The Blind Watchmaker, Cont’d

comments 11
Book Reviews / Reflections / Science

This week I finished the audio version of The Blind Watchmaker by Richard Dawkins, which I discussed in my previous post, and I’d like to give a few responses to the book as a whole now that I’ve completed it.

Dawkins’ overall objective with this work is to describe how the dizzying mélange of organized complexity we find in the biological world can be explained by the properties of non-living matter, the blind mutations of genetic material in organisms, and the power of cumulative selection. He does an admirable job of this, but by the time I’d completed the book I realized I had questions about things that weren’t discussed by Dawkins. Also, the digressions he chose to make at the expense of adhering to the essential components of his explanatory thesis, ultimately disappointed me.

One of the most important assertions of the book is that there’s no point in the process of evolution we cannot explain from the previous. The trick to understanding the stunning complexity we behold in the world around us is to take something bedazzling, like the eye, and then imagine a series of very small steps that might have occurred over a very, very long time to produce it. If you put some sincere effort into it, says Dawkins, you’ll see there is some step-size small enough that you are able to imagine it. I agree this is doable.

What’s hard to envision is how there’s enough time for all of the teeny-tiny steps to have occurred, but Dawkins says that’s not supposed to be obvious. Our brains are not adapted to assess what a million years of time really is, never mind billions. An hour, sure. A lifetime, of course. We have some basic comprehension of a decade, a few generations of family, a century. Maybe a millennium. But a million years? It’s like staring into the sun.

The truth, he says, is that there’s been plenty of time. At known mutation rates, one biologist Dawkins referenced—whose name unfortunately escapes me—has calculated that a mouse could evolve to be the size of an elephant in just 60,000 years, if it weren’t for the obvious push-back such a mouse would receive from the forces of natural selection. In short, time is not the issue.

(There’s also chance enough for the really tricky parts, but that’s for another day.)

Natural selection is a cumulative selection process, meaning that it operates on inherited characteristics, and at each step of the way selects only the best-adapted individuals for reproduction. Because of its cumulative nature, it is more powerful than we may guess. It is sensitive, for instance, to the minuscule advantage delivered by each of the very small steps we envisioned above. In one example—the development of wings—he notes that a small flap of skin between the limbs or fingers of a rodent, no matter how small it might be, would help them navigate gaps between tree branches with just that much more reach and control. As a result, the organisms possessing this dermal oddity would have an advantage.

But how does such a sensitive process work when an organism is developing everything at once—eyes, ears, livers, brains, wings, bones, skin, hair, hearts, digestive tracts, vocal chords, etc.? Do we break all of those into tiny steps and overlay them? Clearly some of these must precede others, so it’s never really all at once. But it’s also clear that organisms evolve as wholes—simultaneously, if you will, with respect to one another, the environment, and the characteristics they already possess, which are already profoundly interwoven. Dawkins never addresses the question of how such a highly sensitive process selects for all of these mutually interacting adaptations concurrently. It might be a bad question on my part—so bad he didn’t think it needed to be addressed—but I’m curious.

In a chapter that comes much later, he provides an explanation for why the fossil record doesn’t show a steady development of particular features. Skull sizes in mammals for instance, would in theory follow some sort of progression. As brains enlarged in response to the increasing complexity of physical organisms and the adaptive advantage of intelligence, skull sizes had to increase to suit. More recent fossils should in general have larger skulls than earlier ones. That’s not always the case, but Dawkins has a reasonable explanation.

At any given time, variations in the expression of a given trait between members of the same species can be fairly significant. The largest human brains are about 40% larger by volume than the smallest ones, for instance. Given how spotty the fossil record is, we shouldn’t expect to discover a particular linear progression. This explanation makes sense in and of itself, but it left me uncertain about how to square the highly sensitive form of natural selection described above with the normal statistical variation for a trait within a species. Is there a contradiction here? I don’t know enough biology to know. It seems that either a tiny difference matters profoundly, or it doesn’t, right? I can imagine that the answer involves the phrase “it’s complicated,” and that both principles apply at different times, based on conditions and context. But Dawkins doesn’t address this. Again, I found myself wanting to understand how biologists may have reconciled these seemingly different conditions.

I’m running out of space here, but two more items were notable by omission. First, consciousness is not addressed at all. This is in keeping with the paradigm of hierarchical reductionism in which Dawkins operates, which he describes in one of the opening chapters. Presumably consciousness is no different qualitatively—in terms of explaining its origin—than mating behaviors or eyeballs. As such, there’s no real need to discuss it. For me, though, it’s an intriguing omission. I’m not sure that the equation of consciousness with rabbit ears rises to the level of common sense just yet.

It’s axiomatic, however, for Dawkins. He postulates (without saying this explicitly) a universe in which consciousness is the product of material evolution, and once he does this, the use of evolutionary theory to prove such a vantage becomes a circular argument. No more or less circular than a theist positing a God and finding evidence for this assertion, but circular nonetheless, so I can see why he leaves it alone. What he does feel obligated to emphasize is that he prefers his own form of circular argument to the ones made by “redneck creationists”—a preference the astute reader has comprehended long before he allows himself the satisfaction of name-calling.

Second, there is very little discussion of how the genome actually relates to the formation of the body itself. There is a really interesting passage on embryological development, but it is very high level. What Dawkins doesn’t address head-on is the need for particular types of relationships between genes and features. His examples throughout the book hinge upon the implication that infinitesimal differences in organisms not only afford them a reproductive advantage, but are inheritable, which means there must be clear genetic distinctions between an organism with a winglet that produces a few square millimeters of skin between two bones and one with a winglet a few square millimeters larger, and one with a winglet a few square millimeters larger than that, and so on.

One can easily imagine a genetic algorithm to produce this—there might be a gene that stores a value for how much skin to produce in a particular location, that could be adjusted like the volume dial on a radio—but that’s hardly the point. Dawkins himself argues that the genetic code is more like a recipe than a blueprint for each and every bit of tissue, so how does this work? Without a hereditary mechanism that is as granular as the countless tiny steps taken by evolution itself, the explanation given by Dawkins would break down. Aside from being a necessary element to deliver a cogent thesis, the question is also a fascinating one, but it is left untouched. We are left to assume that it’s perfectly obvious the genetic algorithms within an organism’s DNA contain the specificity and granularity required to accommodate, uniquely, each digitized step in evolutionary unfolding.

Given the state of genetics at the time the book was written, maybe Dawkins wished he could have said more on the subject, too. I would have much rather he mused on this topic than on the theory of how life might have begun as self-replicating crystals in seasonal streams, or the differences between various forms of taxonomy, or on disputes between various fields of evolutionary research. He included these topics principally to disarm the theists, I believe, not because they’re vital to an explanation of adaptive complexity.

This leads to the central failing of the book as a whole for me: it left intriguing scientific questions central to his argument untouched, and focused on philosophical pursuits instead. The result is a book that suffers for being of two minds. We lose the incandescent thread of wonder at life’s uncanny success, and find ourselves instead as audience to a squabble. For me there’s no squabble per se, so this was a distraction. But the questions raised were enjoyable to ponder, and I look forward to finding other works and authors in the future who may have addressed such questions.


    • Hello Brad, definitely not a book for everyone, but as I said in my previous piece, there are elements of Dawkins’ explication of science that are brilliant and inspiring. It is too bad in my mind that he was caught up in some of the other stuff, as you say. But maybe that was the motivation that brought him to the page, and without it we’d not have the benefit of his other thoughts.


      Liked by 2 people

  1. I’ve never read The Blind Watchmaker, but I’ve read enough Dawkins to see that he can, at times, be a brilliant science communicator. But his preoccupation with religion can limit his effectiveness.

    On the other hand, it’s pretty tough to write a general book on evolution and not get at least a little bit involved with the controversies. Have you ever noticed that many of the fiercest atheists are evolutionary biologists? I think they’ve historically felt the pressure from creationists and IDers in a way that the other sciences haven’t, even ones that really are just as problematic for young earth creationism, such as geology, archaeology, or astronomy.

    I can’t blame him for not addressing consciousness in this type of book. It’s a pretty deep rabbit hole. I’ve read books specifically about the evolution of the brain which avoided the c-word.

    I’m surprised he didn’t at least go over the basics of genetics, how they code for proteins, not necessarily visible features. It’s probably just as well. A lot of what he might have said would have been pretty dated by now.

    Thanks for the review!

    Liked by 2 people

    • Thanks, Mike. Agreed that Dawkins is brilliant. No question. As to the religious stuff, it’s not even that he chose to address the controversy per se, but about how he did it. As I said in a comment to the previous post, I do believe he received hate mail and various unfortunate and unnecessary communications himself. It would be hard to have that stuff in your face and not feel strongly about defending oneself. I get it.

      On the basics of genetics, he did talk about it some as I mentioned. But, as good science writing inevitably does, I suppose, it led to questions. I’m curious to understand more about how the genetic information is structured, such that it is able to both store such a lengthy sequence of minute changes, but also be as effective as it is.

      I did find this paper on the formation of eyes in mammals, and it shows how remarkably complex the process is. I wonder if Dawkins may have inverted the real nature of things to facilitate an ease of understanding on the part of a reader. For instance, he talked a lot about making very small steps in the realized biological structure–such as the eye itself–as a way of showing evolutionary feasibility. (If we can imagine it, then it could be done ostensibly.)

      But… the actual small steps would be in the genes themselves, not in the eye, and it’s not at all clear to me what a particular change in a gene means for the extent of realized change in the eye itself. This is important to understand, if one is to accept Dawkins’ arguments. Given the way genes actually work, it may just not be possible to have a unique set of genes for each step in the formation of an eye that we, the reader, can imagine. It seems like that could, or would, change his argument. When I read that article I linked to, I’m absolutely clueless as to how one would walk this back through time, particularly given that some or many of the genes involved are involved in processes beyond the eye alone. That’s not me saying it couldn’t be done. Only suggesting it remains as fascinating as ever to me in its complexity. Perhaps moreso.


      Liked by 2 people

      • Hi Michael,
        When it comes to genes and observable features, I think we have to see those features as emergent from the genes and all the intermediate layers between. Genes code for proteins, the expression of which can be modified by other DNA sequences. The combination of proteins make up the cellular environment, and really the type of cell. Different cells types produce different types of tissue, which in turn produce organs, which produce observable traits.

        Seen that way, it’s easy to understand just how contingent on environmental factors the macroscopic traits are, and how they’re usually the result of many different genes working in combination. So the macroscopic trait is emergent from the gene / protein / cell / tissue / organ framework.

        So, as mere mortals who can only hold so much in our mind at a time, I think it makes sense to ponder what happens at the molecular level, of genes whose reproductive success is tied by whatever effects they have on the environment (for them the cellular environment and everything about that).

        But when thinking about the traits and what effect they have on the overall organism’s ability to survive and reproduce, I think we have to be aware of the role genes play, but getting too preoccupied with it can cloud the basic issue. A worm with a light detector has an advantage over a worm without one. Another worm with two detectors, allowing it to compute the direction of the light, has an even greater advantage. More light detectors equal yet more advantages, until we have enough for the animal to identify things like food, predators, etc. The ability to swivel the detector surface (retina) and focus light on it are just added capabilities.

        Each of these capabilities come about due to changes in genes or epigenetic factors. It’s important to understand the relationship between the layers. But trying to work with all of them at the same time probably isn’t productive.

        At least that’s how I keep it straight. Your mileage may vary. 🙂

        Liked by 2 people

  2. Thanks, Mike.

    When you say that trying to fathom all the stuff that happens in the background may cloud the basic issue–that basic issue being the benefit an organism enjoys from possessing a certain faculty, trait, or characteristic that others do not–then I’m not sure we’re in agreement on what the basic issue is. Of course, there’s many basic issues, and I certainly don’t disagree with the one you’ve identified.

    In the book, Dawkins notes that the only way complexity could have developed as it has is if a couple of very specific things are true, and I’m taking a bit of liberty here with his exact words. I think basically you need spontaneous differences between organisms to occur, some of which may be advantageous, those differences must be inheritable, and those differences must be selected by giving those organisms the reproductive advantage. I think we agree on all of this, no?

    So, what you’ve described as the basic issue is that some differences between organisms are advantageous, with which I agree. But comprehending that doesn’t eliminate the fact that for the theory to function as Dawkins described it, this is only one requirement. He went to some length to state that large jumps in genetic space–meaning that randomly leaping from one set of genetic values to another–really just doesn’t work. The odds of a good outcome are far, far worse than hoping to win the lottery. So organisms must evolve in small increments, and those increments must not only be advantageous, but heritable. And the only mechanism we have to work with, according to Dawkins, is the genes.

    So I don’t see how we can avoid seeking to understand this. I’m curious about it, too, from a basic desire to understand it. It seems like looking under the hood will lead to lots of interesting insights and new understandings, some of which may have an impact on how we envision this process and ourselves.


    Liked by 2 people

    • We’re in agreement. My only point is that you can think of those things separately. As long as you remember where mutations come from, you don’t have to necessarily think about genes when thinking about whether a particular trait is beneficial or harmful.

      But I certainly wouldn’t say you shouldn’t try to understand it, if you find it interesting.

      Liked by 2 people

  3. This is the question that is the most meddlesome/disruptive to me at the moment in terms of the science as you explained via this Dawkins audio book, and myself on the periphery, have noticed this contradiction as well: the contradiction you point to here,

    “Given how spotty the fossil record is, we shouldn’t expect to discover a particular linear progression. This explanation makes sense in and of itself, but it left me uncertain about how to square the highly sensitive form of natural selection described above with the normal statistical variation for a trait within a species. Is there a contradiction here? I don’t know enough biology to know. It seems that either a tiny difference matters profoundly, or it doesn’t, right?”

    Yes and maybe there’s a metric for which basket to put it in: basket one: variation matters -and basket two: variation matters not. This is amusing to me because unless you observe ALL the possible expressions in a full and complete fossil record, how can you be certain of your statistics on variation?

    It’s been so long since I approached these questions or brushed up on any texts on the topic/s. My understanding of this is also outdated. At some point you have genes and you have genetic expression, or phenotype. There’s complication with alleles… I certain that traits are not one to one, they are complex interrelationships. You toss in the epigenetics (the way the genes environmentally respond as well as within the cell) and you’ve further confounded the issue. It’s stunning how much imagination is still being used here instead of actually mathematically modeling evolutionary biology. Maybe there are models like there are models of climate change, and maybe… my comment has reached its resting place for the evening!

    Thanks for your thoughts, Michael. I appreciate your reviewing this and sharing it.

    Liked by 2 people

    • Haha. Thank you, Ka! I sense from Dawkins’ book there probably is a fair amount of mathematical modeling going on, at least to establish that particular ideas can explain certain outcomes. But all the models Dawkins described were simplified ones that dealt with a few traits at a time, or that were based on a rigid one-to-one relationship between gene and trait, etc. I’m certain there are more complicated models out there. What I’m not certain about is that we’re able to a) model what’s really happening, which means b) the most complex models we make can probably confirm just about anything. I’m being slightly crass there, but once a model contains a bunch of assumptions, ad hoc forces or inputs that are used to simulate something much more complicated that occurs in real life, etc., then they need to be checked against experiment, and that’s a lot tougher to do in biology… Particularly on things that take eons of time to occur…

      But… it’s fascinating! I was digging around the other day and saw that the DNA at the top of a redwood tree is different than the DNA at the bottom, which at first glance sounded like a game-changer to me. I mean, I had this mistaken assumption that an organism’s DNA doesn’t really change–it may very rarely mutate, and when those mutations are captured in the gametes (for animals) they are passed on, and then natural selection can go to work on them–but this article was saying a 2,000 yr old redwood may have all sorts of mutations, and that’s just one tree! But I guess any given organism could have all sorts of mutations in various cells that don’t become systemic, or even get passed on. So maybe that’s just normal and I didn’t realize it before.

      At any rate, thanks for sharing in this moment of wonder, Ka!


      Liked by 2 people

      • The redwood DNA is fascinating. I want to know more about that. DNA is just one component [ to study] of many factors involved in understanding the processes of evolution and development. I was excited when I realized that within my pregnancy, I got to become a chimera (having more than 1 DNA)! Many times fetal cells will even stay in the body of the mother the whole rest of the life. Now, that’s more than a weekend workshop can give in terms of transformation! There’s no end, I think to trying to take on the learning process of where different disciplines such as evolutionary biology and other fields like geophysics interact as well. No doubt there needs to be a foundation for a discussion. Much of what I learned in anthropology in my undergraduate degree was non-focused on the physical aspects of evolution, such as genetics. I took similar survey or entry level courses in cultural anthropology and linguistics. I had, for example, only two classes in physical anthropology and sex and evolution. I also never read up much on it afterwards. Much that I learned was also a very novel scientific argument that we originate from orangutans. I guess that was enough for me to realize that this Harvard guy –Dr. Jeffrey Schwartz– my professor, and any other experts for that matter, have their arguments–and don’t have to agree. There is of course a dominant viewpoint, and I suppose that would be Richard Dawkin’s “Selfish Gene.” Or, may it was. I didn’t even really learn mainstream viewpoints, back then, though I was aware that I was being shown a competing argument. That’s when I became familiar with Dawkin’s name, but I never read (or listened) to any of his work. I agree with you on the mathematical modeling existing; perhaps my point was that despite the existing modeling there’s a lot of imagination involved – and seemed to be emphasized (at least in your review) so thank you for clarifying. Those mathematical models as you said only being one-to-one gene to trait relationship doesn’t necessarily make sense because of the complexity of the alleles and the way certain traits gets activated. So much of what was known when I went to school in the fossil record, is completely blown away. I consider that degree to have been (mostly) useless to me. There’s another point I wanted to make, but I can’t wrap my head around it at this moment without taking a lot of time to be precise with my words. It would be great if I were precise and studied with my words. Why am I commenting, anyways? Because it’s enjoyable. That said, could you help me understand EZ water and the Rainbow and the Worm, by Mae-Wan Ho? I like to share in the wonder, and I need some help getting caught up. Thanks!

        Liked by 2 people

        • Hi Ka,

          Thank you for the lovely ramble! I don’t know much more about the redwood myself. But it intrigued me. I think that some of the things Dawkins took somewhat for granted twenty years ago have already changed quite a bit. That’s part of the beauty of science I think. It keeps moving… It’s the part that’s moving that interests me I think…

          A couple of things I think have changed since The Blind Watchmaker was published are a deepening understanding of epigenetics, the realization that there’s not really “junk” DNA in the human genome, a deepening understanding of horizontal gene transfer, and the realization that due to the relationships that exist between macro-organisms and bacteria, there’s not nearly as bright a line as we thought where one organism begins and another ends…

          As to EZ Water and the Rainbow and the Worm, I’d love to share what I know. Those are a few of my favorite topics, or at least related to them. It’s way more than I could do justice in a comment, though! I’ll do a post or reach out by e-mail, Ka… Thanks for your enthusiasm and interest.

          And yes, it’s enjoyable!

          Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.