Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site rlgvax.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!mcnc!philabs!cmcl2!seismo!rlgvax!guy
From: guy@rlgvax.UUCP (Guy Harris)
Newsgroups: net.physics
Subject: Re: Re: Re: why FTL is illegal (wrt: free will).
Message-ID: <284@rlgvax.UUCP>
Date: Fri, 7-Dec-84 00:48:51 EST
Article-I.D.: rlgvax.284
Posted: Fri Dec  7 00:48:51 1984
Date-Received: Sun, 9-Dec-84 06:35:24 EST
References: <683@gloria.UUCP> <785@ariel.UUCP>  <148@lems.UUCP> <152@talcott.UUCP> <277@rlgvax.UUCP> <171@talcott.UUCP>
Organization: CCI Office Systems Group, Reston, VA
Lines: 43

> > But that just says that the universe just appears random because we haven't
> > looked in great enough detail.  If you assume 1) a deterministic and
> > complete theory of how the universe works, 2) 100% no exclusions complete
> > knowledge of the initial state of the universe, and 3) enough computing
> > ability to crank the model forward from that initial state, you can predict
> > all future states of the universe.
> 
> A big milestone in weather forecasting was when they computed twenty-four
> hours of weather in twenty-four hours.  The thing about a non-linear system
> is that it may take more time than you have to predict its future...
> In any case, we would have to know the position of every molecule in the
> atmosphere, also an impossible task (even in classical mechanics).

A-*HEM*.  We're not talking about whether it's *practical* to predict the
future perfectly; we're talking about whether it's possible in principle.
The original discussion was about "free will", and the original poster asked
whether quantum mechanical uncertainty was responsible for it.  Somebody
else responded saying that considering the brain as a classical non-linear
system could also explain "free will" as well.

Well, just because I don't have every line of code in all the software
running on this machine memorized doesn't mean I think it does things out
of "free will"; I could, if I felt like it (and could summon up the patience),
do a detailed simulation of "rlgvax" running the software now on it and
predict what it does.  (For you CS types out there, think of the difference
between wanting to compute the value of an uncomputable function at some
arbitrary point in its domain (you *can't* - no algorithm exists) and trying
to solve an NP-complete problem (you can, given enough cycles, but there seems
to be no polynomial-time algorithm).  It may not be an important *practical*
distinction in all cases, but it *is* an important philosophical distinction.)

Frankly, I think appeals to QM, non-linear systems, or any other physical
principle to "explain" "free will" are just handwaving.  All statistical
theories say is that you can't predict exactly what a system will do, either
in principle (QM) or in practice.  They don't say that the system can
"choose" what to do - anybody who can put the proposition "This system is
choosing how it will react to a stimulus with its own free will" in terms
amenable to scientific investigation deserves a prize (and if they can
experimentally demonstrate the truth or falsity of that proposition, they
will probably get several).

	Guy Harris
	{seismo,ihnp4,allegra}!rlgvax!guy