• McDonald – The Great Darwinian Paradox

    A reader has specifically asked that I look into the Great Darwinian Paradox as promoted by the authors associated with the Discovery Institute.  I say it this way very specifically, because the only references I can find to the paper that they claim presents said paradox is from Meyer, Behe, and Dembski.

    Of course, Michael Behe’s Darwin’s Black Box (1996) seems to have started it all

    University of Georgia geneticist John McDonald notes a conundrum:

    [ed note: the following is quoted from McDonald] “The results of the last 20 years of research on the genetic basis of adaptation has led us to a great Darwinian paradox.  Those [genes[ that are obviously variable within  natural populations do not seem to lie at the basis of many major adaptive changes, while those [genes] that seemingly do constitute the foundation of many, if not most, major adaptive changes apparently are not variable within natural populations.”

    In Darwinism Defeated (1999), Jonathon Wells says

    Second, in order for natural selection to produce evolution, a population must include suitable variation.  No one doubts that natural populations include variations, but are those variations the kind that can lead to large-scale evolution?  Modern research on the genetic basis of adaptation suggests that they are not.  Geneticist John McDonald considers this “a great Darwinian paradox,” since those genes “that are obviously variable within populations do not seem to lie at the basis of many major adaptive changes,” while those that “seemingly do constitute the foundation of many, if not more, major adaptive changes apparently are not variable within natural populations.

    In Debating Design: From Darwin to DNA, Volume 10 (2004), Stephen Meyer (same author of Darwin’s Doubt) says

    The problem has led to what the geneticist John F. McDonald has called “a great Darwinian paradox”.  He notes that genes that vary within natural populations affect only minor aspects of form and function, while genes that govern major changes – the very stuff of macro evolution – apparently do not vary, or vary only to the detriment of the organism.  As he puts it, “those [genetic] loci that are obviously variable within natural populations do not seem to lie at the basis of many major adaptive changes, while those loci that seemingly do constitute the foundation of many if not most major adaptive changes are not variable…”  (all emphasis in original)

    In Darwin’s Nemesis: Philip Johnson and the Intelligent Design Movement (2006), William Dembski says… well… the same thing.

    This problem has led to what J.F. McDonald has called “a great Darwinian paradox”.  He notes that genes that observed to vary within natural populations do not lead to major adaptive changes, while genes that could cause major changes – the very stuff of macroevolution – apparently do not vary.

    Not the loss of the quoted material and that words like “many major” and “seemingly” and “apparently” have been replaced with “do not lead to” and “do not vary”

    And, finally, we have Meyer’s Darwin’s Doubt  (in both chapters 13 and 16)

    [Chapter 13] This problem has led to what Georgia Tech geneticist John F. McDonald has called a “great Darwinian paradox”.  He notes that the that are obviously variable within natural populations seem to 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.

    Looking at this, we see something interesting.  Behe went from just a straight up quotation.  It’s been modified using almost the exact same language* and continue to add interpretations to the work.  We also see that Meyer is copying his own work from previous years. Why is this an awesome book again?

    Thanks to another reader, I have gotten a copy of McDonald’s actual paper.  Reading it, I wonder how many of the Discovery Institute authors have actually read it.  How many have just taken what one of their fellows said previously and done what most 10th grade students do for research papers (i.e. change a few words in an attempt to avoid plagiarism).

    By adding their own interpretation of the prior authors work (using secondary sources instead of primary sources), quite a bit of error has crept in.  I’m sure it’s just a bit of error, we all know that no creationist would say something that wasn’t true, especially if he is quoting an actual scientist’s paper.

    No, I can’t do it, that level of sarcasm is too much for me.  These people are liars.  Either they are directly lying in order to make someone appear to say something that they actually didn’t or they are the worst researchers ever and shouldn’t be allowed to write non-fiction.  Which is it creationists?

    What I continually find stunning is that creationists find a comment by a scientist (like McDonald or Gould or Darwin for that matter) and try to make it seem like the scientist is saying that there is a fundamental problem with evolution and therefore, design is a valid alternative.  McDonald is a geneticist with Georgia Tech, whose lab is currently doing cancer research.  In the late 90s, he was doing a lot of genome and evolution research.  He’s been doing this research for over three decades.

    So why does a scientist think that there is a great Darwinian paradox and does it really mean that the ID guys have a leg up?

    The paper in question is by John F. McDonald and titled “The Molecular Basis of Adaptation: A Critical Review of Relevant Ideas and Observations”  published in the Annual Review of Ecology and Systematics, Vol 14 (publish 1983).

    Wait… what!?!?  It was suggested to me (a few weeks ago, by a scientist that I greatly respect) that references to things 10 years ago was too long ago.  Even Behe’s first use of this article was over ten years out of date.  We’re now rocking on nearly 30 years since the publication of this article.  At this point, the only way I would use an article like this is for historical purposes or for a piece of data established at the time and not contradicted since then.

    So, Meyer, in Darwin’s Doubt, wants to hang his premise on three words written thirty years ago.  OK, well, let’s look to see what the article actually says.

    The author states that his paper will focus on molecular events that “must have served as the source of adaptive evolution and will continue to do so.”  That’s a rather interesting statement coming from an author who, according to the DI crowd, thinks that there is a “great Darwinian paradox”.

    McDonald continues by asking three questions.

    1. What kinds of changes or variations in the hereditary material will result in the generation of adaptive phenotypes?
    2. What is the origin of these variations?
    3. What mechanisms amplify these variations in populations and species.

    Far from saying that these kinds of things cannot happen, McDonald seems to be intending to explore the concepts and, hopefully, answer the questions that he asks.  (Spoiler alert: He does.)  We haven’t even read a full page of material here and already, it’s not looking good for the creationists.

    McDonald talks a bit about the history of molecular biology, including the substantial levels of genetic variation in populations discovered in the 60s and 70s.

    Then he discusses the first area of concern (I’ll try, but please ask me to explain if you get lost).  The concern, at the time, was that “why are the organisms in a population so variable”.  For example, humans have some 700+ variations of the HLA-B allele.  Why do all those variations exist?  If one of the alleles is “best”, then shouldn’t the others be “less fit” and slowly be eliminated from the gene pool?

    I’ve even heard creationists use this as an attempt to support Young Earth Creationism.  If, they say, evolution is correct, then species will, over time, lose unfit alleles and they will all be very, very similar.  Since this isn’t the case, there must not be enough time for evolution to have acted and therefore the Earth is less than 10,000 years old.  I could drive an aircraft carrier through the holes in that argument.  But, if we ignore the conclusion, it really is a good question.

    The conclusion at the time was that most of the variations were adaptively insignificant or neutral.  There was some uncertainty about whether there was enough non-neutral variation within populations for evolution to occur.

    The argument is this:  If all these variations are effectively neutral (meaning there is no evolutionary significance), then how does evolution function when there’s nothing to work on?

    In the late 70s and early 80s, geneticists began to realize that there was something else going on.  High school (and freshman college) biology rarely gets beyond the “DNA is the code for proteins” level of knowledge.  And this is probably where a lot of the problems lie.

    The scientists of the time where beginning to see genes that didn’t code for proteins.  Instead, they were regulatory genes.  They controlled how and when other genes were activated.

    Think of it this way.  There really isn’t a gene for height.  Tall people don’t have one allele and short people have a different allele.  Or even two or three alleles that combine to make tall, short, or something in between.  There is a human growth hormone produced during development.  And perhaps dozens of other genes that control how much and for how long that human growth hormone is produced.  Variations in the growth hormone DNA will probably result in a protein that doesn’t do what it’s supposed to (resulting in massive dwarfism or death).  However, variations in those regulatory genes won’t prevent the protein from working, just change how much is produced.  Less protein results in a shorter person.

    The data, even in the early 80s, suggested that mutations respond to environmental forcing and that’s where McDonald is going with this.

    McDonald then spends five pages discussing the evidence that supports the “Darwinian paradigm”.  He closes with the statement “In summary, there appear to be significant levels of regulatory gene variation within natural populations.”  He uses this figure to describe the hierarchical structure of gene regulatory networks.


    Basically, mutations in the regulatory genes closer to the protein coding level (in this example R1 and R2) would have less of an effect on the phenotype than a gene like R3, at the bottom of the diagram, that controls many regulatory genes.  If R3 is mutated and doesn’t work, then everything else below it fails.  Genes like R3 are highly conserved within a population.  Where something like R1 might exhibit quite a bit of variation making (say if phenotype 1 is ‘height’).

    Why don’t Meyer and the rest ever talk about those sections of the paper?  Nope, they only interested in one part… the part where Darwinism is called into question.

    McDonald then spends six pages talking about how to determine the adaptive significance of this naturally occurring genetic variation.  There are no fewer than seven examples (keeping in mind that this is from 1983).  Then he talks about the “abundance of regulatory gene variation in natural populations, with potential adaptive significance”.

    Note that this information is very, very different from the “do not vary” claims of Dembski and Meyer.

    If you’re still here, stay with me.  We’re getting close.  I’m going to pull a few passages from the paper and we’ll get to the famous paradox.

    On page 91, McDonald writes

    Based upon the results presently available [ed: 1983], therefore, it is reasonable to conclude tentatively that much of the genetic variation that exists in natural populations has potential adaptive significance, at the least.

    Now, why doesn’t Meyer and the rest mention this?  Because it directly contradicts their claims.  And they don’t want that.

    McDonald explain why the quote I copied above is true, then says this.

    As mentioned above, much of the adaptive genetic potential contained within populations may be conditional.  If this hypothesis is correct, it would present a formidable methodological problem for those attempting to evaluate the adaptive significance of naturally occurring genetic variation.  At the same time, however, the phenomenon of conditional adaptive significance seems to subvert many of the early theoretical objections based upon the concept of genetic load that were raised against the Darwinian paradigm.

    Basically, this is saying that the fitness of an allele is conditional.  Think about sickle cell anemia here.  The sickle cell condition actually has some adaptational advantage, provided it exists in an environment with the malaria causing parasite.

    Although the paleontologists’ objections are directly relevant only to Darwinian conceptions of evolutionary rates, they have indirectly cast doubt on Darwinian conceptions of the mode of evolution as well.  Even closely related species are frequently distinguished by major structural, functional, and/or behavioral characteristics that make them adapted to their particular niches.  however, since the magnitude of interspecific adaptive variation for the most part substantially exceeds the degree of phenotypic variation observed to be segregating intraspecifcially, it seems reasonable to conclude, as Darwin did, that selection gradually brings about major adaptive changes.

    This is something that I didn’t realize had been known in 1983.  It’s certainly a well established fact now.  Basically, there is more variation within a population (think about the population of domestic dogs or people for that matter) than there is between other species (wolf, coyote, and dog for example).

    Again, this is easy to show and it’s how evolution works.  Under certain conditions large dogs might have an advantage.  Under other conditions small dogs might have an advantage.  Which is more evolutionarily fit?  The question makes no sense because we don’t have any conditions to base a decision on.  Say something happens and domestic dogs have to survive without humans.  Big dogs are OK because they can tackle big prey.  Small dogss are OK because they can live on small prey and hide easily.  Medium dogs may get the worst of both worlds.  They are too small to get big prey, but to large to survive on small prey.  They are too small to fight effectively, but too large to hide effectively.  After a few generations, we have, for all practical purposes, a speciation event. We have big dogs, which never breed with small dogs.

     From the perspective of the neo-Darwinian geneticist, these challenges have been more irritating than devastating.  Effective counter arguments have been presented showing that strict gradualism is not an inherent characteristic of modern synthetic theory. Moreover, it has been pointed out that what a paleontologist considers a rapid rate of adaptive change may, in fact, be viewed as a quite comfortable pace by neo-Darwinian geneticists.  This the neo-Darwinian view of evolution, whereby adaptive genotypes are drawn from the store of genetic variation that segregates within species, is not necessarily incompatible with the evidence of rapid adaptive change. [my emphasis]

    This paragraph pretty much dismantles the entire line of argumentation used by Meyer in Darwin’s Doubt.  I wonder if he even knows about?  The evidence suggests that Meyer has never read this article.  If he has, then he is lying about it.

    This information, which really is devastating to Meyer’s claims, have been well known for almost 30 years prior to the publication of his book.  This would be the equivalent of someone writing a book about how there is no way fission could ever be used as a weapon… in 1974.

    McDonald then gives two examples of what he’s talking about before stating

    The problem for the Darwinian view, however, is that there has not been any variation observed within natural populations in those regulatory loci set deep within the control network (Figure 2), as was mentioned above.  In other words, the results of the last 20 years of research on the genetic basis of adaptation has led us to a great Darwinian paradox. Those loci that are obviously variable within natural populations do not seem to lie at the basis of many major adaptive changes, while those loci that seemingly do constitute the foundation of many, if not most, major adaptive changes apparently are not variable within natural populations.

    A devastating problem for Darwinian evolution… unless one has actually been reading up to this point.  McDonald continues in the very next paragraph.

    If the genetic material for major adaptive shifts is not present within species’ gene pools, it must be provided de novo by some sort of mutational events(s).

    He then describes two experiments and provides references to two more than support this view.

    These observations, and similar ones in other organisms, have currently led to the view that rapid genomic restructurings play an important role in the de novo establishment of reproductive isolation in higher eukaryotes.

    As significant as molecular drive mechanisms may be for the rapid formation of prezygotic reproductive isolation, they are only one class of the macromutational processes that are now known to be occurring within the eukaryotic genome.  Recent findings indicate that pieces of DNA may not only become rapidly amplified, but they also may be deleted, rearranged, and translocated to new sites in the chromosome.  Evidence of these macromutations imparts an overall picture of the eukaryotic genome that is quite distinct from the rigid structure envisioned only a few years ago.  The genome is not static; it is dynamic.  Moreover, it may be playing a much more active role in the adaptive process than was previously believed.

    Sorry for the long chunk of text, but this is really central to defeating even modern creationist claims… and it’s from 1983.  Meyer was an idiot for including this in his referenced materials.  Anyone who reads this article for understanding will see that Meyer’s arguments are utterly and completely wrong.  Additional support for these came very recently too.

    Let me put in a few more quotes from the article, then I’ll do a quick sum up.

    In this section, I have been arguing for a renewal of interest in the relative importance of mutation in evolution, since it is the only apparent source of the major regulatory chances seemingly associated with major adaptive shifts.

    In fact, recent evidence suggests that the rates of many mutational events are not always low and constant, but rather that they may increase dramatically during periods of environmental challenge and the consequent organismal stress.  The implications for adaptation of such a scenario are significant; at precisely those challenging moments in evolutionary history when major adaptive shifts are required, genetic mechanisms exist that increase the probability that the appropriate variants will be provided.

    He then provides a number of examples of this.

    What he is saying, and Meyer is explicitly saying doesn’t happen, is that when there are major environmental stresses, mutation rates in the regulatory genes increase.  This provides for greater variation in future generations ensuring (hopefully) that at least some of the population survives.

    Note what has happened in this article.  McDonald describes the history of the concept under discussion.  He describes the issues and the supporting evidence.  Then he describes the solution to the problem.  Meyer and the other creationists have focused on the problem, without mentioning that the solution (and supporting evidence) are contained just a few pages further in the article.

    This is the kind of deceptive practice that most upsets me.  No, that’s not true, concern trolling pisses me off more.

    I will close with one more statement from the paper.

    The result of the analysis is that, although the basic Darwinian tenet of natural selection remains intact, the paradigm nevertheless needs significant revision.  The results of research in evolutionary genetics over the last 20 years indicate that the adaptively significant variation that exists within natural populations may not be the primary source of many, and perhaps most, of the major adaptive shifts that have occurred over time.  This paradox, coupled with recent molecular evidence that many, if not most, mutation events significantly increase in frequency in response to environmental stress, has prompted construction of the following “new wave” evolutionary paradigm.

    According to this approach, species adapted to their particular ecological niche are effectively protected against most changes in their adaptive needs by the store of selectable variation that is present within their gene pools.  Environmental challenges that are stressful to organisms will elicit elevated rates of mutation, consequently increasing the level of genetic variation within the challenged population or species. The periodic introduction of mutational variation into populations has both positive and negative effects.  While the genetic load may be increased, this negative aspect can be more than offset by the generation of new, selectively advantageous alleles.

    And this is just describing McDonald’s solution to the problem of the “great Darwinian paradox”.  It’s certainly not a death knell for evolution and anyone who actually reads the article will understand that.

    Meyer and the other creationists are desperately hoping that no one will read the article and show them that they are no only wrong, but their arguments have been countered for decades before they even made the arguments.

    How much of McDonald’s work here is accurate?  As I mentioned, the research in this is still going on, but it seems to me (from a 30,000 foot overview) that McDonald made some good points and some of the more recent research seems to support him, at least partially.

    Of course, now we must ask Meyer and the others, why McDonald’s one statement, in the middle of this paper is acceptable for them to use to support their views, but the remainder of the paper is not.

    * I’d be wondering about copyright infringement, but as all these authors are working for the same organization, I’m sure that’s not an issue.

    Category: Book ReviewCreationismEvolutionGeneticsResearchScienceSkepticism


    Article by: Smilodon's Retreat