[...stuff cut...]
>> [Aside: I do not believe most biochemists think we can
>> understand everything about an organism by knowing its
>> genome. Many are quite skeptical about genetic determinism,
>> particularly as it has been applied to complex traits such
>> as human behavior. Heck, we're still generally skeptical
>> about the usefulness of results from DNA expresssion
>> profiling...]
David Tyler responds:
> This aside does surprise me. Whilst I can understand why there
> is skepticism about genetic determinism, it does appear to me
> that genetic reductionism is very strong among biochemists.
> This view is reinforced by the almost protesting challenges
> to genetic reductionism that come from those on 'the fringe'.
It's not just the fringes...
I had this discussion previously with Brian Harper. Since then,
I decided to do an (admittedly unscientific) review/survey of
the biochemists/mol. biologists/physiologists (& etc.) who
I know personally. None believe that "genes control everything".
In fact, I've never talked to any geneticist or biochemist
who gave me the impression that they thought genetic reductionism
would solve most of the questions in biology. These people include
hard-core bacterial geneticists/bacteriologists like John Roth,
Fred Neidhardt, Carl Woese, Bruce Ames, John Ingraham & Norm Pace.
Jasper Rine (see below) is also included in this list. Athel
Cornish-Bowden, a hard-core enzymologist & pioneer in metabolic
modelling expressed the same opinion. I'd also include many other
colleagues in developmental neurobiology, immunology, plant
physiology, & etc. We know that most problems in biology
are not simple enough to be reduced completely to genetics.
In one meeting, long ago, I did hear one person -- a computer
scientist helping with the E. coli genome project -- say that
once we had the sequence completed, we would understand exactly
how E. coli works. I looked over at Fred Neidhardt and
watched him chuckle knowingly at the computer scientist's
gaff. Others who managed to stay awake during the talk
and catch the speaker's remark laughed about it afterwards.
So where does this general impression that genetic determinism
and hard-core genetic reductionism enjoys overwhelming support
among biologists arise? I suspect that is has a lot to do with
many writers (scientists) not been too careful with their words
and trying to oversimplify the issues. This seems particularly
true in articles written for the popular press but I've also seen
it in some journal articles. Another factor is the basic tendency
of scientists to use reductionist approaches in their work.
That's just science - Things work *a lot* easier if you can
achieve "a reduction" in explanation. So, even though problems
may not be easily reducible to simpler ones, such an approach
will still be attempted and will always garner the most
interest when successful (read: publications). "Unreduced" problems
don't get much press -- After all, how much is there to say about
a problem you can't crack?
This preference for a reductionist approach to cell biology does
not necessarily lead to the dogma that DNA = destiny. It's more
a matter of trying to do the best with the tools one has. By
far, the majority of biologists know that everything does not
revolve around DNA. It's just the focus of such intense study
because it's something relatively easy to investigate.
[Aside: I too once thought that molecular biologists were overly
dogmatic about genetic reductionism (They were, I thought,
a rather sissy bunch characterized by an almost irrational
fear of physical chemistry courses). But after many IQ-reducing
rounds of beer and becoming one myself (if only briefly), I
came to realize that even they knew what it was that they
didn't know about the problems in biology. And frankly, that's
one of the best complements you can give a scientist.]
> As an example, I'll quote Harold (1995): do you think that most
> biochemists would be comfortable with his view?
Actually, yes, if strictly interpreted. I ran this quote past
a few enzymologists, physiologists and molecular biologists at
work. They agreed too. Maybe Steve Clark can check at U-Wisc.
> "Why do offspring resemble their parents in form and function? The
> answer is not known, not even in principle, for the quest reaches
> deep into the abiding mysteries of organised complexity (p.2765)."
>
> "Most microbiologists look to the genome to play the architects role.
> From where we now stand, this seems to me a rejection of reason.
> All we have learned points to the conclusion that several epigenetic
> layers intervene between genes and form. Morphogenesis cannot be
> orchestrated by the genome, but makes manifest a higher level of
> order, corresponding to the cellular scale of size and order
> (p.2771)."
>
> "The unstated premise that a cell is at bottom a self-assembling
> structure carries reductionism to the point of absurdity (p.2774)."
>
> Reference: Harold, F.M. 1995. From morphogenes to morphogenesis.
> Microbiology. 141, 2765-2778.
I like that reference except for the "rejection of reason" part in
the second paragraph. I think he's stretching things a bit. It
is true that we do not understand in continuous detail how a
father and son may resemble each other. To really claim *complete*
understanding would require a great deal more information and
major advances in many biological fields. Success is some time
off. The example of dog breeds illustrates a similar problem.
Even if we had the complete sequences of two dogs, it is very
unlikely that we would be able to identify which breeds they
were. Dog breeds are simply too similar to each other, and the
detailed genetics of breed characteristics is not understood.
Take a Pembroke Welsh Corgi and a Whippet. Sequence their DNA.
You couldn't tell which sequence came from which dog.
[Aside: In principle, one might find a breed-specific genetic
marker, such as have been used to establish paternity, but that
would be "cheating" -- The marker would probably have nothing
to do with the specific traits we use to identify the breeds.]
Yet, while we can't determine all the traits of dog breeds from
gene sequences alone and given the caveat that we know traits
are the result of multiple interactions (genetic and non-genetic),
this does not mean that many traits do not have a strong genetic
basis or input. For example, Jasper Rine at UC Berkeley is doing
research on the genetic components behind some breed behaviors.
These are developmental traits which produce specific behaviors at
certain defined times in a dog's development. One behavior Jasper
monitors is the stalking crouch that Border Collies exhibit a
number of weeks after birth. Another is an attraction to water
which is displayed in Newfoundlands, again at a particular stage
in their development. These traits are inherited and are very likely
to be the result of the interactions of multiple alleles (as is
suggested by breed records).
[Aside: Rumor has it that Jasper also does this study because it
gives him an excuse to keep a dog in his office.]
Do genes play a role in development and cell biology? Yes. Is
the role of a particular gene (or genes) strongly deterministic?
Sometimes yes, often no. Are there times when a trait cannot
arise without some genes or combination of genes? Definitely.
Do the above opinions make me a genetic determinist or genetic
reductionist, and does my position require a "rejection of
reason"? I suspect not, but YMMV.
> Tim added a second aside:
>> (Aside: Someone here mentioned that genetic determinism was
>> something atheists would like everyone to believe. Odd that
>> he would read your post, David, and skim over Lewontin's
>> religious beliefs ;^)
> Yes. I do find it interesting that there are great divergencies
> of thinking between people who nominally have the same belief
> system. We see it in Christian circles and we see it in Marxist
> circles.
That's because we are genetically programmed to compete. It's
a direct consequence of Darwinian evolution ;*)
> On the other hand, I was very struck by the apparently
> "strange alliance" between Christians, Marxists and Post-moderns
> on the nature of science, and wrote it up as an article.
> http://www.pages.org/bcs/Bcs089.html
> I hope it is of interest.
Interesting, yes. Thanks for the reference.
Regards,
Tim Ikeda
tikeda@sprintmail.hormel.com (despam address before use)