Science as Idea & Meme

From: Dace (
Date: Mon 16 Jun 2003 - 22:17:37 GMT

  • Next message: "Re: Dees' and Dace's war"

    > From: "Reed Konsler" <>
    > Subject: Science and Mechanism
    > From: "Dace" <>
    > Reed:
    > > To say that biochemists can't predict anything is an overstatement.
    > Ted:
    > "The statement is that of a Noble Prize-winning biochemist. And he's
    > Biochemistry is a descriptive science, not an explanatory or predictive
    > science."
    > If I inject plasmids for human insulin production into a yeast cell I can
    > predict the generation of insulin and explain the process, at least at
    > level of detail. My prediction and explanation is limited by current
    > knowledge, yes. [Your] argument is that "biochemists can't predict
    anything" is
    > an overstatement. I don't believe any biochemist would agree with such an
    > extreme view.

    And yet it was a biochemist who expressed it. The key phrase is "at least at some level of detail." Cells are not without mechanisms that predictably perform a given task. But as Paul Weiss points out, every causal chain has a terminus, a point beyond which no further analysis is possible. A true causal chain can be traced all the way back to its source. Between the alleged source, DNA, and its alleged effects in the cell lies a sea of indeterminate micromolecular processes. Here's Weiss, *The Science of Life,* pg 59:

    A [causal] chain that is interrupted and open-ended is functionally no chain at all. Accordingly, whenever a microprecisely determined reaction chain merges into the statistical pool of free systems dynamics, that signifies not only the end of the chain, but also the end of its determinative role: a supposed "minidictator"
    [DNA] is then reduced to the role of member and participant in a patterned macrodetermined operation, in which its separate track gets lost... even if microprecisely programmed chain reactions were set off in separate linear tracks as "determiners" for organized end products, their aim would get blurred on the way in the trackless stretches of the network of systemic anastomoses [the meeting-points of branches].

    Weiss is refering here to the (literally) incalculable complexity of the cell at the molecular level. No causal chain has ever been traced from DNA straight through to supposed "end-products" in the cell. This is not to imply that an experiment can't demonstrate predictable, general outputs resulting from a specific input, as you describe above.

    > Ted:
    > "Ultimately a holistic, systems approach will be required to get beyond
    > description."
    > That is a prediction, not a fact. My stance is that one cannot know if a
    > model is useful until it is used. If it works, good; if not, then one
    > again. The future is obviously uncertain, save for the dogmatic.

    Systems theory is precisely the effort to get beyond our dogmatic attachment to genetic reductionism.

    > "Somehow, despite the chaos and the general lack of mechanical or chemical
    > determinism..."
    > I don't know what you mean by that.

    I mean that outcomes are not generally necessitated by physical principles. Back to Weiss, page 39:

    A cell works like a big industry, which manufactures different products at different sites, ships them around to assembly plants, where they are combined into half-finished or finished products, to be eventually, with or without storage in intermediate facilities, either used up in the household of that particular cell or else extruded for export ot other cells or as waste disposal. Modern research in molecular and cellular biology has succeeded in assigning to the various structures seen in micrographs specific functional tasks in this intricate, but integrated, industrial operation. There is a major flaw, however, in the analogy between a cell and man-made factory. While in the latter, both building and machinery are permanent fixtures, established once and for all, many of the corresponding subunits in the system of the cell are of ephemeral existence in the sense that they are continuously or periodically disassembled and rebuilt, yet always each according to its kind and standard pattern. In contrast to a machine, the cell interior is heaving and churning all the time; the positions of granules or other details in the picture, therefore, denote just momentary way stations, and the different shapes of sacs or tubules signify only the degree of their filling at the moment. The only thing that remains predictable amidst the erratic stirring of the molecular population of the cytoplasm and its substructures is the overall pattern of dynamics which keeps the component activities in definable bounds of orderly restraints. These bounds are not to be viewed as mechanically fixed structures, but as "boundary conditions" set by the dynamics of the system as a whole.

    It's because of this "heaving and churning," this continual disassembly and reassembly of components, that biochemists, in the words of Nobelist Alfred Gilman, "have absolutely no ability to predict anything" even when they've got "a map of all the components in a cell [with] all the proper arrows connecting them."

    > "How could a theory mechanistically relate causes and effects in living
    > when they don't behave according to simple, mechanistic determinism until
    > they die?"
    > I believe that they do behave according to laws, to the extent that
    > in the universe does. I wouldn't call it simple, though.

    So you're not just a nose-to-the-grindstone pragmatist. You do have beliefs, and these beliefs are not based on known facts. It's easy enough to explain what happens in cells after they die in accord with physical principles. It's the inability to do so in regard to living cells that defines them as living. Ultimately, you can't understand the organism by killing it. It must be taken on its own terms.

    > However they actually are, a theory must propose some relationship between
    > causes and effects. Even a "holistic theory" must establish some kind of
    > mechanism. Otherwise, what is it saying?

    Sheldrake's theory could just as easily be known as "morphic mechanics" which, like quantum mechanics, is probabilistic. Nothing definitively determines the course taken by the cell. But the *likelihood* is that it will fall under the influence of past cells that are morphically similar to it, as opposed to cells that are less similar. The mechanism by which morphic information is transfered is similarity to previous organic forms
    (which can only be holistic).

    > Ted:
    > "Right. The article went on to point out that in order to produce useful
    > models, researchers must abandon the notion that organic forms are caused
    > genes. This is why researchers are questioning the "central dogma." It
    > just doesn't work in the creation of accurate, mathematical models."
    > I don't think that this is what the article said.

    According to "Cybernetic Cells" (W. Wayt Gibbs, Aug 2001 Scientific American, page 54), "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."

    In other words we won't have a useful, i.e. "powerful", explanatory model until we develop one that doesn't reduce cellular regulation to genes.

    > "...Needless to say, lack of imagination is not the most reliable
    > for a theory."
    > Agreed. I would say the utility of a theory is it's foundation.

    Do you mean utility at generating reliable predictions? Endy and Yin failed to predict whether a given virus would develop resistance to RNA-based therapies. Why? Because their model assumed simple, genetic causality. Their next model won't include that assumption.

    > "Field theory is absolutely not reductionistic. Fields are spatial, not
    > material. A field cannot be regarded as a fancy way of saying "a
    > of particles." The field is simply the spatial aspect of the particle,
    > which is the material aspect of the field. They are one and the same and
    > therefore mutually irreducible."
    > This doesn't make sense to me.

    That a magnetic field cannot be reduced to a magnet is demonstrated by the fact that the field cannot be divided into two discrete halves, even when the magnet it's associated with is cut in half. Instead of two halves of a magnetic field, you get one whole field for each half of the magnet.

    Same goes for quantum theory. An electron can be viewed as either a "wave" or a "particle." These are simply different ways of looking at the electron, one from the spatial point of view, the other from the material point of view. Neither is reducible to the other; both merely describe the electron in different ways.

    > "In other words, current embryos "resonate" with past
    > embryos on the basis of "morphic" similarity. This hypothesis is highly
    > testable, and Rupert Sheldrake has had a great deal of success in
    > demonstrating the effect."
    > If that theory allowed one to DO anything, it would be more widely

    It allows us to make predictions, such as the fact that members of a given species will perform better at a given task if many other members of that species have already successfully performed it. Read *The Presence of the Past* for details of numerous experiments, from a wide variety of fields, that demonstrate this effect.

    > > How do you prove something to be true?
    > "Hey, who said science is easy? Nonetheless, science has uncovered many
    > truths, such as the fact that the earth revolves around the sun while
    > rotating on its axis. I'd say this has been satisfactorily proven."
    > You're mistaken. Present evidence is consistent with that theory.
    > in science is ever proven.

    For all intents and purposes it's a proven fact. Don't go philosophical on me.

    > "Not at all. But biochemistry alone can't tell us what life is."
    > Well of course not. We have art, religion, and other aspects of culture
    > that are vastly more important to answer the question: "What is life?".
    > Biochemistry is confined to explaining how biological systems function.

    Life is the way biological systems function. Biochemistry cannot help us understand life until it develops a model that doesn't depend on genetic reductionism.

    > "We've been led to believe that DNA is the secret or essence of life."
    > Straw man.

    You might want to take a look at *DNA: The Secret of Life,* by James D. Watson and Andrew Berry, Knopf, 2003. Watson has been peddling this nonsense for 50 years now.

    > Reed
    > > But that recognition won't ever happen if there isn't a material reason
    > > make it.
    > "The material reason is that a holistic theory must be compatible with
    > evolution. As long as they're based on inherent memory rather than
    > equations, fields can evolve right along with the organisms they both
    > reflect and regulate."
    > Whatever. I mean "material reason" as in useful application. When it
    > something, then it will be worth adopting.

    Science, the idea that we can obtain knowledge of nature, has been corrupted into the memetic idea that we can obtain data that allows us to *manipulate* nature. Unlike the original idea, which resulted from reasoned reflection on the part of early scientists (and continues to be renewed by many contemporary scientists), the meme is perpetuated precisely through a *lack* of reflection as it replicates from mind to mind. That science is a blind mechanism churning out technological applications, i.e. "machine science," is a deeply ingrained culturally-shared habit of thought bearing no relation to reality and which is perpetuated strictly through its own momentum-- much like a personal habit-- rather than by appeal to reason.


    =============================================================== 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 : Mon 16 Jun 2003 - 22:23:39 GMT