Newsgroups: comp.ai.alife
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!agate!newsxfer.itd.umich.edu!zip.eecs.umich.edu!newshost.marcam.com!news.mathworks.com!news.duke.edu!godot.cc.duq.edu!newsfeed.pitt.edu!dsinc!ub!galileo.cc.rochester.edu!prodigal.psych.rochester.edu!stevens
From: stevens@prodigal.psych.rochester.edu (Greg Stevens)
Subject: Re: "What is Life?"
Message-ID: <1995Feb2.161422.3233@galileo.cc.rochester.edu>
Sender: news@galileo.cc.rochester.edu
Nntp-Posting-Host: prodigal.psych.rochester.edu
Organization: University of Rochester - Rochester, New York
References: <jhansen-120195102007@cetq10.coe.uga.edu> <3gbhj1$klg@engnews2.Eng.Sun.COM> <D38ACu.t7F@ns1.nodak.edu> <1995Jan30.224556.1414@galileo.cc.rochester.edu> <1995Jan30.224310.1099@galileo.cc.rochester.edu> <3gmelc$j40@engnews2.Eng.Sun.COM>
Date: Thu, 2 Feb 95 16:14:22 GMT
Lines: 95

In <3gmelc$j40@engnews2.Eng.Sun.COM> holtz@netcord.Eng.Sun.COM (Brian Holtz) writes:
>In article <1995Jan30.224556.1414@galileo.cc.rochester.edu>,
>Greg Stevens <stevens@prodigal.psych.rochester.edu> wrote:
>>(Brian Holtz) writes:

>>>an individual can 'change' or 'grow' but not 'evolve'.

>>Both "change" and "grow" are the same as "evolve" in the sense that they
>>are adaptations -- just on different time-scales.  Why is it necessary that
>>something go through generational change?

>The requirement of generational change excludes as non-living a lot of
>phenomena -- galaxies, stars, continental plates, weather systems,
>fire -- that do not seem very alive.  

Well, that seems kind of ad hoc to me.  "Well, if we take definition X,
and it includes things we don't want, we'll just add conditions Y and Z
until they are excluded, too."  It seems the kind of process that gives
rise to that bulky and inelegant list of "respiration, growth, metabolism,
etc etc" that people learn in 9th grade.

>However, I'm not sure this
>requirement is "necessary" for our definitional purposes in the sense
>that a deeper distinction could not be made that still excludes all
>these biologically uninteresting kinds of change.

This is how I feel -- it makes sense to me that something in the STRUCTURE
of the organism could be distinguished in determination of whether it is
alive, without resorting to the criterion of evolution, which seems kind
of post hoc to me (we had notions of life long before notions of evolution).

>>Is something has the right structure
>>to adapt and maintain some kind of homeoregulation (maybe homeodynamisis
>>is a closer term than homeostasis -- or you could adopt the notion of
>>"autopoiesis", but in any case, reproduction and evolution are SUBSETS
>>of these more general activities) but that never explicitly reproduces
>>alive?

>If it could reproduce but simply happens never to do so, I would still
>consider it alive.  If it used to reproduce but now just self-modifies,
>and if in fact all members of the population have somehow forever lost
>the ability to reproduce, then it's still alive because it was alive
>when it was born (i.e., the species could still reproduce back then)
>and it hasn't died yet.

>The hard case is some kind of immortal self-modifying entity which is
>truly irreproducible and one of a kind because all attempts to
>build/grow a similar one have been abject failures....

I wasn't even thinking of irreproducABLE, but just not reproducing, not
intended to reproduce, not ever having reproduced.  Perhaps what made it
was a different thing, so in the creation of this thing no REproduction
was going on.  And maybe it can make another one, but just doesn't.  Or
won't.  In a case like that, is this non-reproducing eternal entity alive?

It can't be alive by the argument that it's species NORMALLY reproduces,
because it doesn't.
It can't be alive by the argument that is was created by reproduction,
because it wasn't.

But let's say it is self-building, self-modifying, adapting and maintaining.
Let's say it responds to stimuli in a way such that its structure alters
to maintain a homeodynamisis, and undergoes permanent change so that its
structure reflects its experiential history.  Lets say it repairs itself
and grows even.

Is it alive?

I'd say yes, I think.

>>In some ways I agree more with the general thrust
>>of the autopoiesis / autonomous systems AL camp (more European) than the 
>>reproduction/evolution camps (moer SFI/American).  To them it is the 
>>functional relations within the structure that embody the characteristics
>>of life  -- to them, something that lived forever and never reproduced
>>could still be alive if it had the correct structural properties.

>To me, the appeal of life isn't simply in its structural or functional
>complexity.  To me, the appeal of living things is in what can happen
>after you toss one or two into a dish.  In a few generations the dish
>can be overflowing with 'em.  And in a few epochs they might come back
>and toss _you_ in a dish.  Reproduction and evolution.

Well, I guess, than, different people will define "life" differently
depending on what they find "interesting" about life.  Reproduction
is neat, but life seems to be some much more, in terms of individual
adaptation maintanence and change.  So that's my interests, which is
consistent with how I define life, just as your personal interests
are consistent with how you define it.  Maybe we should make two different
words.....

Greg Stevens

stevens@prodigal.psych.rochester.edu

