On Wed, 4 Jun 2003 04:09:42 +0000 (UTC), jorn{at}enteract.com (Jorn
Barger) wrote:

>This message is xposted to sci.bio.evolution, comp.ai.alife,
>and comp.ai.nat-lang.  S.b.e is moderated, so the comp groups
>may want to trim followups.
>
>The main audience I'm addressing is the alife community, and
>my question for them is a survey:  How many of you know, off
>the tops of your heads, which group of animals was the first
>to evolve nervous systems?  (Hint: despite my subjectline,
>it's the next group _after_ sponges.)
>
>I'll answer below, but first let me explain where I'm coming
>from.  One way to describe my life's work is that I've been
>searching for an improved classification scheme for Roget's
>1852 Thesaurus.  More than half of Roget's categories were
>explicitly psychological, so these have been the focus of
>my search.  Surprisingly, there have been almost no efforts
>along these lines, even though Darwinian thinking should
>have revolutionised it long ago.
>
>I see my approach as A.I.-like, with Lenat's Cyc Project [1]
>as the closest approach... but Lenat has explicitly omitted
>the problematic psych-categories from his ontology.  But
>while discussing this last week in a comp.ai thread [2] I
>coined a new word "etymogeny" for the idea of re-arranging
>Roget in the Darwinian sequence by which those categories
>historically evolved.
>
>Along with Roget, another of my touchstones is Polti's
>1900 "36 Dramatic Situations"-- Polti was a French drama
>critic who sorted out the great themes of the theater into
>36 classes, which like Roget are obviously inadequate, but
>never bettered.  So the etymogenic approach should also
>ask, for each stage of evolution, what 'dramas' are played
>out by the creatures at that level?
>
>Starting with the simplest bacteria, one major drama was
>"to eat or not to eat".  Some molecules were okay to let
>thru the cell-membrane, but others were poison.  Great
>hunger increased the drama, and mixtures of good and bad
>created special problems.  Even at this pre-nervous level,
>there was an emerging sense of 'personality'-- some were
>cautious and some took risks.
>
>A related drama was 'eating' DNA-- in the sense of adding
>it to your own genome, rather than digesting it.  Perhaps
>the immune system originated here?
>
>The evolution of cilia and other forms of locomotion added
>the new dramas of approach and avoidance.  Ciliated
>_multicellular_ structures brings us to the sponges, who
>use their cilia for catching food.  Oddly, their dramas
>also included differentiation of sperm and eggs, and the
>challenge of avoiding self-fertilisation because they
>were hermaphroditic (the first taboo?).
>
>And finally we get a nervous sytem with coelenterates,
>eg the jellyfish, allowing fine control of muscle
>contractions, and the evolution of sense organs.
>
>The digital nature of nerves meant muscles could be
>fired in any chosen sequence, which introduced the
>whole A.I.-domain of planning, and expectations, and
>modelling of the world. (Cf my 'cichlid-challenge' to
>the comp.ai groups: [3] below.)
>
>My understanding is that at this point the line split
>into crawling creatures and swimming creatures.  The
>former migrated their nervous sytems to their bellies,
>and kept their basic body-structure flexible, with
>shells added on the outside-- the invertebrates.
>
>The swimmers discovered that swimming was most
>efficient if they had a stiff rod running down their
>length, and migrated their nervous system to run
>inside it-- the notochord.  This body plan dominated
>the seas, but was beaten at first on land by the
>invertebrates.
>
>Even before crawling out of the seas, the fishes
>seem to have explored courtship behaviors and
>possibly territoriality.  Once on land, the reptiles
>began caring for their eggs, and the mammals took
>this drama in many new directions by caring for the
>youngsters themselves.
>
>>From here, complex social behaviors evolved, and
>eventually language... that's the outline of my
>proposed 'etymogeny'.
>
>
>[1] Cyc Project: http://www.robotwisdom.com/ai/cyc.html
>[2] Longish comp.ai thread on Semantic Web, etc:
>http://groups.google.com/groups?threadm=bb9gpu%246bg%241%40mulga.cs.mu.OZ.AU
>[3] Short cichlid challenge thread:
> http://groups.google.com/groups?th=b899fd4891d44838

Your scenario makes for some interesting reading but unfortunately it
is quite incorrect -- maybe plausible sounding but not in keeping with
the facts.  If you carefully pick and choose selected types of
organisms, your story seems appropriate but you ignore the innumerable
counter examples.

Your ascribing human characteristics to bacteria (anthropomorphisms)
is a nice literary conceit, but not terribly good science.  Whatever
the degree of "risk-taking" among bacteria, it seems now that there
was a tremendous amount of DNA sharing -- horizontal gene transfer as
it is called.  This would have been long before the immune system.  In
any event, I don't see how you find anything immune-system-like in
assimilating consumed DNA.  

You have a direct linear trail from bacteria to cilia, then ciliated
multicellular organisms (sponges) with differentiated sperm and egg,
to nervous systems in coelenterates (now called Cnidaria) which are
purely digital, then a split between the vertebrate line of swimmers
and the invertebrate line of crawlers and so on.

The problem is all the side lines.  Motility has three very different
sources.  Bacterial flagella are quite distinct from eukaryote cilia
and flagella and other tubulin forms of motion (spindle apparatus)
which are quite distinct from muscle contraction and other actin forms
of motion.  Many single celled organisms use cilia/flagella for
feeding. Many organisms (including all plants!) have differentiated
sperm and egg often with mechanisms to avoid self-fertilization.  The
ability to sense (to detect the environment), to respond appropriately
(to act), and to connect the two with some sort of "signaling
apparatus" (to coordinate and control) long predates multicellular
animals.  Nervous systems are not fundamentally digital -- the primary
processes of cell signaling which form the basis of neural computation
are all analog.  Probably the original nervous systems were purely
analog and the digital action potential came later.  Even today, a
tremendous number of the cells in our own brains do not make action
potentials and function purely using analog computation. And, of
course, the split between swimmers and crawlers is quite incorrect.
The earliest chordates (our own kind) were quite innocuous creatures
who couldn't even swim. They just attached themselves to a rock and
sucked in water.  Many animals besides reptiles care for their eggs or
for their young.

It is interesting and perhaps even useful to try to organize complex
ideas (like a thesaurus) in systematic ways. Just don't try to claim
that biology plays a real role in your system of classification.
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 6/5/03 10:57:08 AM