Newsgroups: comp.ai.alife
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!bloom-beacon.mit.edu!news.media.mit.edu!minsky
From: minsky@media.mit.edu (Marvin Minsky)
Subject: Re: The Meaning of Life
Message-ID: <1994Sep11.025940.21054@news.media.mit.edu>
Sender: news@news.media.mit.edu (USENET News System)
Organization: MIT Media Laboratory
References: <34qm2c$ei2@scratchy.reed.edu> <SWRA01.94Sep11122615@cs19.cs.aukuni.ac.nz>
Date: Sun, 11 Sep 1994 02:59:40 GMT
Lines: 58

In article <SWRA01.94Sep11122615@cs19.cs.aukuni.ac.nz> swra01@cs19.cs.aukuni.ac.nz (Stephen David  Wray) writes:
>
>>    Why is a definition of life important?
>
>Well, that depends very much on who you talk to :)
>
>I recently gave a seminar on the need for real efforts to be made in
>the development of genuine AI's, given the great rate of growth in
>complexity of computer systems, and the sensitivity of the tasks
>allocated to those systems.
>
>Among my suggestions were that learning systems were very important,
>and that emergent properties of personality and emotion should be
>sought, since these would (hopefully) lead to some kind of
>responsibility and/or culpability on the part of the artificial system
>itself, rather than on the part of its creator.
>
>Anyhow, I had one major critic -- most of the rest of my audience
>seemed to appreciate the  point I was making. The critics comments
>were entirely definitional -- "Give me a definition of what you mean
>by intelligence", "How do you define consciousness" etc -- all this
>*despite* the fact that I had already cited the Turing Test and
>Descartes Dreamer as examples of situations where objective criteria
>cannot be applied, yet where decisions must be made -- and hence,
>where  subjective appraisals should be seen  as valid...
>
>Anyhow -- this person turns out to be, basically, an ARISTOTELIAN.
>In a computer science department... wonders will never cease.

This happens to me all the time.  And it is ruining this originally
promising group!  The idea that "life" is important to define comes
from, I think, the idea that there's really some very special thing --
essence, spirit, vital force, etc., that animates the objects that we
call 'alive' in informal language.  This theory, called "vitalism",
was endemic until around the time of Pasteur, and no longer plays any
role in scientific research, either in biology or anywhere else.
Perhaps the trouble is that all the "alive" things we've learn to
accept (until the dawn of a-life, that is) did indeed have quite a lot
of common properties -- e.g.,
	metabolism (carbon-based, of course)
	self-reproduction (at least on the part of their parents)
	reasonable adaptivity (except for the non-survivors)
	etc.

However, these features are not essential in describing other
interesting features of the instances of terrestrial biology.
Instead, they're mere consequences of the common origin of the
terrestrial evolutionary scene.  And, obviously, are quite beside the
point for building new forms of a-life, or artificial inteliigence,
etc.  So please, please, stop wasting so much time.  (By the way, there was a
nice short discussion of this in the introduction of my old textbook,
"Computation: Finite and Infinite Machines."   I'll upload a copy of
that section, if I can find a copy.  Prentice-Hall seems to have let
it go out of print.)

-- marvin minsky

