Newsgroups: comp.ai.alife
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!uhog.mit.edu!bloom-beacon.mit.edu!eru.mt.luth.se!news.luth.se!sunic!news.chalmers.se!news.gu.se!gd-news!d6203
From: sa209@utb.shv.hb.se (Claes Andersson)
Subject: Re: Lamarckian Evolution
Message-ID: <1995Feb16.182951.12856@gdunix.gd.chalmers.se>
Sender: usenet@gdunix.gd.chalmers.se (USENET News System)
Nntp-Posting-Host: d6203.shv.hb.se
Organization: Dept. of economy and computer science.
X-Newsreader: News Xpress Version 1.0 Beta #2.1
References: <1995Feb8.100831.10995@gdunix.gd.chalmers.se> <3hc1u2$dl0@laplace.ee.latrobe.edu.au> <1995Feb12.154650.7235@gdunix.gd.chalmers.se> <3hms5m$jor@laplace.ee.latrobe.edu.au>
Date: Fri, 17 Feb 1995 00:49:23 GMT
Lines: 112

khorsell@ee.latrobe.edu.au (Kym Horsell) wrote:
>In article <1995Feb12.154650.7235@gdunix.gd.chalmers.se> sa209@utb.shv.hb.se (Claes Andersson) writes:
>> I don't doubt that the envirinment affects the DNA but this has nothing
>>to do with the design of new sequences. It has to do with that a simple
>>difference in the cytoplasma can affect the transciption of DNA but it
>>do in no way CALCULATE how to do it and will always be constrained
>>to a couple of solutions. What do you think would happen if you encountered
>>the E. Coli's with something else? Would in "calculate" the appropriate
>>gene sequences for dealing with other chemicals as well. If it does THAT
>>would be strange.
>
>"I don't doubt that environment affects DNA" is a start.
>I think you are maintaining that such DNA changes can't possibly be directed.
>Perhaps this is the only difference between our positions at present.
>I would say that you can't rule out the possibility of "directed mutation",
>despite whatever experimental work from the c1940 that might say mutations
>are  invariably of this type. ;-)
>
>An interesting idea along the lines of "directedness of  mutations"
>is to write one of those simple self-replicating programs. How many
>"mutations" that can be made to the program or to its "chromosome(s)"
>immediately causes death in the given cell or eventually sterility
>in its children? From simple calculations on a few simple such programs
>it seems that most mutations are immediately fatal or lead to sterility.
>In a sense random mutation _must be_ directed in that only a relatively few
>are viable.
>
>If you then place a "real" organism in a changing enviornment it's hard
>to say that some mechanism hasn't evolved to direct errors away from the even
>more likely possible fatal mistakes. E.g. we know that error avoidance (i.e.
>redundancy at various points) and DNA repair exists and _this_ had to evolve
>(we presume). Did evolution of the mechanism stop there? I have already
>tried to show there is an _advantage_ to Lamarckian evolution in at least
>one case...

 I agree that some sort of directed mutations very well can exist but it has nothing
to do with Lamarck. Lamarck="Stretch your neck, and a longer neck will be ancoded
in your genes." That is mathematically impossible. It is just as possible as that a
whale would materialize in orbit around Earth. Why? How do what know what happens
when you stretch your neck, or your back, or bend you legs etc. etc. Something that
encoded such things from the stimuli: the stretching, to the reaction: a genetic sequence
for a longer neck. The correlation is so tremendously complex that even the brain which
is very prominent in size, has a hard time to figure it out.


>
>>I also understand in which way you suggest that a Lamarckian
>>mechanism would appear and I am sure it would, if it would be possible.
>>But I can assure you, it isn't. I refer to what I've written above.
>
>I'm not sure your assurance isn't based on circular reasoning.

 No it isn't, it's based on the complexity of the task. It would be stupid to
suspect that such a mechanism exists since nothing points towards it.

>
>> Do you suggest that when you stretch your back, the intricate little
>>effects this has on each cells, a HUGE amount of cells, is interpreted in
>>such a dynamic way that it can see the difference between when you
>>stretch your back and when you bend to the left and encode DNA
>>accordingly. This would be a calcualtion of tremendous size, and you
>>suggest that there wouldn't have to be any central processing unit
>>that is visible? If the brain, which can in no way handle such operations
>>is very noticeable with its great volume, how big is the chance that
>>somthing of even higher computational power would be microscopic?
>
>Rather a large straw man. (In any case there need be no "central
>processing unit".)

 it would, something would have to understand that the back was stretched
and the design the appropriate genes. No straw man, that's what Lamarckism
is about. If you support it, you're not only probably a fossil, you have also
failed to see why it was discarded long, long time ago.

>
>Your position is that "Lamarckian evolution is impossible" either logically
>or theoretically. All I have to do is argue that it exists somewhere, no
>matter on what scale, to disprove that. Since the main work is done on bugs,
>let's stick to that. I understand even the biology and genetics of single
>eukaryote cells isn't terribly well understood in any case, let alone
>complete organisms or populations thereof.

 Yes, just like you get stuck when discussing with creationist and they say:
"You can't attack me because you can't falsify a deity! The result is: I can
attack you but you can't attack me and it doesn't matter how utterly weak
my attacks are, if I go on long enough I will defeat you!"

 The problem is that this isn't the way science works... You don't verify something
but saying that it is unfalsifiable. That has nothing to do with truth, it has
something to do with someone wanting to believe in something.

>
>If Lamarckian evolution exists at the cellular level then there
>may be an observable effect for an entire multi-celled organism. But I'm
>not sure Lamarck's original examples or others of the same "non-small"
>type are relevant. If there is nothing observed along the lines of back
>arching (or whatever, as per above) it doesn't disprove the case for
>Lamarckian mechanisms. Since starvation responses appear to be
>a basic case, I would argue it would be more likely that Lamarckian
>evolution effects metabolic rates or reproduction rates than the length of an
>animal's neck.

 Your'e on the right road.. There is a difference you see.. It's probably possible
to "toggle" as gene for controling metabolism levels etc. And I don't doubt that
this could be made in a "Lamarckistic" way. But reconstructing the body is
soething else and without no hesitation whatsoever I can tell you: Lamarckistic
evolution as Lamarck meant it don't exist anywhere.


Regards,

Claes Andersson. University of Bors. Sweden
