Home » Intelligent Design » Someone’s wrong. Who is it?

Someone’s wrong. Who is it?

P. Z. Myers, writing over at Pharyngula:

Neo-Darwinism does not predict that early development will be conserved.

Neo-Darwinism does not predict that early development will be conserved.

Neo-Darwinism does not predict that early development will be conserved.

…That early stages should be more resistant to change is not a prediction of evolutionary theory; it’s an inference from molecular genetics, that genes at the base of a long chain of essential interactions ought to be less likely to vary between species. What that doesn’t take into account is that genes are part of the great cloud of environmental interactions that go on to generate a selectable function, and that if the environment in which the gene is expressed changes, it can enable great changes in the activity of the gene.

Kenneth Miller and Joseph Levine, in their textbook, Biology: Discovering Life, p. 162 (2nd Ed., D.C. Heath, 1994):

Why, then, should the embryos of related organisms retain similar features when adults of their species look quite different? The cells and tissues of the earliest embryological stages of any organism are like the bottom levels in a house of cards. The final form of the organism is built upon them, and even a small change in their character can result in disaster later…

The earliest stages of the embryo’s life, therefore, are essentially “locked in,” whereas cells and tissues that are produced later can change more freely without harming the organism. As species with common ancestors evolve over time, divergent sets of successful evolutionary changes accumulate as development proceeds, but early embryos stick more closely to their original appearance.

Look. I really don’t have a dog in this fight. I happen to accept common descent anyway. I also realize that ontology doesn’t recapitulate phylogeny, as Haeckel mistakenly believed. But when I attended high school in the 1970s, the similarity of [vertebrate] animal embryos in the early stages was touted as one of the four main arguments for evolution – the other three being comparative anatomy, fossils and convergent evolution. (I guess I’m showing my age here.) All I’d like to know is: is this similarity in the early stages still considered by biologists to be a valid argument for evolution? Because if Professor Miller is right about embryo development, then it could reasonably be taken that way, but if Professor Myers is right, then I really don’t see why embryological similarities at any stage (early, middle or late) prove anything about evolution, one way or the other.

Color me confused. I think the good professors need to talk. It sure helps to get your story straight.

  • Delicious
  • Facebook
  • Reddit
  • StumbleUpon
  • Twitter
  • RSS Feed

18 Responses to Someone’s wrong. Who is it?

  1. I don’t have a copy of Miller & Levine, so I am not sure of the context. At least, on a superficial read of that quote, I don’t see any obvious disagreement.

    PZ was saying that it is not conserved across all species. Miller & Levine seem to be talking about a single species.

  2. Hi Neil. I realized that my original quote from Miller and Levine was a bit too short, so I’ve added a couple of extra sentences from page 162 of their book, in order to provide the full context. I think it should now be clear that Miller and Levine are talking about multiple related species. Sorry for the brevity of my original quote.

  3. 3
    Elizabeth Liddle

    My guess is that Miller is talking about really really early stages – the conceptus and morula (“embryo” isn’t the best term here, but it’s become widely used), and Myers is talking about embryos as illustrated in Haeckel-esque drawings.

    Miller actually uses the word “earliest”.

    And he’s right – the first stage of differentiation involves the expression of genes that essentially decide which side is top and which bottom, then which side is head and which tail. These really are highly conserved genes and from that stage, the differentiation is pretty well “locked in”. One a cell is designated “head end” it can’t become “tail end”, and vice versa.

    However, when it comes to the rest of embryology, what happens in what order is going to vary hugely from species to species, as the expression of even of genes that are widely conserved across species are governed by different regulatory genes, affecting the developmental timetable.

    So it’s interesting that in some species, certain features that are shared with related species show up at a particular embryological stage, and are then reabsorbed (occasionally not quite fully, like the hindlimbs of whales) later.

    But given what we know of the evolution of developmental processes (whatever mechanism you want to postulate for the origins of genetic variance!), then there is no terribly strong reason to expect that “ontogeny will recapitulate phylogeny” although of course sometimes it will, given that at least some time-tabling genes are also conserved.

    Does that make sense?

  4. I don’t see what M & L saying as being a prediction of neo-darwinism.

    I see it as an post hoc, ad hoc explanation.

  5. Thanks for extending that quote.

    I am still doubting that there is a contradiction. If you watch that video of PZ, he does say that the pharyngula stage is similar for many species, which seems to agree with the Miller quote.

    PZ’s comments were centered around that discredited recapitulation idea, which claimed that there was a lengthy period of early development that reflected evolutionary history. Miller is saying that small changes in early development would cause large differences in the fully developed organism. I don’t see them as disagreeing.

    All I’d like to know is: is this similarity in the early stages still considered by biologists to be a valid argument for evolution?

    I’m not a biologist, so I can’t speak for them. It does seem to me that the observed similarity is consistent with evolution, so supports the theory in that sense of being consistent. It is not a knock-down proof of evolution.

    Much of the evidence that is consistent with evolution is also consistent with an intelligent designer following an evolving plan. It’s a bit of a mystery to me why ID proponents chose to pick a fight with science.

  6. Hi Elizabeth and Neil,

    Thank you very much for your posts. I’m well aware that ontogeny doesn’t recapitulate phylogeny (that was drummed into me even when I was in high school), and I’ve just added a short sentence to my post to clarify that point, so Professor Myers doesn’t get the wrong idea about my question.

    Elizabeth, you attempt to reconcile Miller & Levine with Myers by proposing that during the very earliest stages (conceptus and morula) there really is a high degree of conservation, but that at later stages, a great deal of variation between species is possible. But Professor Myers actually wrote in his earlier reply to Jonathan M. on June 17:

    It is not a tenet of evo-devo that primitive structures must follow identical ontogenetic pathways. We actually understand that divergence can occur at all stages of development.

    Neil, you point out that Professor Myers states that the pharyngula stage is similar for many species. But the pharyngula stage follows the blastula, gastrula and neurula stage, and Elizabeth was arguing above that we should observe a high degree of similarity up to the morula stage.

    Another thing I’d like to know is this: how do you measure “similarity”? There must be a metric for it. Without a way of quantifying it, all this talk about degrees of similarity is hand-waving nonsense. Using the eyeball test, it doesn’t seem to be true that the earliest stages are the most similar. A quote from CreationWiki illustrates this point:

    Differences among organisms have traditionally been assumed to become more significant as development proceeds, and therefore earlier embryonic stages should resemble one another more than latter stages. This view can be found presented in popular biology textbooks.[19] Although these assumptions have persisted into modern times, it has been recognized by embryologists since Darwin that this doctrine of developmental biology “is not in accordance with the facts of development“, according to British embryologist Adam Sedgwick (1894) who also stated: there is no stage of development in which the unaided eye would fail to distinguish between them with ease“.[22] Modern embryologists confirm this as well, and have further challenged the erroneous nature of these assumptions. Recent work has revealed that embryos are not more similar during the earlier stages of cleavage and gastrulation, but in fact, are most alike at the later stage illustrated by Haeckel and recently photographed by Richardson. In the gastrulation stage, fish show dramatic differences when compared to amphibians, and neither are similar to reptiles, birds, or mammals. Organisms begin life very dissimilar, then become somewhat alike during embryo development, then diverge again as adults. Evo-devos (evolutionary developmental biologists) refer to this pattern as the developmental hourglass. [23] William Ballard adds that only by semantic tricks and subjective selection of evidence can we claim that “gastrulas” of shark, salmon, frog, and bird are more alike than their adults.[24] (Bold emphases mine – VJT.)

    So what I’d like to ask is: is the hourglass real, or isn’t it? Did CreationWiki get this right or wrong?

    Here’s another damaging admission by Professor Myers:

    [W]what we see despite differences between species is a widely conserved molecular homology — that there is an interplay between BMP and Dpp in defining the prospective nervous system in flies and vertebrates. These deep homologies in organization were not expected and not predicted by evolutionary biology, but their presence does imply evolutionary affinities. (Emphasis mine – VJT.)

    Hang on a sec! If evolutionary biology didn’t expect or predict these deep homologies, and if it’s consistent with their absence, than how can they be used as evidence for evolution?

    Finally, I’d like to reiterate that I have no problem with embryology being used as an argument for common descent. I’d just like to know what the argument actually is. If I can’t even state it coherently to myself, how can I possibly hope to convince anyone else of it?

  7. 7
    Elizabeth Liddle

    I think the arguments for common descent from homology embryology are no different from the arguments from any other homology – the best argument is the nested hierarchy of heritable features, and those now include genes.

    Evo Devo is interesting in that it provides a mechanism by which a small adjustment to a regulatory gene can result in a fairly large, functionally coherent phenotypic effect.

    But it’s not an argument for Darwinian evolution over ID, just an argument for, as you say, common descent.

  8. Elizabeth:

    Thanks for your post. It helped to clarify matters. I especially liked your comment:

    … the best argument is the nested hierarchy of heritable features, and those now include genes.

    and also:

    But it’s not an argument for Darwinian evolution over ID, just an argument for, as you say, common descent.

    Thanks again.

  9. I think the arguments for common descent from homology embryology are no different from the arguments from any other homology – the best argument is the nested hierarchy of heritable features, and those now include genes.

    Genes nested within genes?

    If evolution is true, where are the gene hierarchies?

    Ex: G0 = ancestral Gene

    G0a = descendant of G0
    G0b = descendant of G0
    G0a1 = descendant of G0a

    etc.

    Genes should be nested into phyla, classes, orders, etc. too, right?

  10. The reason PZ thinks that Neo-Darwinism doesn’t predict that early development will be conserved is that he’s an incompetent logician (yes, I know, shocker).

    Think about building any complex entity, like say a house.

    What would happen if you started making random changes to early stages of the building plan, while leaving the later stages unchanged – changing the layout of the foundation, for instance, without changing the layout of framing that goes on top of it in later stages? You’d have a house getting placed on thin air, which of course wouldn’t work at all and would fall apart immediately. Making changes in the early, foundational stages of a building process requires making deliberate, corresponding changes to the later stages, or the whole thing goes to pot.

    On the other hand, if you make random changes in the later stages once all the foundational stuff is in place (like the placement of the siding) or if you make random additions to a finished house, you will likely screw something up, but you’ll probably still have a standing house.

    Hence, absent some very specific reason to the contrary, a rational person would predict that a mindless, tinkering process like Darwinian evolution would work primarily by adding to or tinkering with a finished or mostly finished product.

    PZ hasn’t given any reason why that expectation is wrong (and no, “because we don’t observe it” doesn’t count), but then I doubt he understood the reasoning behind the prediction in the first place.

  11. vjtorley (#6):

    Another thing I’d like to know is this: how do you measure “similarity”? There must be a metric for it.

    You have touched one of my pet peeves.

    Philosophy, behaviorist learning theory, machine learning theory from AI; they all depend on similarity. The child is said to hear the word “doggie” in a number of similar situations, and because of that similarity, he learns the meaning of the word.

    If you ask the philosophers or the behaviorist psychologists or the AI learning people what they mean by “similar”, they quickly become evasive. The closest they can come is the compare instruction in a computer, but that does not do what is needed for the way that “similar” is used.

    As best I can tell, there is no suitable account of “similar.” I don’t have a reference, but I seem to recall reading a paper or book by Quine, where he admitted that there is no satisfactory philosophical account of “similar.”

    Without a way of quantifying it, all this talk about degrees of similarity is hand-waving nonsense.

    Yes, I agree. That’s part of the reason for the name I chose for my blog.

    Roughly speaking, all of philosophy is suspect. All of philosophy of science is suspect. Science, for the most part, is safe because it can at least fall back on “it works.” However, some scientist’s explanations are suspect.

    As for the eyeball test of similarity – I don’t have a problem with that. I agree, it is about what we mean. But before you can use that, you need an account of perception (of getting those eyeballs to work) which does not itself depend on that eyeball test notion of similarity.

    So what I’d like to ask is: is the hourglass real, or isn’t it?

    I’m not a biologist. But the explanation that PZ gave in his recent Edinburgh talk at least seemed plausible. His view was that the early differences (at the gastrula stage) are mostly due to differences in egg structure. As the embryo develops toward the pharyngula stage, those differences begin to disappear. But, following the pharyngula stage, differences in development toward different body shapes and structures begin to show up.

    In any case, I agree with you that there is probably a lot of hand waving (i.e. vague, unconvincing argumentation) coming from all sides in this particular dispute.

  12. Deuce

    Think about building any complex entity, like say a house.

    What would happen if you started making random changes to early stages of the building plan, while leaving the later

    stages unchanged – changing the layout of the foundation, for instance, without changing the layout of framing that goes

    on top of it in later stages? You’d have a house getting placed on thin air, which of course wouldn’t work at all and

    would fall apart immediately.

    (Warning, may contain some handwaving)

    Now, I’m not a logician, a builder, a developmental/evolutionary biologists or even necessarily completely sober at the moment, but I like to think of your basic blank-canvas house as the pharygula stage of housing (hint I definitely am making this up as I go along).

    Ok, your basic blank-canvas house should have the following: It should be supported by the ground — not be floating in the air; It should also have surrounding walls, with apertures that may later become doors and windows; an electricity supply; an optional gas supply; access roads, drainage nearby; and so on. It should contain: an area suitable for general living/sleeping; another area with access to drainage/water for bathroom / powderroom / ladies’ and/or gentlemens’ room / unmentionables room; an area with drainage/water suitable for preparing food. And those last two must be different rooms.

    It doesn’t matter how you got to that state. Maybe you started with the roads, then the drains, then the electricity, or maybe you started on the house first (so you had to overcome some problems delivering the building materials) then waited for the other stuff. Maybe you had one big delivery of bricks at the beginning which you had to work around, or perhaps you built a mini train set so that your mother could deliver exactly the right brick to you at precisely the right time. OTOH, maybe you just slotted together some prefabricated components. At various times along the way an onlookers may observe significantly different scenarios.

    It doesn’t matter: What matters is that all efforts converge on a basic ISO-XXXX house structure that the various busybodies and unions are able to tolerate and work upon.

    *** Pharyngula stage ***

    Now you can go about customising it: Your interior designer can deck it out with fur, mock tudor beams, MDF, Formica, Flock wallpaper etc.

    And your electrican can fit light bulbs hanging from electical wiring, or use subtle indented lighting, dimmers, flashing LEDs, blinding lasers and so on, or you can decide to go with candles and/or darkness.

    Your kitchen outfitter may install a microwave, an Aga, a gas/electric oven, a walk-in refridgerator, a telephone with preprogrammed local fast food suppliers on speed dial or any combination of the above. Whatever. In short, things start to diverge again.

    And during or after all of this you can grow your kitchen to ten times the normal size (or half of it), gradually change the windows from rectangular to ellipsoids, flatten the roof (or turn it into into some sort of dome), split the bedrooms/living area into two or more parts – all of which end up much larger than the original. Naturally, the plumbing and electrics will adapt because they are not static artifacts but systems that can grow and change – and, and, … Hmmmm, maybe growing houses isn’t the same as growing tetrapods in some ways.

  13. Isn’t there anything from the fruit fly literature to tell us about what changes to genetic material that are involved in early development, do to it.

  14. Elizabeth:

    I think the arguments for common descent from homology embryology are no different from the arguments from any other homology – the best argument is the nested hierarchy of heritable features, and those now include genes.

    How are yo defining “nested hierarchy”? I ask because the ToE does not predict such a pattern to arise from descent with modification.

  15. Joseph:

    ToE does not predict a nested hierarchy? That´s an interesting claim. Maybe you can explain in a bit more detail what you mean by this?

  16. Still waiting for that list of gene species and the nested hierarchies they fall into.

    If evolution takes place at the level of the gene, we should see it.

    Or do genes themselves not evolve by descent with modification?

  17. Maybe you can explain in a bit more detail what you mean by this?

    See @16.

  18. You know, I just love it when evolutionists argue at cross purposes.

    Either they do not understand their own theories, or it is in fact the case that their theories are contradictory.

    One one hand we have the evo supporter claiming that the nested hierarchy is not a prediction, but rather an pattern that the theory attempts to offer an explanation for.

    On the other hand we have the evo supporter evincing astonishment to hear that the nested hierarchy is not a prediction of the theory.

    Now, shouldn’t they be arguing with each other? Why do they save their objections for the ID critic?

    Darwinism – it’s whatever you need it to be.

Leave a Reply