Caught in Contradictions, PZ Myers Claims "Evolutionary Theory Predicts Differences as well as Similarities" (and Therefore Predicts Nothing) (Updated)
In response, PZ did not contest Jonathan M.'s claim that vertebrate embryos have significant differences, even in their early stages. Instead, PZ claimed that none of this challenges evolution, conceding to Jonathan M. that "[v]ariation can and does occur at every stage of development; early and late stages vary greatly." This leads us to our first of three contradictions from PZ:
PZ Contradiction 1: Do Vertebrate Embryos have "Superficial" Differences or do they show "Wide Variation" and "Vary Greatly"?
PZ might not realize it, but when Jonathan M. forced PZ to admit that "[v]ariation can and does occur at every stage of development; early and late stages vary greatly" and say,"it is trivially observable that there is wide variation in the status of the embryo at fertilization," PZ was contradicting himself, as well as PZ's fellow evolution blogger Carl Zimmer. In 2006, Zimmer stated in response to me: "As fellow scienceblogger PZ Myers has clearly explained, the differences in the earliest stages are superficial."
But don't blame Zimmer for the contradiction, because Zimmer was simply citing a post from PZ where PZ did in fact did say: "The differences are superficial." That sure doesn't sound much like the admission that Jonathan M. recently forced out of PZ that "[v]ariation can and does occur at every stage of development; early and late stages vary greatly" and that there is "there is wide variation in the status of the embryo."
So which PZ is correct: the old PZ or the new PZ? If you're wondering, it's the new PZ. As a 2000 paper in Systematic Biology states:
Recent workers have shown that early development can vary quite extensively, even within closely related species, such as sea urchins, amphibians, and vertebrates in general. By early development, I refer to those stages from fertilization through neurolation (gastrulation for such taxa as sea urchins, which do not undergo neurulation). Elinson (1987) has shown how such early stages as initial cleavages and gastrula can vary quite extensively across vertebrates.
(Andres Collazo, "Developmental Variation, Homology, and the Pharyngula Stage," Systematic Biology, Vol. 49:3 (2000) (internal citations omitted).)
Likewise, a 2010 paper in Nature observed that early stages of embryos can vary widely:
Counter to the expectations of early embryonic conservation, many studies have shown that there is often remarkable divergence between related species both early and late in development, often with little apparent influence on adult morphology.
(Kalinka et al., "Gene expression divergence recapitulates the developmental hourglass model," Nature, vol. 468:811 (December 9, 2010) (internal citations removed).)
Contra Carl Zimmer and the old PZ, it sure doesn't sound like differences between early embryonic stages are merely "superficial." Thankfully, PZ now has his story straight and is acknowledging that early vertebrate embryo stages can "vary greatly" or show "wide variation." But there are still more contradictions from PZ to analyze.
PZ Contradiction 2: Evolutionary Biology Predicts Both Similarities and Differences
As the above-quoted Nature article suggested, evolutionists had "expectations" that early stages of embryonic develoment would be conserved. But as the article explained, "Counter to the expectations of early embryonic conservation, many studies have shown that there is often remarkable divergence between related species both early and late in development." So how does PZ deal with the fact that early stages of embryo development can vary greatly -- "counter to the expectations" of many evolutionary scientists? Well, he purports to deny that evolution makes any particular predictions about whether early stages of vertebrate embryo development will be conserved. Here's the rest of the story:
An old adage says that 'the theory which explains anything really explains nothing.' These are words worth remembering as we consider the following comment from PZ in response to Jonathan M.: "I wish I could get that one thought into these guys heads: evolutionary theory predicts differences as well as similarities."
To reiterate comments from my previous article, this is rich: If PZ is correct that evolutionary biology predicts both similarities and differences among embryos, then evolutionary biology makes no predictions and is unfalsifiable regarding the similarities and differences in vertebrate development. According to PZ, evolutionary theory predicts whatever it predicts, conserves whatever it conserves, and modifies whatever it modifies. Some theory.
Such logic might help in saving your theory from falsification, but it doesn't help to construct a robust theory that makes testable predictions.
Let's pause to consider what we've now witnessed.
PZ will probably never admit this, but Jonathan M.'s debate with PZ Myers has now forced a couple significant admissions out of the Professor. First, PZ has admitted that early stages of development can "vary greatly" and show "wide variation," contradicting his claims in the past that the differences are merely "superficial." Second, because of that wide variability, PZ has been forced to put evolutionary biology in an unfalsifiable position to save common ancestry from falsification with respect to the great differences between early stages of embryonic development.
Most recently, PZ has replied to Jonathan M. by castigating him for claiming that evolutionary biology predicts that early stages of vertebrate development ought to be conserved. If you've been reading closely, then you should already be aware that PZ is wrong and that Jonathan M. was not misrepresenting evolutionary thinking: We already saw from the above-quoted 2010 Nature paper that evolutionary biologists had "expectations of early embryonic conservation." So Jonathan M. is inventing no straw man argument of evolutionists here.
To further clarify the arguments from PZ and Jonathan M., this particular debate has now gone through the following 4 stages:
(i) Jonathan M. first observes that early vertebrate development varies greatly, contradicting common claims that similarities in early stages of vertebrate development show common ancestry. In essence, Jonathan M. is rebutting what a recent article in Nature Communications called the Funnel-like model:
"The funnel-like model predicts conservation at the earliest embryonic stage. During embryogenesis, diversity increases additively and progressively. This model is based upon the extreme case of developmental burden or generative entrenchment, in which the viability of any developmental feature depends on an earlier one."
(Image and quotation from Naoki Irie & Shigeru Kuratani, "Comparative transcriptome analysis reveals vertebrate phylotypic period during organogenesis," Nature Communications, Vol. 2:248 (2011).)
As that above-quoted article explains, this model is different than classical Haeckelian recapitulation.
(ii) PZ predictably replies by claiming Jonathan M. is promoting Haeckelian recapitulation, stating "Evolution does not predict that development will conserve the evolutionary history of an organism." This is of course followed by various accusations and attacks from PZ that Jonathan M.'s comments are "stupid," etc. (link). (iii) Jonathan M. replies by explaining that he wasn't claiming that evolution must conserve the evolutionary history of an organism a la Haeckelian recapitulation. Rather, Jonathan M. was talking about the funnel-like model which claims that evolution conserves early stages of development (e.g. a pattern still cited in many textbooks as evidence for evolution--which is different from Haeckel's recapitulation theory). Specifically, Jonathan M. says in reply: "Myers thinks I'm worried about Haeckelian recapitulation. But that's completely wrong." (iv) PZ then replies by again reiterating that (a) "Evolution does not predict that development will conserve the evolutionary history of an organism" because (b) "Neo-Darwinism does not predict that early development will be conserved." And of course, there are more of PZ's customary attacks that Jonathan M. is "completely ineducable," "appallingly obtuse," etc. (link)
The problem is that PZ has conflated classical Haeckelian recapitulation (iv-a) with the funnel-like model (iv-b). Jonathan M. never claimed evolution requires iv-a, and it wasn't the topic of his challenge. Rather, Jonathan M. challenged iv-b, which is not Haeckelian recapitulation. Thus, as I explained in my previous article, PZ is constantly misconstruing ID arguments as if we're attacking Haeckelian recapitulation, when we aren't.
The core argument behind the funnel-like model is that there are certain early stages of development which are crucial to later developmental stages, and thus are said to be resistant to evolutionary change. These unchanged early stages of development are said to reflect crucial developmental processes inherited from a common ancestor. While PZ rejects the funnel-like model, he does adhere to its core argument.
So just where would Jonathan M. get the idea that evolutionary biology cites similarities among early stages of vertebrate embryos as confirming evidence for common ancestry (e.g. argument iv-b)? As we'll see in to response PZ's third contradiction, Jonathan M. might have gotten this idea from PZ himself, who argues that conserved similarities between vertebrate embryos are evidence for common ancestry.
PZ Contradiction 3: No Wait, Evolution Really Does Predict Embryo Similarities After All!
A couple years ago, PZ posted a response to me writing: "Vertebrate embryos at the phylotypic or pharyngula stage do show substantial similarities to one another that are evidence of common descent. That's simply a fact."
Likewise, in response to Jonathan M., PZ argued that "there is an interesting and real convergence on the broad, general outlines of the body plan at one point in development that needs to be explained." In another post, PZ claimed that only common ancestry can answer the question: "Why should animals that differ in appearance as adults exhibit similarities as embryos?"
PZ answered that question in response to Jonathan Wells, making an evolutionary argument for why the phylotypic stage must be conserved: "The specific morphology of the phylotype really does represent a literal foundation upon which the rest of development proceeds, and is resistant to evolutionary change, because there are too many later events that are dependent on it." After discussing some shared similarities between vertebrate embryos, PZ contended that "[t]he best explanation for these phenomena is that they are a consequence of a common heritage."
Hold on! Didn't PZ just tell Jonathan M. that "evolutionary theory predicts differences as well as similarities" and that "Neo-Darwinism does not predict that early development will be conserved"?
Yes, he did. So how is PZ now citing similarities alone as evidence for common ancestry? Here's what I think is going on:
Simply put, PZ is contradicting himself once again. PZ wants to have it both ways where the differences between early stage of vertebrate embryos don't count as evidence against common ancestry, but the similarities do count as evidence for common ancestry. Thus he writes that "substantial similarities" in "[v]ertebrate embryos at the phylotypic or pharyngula stage" serve as "evidence of common descent," and "[t]he best explanation for these phenomena is that they are a consequence of a common heritage."
And PZ isn't the only authority which could have given Jonathan M. the notion that modern evolutionary biology continues to cite similarities between vertebrate embryos as "evidence of common descent." Just about every mainstream biology textbook makes virtually the same argument:
"Similarity among chordate embryos. These diverse organisms appear very similar in the first stages of development (shown in the top row), evidence that they share a common ancestor that developed along the same pathway." (Colleen Belk and Virginia Borden Maier, Biology: Science for Life, p. 234 (Benjamin Cummings, 2010).)I could go on and on with additional examples. PZ may claim (for the dual purpose of saving common ancestry from falsification and berating Darwin-doubting students like Jonathan M.) that "Neo-Darwinism does not predict that early development will be conserved," but many textbooks (and PZ himself!) state otherwise.
"Anatomical similarities in vertebrate embryos. At some stage in their embryonic development, all vertebrates have a tail located posterior to the anus, as well as pharyngeal (throat) pouches. Descent from a common ancestor can explain such similarities." (Neil. A. Campbell and Jane B. Reece, Biology, p. 449 (Benjamin Cummings, 7th ed., 2005).)
"Early in development, the human embryos and the embryos of all other vertebrates are similar. These early similarities are evidence that all vertebrates share a common ancestor. ... They embryos of different vertebrates are very similar during the earliest stages of development." (Holt Science & Technology, Life Science, p. 183 (Holt, Rinehart, and Winston, 2001).)
"Significance of developmental similarities. At this comparable developmental stage, a chick embryo and a pig embryo have many features in common, which suggests they evolved from a common ancestor." (Sylvia S. Mader, Essentials of Biology, p. 226 (McGraw Hill, 2007).)
"Developmental Homology: Structures that Appear Early in Development are Similar. The early embryonic stages of a chick, a human, and a cat, showing a strong resemblance." (Scott Freeman, Biological Science, p. 500 (Prentice Hall, 2nd ed., 2005).)
"The early embryos of vertebrates strongly resemble one another because they inherited the same ancient plan for development. ... From comparative embryology, examples of evidence of evolutionary relationship among vertebrates. (a) Adult vertebrates show great diversity, yet their early embryos retain striking similarities. This is evidence of change in a shared program of development." (Starr and Taggart, Biology: The Unity and Diversity of Life, pp. 316-317 (Brooks/Cole, 2001).)
"In their early stages of development, chickens, turtles, and rats look similar, providing evidence that they shared a common ancestry." (Kenneth R. Miller and Joseph Levine, Biology, pp. 385 (Prentice Hall, 2008).)
"Scientists also make inferences about evolutionary relationships by comparing the early development of different organisms. Suppose you were asked to compare an adult fish, salamander, chicken, and opossum. You would probably say they look quite different from each other. However, during early development, these four organisms are similar, as you can see in Figure 8. For example, during the early stages of development all four organisms have a tail and a row of tiny slits along their throats. These similarities suggest that these vertebrate species are related and share a common ancestor." (Prentice Hall, Life Science, p. 183 (Prentice Hall, 2005).)
And PZ is not alone in making these predictions. In fact, embryology authority Michael Richardson agrees with PZ that vertebrate embryos can be highly divergent in both their early and late stages. (As we'll see in my next article, Richardson goes much further than PZ would prefer, suggesting that vertebrate embryos are so diverse that the pharyngular stage may not even exist.) But Richardson does make an important point which exonerates Jonathan M. from many of PZ's attacks of misunderstanding evolutionary embryology. Richardson acknowledges:
Many biologists assume, as Darwin did, that natural selection acts mainly on late embryonic or postnatal development. This view is consistent with von Baer's observations of morphological divergence at late stages. It is also suggested by the conserved morphology and common molecular genetic mechanisms of pattern formation seen in embryos.
(Michael Richardson, "Vertebrate evolution: the developmental origins of adult variation." BioEssays, Vol. 21:604-613 (1999).)
Richardson notes that "many biologists" adhere to something like the funnel-like model, where earlier stages of development are more crucial to development of the organism and thus more resistant to evolutionary change. This corroborates the aforementioned Nature paper which noted that evolutionary embryologists have had "expectations of early embryonic conservation." Indeed, PZ himself makes precisely this class of argument with regards to the phylotypic stage, stating that "the phylotype really does represent a literal foundation upon which the rest of development proceeds, and is resistant to evolutionary change, because there are too many later events that are dependent on it."
This rebuts PZ's response to Jonathan M. that biologists don't think like this anymore. Given that many textbooks make such claims, it would seem that Jonathan M. is not so "ineducable," but rather has learned evolutionary biology quite well.
Like the new PZ, Richardson's BioEssays piece persuasively argues against the funnel-model on the basis of observations that there is wide variation among vertebrates in their early stages of development.
Also like the new PZ, Richardson reinterprets the predictions of evolutionary theory, arguing that there's no reason for evolution to conserve early stages. What you end up with is something like PZ Myers' unfalsifiable version of evolution: "evolutionary theory predicts differences as well as similarities."
Perhaps that's correct, but it doesn't make for a potent theory. Moreover, it directly contradicts PZ's previous evolutionary argument for conserved similarities a la "the phylotype really does represent a literal foundation upon which the rest of development proceeds, and is resistant to evolutionary change, because there are too many later events that are dependent on it."
Thus, PZ should not fault Jonathan M. for operating under the assumption that early vertebrate embryo similarities are commonly cited as evidence for common ancestry, since textbooks and "many biologists" (including, at times, PZ himself!) make this same argument.
In fact, some data suggests that organisms do resist changes to genes expressed early in development. A recent paper in PLoS Genetics confirms that certain changes to genes expressed early in development tend to wreak havoc with embryogenesis:
We show that, in both species, genes expressed early in development (1) have a more dramatic effect of knock-out or mutation and (2) are more likely to revert to single copy after whole genome duplication, relative to genes expressed late. This supports high constraints on early stages of vertebrate development, making them less open to innovations (gene gain or gene loss).This could suggest that unguided evolution would have difficulty explaining the observed wide variation in early embryonic development that is acknowledged by PZ and Michael Richardson.
(Julien Roux, Marc Robinson-Rechavi, "Developmental Constraints on Vertebrate Genome Evolution," PLoS Genetics, Vol. 4(12) (December, 2008).)
The Old PZ vs. the New PZ
In closing, let's contrast PZ's old arguments with his new ones:
The Old PZ: Differences between early stages of embryos are "superficial," and similarities during the pharyngular (or 'phylotypic') stage "are evidence of common descent." According to PZ there is "conservation of the morphology at the phylotypic stage," and "the phylotype really does represent a literal foundation upon which the rest of development proceeds, and is resistant to evolutionary change, because there are too many later events that are dependent on it." The New PZ: Early stages of embryos "vary greatly," showing "wide variation," and "evolutionary theory predicts differences as well as similarities" because "Neo-Darwinism does not predict that early development will be conserved."
The new PZ further contradicts himself by reverting back to the old PZ, suggesting that evolutionary theory does predict similarities between vertebrate embryos after all, since he continues to cite "convergence on the broad, general outlines of the body plan at one point in development" as evidence of common descent in response to Jonathan M. In the next article, we'll assess whether this alleged convergence--the "pharyngular" stage--actually exists.