" or "learn" to an ordinary computer running a
> >> >>typical program, than we would apply the term "free will".
>
> >> >Neil Rickert.
>
> >> >[stuff deleted]
>
> >> I note that you have snipped most of what I wrote. Perhaps you found
> >> it non-relevant. That's fine. I point it out since both you and Jim
> >> were critical when I snipped parts of your earlier posting that I
> >> found non-relevant.
>
> >I wrote
>
> > BTW, it is blatantly dishonest to snip Daryl's justification for his
> > question:
>
> >might not be an issue, but you snipped the context when answering
> >the question of why it is an issue. That has nothing to do with mere
> >snipping, which could simply be that Daryl has no dispute with what
> >you wrote or has already responded to it. And you know this full well,
> >jackass.
>
> This is much like the fact that I had no dispute with the part that I
> had snipped. My response to Daryl in no way depended on that.
The point of his question depended on it, you jerk.
--
<J Q B>
Jim Balter
>
> "Jim Balter" <j...@digisle.net> wrote in message news:3AB05CE9...@digisle.net...> Pierre-Normand Houle wrote:
> > >
> > > ----- Original Message -----
> > > From: "Jim Balter" <j...@digisle.net>
>
> > > > rather than being the sort of free-floating emotional "I want to
> > > > smack the one to blame" medieval anti-intellectual retributional
> > > > primitive reaction that is the alternative.
> > >
> > > This seems to be a false dichotomy to me --
> >
> > *You* are the one who used the word "different", claiming a dichotomy.
>
> the two contemplated alternatives are not exhaustive.
No, it's to submit that they aren't mutually exclusive.
In this case, since I am talking about how *I* take
"blame", your charge is nonsensical.
I'm going to skip the rest because I frankly find statements
such as that one has responsibilities "because I am social
rational animal" to be akin to "colorless green ideas sleep furiously".
You and I apparently inhabit very different conceptual worlds.
--
<J Q B>
Neil W Rickert
>> Jim Balter <j...@digisle.net> writes:
>> >Neil W Rickert wrote:
>> >> >[stuff deleted]
>> >I wrote
Just as the point of my more recent message depended on the context
that Daryl snipped. But you are too dishonest to admit that
possibility, even to yourself.
Pierre-Normand Houle
"Jim Balter" <j...@digisle.net> wrote in message news:3AB33330...@digisle.net...
> Pierre-Normand Houle wrote:
> >
> > "Jim Balter" <j...@digisle.net> wrote in message news:3AB05CE9...@digisle.net...> Pierre-Normand Houle wrote:
> > > >
> > > > ----- Original Message -----
> > > > From: "Jim Balter" <j...@digisle.net>
> >
> > > > > No, it's not, at least not for me -- for me, blame applies to causes,
> > > > > rather than being the sort of free-floating emotional "I want to
> > > > > smack the one to blame" medieval anti-intellectual retributional
> > > > > primitive reaction that is the alternative.
> > > >
> > > > This seems to be a false dichotomy to me --
> > >
> > > *You* are the one who used the word "different", claiming a dichotomy.
> >
> > To charge someone of making a false dichotomy is to submit that
> > the two contemplated alternatives are not exhaustive.
>
> No, it's to submit that they aren't mutually exclusive.
> In this case, since I am talking about how *I* take
> "blame", your charge is nonsensical.
Don't take my word for it. Look up
http://mind.ucsd.edu/syllabi/98-99/logic/falsedichotomy.html
or
http://www.ucs.mun.ca/~alatus/Intro/CTFallaciesExamples&Solutions.html
or do a Google search for "false dichotomy" and "fallacy" to find
other references.
Daryl McCullough
>
>da...@cogentex.com (Daryl McCullough) writes:
>
>>>But what would causality do for us without free will.
>
>
>And I guess I don't know what you mean by causality.
You uttered the sentence
"But what would causality do for us without free will."
which contains the phrases "causality" and "free will". I
assumed that you actually had something in mind as the meanings
of those two phrases.
I was only using causality in the loose sense of
being able to reliably predict future behavior of
a system from a knowledge of its current state and
its environment. My point is that if such prediction
is possible, then you don't need to assume anything about
"free will" in order to "trust" the system.
Daryl McCullough
[Criteria for a meaningful discussion:]
>>>> 1. You are hoping to learn something about the discussion topic.
>>>> 2. You are hoping to learn something about the other participants.
>>>> 3. You are hoping to influence the future behavior of the other
>>>> participants.
>>>> 4. [Insert other reasons here...]
>>>In case you haven't worked it out, I think all three of your reasons
>>>presuppose free will.
>
>>Then I really don't know what you mean by "free will". Can a system
>>whose actions are completely determined by a program have free will?
>
>I guess it depends on what you mean by "completely determined by a
>program." In my opinion, the type of system envisaged by GOFAI, for
>example, cannot have free will.
Then free will is not necessary to meet criteria 1-3.
>>Such a system can definitely meet criteria 1-3, though.
>
>Again, assuming that we are talking about the type system known as
>GOFAI, there is zero evidence that "such a system can definitely meet
>criteria 1-3."
Let's take them one at a time:
1. You are hoping to learn something about the discussion topic.
You don't think that a GOFAI program can print out something
you didn't already know about a topic?
2. You are hoping to learn something about the other participants.
You don't think that you can learn something about a computer
program by interacting with it?
3. You are hoping to influence the future behavior of the other
participants.
You don't think that the inputs to a GOFAI program can influence
its future behavior?
Daryl McCullough
I would like to see that explained. It seemed to me that
the point of your message was
In case you haven't worked it out, I think all three of your reasons
presuppose free will.
which I did not snip. In what way did I miss the point of what
you were saying?
Jim Balter
I never defended Daryl of anything, you stupid dishonest swine.
> But you are too dishonest to admit that
> possibility, even to yourself.
You're too much of an idiot to understand that a tu quoque
argument is fallacious, and stupidly so, since it is
self-reflective.
--
<J Q B>
Jim Balter
Ok, then I've misunderstood the definition. But a) there are examples
there of both types; e.g., "either I keep smoking or I'll get fat"
is not a matter of non-exhaustion but rather of non-mutual exclusivity
and b) your charge is *still* nonsensical -- no claim of exhaustion
was made, I merely contrasted my process of blame with what is common.
And the "alternative" you offered, blaming as a matter of
denying entitlement, is quite the sort of reaction I wrote of,
even if couched in high faluting moral philosophical terms.
--
<J Q B>
Jim Balter
You're too much of a cantankerous idiot to understand that a tu quoque
Neil W Rickert
>You're too much of an idiot to understand that a tu quoque
>argument is fallacious, and stupidly so, since it is
>self-reflective.
Let's set the record straight. Balter's accusation that I
acted in bad faith was an assertion about my intentions. Balter
cannot read my intentions. He can at best guess from the public
record. Since his guess was clearly wrong, Balter's accusation
was clearly wrong.
Similarly, I cannot directly read Balter's intention. I can only
guess from his public behavior, such as his repeated insistence
on making false accusations.
I will not waste any more time trying to point out Balter's errors.
Jim Balter
Good idea, since you are so bad at it.
--
<J Q B>
Pierre-Normand Houle
From: "Jim Balter" <j...@digisle.net>
> I merely contrasted my process of blame with what is common.
> And the "alternative" you offered, blaming as a matter of
> denying entitlement, is quite the sort of reaction I wrote of,
> even if couched in high faluting moral philosophical terms.
But if you don't couch them in disparaging terms there is nothing
intrinsically wicked about attitudes of denying entitlement.
In our societies, people are denied entitlement to park their
cars in some areas or to enter theatres without paying a ticket.
In rational debates people can be denied entitlement to endorse
claims that are incompatible with other claims they are committed
to. A chess players is denied entitlement to castle when his
king is in check.
You might be inclined to explain these attitudes in term of rule
utilitarianism. I prefer to see them as forms of behavior
that are primitive (although not exempts from demands of
justification). They are primitive because the sort of
rationality that is necessary to assess claims of utility is not
even possible without the capability to express these attitudes.
For instance, you might have as a goal to win a game of chess
(You see it as useful for improving your social status, say)
but you have no mean to accomplish that goal if you do not
have a grasp of what moves you are entitled to do and what
moves you are not.
Neil W Rickert
>In article <98u196$8...@euclid.cs.niu.edu>, Neil says...
>>da...@cogentex.com (Daryl McCullough) writes:
>>>>But what would causality do for us without free will.
>>>I guess I don't know what you mean by "free will".
>>And I guess I don't know what you mean by causality.
>You uttered the sentence
> "But what would causality do for us without free will."
>which contains the phrases "causality" and "free will". I
>assumed that you actually had something in mind as the meanings
>of those two phrases.
Right. But you evidently disagreed, and wondered whether I might
have a strange notion of "free will". In turn, I wondered whether
your disagreement might indicate that you had a strange notion of
"causality.
>I was only using causality in the loose sense of
>being able to reliably predict future behavior of
>a system from a knowledge of its current state and
>its environment. My point is that if such prediction
>is possible, then you don't need to assume anything about
>"free will" in order to "trust" the system.
Then you don't have a strange notion of causality. But your
explanation only makes my point. For the term "predict" is itself a
reference to an exercise of free will. We do not normally use the
term "predict" to describe temporal correlations. Rather, a
prediction is an autonomously given willful assertion by the person
making the prediction. It is not part of the meaning of "predict"
that the prediction be accurate. A few months ago we had some people
predicting that Bush would win the presidential election, while
others were predicting that Gore would be the winner. It is part of
making a prediction, that the predictor accept responsibility for
that prediction. This is the same sort of responsibility that we
associate with free will.
Neil W Rickert
Sure it can. But printing out something does not make for a
discussion. We can learn from books too, but we don't say that the
books are involved in a discussion.
> 2. You are hoping to learn something about the other participants.
>You don't think that you can learn something about a computer
>program by interacting with it?
Sure. But this is a change of subject. When we talk of learning
something about the other human participants in an ordinary
discussion, we are not talking about learning of their neural
mechanisms.
> 3. You are hoping to influence the future behavior of the other
> participants.
>You don't think that the inputs to a GOFAI program can influence
>its future behavior?
Sure. But the GOFAI is not a participant in anything we could call a
discussion. I am not having a discussion with my word processor when
I use it to write a memo.
We see discussion with computers depicted in movies (Hal, R2D2 for
example). I am not aware of anything that you could really call a
discussion. The programs submitted for the Turing contest probably
come closest. But they are not good examples of your requirements
(1), (2) and (3).
Daryl McCullough
of free will in having a serious discussion is
this: Are you saying that
(1) Free will is a *prerequisite* to such discussions, or
(2) Any successful effort to create a system capable of
having meaningful discussions will create a system with
free will as a side-effect?
If it is (2), then I don't think that we need to be concerned
about free will in AI---just worry about getting the behavior
right, and the free will comes as part of the package. If it
is (1), then it's necessary to say what you mean by "free will".
It doesn't mean anything much to me, except as part of the
folk psychology of desires, beliefs, intentions, etc.
Daryl McCullough
>
>da...@cogentex.com (Daryl McCullough) writes:
>>I was only using causality in the loose sense of
>>being able to reliably predict future behavior of
>>a system from a knowledge of its current state and
>>its environment. My point is that if such prediction
>>is possible, then you don't need to assume anything about
>>"free will" in order to "trust" the system.
>
>Then you don't have a strange notion of causality.
Okay, so now that you understand what I mean by "causality",
what do *you* mean by "free will"?
>But your
>explanation only makes my point. For the term "predict" is itself a
>reference to an exercise of free will.
I think that you are mixing up the issues. We're talking
about a trust relationship: Person (or system) A trusts
person (or system) B. I claimed that A doesn't need to
assume anything about whether B has free will in order
to be able to trust him (or it).
>We do not normally use the
>term "predict" to describe temporal correlations. Rather, a
>prediction is an autonomously given willful assertion by the person
>making the prediction. It is not part of the meaning of "predict"
>that the prediction be accurate. A few months ago we had some people
>predicting that Bush would win the presidential election, while
>others were predicting that Gore would be the winner. It is part of
>making a prediction, that the predictor accept responsibility for
>that prediction. This is the same sort of responsibility that we
>associate with free will.
We're not talking about whether the *predictor* has free will.
We were talking about whether Neil Rickert (who, for the sake
of argument, we will assume has free will, whatever that is)
can trust David (who, we will assume, also for the sake of
argument, doesn't have free will).
Neil W Rickert
How about:
(3) It is not possible to have a meaningful discussion
with a rigidly deterministic stimulus-response system.
That avoids the term "free will", which is apparently causing
confusion. Note, however, that this is a more limited assertion.
There are systems which are not rigidly deterministic
stimulus-response systems, which we would nevertheless not credit
with having free will.
Neil W Rickert
>In article <991cqv$f...@euclid.cs.niu.edu>, Neil says...
>>da...@cogentex.com (Daryl McCullough) writes:
>>>I was only using causality in the loose sense of
>>>being able to reliably predict future behavior of
>>>a system from a knowledge of its current state and
>>>its environment. My point is that if such prediction
>>>is possible, then you don't need to assume anything about
>>>"free will" in order to "trust" the system.
>>Then you don't have a strange notion of causality.
>Okay, so now that you understand what I mean by "causality",
>what do *you* mean by "free will"?
>>But your
>>explanation only makes my point. For the term "predict" is itself a
>>reference to an exercise of free will.
>I think that you are mixing up the issues. We're talking
>about a trust relationship: Person (or system) A trusts
>person (or system) B. I claimed that A doesn't need to
>assume anything about whether B has free will in order
>to be able to trust him (or it).
I take back my earlier comment. You still might have a strange
notion of "causality".
Reagan used to say (with reference to USSR politicians) "trust but
verify".
There isn't much difficulty in trusting that which you can verify. I
trust my wristwatch, because I expect to be able to easily notice if
it has failed. I trust the computers I administer, but I still
regularly check the logs to make sure that nothing is amiss.
Trusting what you can verify is a rather weak form of trusting.
Generally, when we use the term "trust" for relations between people,
we are using a stronger form of trust, one which does not depend on
verification.
Returning to causality. If I *know* the causal structure of a
system, I will trust it more than in the absence of such knowledge.
But, as I understand the term, "causality" is a claim that a causal
structure exists, not a claim that a causal structure is *known*.
My earlier statement "but what would causality do for us without free
will," was relevant there. The point is that existence of causal
relations does nothing for us. We would need knowledge of those
causal relations. Yet, without free will, it would seem that we
would have no means for acquiring such knowledge.
>>We do not normally use the
>>term "predict" to describe temporal correlations. Rather, a
>>prediction is an autonomously given willful assertion by the person
>>making the prediction. It is not part of the meaning of "predict"
>>that the prediction be accurate. A few months ago we had some people
>>predicting that Bush would win the presidential election, while
>>others were predicting that Gore would be the winner. It is part of
>>making a prediction, that the predictor accept responsibility for
>>that prediction. This is the same sort of responsibility that we
>>associate with free will.
>We're not talking about whether the *predictor* has free will.
>We were talking about whether Neil Rickert (who, for the sake
>of argument, we will assume has free will, whatever that is)
>can trust David (who, we will assume, also for the sake of
>argument, doesn't have free will).
If "David" is the name for my computer, or my wristwatch, or some
other such system, then I can trust it only to the extent that I am
able to verify. As far as I can tell, it would be irrational for me
to trust such systems with the stronger sense of trust that is
applicable between people.
Daryl McCullough
>How about:
>
> (3) It is not possible to have a meaningful discussion
> with a rigidly deterministic stimulus-response system.
I'm not sure what you mean by "rigidly deterministic
stimulus-response" system. Do you mean that (A) the response
is a deterministic function of the previous stimulus,
or that (B) the response is a deterministic function of the history
of all stimuli so far?
If the former, then sure, you can't have a meaningful
discussion with such a system. If you mean (A), then
I would certainly agree, but I would say that GOFAI
is not limited to making such systems. If you mean (B),
then I think you are seriously mistaken. There is no
observational difference between a system of type (B)
and a nondeterministic system.
Daryl McCullough
>How about:
>
> (3) It is not possible to have a meaningful discussion
> with a rigidly deterministic stimulus-response system.
Well, I think that's false. It's certainly not manifestly
true, so if you're going to claim that, you need to provide
an argument.
There is no *observational* difference between a deterministic
system and one that's not deterministic.
Daryl McCullough
>>Okay, so now that you understand what I mean by "causality",
>>what do *you* mean by "free will"?
[stuff deleted]
I'm waiting to hear an explanation of what *you* mean
by "free will".
Neil W Rickert
>Neil says...
>>How about:
>> (3) It is not possible to have a meaningful discussion
>> with a rigidly deterministic stimulus-response system.
>Well, I think that's false. It's certainly not manifestly
>true, so if you're going to claim that, you need to provide
>an argument.
>There is no *observational* difference between a deterministic
>system and one that's not deterministic.
Sure there is.
If you have a piano which is a rigidly deterministic stimulus
response system, you will never have to call the piano tuner.
Neil W Rickert
>Neil says...
>>How about:
>> (3) It is not possible to have a meaningful discussion
>> with a rigidly deterministic stimulus-response system.
>I'm not sure what you mean by "rigidly deterministic
>stimulus-response" system. Do you mean that (A) the response
>is a deterministic function of the previous stimulus,
>or that (B) the response is a deterministic function of the history
>of all stimuli so far?
I mean (B).
>If the former, then sure, you can't have a meaningful
>discussion with such a system. If you mean (A), then
>I would certainly agree, but I would say that GOFAI
>is not limited to making such systems. If you mean (B),
>then I think you are seriously mistaken. There is no
>observational difference between a system of type (B)
>and a nondeterministic system.
Then why is it that there are highly respected physicists publishing
research which reports observing quantum indeterminism? Why are
there highly respectable refereed physics journals that accept such
reports?
Neil W Rickert
>Neil says...
>>da...@cogentex.com (Daryl McCullough) writes:
>>>Okay, so now that you understand what I mean by "causality",
>>>what do *you* mean by "free will"?
>[stuff deleted]
>I'm waiting to hear an explanation of what *you* mean
>by "free will".
To a first approximation:
The ability of a system to autonomously choose actions.
The ability of a system to act in ways that are not rigidly
dictated by the surrounding environment.
The above is not quite right, since there is also an expectation that
the actions chosen will have some degree of rationality.
Daryl McCullough
>>I'm waiting to hear an explanation of what *you* mean
>>by "free will".
>
>To a first approximation:
>
> The ability of a system to autonomously choose actions.
What does it mean to "autonomously" choose?
>The ability of a system to act in ways that are not rigidly
>dictated by the surrounding environment.
By "rigidly dictated by the surrounding environment", do you
mean that the system's actions can't take into account internal
system state? Or do you mean rigidly dicated by the combination
of system state + surrounding environment?
If the latter, then I don't think that we have any conclusive
evidence that any of our choices are the result of "free will"
in that sense.
Daryl McCullough
That doesn't follow. It may go out of tune as a deterministic
function of the pattern of playing it receives.
Daryl McCullough
Because a nondeterministic theory can be much simpler and
more elegant than a corresponding deterministic theory
that makes the same predictions. But for every nondeterministic
theory, there is always a corresponding deterministic
theory that makes the same predictions. For instance,
the Bohm interpretation of quantum mechanics is deterministic,
yet observationally indistinguishable from the Copenhagen
interpretation.
Neil W Rickert
>>>>How about:
>>I mean (B).
Nice evasion.
The issue was about whether the world is deterministic, not about
whether scientific theories that attempt to describe that world are
deterministic. These are far from the same.
Neil W Rickert
>>da...@cogentex.com (Daryl McCullough) writes:
>>>Neil says...
>>>>How about:
>>Sure there is.
I doubt it.