From: Chris Taylor (christ@ebi.ac.uk)
Date: Thu 19 Feb 2004 - 11:50:58 GMT
>> From: Chris Taylor <christ@ebi.ac.uk>
>> Reply-To: memetics@mmu.ac.uk
>> To: memetics@mmu.ac.uk
>> Subject: Re: memetics-digest V1 #1480
>> Date: Wed, 18 Feb 2004 09:52:24 +0000
>>
>> And let's not forget that organismal form is only the result of the
>> interaction between genome (plus info in cellular components) and _the
>> environment_ which is subject to change, confusing any attempt to
>> preserve form.
>>
>> The genome is an important environment, for genes -- foreign DNA from
>> retrowhatevers, conflicts in silencing between the sexes, the constant
>> trade-offs between suboptimal compromises (neutral evolution at the
>> phenotypic level -- shorter legs, but longer neck), fiddling with
>> expression through chromosomal rearrangements, accidentally finding
>> yourself part of a new species because of a Wolbachia infection. All
>> these things (and a metric tonne more) confuse the preservation of form.
>>
>> What you get out of evolution is a real mish mash, that works. There
>> are about a zillion constraints, most we'll never know about, and yet
>> some organisms still manage to do flashy things like encode genes on
>> both of the complementary strands of the same stretch of DNA, or
>> overlaid in more than one reading frame.
>>
>> Lesson One: This is really goddamn complicated, and we have a
>> measurement problem. Sounds a lot like memetics.
>>
>> Fact we can borrow from biology #1: Selection on individuals is not
>> appropriate, nor selection on kinds (group/species selectionist);
>> evolution happens (to repeat Scott) in populations. For instance I'm
>> sure when talking about French, Canuck French, French creoles etc. it
>> is a lot easier to talk about populations of speakers than of
>> individuals, or of 'French'.
>>
> For biological evolution I tend to be biased towards individuals as the
> target of selection, a bias which I think I share with Mayr. Dawkins is
> a genocentrist so he would opt for the gene as the target of selection.
> Others have argued for group and species level selection.
>
> Genocentrists would, if they followed a Kantian approach to the
> argument, point to individuals as fleeting appearances and the "immortal
> coil" genes as *ding an sich* or the true reality of the matter. To put
> it in Schopenhauerian terms, organisms are a Mayan illusion and the
> selfish genes are the Will.
>
> Nonetheless selection "looks at" organismic phenotypes and is more or
> less blind to genes.
I think I'd try to avoid being pinned on this one -- mostly I see
populations of genes as the 'stuff' to measure, however epistasis and
pleiotropy constitute one of the most important features of having
individual organisms, demarcating what is possible, and are the main
reason why you have to take The Selfish Gene with a big pinch of salt.
Genes may have cover through n > 1 ploidy, which is also important,
especially for neutral evolution, and of course the packaging into
individuals, to finally be selected, as a whole, is important for a
second time; but as you say there are many levels (chemistry -> ecosystems).
(Nearly) neutral drift is a huge thing -- populations drift through
(nearly) neutral networks, by diffusion (neutrality also occurring on
several levels), allowing 'background' change to prepare for rapid
'visible' change (punctuated model allowing rapid change from genes
mostly of small effect), or the accumulation of variation as a resource,
or even speciation and stuff.
The reason I mention such drift is because, in addition to being the
main reason why eyes go eventually in caves (selection for efficiency is
a slightly fallacious idea here, because drift _is_ the process that
produces the mutations that would be selected for, so drift versus
selection against is just a matter of degree -- I think if having eyes
was an actual hazard, things would go quicker than they would have
anyway, by drift), it has another important consequence for EP
If memetic transmission is good enough to pass on infomation between
generations, then any programmed behaviour that could be learned would
no longer need to be genetically transmitted. So basically anything that
isn't a mid-brain thing is unlikely to be instinctive, thanks to erosion
by drift after sleection to maintain genetic transmission was relaxed.
Some stuff is left for the EPers under this picture, but not much...
Cheers, Chris (sorry for waffling and stating the obvious in equal
portions there -- no time to edit).
> Memeticists would probably be memocentrists, opting for the meme as the
> target of selection during cultural evolution.
Pleiotropy and epistasis. I will not believe the advice of someone who
believes that frogs are actually alien observers, and that Bush and
Kerry are actually amphibians with hidden gills (always in high shirt
collars -- think about it).
> My main point was to point out to Dace that individuals do not evolve, a
> point which Keith was trying to get across to him.
Yeah -- individuals are the things that get killed off, but the
heritable variety is decided at conception, by drawing from the
_population's_ gene pool, and thereby affecting gene frequencies in it.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Chris Taylor (christ@ebi.ac.uk)
MIAPE Project -- psidev.sf.net
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
===============================================================
This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)
see: http://www.cpm.mmu.ac.uk/jom-emit
This archive was generated by hypermail 2.1.5 : Thu 19 Feb 2004 - 12:02:16 GMT