A bit more on natural selection

After publishing my last post I noticed I’d downloaded from somewhere (maybe the Sanford article) a 2008 paper by Austin L Hughes about the methods that are used to detect genes that have been subject to natural selection.

Hughes is a professor of biology at South Carolina University with 35 years of experience, a mass of well-cited articles to his credit and a wide range of research interests: Evolutionary Studies, Conservation Biology, Computational Biology, Vertebrate Biology, Insect Biology and BioInformatics.

The paper’s quite short, and although a little technical in the middle is well-worth reading. One of the things, in discussions on sites like BioLogos, that intimidates the non-professional who has doubts about the sufficiency of natural selection is the citing of many examples of genes that have been subject to selection. There’s clearly a tool, or tools, available to biologists-who-understand-evolution that removes all doubt about the matter. However, Hughes’s article not only casts reasoned doubt on the methods being employed for this purpose, but shows they are fallacious by a specific and clear example from the literature (the opitimisation of eye pigments). Here’s an important, and challenging, summary statement:

Contrary to a widespread impression, natural selection does not leave any unambiguous “signature” on the genome, certainly not one that is still detectable after tens or hundreds of millions of years. To biologists schooled in Neo-Darwinian thought processes, it is virtually axiomatic that any adaptive change must have been fixed as a result of natural selection. But it is important to remember that reality can be more complicated than simplistic textbook scenarios.

Hughes goes on to make the same point I raised in my post: that the evidence seems to point to nearly all biological change being, according to the evidence, due to near-neutral chance mutations not subject to positive selection. This does raise once more the spectre of a complete lapse into implausibility – if evolution by natural selection made it intellectually viable to be an atheist, neutral evolution throws one back either on many-world multiverses or miracles. And the paper makes no effort to resolve that, having demolished (once again) the foundation for Neodarwinism. It’s no wonder that, despite neutral theory’s increasing dominance, we’re always tossed the bait of Darwinian natural selection – and gradualist selection at that – to assuage any qualms we might have about getting up Mount Improbable in one piece.

But I suspect Hughes may have a more sophisticated and interesting take on these matters from another article I’ve just found by him, which is an able critique of scientism in The New Atlantis. The guy is clearly widely-read and a serious thinker – and scientists like that seem to buck the mainstream trends more, sadly, than most Christians who are scientists do, if BioLogos is any indication. I don’t propose to review this article, as it’s quite absorbing enough in itself – read it and get back to me with your thoughts.

Avatar photo

About Jon Garvey

Training in medicine (which was my career), social psychology and theology. Interests in most things, but especially the science-faith interface. The rest of my time, though, is spent writing, playing and recording music.
This entry was posted in Creation, Science, Theology. Bookmark the permalink.

15 Responses to A bit more on natural selection

  1. pngarrison says:

    Jon, I don’t know why the quote above is supposed to be a surprise. I’ve never seen anything but statements agreeing that the signs of selection are obscured by recombination after a short time, in evolutionary terms. That doesn’t mean that selection didn’t occur.

    Did you look at the exchange Lou Jost pointed to between Paul Nelson and Jerry Coyne at WEIT? Nelson quoted a list of evolutionary luminaries including Lynch to the same effect that you are, and every one of them repudiated Nelson’s interpretation of what they said in the strongest possible terms. You can believe what you want (and I don’t know enough to have a definite opinion), but I don’t think you have the support of the experts that you think you do.

    • James says:

      Hello, PNG.

      Nelson wasn’t saying that selection “didn’t occur.”

      Coyne is hardly an impartial debater, and of course is going to frame the quotations to suit his ends. But note that Larry Moran, of all people — no more a friend of ID than Coyne — *defended* the rough sense of Nelson’s point, if not Nelson’s exact phrasing. You can find a long discussion of this at:


      It’s always good to get both sides of the story before deciding who’s right.

      Regarding Jon’s remarks, I took him to be saying, not that natural selection does not occur, but that much of evolution takes place “outside of selection” so to speak. And this is not particularly an ID or creationist position, but has been put forward by many mainstream evolutionary biologists who have no religious commitments.

      To say that much of evolution takes place outside of selection is of course not an anti-evolutionist position. It’s merely a rebalancing position, contradicting the extreme selectionism of people like Richard Dawkins, regarding the various *causes* of evolutionary change.

      I grew up on popular expositions of neo-Darwinism. Some of them were written not by mere journalists but by people with Ph.D.s in the various relevant sciences. And the dogma then being pushed was that “random mutations” were filtered by “natural selection” and that these two factors together accounted for all biological innovation and all stabilization of innovations. *Every* change was under selection. This dogma was based on a crude conception of genes, of proteins, of genotype-to-phenotype correspondence, etc. It has been steadily under attack since the days of the Wistar Conference in 1966, and evolutionary theorists have been modifying this crude dogma in light of empirical evidence that evolution is much more complicated than that.

      Many biologists have since argued that “random mutation” is not nearly as important a contributor to biological novelty as was supposed in the days of the Modern Synthesis. Margulis argued that genome mergers were much more important, and that individual random mutations were almost negligible in importance. Others have argued for inherent structural constraints determining form that operate outside of selection (Newman, etc.). Others argue for a quasi-Lamarckian ability of organisms to re-engineer their own genomes (e.g., Shapiro). The whole point of the Altenberg conference — attended by non-religious scientists whose special field is evolutionary biology — was that the Modern Synthesis appears to have some serious inadequacies and may need major extensions and modifications.

      I’m not competent to enter into detailed discussion of evolutionary mechanisms; I’m merely reporting on what I read. And what I read tells me that the popular, simplified, pure Modern Synthesis conception of evolutionary change — pushed for years by Sagan, Dawkins, etc., and now championed by TEs like Ken Miller and Dennis Venema — is increasingly considered naive by many research biologists whose special field is evolutionary mechanisms. But pure geneticists like Venema, or pure cell biologists like Miller, who don’t actually work in evolutionary theory per se (their c.v.s show zero [non-popular] articles in the field), appear to be almost completely oblivious of these developments. This is why I think that BioLogos is actually doing harm to the very cause it champions. By trying to harmonize Christianity with a form of evolutionary theory that the cutting-edge theorists are increasingly questioning, it runs the risk of exposing Christianity to ridicule when science changes.

      BioLogos desperately needs a regular columnist who is both Christian and a front-line researcher in evolutionary biology, who goes to all the conferences, publishes in all the journals, and keeps up with the field. Until they get such a person, their writing on evolution and faith will be of very little value. Of course, if their only goal is to convince fundamentalists that evolution in fact occurred, that doesn’t matter much. But if their goal is to present themselves to the world as *serious* thinkers in the field of “theology and science,” it matters very much. A version of evolutionary theory that pretends that the objections of Gould, Margulis, Newman, Jablonka, Shapiro, Gunter Wagner, etc. don’t exist is a version of evolutionary theory that has no academic credibility.

      • Avatar photo Jon Garvey says:


        Granddaughter duty this weekend, so will have to try and get a timely and inadequate reply in rather than attempt to fully document. Accordingly, I’ll follow the Nelson-Coyne links later – after all, I haven’t cited or even read Nelson at all for my piece, and Eddie, as so often, picked up the nuances of what I did write. Meanwhile, I am sympathetic to GD’s suggestion that a question along the lines, “Are you for science or do you support creationists?” isn’t always going to clarify the views of, say, Lynch that much.

        Over on BioLogos I made a simple suggestion to you that truly “neutral mutations” are really not common, and that most are slightly deleterious. (as usual melanogaster, I see, has read it differently and started an inquisition – I’ve never found it productive replying to him, and even less so now). “Towards slightly deleterious” seems to be the trajectory the neutral theory took after Ohta, and not only according to Sanford et al. Wikipedia says under “Mutation”:

        “A nearly neutral mutation is a mutation that may be slightly deleterious or advantageous, although most nearly neutral mutations are slightly deleterious.”

        So was my point to you mistaken, or true?

        I should add that we now know a lot more about overlapping genes, alternate splicing etc, so synonymous mutations, always assumed to be neutral, are likely to be more deleterious than hitherto thought.

        Then I read in Eyre-Walker et al, Quantifying the Slightly Deleterious Mutation Model of Molecular Evolution that they find most mutations to be strongly deleterious (and so subject to purifying mutation) but:

        “Slightly deleterious mutations, those mutations with effects close to 1/Ne, could be a substantial fraction of fixed mutations because we found evidence for differences in selective constraint and the kinds of amino acid that are fixed between species with different recent effective population sizes.”

        (They are talking about small population sizes, which I’ve assumed throughout because that’s the sutuation with most higher vertebrates). So regarding molecular mutation at least, selection clearly has some issues. As Hughes writes (from Wikipedia again):

        “Evolutionary biologists typically distinguish two main types of natural selection: purifying selection, which acts to eliminate deleterious mutations; and positive (Darwinian) selection, which favors advantageous mutations. Positive selection can, in turn, be further subdivided into directional selection, which tends toward fixation of an advantageous allele, and balancing selection, which maintains a polymorphism. The neutral theory of molecular evolution predicts that purifying selection is ubiquitous, but that both forms of positive selection are rare, whereas not denying the importance of positive selection in the origin of adaptations.” In another essay, Hughes writes: “Purifying selection is the norm in the evolution of protein coding genes. Positive selection is a relative rarity — but of great interest, precisely because it represents a departure from the norm.”

        I gather neutral theory, a la Kimura at least, maintains phenotypical selection. Koonin’s summary seems less sympathetic to that, but in any case I note how seldom the two are distinguished at, say, BioLogos, where it’s genetic selection that has been discussed, not least on the recent thread: one changed gene and the giraffe gets a new neck.

        On to this background I have discussed a couple of papers by Sanford et al, and (vis a vis GD’s other point) these werenot review arrticles interpreting Lynch, etc, but were original research using a new software platform to quantify deleterious near-neutral mutation loads more exactly. From what I’ve read in the articles they cite, they have not misrepresented the views of those papers – I can’t speak for Nelson, and neither should I need to. But even if these authors had misunderstood them, their results should be assessed on their own merits.

        Certainly I cited Koonin 20o9, having read his paper myself, and it’s hard to take this quote as a full endorsement of natural selection:

        “Equally outdated is the (neo)Darwinian notion of the adaptive nature of evolution: clearly, genomes show very little if any signs of optimal design, and random drift constrained by purifying in all likelihood contributes (much) more to genome evolution than Darwinian selection.”

        Similarly James Shapiro http://www.huffingtonpost.com/james-a-shapiro/does-natural-selection-evolution_b_1769524.html mentions that the idea that NS is a stabilising force goes back to Wallace, and goes on:

        “Darwin was correct in believing that natural selection could only play a creative role if variation was continual and small. Based on the ‘Uniformitarian’ idea that change had to occur in the small steps we observe under normal conditions, he originally insisted that evolutionary innovation had to be gradual.”

        So if gradual changes, as the work I’ve cited suggests, actually passes under the radar of natural selection, then how can natural selection work at the genetic level – or even, without some new understanding, at the phenotype level?

        I suppose what gets my goat a little is seeking quite hard to get the sense of what’s going on, reading about a bitter internal wrangle over adaptationism which still rumbles today, and then being told over and over again that nobody in science is disagreeing at all, and it’s all to do with creationists deliberately misunderstanding the research. I read the research, and not the creationists (in this case Paul Nelson) … and still get bracketed with him rather than with Koonin or Shapiro, whom I have read.

          • Avatar photo GD says:

            I have read this article and now have to scratch my head even more… statements such as …”It shields the cell from the harm of the mutation, allowing the mutation to get passed down to the next generation and spread throughout the population.” leave me asking more questions than ever before, esp. regarding the ubiquitous law of natural selection. I know this is not what such a statement may mean, but it highlights the difficulty …”does the cell sit down and decide what it should shield? Does something else do it for the cell? Or is this another point for endless argument?

            I am glad that people are examining various aspects of Darwinian thinking, but I am more convinced now (then before) that a lot more is needed before we can accept the claims of many evolutionists. I am looking through papers that discuss evolutionary ethics, aesthetics and similar subjects. I must confess that as a scientist, I feel staggered at the hubris shown by atheists and evolutionists. I think that Darwin may fit in with the (relative term) thinking of the likes of Dalton and Boyle, but his followers have exaggerated the scientific merits of his ideas.

          • Avatar photo Jon Garvey says:

            Thanks for posting that PNG. Zimmer’s article half-correlates with a quote from the 2009 Koonin paper (he’s basically a neutral theory guy too):

            There is no consistent tendency of evolution towards increased genomic complexity, and when complexity increases, this appears to be a nonadaptive consequence of evolution under weak purifying selection rather than an adaptation.

            So we have in him the “complexity arises free of selection” story, but also an observational claim that complexity hasn’t increased much, which seems to run counter to the “first law of biology.”

            It all shows that natural selection isn’t quite the monolith it was, yet there still seems to be a lack of flesh on the bones, as in other “self organisation” theories. In particular, as critics like Erwin seem to be saying, “stuff tends to get more complicated” is fairly contentless as a recipe for coordinated systems.

            In nature, snow cornices get bigger and more complex from the wind’s action – until they collapse completely. Or add more massive bodies to a Newtonian system and at some stage it will usually become unstable.

            In human affairs, machines tend to get more complicated because a lot of brainwork is done to hold everything together. Conversely, prog rog or big band music – or buraeucracies like the EU – get more complicated (perhaps by lack of design constraint!) until they just become top heavy and go extinct.

            How can complexity itself be a mechanism? Given the vast possible search space, why should there ever be a neutrally-formed enzyme handy to cope with a new mutation, let alone the billions that actually have arisen in the history of life?

            Such ideas are interesting to me, though, (a) because NS seems inadeqate and (b) because the evidence seems to suggest a lot of non-selected change is actually going on. But do we have a real plausible mechanism? As GD says below, it’s not quite proven – and is getting quite far from Darwinian processes too.

        • pngarrison says:

          As to the point about what percent of mutations are really neutral or deleterious, after looking at some of your references I think the confusion is that they are talking about just mutations in protein coding regions, while I was talking about all mutations. The ref you gave on Biologos about mice gave a total deleterious rate per genome per generation of about 0.5. If the total mutation rate in mice is comparable to humans (it shouldn’t be too different), ~60 per diploid genome per generation, the deleterious/total rate would be 1:120. We were talking about 2 different things. Not surprising that the % deleterious in coding regions only would be much higher.

          It may be that the resolution (partly) to the larger question is that while selection won’t work for very weakly positive mutations (s ~ +1/N), you can still get gradual evolution with mutations that have fairly low s, but still substantially above 1/N.

          In the Koonin quote above, he seems to be talking about genome evolution, not organismal evolution. It’s my impression that that has been the point of much of Lynch’s work, that genome evolution, like the expansions with transposons, rearrangements, etc. has been largely near neutral. Genomes aren’t organisms.

          I have read some of the papers on accumulation of slightly deleterious mutations in species with small populations (Kondrashov, Crow). I don’t know what the resolution is, but there must be one, since we aren’t extinct.

          Also, I’m confused. Does James = Eddie? Are they right and left hemispheres?

          • Avatar photo Jon Garvey says:

            I suspect from some of my own reading that the difference between total-genome and protein-coding regions is significant as you say, though it gets above my pay grade very rapidly at that point. Maybe that’s partly because of how often the organism is spoken of as really nothing more than a hard copy of the genome.

            I can get my head round natural selection working on gross traits produced by the sum-total of genetic output – though that still seems a blunt tool compared to Darwin’s concept that can cope with near-infinite numbers of variations.

            But how does one distinguish “neutral” from anything else in the non-coding genome? Surely that depends on very variable degrees of (known) function, and interpretation will change with each new function found.

            For example, while a pseudogene is not functionality can presumablt be a complete Frankenstein and yet “neutral”. But the recent work that suggests some pseudogenes function as rate-limiters for expression of real genes would mean that the same change would be highly deleterious, and consideration of “mildly deleterious/beneficial” would be similar to coding genes.

          • Avatar photo GD says:

            I have looked up some references and also did a quick lit search for positive mutations and beneficial mutations in humans – the best that I have come up with seems to be either a positive selection (which seems another version of natural selection) which may favour beneficial mutations, and some discussion which may or may not be beneficial (e.g. change of pigment). I don’t think that searching for more papers would help, as it seems that statistical methods have been used to locate regions that are equated with mutations over time, and these seem ‘a needle in a haystack’ exercise.

            Whatever the results however, I cannot find anything that would amount to a clear and positive proof that mutations and natural selection amount to a simple and ‘astoundingly’ elegant scientific account of variations over time. Perhaps you may have a reference that would provide such a clear account.

          • Avatar photo Jon Garvey says:

            PS – I think James is that “Jon Garvey” bloke over at BioLogos, but I’m not sure who’s me here.

  2. Avatar photo GD says:


    Two points regarding the ‘e-mails of denouncement’ referred to in the link in the previous post regarding arguments about natural selection.

    The first is somewhat humorous, because as I read the emails and responses on that blog, I could not help but smile, as I remember that Mao had his red book, and all Chinese had to perform some type of swearing to uphold the thoughts of Mao. All that is left for Darwinian chaps is public pronouncements swearing belief and adherence to the thoughts of Darwin.
    The second point is the disservice done by anti-evolutionists. By this I mean the use and misuse of published material, in which they strive to give another interpretation of people’s scientific research. I deplore such an approach – if a Darwinist performs research and reports in within the overall framework of Darwin, it is wrong to think the reader would have a better understanding of that scientists work. If anyone takes such an approach, they must obtain their own data and discuss it to show how their data would oppose the discussion provided by the previous research scientist.
    It is important to distinguish between criticisms of the framework (what I often term Darwinian thinking, especially regarding natural selection) and the actual work and data published. I recall my remarks on the Lenski paper discussed on Biologos, I pointed out that the argument of simultaneous and sequential events was addressed in this paper (including attempts at quenching the events so as to ascertain a sequence). The authors concluded they could not give a definitive result, but noted that three events were required before they could obtain data on the fourth. The Biologos chap decided not to clearly state this, while the IC people decided to interpret this to suit their purpose. This is an example where both approaches are wrong.
    The valid criticisms(s) on Darwin is the inadequacy in modelling the bio-world, and this is because of the extraordinary broad approach and an assumption that the entire bio-world (and for some people, the origin of everything, including life, species, the Universe and so on), is scientifically explained by the Darwinian view. This blatantly absurd; this is especially so when a ‘law’ of natural selection is invoked.

    The point is to show the difference between criticism of the overall framework, and that of specific work by scientists. It is also important to realise that valid scientific research can be carried out within an inadequate framework – the history of science is filled with examples of this.

  3. James says:


    I posted as James R over on BioLogos, for several months. Then, one day, I was inexplicably banned. There was no comment from the moderators on any of my posts, at any point. No public comment. No private comment via e-mail. No communication of any kind indicating that I had violated any rule of discourse there.

    In fact, my banishment surprised several commenters there, including a couple, I think HornSpiel and Merv, with whom I was having vigorous — but very polite — intellectual disagreements. One of my opponents called for the moderators to explain James R’s banishment, but no one from BioLogos said anything.

    Since, if I had violated some rule of discourse, BioLogos could easily have offered a public and/or private statement: “James has / you have been banished due to the un-Christian comments made in posts # xxx and #yyy,” and since no such justification appeared, it is a safe bet that no violation had occurred.

    Since there was no violation of the rules of discourse in what I was posting, the best guess is that the BioLogos people did not like the *contents* of what I was posting. The last post that they published, before I was banned, compared ID and TE leaders in regards to their philosophical training. I noted that Dembski, Richards, Meyer, Nelson and others had Ph.D.s in philosophy or philosophy of science or had otherwise indicated awareness of the philosophical questions, and that this was comparatively rare among the leaders of TE. I was frozen out after that — as I say, without a word of explanation. I would guess that the cause was wounded *amour propre*, but of course, that is just speculation.

    What is interesting is that, despite BioLogos’s concern that all comments should be offered in a constructive spirit that typifies Christian discourse, it has always allowed thuggish and insulting commenters to last there indefinitely and without reprimand — provided that they direct their un-Christian attacks against ID people or creationists. Thus, we see that ID and YEC people have been called “cowards” “lacking in faith” “violators of the commandment” (not to bear false witness), “hypocrites,” etc., with the moderators looking the other way; while, e.g., a statement from and ID/YEC commenter that Dobzhansky was far from an orthodox Christian in practice (whatever he was on paper) was jumped on by the management with alacrity, in a chastising public comment — even though that statement appears to be entirely true, as there is very little documentary evidence that Dobzhansky practiced a typical churchgoing Christianity, and his student Ayala stated directly that Dobzhansky leaned toward something close to pantheism.

    The continuing strange alliance between TE and atheism, whereby Behe and Meyer are targeted by BioLogos far more often than Dawkins and Coyne are, and whereby ID/YEC commenters at BioLogos often feel (based on many precedents) that they are on the edge of being banned, whereas atheist commenters never seem to feel that way, is certainly remarkable.

    I think I would not likely have been banned under the new management, which seems to be less heavy-handed regarded policing. But in any case, I came here and found a number of people, including Jon Garvey and penman, who formerly posted on BioLogos, and have found this a site where the issues that are avoided, or at best skirted, by BioLogos, are dealt with here head-on. BioLogos tends to “hide behind” technical defenses of evolution, and its regular columnists very rarely tackle theological issues arising out of evolution in a head-on way. I think that part of the cause of that is that so many of the leading TEs there are trained in the life sciences and are not comfortable dealing with theological or philosophical questions at a high level. But of course, saying that out loud on the BioLogos site is, I suppose, like insulting one’s host when one is a dinner guest. Even if it is said politely.

    In any case, for my money, Hump of the Camel does theology/science discourse better than either BioLogos or Uncommon Descent. It is too bad this site is less frequented than either. It deserves a wider readership. I have said favorable things about it to ID folks I know, but I think I should step up my promotional efforts.

  4. Avatar photo Jon Garvey says:


    Late comment – relating to my low pay grade above. This paper http://www.pnas.org/content/104/30/12235.full by Masatoshi Nei overtly suggests that phenotypic, not only genotypic evolution, occurs primarily by neutral means, with natural selection taking a secondary place.

    The phraseology even of the abstract places clear water between his own neutralist (I think he would say “mutationist”) position and the Neo-darwinian paradigm.

    From the discussion: “The prospective view of evolution suggests that evolution occurs without purpose by mutation and adaptation to new environmental conditions, and therefore it is intrinsically unpredictable. “

    The following paragraphs detail what he sees as the shortcomings of much evidence for phenotypic selection. It’s hard for me to believe that everyone truly is singing from the same hymnsheet, despite the apparent united front against “creationists”.

    I was unaware of this work when I wrote these two pieces, but it seems to suggest I wasn’t greatly misrepresenting this strand of current evolutionary thinking in them. Much evidence seems to support it, but as I said it leaves nature without any truly creative mechanism – unless mutation itself is not actually random and “without purpose”. “Unpredictable” it may be, to us, but it seems possible to predict that it always turns up beautifully functional forms – how does that work, without selection as the main mechanism powering adaptation?

Leave a Reply