Re: Dawkins gives incorrect answer

From: Michael Ragland (ragland37_at_webtv.net)
Date: 09/08/04 

Date: Wed, 8 Sep 2004 17:08:19 +0000 (UTC)

Michael,
[snip]
in article chjf65$1oih$1@darwin.ediacara.org, Michael Ragland at
ragland37@webtv.net wrote on 9/6/04 9:58 PM:
GH:
Absolutely. This is the essence of empirical reductionism. There is
nothing wrong with it IMHO. I just see it as incomplete and lacking an
appreciation of the physical emergence of macroscopic phenomena and the
necessity of understanding phenomena at the scales in which they are
manifested. This understanding requires looking upward in scale to see
the top-down effects on macroscopic phenomena (e.g., mass action
effects, entrainments, ...) that are fundamental to their organization
(IMHO).

MR:
I notice you state "appear to degrade". It seems to be zooming in on
such finer scales of observation reveals the complexity of complex
systems but it may make the coherence of the internal dynamics of a
complex system more difficult to grasp.

GH:
Well said. I think this is universally true.

GH:
Yes to the latter part. In contrast to the former part, I would have
said that zooming in obscures the complexity of a system while also
making it appear more complicated. Remember the dictum that "complexity
breeds simplicity." I think we treat systems as more complicated than
they really are when we ignore their complexity (the integrated nature
of their dynamics).

MR:
In this context how would you address the "concept of the gene"? It does
seem at least in the media and popular culture to be generally
oversimplified. A gene for aggression, a gene for homosexuality, a gene
for schizophrenia, etc. Since most human genes are pleitropic, polygenic
and some epistatic wouldn't it be more accurate for scientists to state
they had discovered a possible or actual "allele" for schizophrenia,
bipolar, aggression, homosexuality, etc.?

GH:
I think it would be more accurate to say that we have discovered a gene
in which the existing variation influences (perhaps strongly) the
variation in some human trait. This would allow for the same genetic
variation to have different phenotypic effects in other species, which
can happen. It also allows for other (probably unknown) genes to have
(perhaps even greater) effects on the same trait, because the statement
of effect is implicitly contingent on both genomic background and
environmental effects.

MR:
I agree with you. The problems I have are in the social, ethical and
legal arenas. I know sci.bio.evolution is technically about about
biology and evolution but the social, ethical and legal implications
require just as much attention..even though scientists can't be the only
possible participants in this process they do need to be involved.
Society needs to be more aware of these issues. I remember years ago
when Surgeon General C. Everett Koop developed a plan to mail literature
on AIDS to every household in America. The U.S. federal government and
state government needs to do more about addressed these issues but there
are vested interests which would oppose that.

You write, "I think it would be more accurate to say that we have
discovered a gene in which the existing variation influences (perhaps
strongly) the variation in some human trait. It also allows for other
(probably unknown) genes to have (perhaps even greater) effects on the
same trait, because the statement of effect is implicitly contingent on
both genomic background and environmental effects." I agree. The problem
is without at least some anti-genetic discrimination and genetic privacy
laws at (at both the federal and state levels) insurance companies,
employeres, realtors, etc. aren't going to be concerned about the subtle
genetic interactions being implicitly contingent on both genomic
background and environmental effects. If they discover a person has a
genetic predisposition for a disease (high or low) or one which may
never even express itself they are going to use that information for
their fat pocketbook and genetically discriminate against others. That
is already happenning on a limited scale and without anti-genetic
discrimination laws and genetic privacy laws (however imperfect they may
be) it will only increase. This idea of discriminating against a person
on the basis of a genetic predisposition for a disease (which may be
low-high or never even will express itself and has not been clinically
diagnosed by symptoms by a physician) is pernicious and false as
labeling and categorizing a person by "race" and referring to an
African-American as a "***", a Hispanic-American as a "***", a
Jewish American as a "***" and all the deragotory traits these terms
imply. I'm not saying the two are identical but if public policy is
based on your statement, "I think it would be more accurate to say that
we have discovered a gene in which the existing variation influences
(perhaps strongly) the variation in some human trait. It also allows for
other (probably unknown) genes to have (perhaps even greater) effects on
the same trait, because the statement of effect is implicitly contingent
on both genomic background and environmental effects" I think it will be
abused by many elements in society without at least some pretense of
safeguards and enacted legislation. I'm not against the scientists here.
I'm concerned about how society will unscientifically and
inappropriately use that information.
  
MR:
I don't consider myself a dummy but there was a time in the past when I
believed in the monolithic concept of the gene and actually thought
there might be a "gene" for aggression and other complex traits. I'm not
scientific minded really but many people aren't. How many more people
out there think there is a "gene" for aggression, alcoholism,
schizophrenia, homosexuality, bipolar, etc.? The media perpetuates this
and so does the scientific community (when they are quoted..assuming
they were quoted correctly) by stating "schizophrenia disease causing
gene discovered" or "gene for homosexuality discovered".

GH:
Right. I think this is a travesty. The scientists should know better. We
are letting the tail wag the dog when we use language like that in order
to please the media or the masses.

MR:
Sometimes I can tell its the media's fault and many scientists are
reluctant to be interviewed because they are quoted out of context and
what they said was distorted. Or, a media outlet will pick up the story
from somewhere else and distort it. But I have come across some
scientists who apparently do use such misleading terminology.
Unfortunately, I can't say much positive about the media except for
scientists to really investigate the reporter and the media outlet
extensively before granting an interview.

MR:
Yes, there is often a disclaimer at the end of these articles that more
than one gene is involved but what is going to make a deep impression
upon the average reader is there is a gene which causes a particular
disease or trait. That's not the case. Instead IMO geneticists should be
referring to an "allele" associated with a particular disease or trait
and making it clear the "gene" contains many products.

GH:
Yup.

MR:
The current status quo is potentially dangerous in my view. Let's say
scientists discover an allele associated with schizophrenia but there
are other alleles which are involved in the disease. Let's say genetic
screening/testing is able to test for this allele associated with
schizophrenia and make a prediagnosis. If the pregnant mother tests
positive she may have the option of terminating her pregnancy. Now
perhaps this allele associated with schizophrenia is shared by people
who don't have the other alleles for the disease. In other words, in
this hypothetical genetic screening and testing would be done without a
full understanding of the genetic mechanisms of the disease and possibly
result in termination of pregnancies which would have been normal e.g.
didn't have schizophrenia. I find such a scenario nightmarish.

GH:
Your scenario is a very real possibility in today's practice of genetic
counseling. However, I suppose that people would differ on whether they
thought it was "nightmarish" or practical. I would rather not get into
that political hot-potato here.

MR:
Well I guess it will be trial and error. Will have some astounding
successes and dismal failures. My own view is of informed consent where
all the possible risks, benefits, chances, etc. are opently discussed
with prospective parents and they are allowed to make their own
decision. But this is the ideal and often those in control have an
agenda or the events take place in a particular setting which influences
informed consent.

MR:
Human genes are pleitropic, polygenic and some epistatic and act in
concert with the environment. Yet when it comes to genetic engineering
of animals such as mice the environment doesn't seem to figure in much.

GH:
Note that these studies use highly inbred strains of mice, which
controls for variation in genomic background, and they standardize as
much as possible the environments of the mice to control for that source
of variation in phenotypic effects. Nobody doubts that genes influence
phenotypes, so when you isolate those effects they naturally appear to
be in control. No rational scientist would take results from such
studies to suggest that the particular gene under study was the most
influential gene (compared with those that were not studies and may
remain unknown), independent of its existing allelic variation, or that
the environment has less influence on the phenotype than genetics.

MR:
True, nevertheless the molecules the scientists studied in the
genetically engineered mice were associted with high endurance. The
scientists of this study may not consider the "molecules" they used to
genetically engineer these mice as the most influential in endurance but
they were certained linked.
 
MR:
Since human genes are pleitropic, polygenic and some epistatic and act
in concert with the environment I'm wondering how this will work out
when it comes to genetically engineering people. I recently did a post
on "epigenotype" which was advocated by Dr. Feinberg who stated (I'm
paraphrasing) that only by having a full understanding of both genotype
and epigenotype (different tissue samples from different people) and
superimposing one on the other will we have a understanding of the
healthy functioning gene and the diseased gene. Would you agree with
that?

GH:
I'm not sure I understand what Dr. Feinberg has suggested, but I suspect
that the term "FULL understanding" is something of a red herring.

MR:
What, you think he has his own agenda and is just using the concept of
epigenotype to advance it? My impression was he was referring to
cellular processes in different tissues of different people which are
the product of the environment (epigenetics) and that a better
understanding of how epigenotypes and genotypes interact with each other
is necessary for "fully understanding" the healthy functioning gene and
the diseased gene. The article also mentioned compared to genotyping
epigenotyping currently doesn't have the high technology to rapidly
epigenotype. Since you acknowledge the role of the environment in
genetics I thought you might be favorable to the idea.

MR:
I'm sorry for going so off course but when you stated, "I think we treat
systems as more complicated than they really are when we ignore their
complexity (the integrated nature of their dynamics)" I couldn't help
but think of the concept of the gene. There are so many different fields
such as genomics, proteomics, biotechnology, etc. How do you put it in a
holistic framework? I'll tell you right now I'm not satisfied with the
way things stand and genetic engineering of humans, unless it is done
cautiously and carefully, stands to massively disrupt any holistic
framework.

GH:
While I agree that we know very little about the engine we might tinker
with, I am not as pessimistic as you about the chance for "massive"
disruption. xComplex adaptive systems in general, and organismal
functioning in particular, are remarkably resilient and able to remodel
their structures (self-reorganize) in adaptive ways. If the bottom-up
effects of tinkering with a gene here and there has significant effects,
I would expect the rest of the engine to respond in ways accommodating
those manipulations.

MR:
Well, genetic engineering of humans is still a ways off. We've created
genetically engineered knockout mice lacking nitrous oxide and the
result was predatory savagely aggressive male mice who killed other not
genetically engineered mice when put in a cage with them and continously
attempted to mount female mice for hours even when they weren't in
estrus. It wouldn't surprise me if the military had scientists doing
genetic research aggression on animals with the long term goal of
creating a super aggressive (yet obediant and average intelligenced)
soldier.

You state, "Complex adaptive systems in general, and organismal
functioning in particular, are remarkably resilient and able to remodel
their structures (self-reorganize) in adaptive ways. If the bottom-up
effects of tinkering with a gene here and there has significant effects,
I would expect the rest of the engine to respond in ways accommodating
those manipulations." We can argue about what is considered "adaptive".
I guess at a minimum a species capable of surviving and reproducing. By
that measure most would agree the human species is adaptive but I have
argued before I don't think we are totally adaptive as a species. You
state, "If the bottom-up effects of tinkering with a gene here and there
has significant effects, I would expect the rest of the engine to
respond in ways accommodating those manipulations." I think you're
probably right. The question mark is how will the rest of the engine
respond in ways accomodating these manipulations? I would argue such
scientific efforts, particularly with complex traits such as
aggression, would lead to several failed models even if the engine
responded in ways accomodating those manipulations. Just "how" would the
engine "accomodate" the manipulations? The question also has to be asked
that even if the engine did accomodate the manipulations would the
animal (species) be adaptive to its current enviroment? These are
questions which will only be answered slowly and many mistakes and
through trial and error. I think our best hope is animal research and
once we have enough knowledge and have advanced in genetic engineering
and genetics of aggression using chimpanzees/bonobos. After that some
scientist is going to try it out on a human subject. I don't think
genetic engineering necessarily creates "monsters". In fact, even though
I have reservations about how it proceeds and is used and how it might
make things initially worse in the world, I think it is ultimately the
only hope left for saving mankind.
 

"It's uncertain whether intelligence has any long term survival value.
Bacteria do quite well without it."
 Stephen Hawking