Newsgroups: comp.ai.alife
From: stevem@comtch.iea.com (Steve McGrew)
Subject: Re: Proposed test for life
Organization: New Light Industries, Ltd.
Distribution: world
References: <32250D9F.5268@gate.net> <507nfs$it0@plains.nodak.edu> <ncWZXBAyFBTyEwdd@wandana.demon.co.uk>
X-Newsreader: News Xpress 2.0 Beta #2
Date: Mon, 30 Sep 96 13:27:32 GMT
NNTP-Posting-Host: spk0a-1.iea.com
Message-ID: <324fcab6.0@news.iea.net>
Lines: 71
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news4.ner.bbnplanet.net!cam-news-hub1.bbnplanet.com!news.mathworks.com!newsfeed.internetmci.com!news.iea.net!sluggo

>In article <324B692B.524D@gate.net>, Howard Gardner <hgardner@gate.net>
>writes
[clip]
>More importantly, I wish to ensure that a virus will pass the life test
>but a bush fire will fail to pass.
[clip]
>I am going back to repro and metabolism,... If you can prove the
>existance of these things then you have life.

        Reproduction is too easy: lots of computer programs are able to 
self-reproduce; there are plenty of artificially constructed molecules that 
can self-reproduce.  If metabolism is taken to mean that energy (or something 
analogous thereto) is extracted from transformations induced in elements of 
the environment, and/or that elements of the environment are transformed in 
order to carry out the process of reproduction, then there are computer 
programs pass that test as well.

>I now want to concentrate on WHAT is the little bit of something that
>makes me KNOW that the pixels and memory variables in my ant simulation
>is NOT life.. because I sure as hell cannot believe that it IS life.

        It's not fair or reasonable, is it, to start with a distinction then 
seek a definition that makes the distinction?  It's like starting with the 
assumption that the Earth is the center of the universe and looking for a 
description of the planetary motions consistent with the assumption.  Sure 
enough you can come up with a description, and even a physical theory that 
predicts those motions.  But the theory won't pass the test of Occam's razor.  
It turns out to require far too many a priori constants.  Newton's theory of 
gravity started with a very simple model/definition that ended up predicting 
the planetary motions with high accuracy, with essentially two constants: the 
Sun's mass and the gravitational constant.

        In seeking an answer to the question of "what is life?", I think it is 
best to look for simple definitions of life, and accept a definition that says 
*most* of the things we agree are alive are indeed alive, and distinguishes 
between those "certainly alive" things and the things we agree are *not* alive 
(e.g., a free, noninteracting photon).  The distinction can be a matter of 
degree, or it can be a binary classification -- but the basis for the 
distinction *must* be very simple or it will be very weak.

        I personally lean towards a definition something like this:  

        "Life is the capability of a system for dynamic persistence."  

        Dynamic persistence would mean the tendency to continue existence and 
functionality in the presence of a changing environment.  A non-reproducing 
organism that metabolizes (say, a castrated male frog) would be alive even 
though it can't reproduce.  A frozen bacterial spore would be alive even 
though it neither reproduces nor metabolizes, because it persists and is ready 
to begin functioning again as soon as it thaws out and finds itself in the 
presence of water and nutrients.  A brush fire or a prion would have a much 
lesser capability for dynamic persistence, and would therefore be less alive.

        A population of organisms capable of reproduction, metabolization and 
evolution would be *more* alive because it is more dynamic and because it is 
able to adapt to radically changing environments.

        I know that a lot of folks are vehemently opposed to any definition of 
life that would attribute "degrees" of life to things, or would attribute any 
degree of life to a brush fire.  But I think their opposition is based more on 
something like religion than on scientific reasoning.  They seem to be 
starting out with a severe, unfounded constraint: "Whatever the definition of 
life turns out to be, it has to classify computer programs and brush fires as 
not-alive."

=========================================================================
| Steve McGrew               |          stevem@comtch.iea.com           |
| New Light Industries, Ltd. |          Phone: (509) 456-8321           |
| 9713 W. Sunset Hwy	     |          Fax: (509) 456-8351             | 
| Spokane, WA 99204 USA	     | 		http://www.iea.com/~nli         |
=========================================================================
