Newsgroups: alt.christnet.philosophy,comp.ai.philosophy,sci.philosophy.meta,talk.philosophy.misc
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!Germany.EU.net!EU.net!sun4nl!cwi.nl!olaf
From: olaf@cwi.nl (Olaf Weber)
Subject: Re: The Search For Truth
Message-ID: <DA42K4.2Ar@cwi.nl>
Sender: news@cwi.nl (The Daily Dross)
Nntp-Posting-Host: havik.cwi.nl
Organization: CWI, Amsterdam
References: <mike.799620809@mik.uky.edu> <3qnqmmINNljl@oasys.dt.navy.mil>
	<fXe0vs9YeDYO083yn@io.org> <rjk.802389627@laraby.tiac.net>
	<3r2i32$iea@charnel.ecst.CSUChico.EDU> <3r2pcp$7rr@news.nd.edu>
	<Pine.LNX.3.91.950607085100.959B-100000@linux1.ph.utexas.edu>
	<D9wCDH.B6L@cwi.nl>
	<Pine.LNX.3.91.950609092302.14087A-100000@linux2.ph.utexas.edu>
	<DA25H9.M7q@cwi.nl>
	<Pine.LNX.3.91.950612093737.579A-100000@linux1.ph.utexas.edu>
Distribution: inet
Date: Tue, 13 Jun 1995 12:26:33 GMT
Lines: 140
Xref: glinda.oz.cs.cmu.edu comp.ai.philosophy:28809 sci.philosophy.meta:18788

In article <Pine.LNX.3.91.950612093737.579A-100000@linux1.ph.utexas.edu>,
Stephen Froehlich <froehlik@physics.utexas.edu> writes:
> On Mon, 12 Jun 1995, Olaf Weber wrote:

I get the impression that you are equivocating between two points of
view.

The first is an "interface theory" which posits the will as a cause of
quantum fluctuations.  The second is merely wanting non-determinism in
the way the brain works.

My point is that in the first case there should be a difference
between the probability of quatum fluctuations happening depending on
the presence or absence of will.  Thus, such an interface theory is in
principle testable.

The main problem with interface theories is that they do not explain
where will itself comes from.  Once you start to address that issue,
you end up again with thying to explain will as a product of causation
and chance.

> What I've been trying to say all along, is that something that might
> look random to our experiments might not be.

And what I'm trying to say is that if something appears to be random
to _all_ conceivable experiments, it _is_ random, and that if some
process has more structure than our theories predict, that experiments
can be designed that demonstrate this.

>> Suppose we know enough of the brain to determine that fluctuation F
>> has to be "magnified" for a subject to think thought T.  We repeatedly
>> ask the subject to think T.  He could only do that reliably if he can
>> influence the probability of F.

> That is assuming that the only cause of thought T is fluctuation F.

You can drop this assumption and come up with two cases: T occurs
because of F, and T occurs because of stimulation by other neurons.
You then have to set up the experiment so that only the first cases
are counted.  Difficult, but not a priori impossible.

>> A more feasible experiment would involve quantum fluctuations
>> outside of the brain, on the hope that the ability to influence
>> events in the brain doesn't stop at its limits.  You take a random
>> number generator that depends on quantum fluctuations for its
>> functioning, and calibrate it to produce a certain distribution of
>> random numbers.  The experiment then consists of seeing if someone
>> can change that distribution by "wanting" to do so.  (Some
>> parapsychological experiments are of this form.  The results so far
>> are equivocal at best.)

> Exactly what kinds of patterns would you be looking for?

In this case, a deviation of the observed distribution of the random
numbers produced by the random number generator from the distribution
it was calibrated to produce.

> The distribution of quantum fluctuations is mathmatically white
> noise.  If you change a parameter on white noise, you still get
> white noise.  I doubt we'd see any kind of spectrum changing event
> as a function of the will.  What kind of change were you referring
> to?

Either a spectrum changing event or some other change in the
probability distribution as a function of the will.  After all, what
does it mean to give the will as a cause for quantum fluctuations, if
not that in the absence of will they occur in a different pattern?

And yet you also state that you doubt to see such changes, which is
why I consider you to be equivocating on the issue.

>> Could you be clearer on the "upredicated" bit, I'm not sure I
>> understand what you mean.

> I meant that the will is a causam sunam, or an ultimate cause.  The
> will is capable of acting without provocation, though it is usually
> in a mode where it is reacting to circimstances.

This seems again to argue for an interface theory: will as an uncaused
(or at least unexplained) causer.

If all you argue for is non-determinism in the working of the brain,
then you have to accept that will, as a product of those workings, is
the product of causation and chance.

>> I do not think that determinism is incompatible with an ability to
>> choose, although it might violate some of your intuitions about
>> choice.

> This is intriguing, can you explain further?

How could a (deterministic) algorithm for making choices work?  The
crucial observation is that since some choices are better than others,
the mechanism behind making choices should systematically favour the
better ones.

Suppose we have have several courses of action to choose between.
Some possibilities are more "attractive" than others: if there is a
clear winner, that's the one chosen, if there isn't, some kind of
tie-breaking must be done.  The tie-breaking algorithm could have some
randomness in it, but the equivalent of a pseudo-random number
generator would perform as well.  Randomness in other parts of the
algorithm would be detrimental: it would increase the chances of a bad
choice being made.

The "product" of the algorithm would be the will to follow a certain
course of action.  So in this view, will doesn't determine choice;
instead choice focuses the will.

The "attractiveness" of a course of action depends on how well it
seems to fulfill the desires of the chooser.  These desires in turn
derive from its (biological) needs, combined with some information on
how well these needs are currently satisfied.

Possible courses of action are suggested by the needs: look for food,
or water, or a mate.  The more intelligent the chooser is, the more
possible choices will it come up with.  From the chooser's point of
view, this increases its freedom of action, even if the actual choice
is made by a deterministic algorithm.

>> The brain has to be it, granted, but there is little reason to
>> believe that the brain actually has this ability.  Neurons seem
>> fairly impervious to quantum effects.

> I really am just speculating here, but I'd like to run the numbers
> some day.  What kind of voltages does it take to activate the most
> sensitive neurons, especially when these are applied only across the
> nerostimulant receptors at the dendrite end of things?

Communication across the synaptic gap is done by an exchange of ions.
This is basically a (bio)chemical reaction.  AFAIK the number of ions
that has to be "received" is fairly large, and as soon as that number
is significantly more than 1 we have that the probability of a quantum
fluctuation setting off a neuron by itself is exceedingly small.

That the action of neurons is insulated from such quantum effects
should not come as a surprise: no animal can afford that its neurons
fire at random.

-- Olaf Weber
