Re: Science and Mechanism

From: Dace (
Date: Thu 12 Jun 2003 - 23:07:20 GMT

  • Next message: Ray Recchia: "Re: definition of meme"

    > From: "Reed Konsler" <>
    > >The mechanics of protein-formation are still not known, and it's not at
    > >clear that protein-formation, particularly at the quaternary level, is
    > >forced into place by purely chemical and mechanical factors.
    > Biochemistry is based on the premise that macroscopic events as simple as
    > crystallization and as complex as embryonic development are a result of
    > microscopic structure. It is a mechanistic way of viewing the world.

    Hi Reed,

    As Paul Weiss pointed out many times, organisms are determinant at the macro level and indeterminant at the micro level. This is why biochemists, such as Alfred Gilman, can't predict anything within the simplest of microorganisms, even when they've got a map of all the cell's components and interconnections.

    > A scientific theory requires several things. This is more or less in
    > from primary to peripheral
    > Mechanism: A theory must relate causes and effects.

    This is impossible at the micro level. Researchers consistently find a high degree of freedom from mechanistic determination within cells. Here's another passage from Gibbs' August, 2001 Scientific American article:

    In 1994 [Drew Endy of UC Berkeley] and John Yin of the University of Wisconsin-Madison began programming a computer model that would incorporate virtually everything known about the way that a certain virus, T7 bacteriophage, infects Escherichia coli bacteria that live in the human gut. The virus looks like a lunar lander. It uses clawlike appendages to grasp the outer wall of a bacterium as the phage injects its DNA into the cell. The genetic material hijacks the cell's own reproductive apparatus, forcing it to churn out bacteriophage clones until it bursts.

    Endy and Yin's model simulated mathematically how all 56 of the virus's genes were translated into 59 proteins, how those proteins subverted the host cell and even how the viruses would evolve resistance to various RNA-based drugs. That seems impressive. But peek inside the equations, Endy says, and you'll find that despite including measurements from 15 years of laborious experiments, "there are still a tremendous number of degrees of freedom." The equations can be tweaked to produce almost any behavior. "A useful model must suggest a hypothesis that forces the model builder to do an experiment," Endy says. This one didn't.

    > Empirical Falsifiable: A theory must engender experiments or directed
    > observations.

    As the late Stephen Jay Gould observed in *The Structure of Evolutionary Theory,* though Gould himself supported it, reductionistic theory *clearly* does not pass this test. Gould noted that reductionism, since it was devised by August Weismann, has always relied purely on logic. Weismann insisted that we accept the reduction of evolution and development to "germ plasm," not because we will ever be able to devise tests that verify it (an impossibility due to the extreme complexity of cellular activities) but simply because there's no other logical possibility. He claimed that there are only two possibilities by which morphological traits are passed on from parent to child: Lamarckism and Darwinism. Since germ plasm (i.e. DNA) can't be influenced by acquired traits, that leaves Darwin. Ironically, Darwin himself totally rejected the view that acquired traits have no bearing on evolution. Had he lived long enough to see the emergence of Weismann's theory, he might have suggested that acquired traits are passed on without the need for modification of germ plasm. In *The Origin of Species,* Darwin even pointed to the then-recent development of field theory in electromagnetics as an example of how science can get beyond conceptual logjams through creative, new approaches.

    > This isn't so much to confirm or refute the theory itself
    > (despite arguments, no single experiment has ever lead a person to chuck
    > otherwise useful theory). Rather, falsifiable experiments allow a theory
    > be developed and refined. Without empirical restriction a theory will
    > either remain amorphous or develop in an idiosyncratic way unlikely to be
    > useful.

    Due to its purely theoretical nature, gene-based mechanism remains amorphous and untestable.

    > Application: A theory that doesn't do something for the community at
    > will struggle and wither as resources are directed towards more useful
    > endeavors. Actually, application often comes first: something useful is
    > discovered by accident (like an antibiotic) and then scientists work to
    > understand it.

    Yes, and genetics has had tremendous applications in medicine. This causes people to assume its underlying theory is correct, despite the total lack of direct evidence.

    > Truth is not a factor.

    Science (scientia) is Latin for knowledge. Knoweldge means truth. If truth is not a factor, then science needs to come up with a new name.

    > Exactly how a protein folds...indeed, if they even
    > exist at all...isn't relevant. The question is: if we believe and act on
    > that theory, what are the consequences? The explosion of biochemistry and
    > biotechnology is a result of the fact that I can teach the basics of the
    > theory in high school, thousands of people can find work within the field,
    > and insulin can be manufactured cheaply in massive quantities.

    You reduce science to a glorified form of engineering.

    > Proteins might not fold according to a "mechanical" mechanism. But, it's
    > pointless to argue that they fold according to no mechanism at all.

    The question is whether they fold according to a linear causal sequence reaching back to DNA (which can't be shown due to the complexity of cellular activities) or if they fold according to a holistic model of the protein based either on eternal equations or inherent memory.

    > Which theory is most useful, of all that have ever been proposed? I would
    > argue, at present, it is the biochemical model.

    Useful for what? Generating technologies are describing life as it actually is?

    > From: "Reed Konsler" <>
    > Subject: Morphogenic Fields
    > Dace:
    > "Okay. And what determines the "homologous pathways of development?"
    > or fields? Whatever answers researchers come up with cannot help but be
    > provisional. Every answer merely moves the question back a step.
    > Ultimately, the information is either particulate or holistic.
    > is either bottom-up or top-down. Since biochemists are no closer than
    > were forty years ago to providing a detailed picture of how genes build
    > bodies, there's no reason not to explore other possibilities. Indeed,
    > is the general trend of contemporary biology."
    > First: "...are no closer..."? I'm not going to argue with you about your
    > opinion, but there are conferences full of molecular biologists that have
    > the impression that they know more than they did forty years ago.

    Of course. They know plenty more about the mechanics of cell function. But what if organic form is not primarily a question of mechanics? In that case, all their progress has brought them no closer to the basis of organic form.

    > Anyway, it is my experience that "contemporary biology" is shifting, as a
    > general trend, further in the direction of the "bottom-up" biochemical
    > approach.

    You're about 75 years behind the curve.

    > From: "Reed Konsler" <>
    > Subject: Biologists and Wishful Thinking
    > Here's an illuminating excerpt from the August 2001 issue of Scientific
    > American ("Cybernetic Cells," by W. Wayt Gibbs):
    > "[M]ost biologists still use computers as little more than receptacles for
    > the surge of data gushing from their robotic sequencers and gene chip
    > analyzers. The 'models' they publish in their journal articles are
    > caricatures based on the best theory they have: the central dogma that a
    > gene in DNA is converted to an RNA that is translated to a protein that
    > performs a particular biochemical function."
    > That would tend to confirm that most biologists are "bottom-up."

    Talk about taking a quote out of context! Here's the very next sentence:

    "But the past few years have seen a growing movement among mathematically minded biologists to challenge the central dogma as simplistic and to use computer simulation to search for a more powerful theory."

    The trend is clearly toward top-down. The next step after this is to recognize that a holistic view doesn't necessarily mean mathematical idealism. The "fields" or "systems" determining organic structure might result from inherent memory rather than timeless equation. This is more in accord with evolution, as equations don't evolve.

    > "We're witnessing a grand-scale Kuhnian revolution in biology,"
    > avers Bernhard O. Palsson, head of the genetic circuits research
    > group at UC San Diego.
    > "We are so going to get laid by those chicks" avers Sean,
    > self-proclaimed 'ladies man' of the Druid pub in Inman Square.

    It should bother you that Palsson is a highly respected researcher. That you don't find his statement the least big significant suggests you're beholden to a deeply ingrained meme.

    > From: "Scott Chase" <>
    > Subject: Re: Biologists and Wishful Thinking
    > >From: "Reed Konsler" <>
    > >
    > >"We're witnessing a grand-scale Kuhnian revolution in biology,"
    > >avers Bernhard O. Palsson, head of the genetic circuits research
    > >group at UC San Diego.
    > >
    > >"We are so going to get laid by those chicks" avers Sean,
    > >self-proclaimed 'ladies man' of the Druid pub in Inman Square.
    > >
    > >
    > A buzzphrase like "grand-scale Kuhnian revolution" is usually enough for
    > to run, not walk, toward the nearest exit and establish a cordon
    > Even if a concept has merit, parading it as a "Kuhnian revolution" tends
    > raise the red flags.

    We might call it the "anti-Kuhnian" meme.


    =============================================================== 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:

    This archive was generated by hypermail 2.1.5 : Thu 12 Jun 2003 - 23:14:14 GMT