% The Flaming Right by paul murphy

Raging Speculations: Replying to David Gelernter

In his recent Claremont essay: "Giving up Darwin",Dr. David Gelernter of Yale University's Computer Sciences faculty argues that Darwin's theory of evolution is elegant, simple, and widely accepted; but, like Newton's understanding of motion, inadequate beyond a minor subset of the phenomena it purports to describe.

In the essay Dr. Gelernter describes coming to this conclusion despite the confounding effects of religious belief and on the basis of arguments and observations mainly from paleontology, biology, and elementary probability theory.

In this response I want to offer a highly speculative alternative defending the generality of Darwin's theory as usually understood: i.e. that once life starts, both speciation and specialization are driven by environmental change.

That weasel worded formulation is defensible in the context set by Darwin and accepted by Dr. Gelernter because Darwin's work is about the Origin of Species, not the Origin of life; but the distinction may not actually be necessary because the key speculation on which this response is based can be seen as requiring that Darwinian evolution started when the first proteins formed.

This, highly speculative, model considers four statements to be true or mostly true:

  1. universal physics combines with external conditions to fully determine how atoms form molecules and how those molecules interact.

    Expose a known set of atoms or molecules to known conditions and what happens is predictable because fully determined by the combination of materials and conditions. Basically, chemistry is a science, not magic, and works the same way every single time: things work they way they do because that's how things work --and the physics people are getting closer and closer every day to understanding how that could come about from essentially nothing.

    Notice that no external initiator is needed for chemistry to work - and, because the model says biology is a subset of chemistry, we don't need one for life either.

  2. adaptive change occurs in cellular life forms during DNA transcription and translation and are necessary (in the deterministic sense of claim one, above) responses to external factors affecting chemical reactions between segments of DNA, various forms of RNA, and surrounding proteins.

    Not, in other words, through random mutation of DNA but through fully determinable reactions driven by the conditions under which the translation/transcription processes take place.

    This claim predicts both limited (thicker wool) and far ranging (new species) adaptive evolution: given a consistent, multi-generational, change in external conditions you expect to see increasingly consistent change in the transcription/translation processes that will look purposeful (and correspondingly volitional) to a much longer lived observer without actually being purposeful.

    This claim is the most outrageous one made here but there is actually considerable experimental support for it. A recent report, for example, on adaptation in yeasts shows that the process can be tracked while admitting of no simpler explanation.

  3. the DNA/RNA and related strands found in a cell or simpler life form constitute all combinations of some relatively small length formable in the order of increasing complexity using the materials available under the conditions obtaining during strand formation.

    Claim two of the model says that external conditions cause changes in the replication processes whose effects are to swap (activate/deactivate) DNA segments between the controlling and other parts of the strands involved. We know this happens, but need to explain the happy coincidence that the segments needed for the reactions to complete are usually available - and the guessed-at answer forming claim three here is that we'll eventually find that the unique segments of length greater than some quite small number in the combined DNA/RNA available during translation/transcription exhaust, or exhaust for all but special cases, the more limiting of either the materials available or the list of all amino acid sequences feasible (in the deterministic sense of claim one, above) under the conditions and constraints obtaining during strand formation.

    Claim three is a guess developed in part in response to the claim, put forward by Meyer et al and supported by Dr. Gelernter in the essay this responds to, that randomly picking the next one in a string from 37 (20 main amino acids plus 17 known substitutes) bins 250 times implies that there are 37250 possible polypeptides of length 250. In reality the analogy to picking balls from a jar simply does not hold: when two or more amino acids come into contact and conditions are right to form a larger polypeptide, strong constraints imposed by (at least) proximity, alignment, surface topology, endedness, basic chemistry, surface charge states, and the presence of catalysts (and/or shield proteins) then limit what can get added next and where.

    Claim three also responds to a problem in the analysis of polypeptides found in cells: we don't always see the same sub-sequences in the same place, the same order or embedded in the same longer sequences, but overall what's there suggests resource exhaustive processes that start from similar but not identical conditions and therefore don't always follow the same paths to similar ends.

    Notice that this seeming unpredictability extends the number and range of RNA and pre-RNA segments available for cell formation while adding more conditions to that process as the strings get longer - thus the conditions under which two near proteins fit together to produce a more complex structure are much less restrictive than the conditions under which one unicellular organism will absorb DNA from another to form a third - and by the time evolution gets to motile, multi-celled, organisms the list of imaginable polypeptides formable from the library of imaginable segments on the Meyer model is almost certainly much more than two hundred of orders of magnitude larger than the list of polypeptides that can actually be formed under natural conditions.

  4. for recognizable life of any but the lowest possible level of complexity, the messenger for external conditions either is, or is produced by, a lower (i.e. less complex) form of life.

    Claim four answers another obvious question: how can conditions inside the cell or a complex organism containing the cell, be affected by conditions external to the cell or organism?

Notice that there are many unknowns in this model, but no apparent unknowables. We do not know, for example, exactly what the required conditions are for each peptide to be added/dropped or how these manifest, but there is no reason to think we can't eventually describe and list all of them. Similarly we do not know what drives either the creation or the apparent motion of polypeptides within a cell, but it is reasonable to believe that neither can be volitional and therefore that a full understanding of what drives, and what limits, the mechanisms involved will eventually be developed.

This model is obviously "imaginative", but both mostly consistent with what we think know and inconsistent with every argument Dr. Gelernter cites in support of his decision to abandon Darwin - and that's the point: not that this model is real science, but that we can easily conjure up testable, and fully deterministic, theories explaining what we see around us without invoking third parties, magic, or unknowable forces.

The clearest argument Dr. Gelernter brings forward from work by Meyer et al is the claim that the odds are rather strongly against positive random mutation in DNA. This is undoubtedly correct (although the analogy they rely on is actually inapplicable: the probability of a contextually impossible combination forming is zero, not 1/20) but the argument is an attack on neo-Darwinism's attempt to account for species change through the selective perpetuation of random change, and thus an attack on an extension to Darwinian evolution made necessary by its apparent failure in light of new information and not an attack on Darwinian evolution itself.

Neo-Darwinism can be dismissed (and so the Meyer et al numerical argument with it) because, numbers and mechanisms aside, it is rather obviously falsified by everyday experience. Thus Muller's fruit flies may really have been affected by heavy doses of X-rays, but experiments using gene insertion don't have much to do with natural evolution either, and neither extensive short term radiation as in Hiroshima, Nagasaki, Chernobyl, or any of a dozen major test sites worldwide, nor the low level, but very long term, radiation associated with life in environments dominated by large quantities of granite has yet produced King Kong - or even as much as a mutant sunflower. Conversely, and more directly, people watching genetic adaption in everything from bacterial colonies competing for a food source to weeds adapting to new herbicides normally see contextually positive mutations happen nearly concurrently across entire organisms or colonies and then become self-perpetuating very quickly - no cosmic rays affecting single strands here or there, no massive radiation, no third party engineering, and only a few generations, required.

Dr. Gelernter strongest looking argument is that Darwin's theory explains small changes within a species, but not the origin of species. Here's how he expresses that, first near the beginning and again near the end of the essay:

There's no reason to doubt that Darwin successfully explained the small adjustments by which an organism adapts to local circumstances: changes to fur density or wing style or beak shape. Yet there are many reasons to doubt whether he can answer the hard questions and explain the big picture -not the fine-tuning of existing species but the emergence of new ones. The origin of species is exactly what Darwin cannot explain.

...

There is a general principle here, similar to the earlier principle that the number of useless polypeptides crushes the number of useful ones. The Georgia Tech geneticist John F. McDonald calls this one a "great Darwinian paradox." Meyer explains: "genes that are obviously variable within natural populations seem to affect only minor aspects of form and function -while those genes that govern major changes, the very stuff of macroevolution, apparently do not vary or vary only to the detriment of the organism." The philosopher of biology Paul Nelson summarizes the body-plan problem:

Research on animal development and macroevolution over the last thirty years -research done from within the neo-Darwinian framework -has shown that the neo-Darwinian explanation for the origin of new body plans is overwhelmingly likely to be false -and for reasons that Darwin himself would have understood.

Darwin would easily have understood that minor mutations are common but can't create significant evolutionary change; major mutations are rare and fatal.

What this line of reasoning boils down to is the argument that the evolutionary mechanisms proposed by the neo-darwinists are generally inadequate to describe the degree of change needed for species differentiation and specifically inadequate to describe the speed and enthusiasm with which this appears to have happened during the Cambrian species explosion. Again, this is obviously correct, but is an argument against Darwinian evolution only if no other, vastly more capable, evolutionary mechanism can reasonably be adduced.

Assume, in contrast to the neo-darwinian focus on random DNA mutation, that genetic change occurs when external conditions cause the transcription/translation processes building an organism's DNA to use different segments drawn from the full list of physically possible combinations, and two things immediately become obvious:

  1. first, that the extent to which external conditions should affect DNA change across many cells in an organism or colony simultaneously or nearly so depends on the prevalence of those conditions; and,

  2. second, that the kind of selective breeding that leaves a Mexican hairless chihuahua every bit as much genetically a dog as my Alaskan Malamute can be seen as part of a continuum of possible change that starts with minor adaptations amounting to little more than shifts in emphasis and extends to the species level differences Darwin both saw and predicted.

Notice, in this context, that the rapidity and scale of RNA initiated DNA change eliminates the Cambrian species explosion as a cause for concern in defending Darwin's theory. We don't, of course, know why it happened - History Channel buffs can imagine meteorites full of DNA from an exploded fourth planet fighting a brave, but ultimately hopeless, battle to adapt to earth conditions while fans of Conway's Game of Life can point to the Cambrian-like species explosions and extinctions you can get when two long running "life" forms collide - but we don't really need to know. All we need to know is that because it did happen it can happen - and the rapidity and extent of change possible under the current hypothesis removes the mystery from this fact.

Bottom line? this particular evolutionary model or hypothesis is pretty much guaranteed to be utter bunk since I know very little about biology and invented the hypothesis solely for this response to Dr.Gelernter essay - but the point is that a purely deterministic view of cellular adaptation is possible, internally consistent, and sufficient to let Darwinian evolution trump all forms of the intelligent design hypothesis.

(A further note: nothing in this essay should be understood as "debunking the God hypothesis". Plato (along with Avicenna in the 11th century and George Lucas more recently) argued that an evolving life force would co-evolve God, and that's undoubtedly true in terms of our perception of God, but begs the origin question by convolving God with life. Some people claim that physics asserts, or is about to assert, that everything came from nothing - and that may be true too, but doesn't answer the origin question either in that any three year old will ask you where the nothing came from. Thus the bottom line is simply that evolution says nothing about God because arguments about the existence or otherwise of God are arguments about first causes, and evolution has nothing to do with that.)


Paul Murphy, a Canadian, wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specializing in Unix and Unix-related management issues.