Tree? What tree?

Max Telford, who is Professor in the Department of Genetics, Evolution, and Environment at University College, London, has written an engaging book, The Tree of Life: Solving Science’s Greatest Puzzle, which has stirred up reactions at the Discovery Institute site Science And Culture Today. (I have not yet read Telford’s book, except to find a passage where he mentions me, saying that when he attended a lecture by me years ago he found me terrifying. Sorry about that.)

The conclusion that there is a treelike genealogy of life is about 250 years old, starting to gain acceptance even before Darwin’s work. It is central to modern biology.

But this is not the impression you’d get if all you read was the Discovery Institute’s website Science and Culture Today. Instead you’d find Emily Reeves, who describes herself as “a biochemist, metabolic nutritionist, and aspiring systems biologist” questioning the conclusion that there is such a genealogy, in two posts responding to Telford’s book. And even before that, posts by Rob Stadler, a biomedical engineer, concluded that the evidence from homology did not support the conclusion that it came from common ancestry.

What is particularly interesting is that this seems to signal the Discovery Institute’s shifting from arguing that Intelligent Design is detectable in evolution, to arguing that common ancestry is not detectable.

Here are some indications:

Emily Reeves’s argument from design

Emily Reeves has two recent posts responding to Telford’s book. In the first of them, here she argues that the same evidence that Telford takes as supporting an evolutionary tree could also be indicating design instead:

Telford is acknowledging that nested hierarchies or “evolution” arise from non-genetic processes and can be represented by tree structures. Yet, he is also asserting that the genetic nested hierarchy we observe in living organisms constitutes definitive proof that “everything alive today arose from one single origin” (p. 92). Therein lies a logical problem.

She uses as an example the historical tree of languages, calling it the result of “intentional design”. Whoa there. Italian and Spanish did not become different by having a joint design committee choose different words. Italians and Spaniards blundered their way into different words, pronunciations, amd usages, as all peoples do. This makes the process much closer to biological processes of evolution.

Reeves also seems to assume that design processes will inherently lead to a hierarchical pattern of similarities:

he acknowledges that intelligent agents can produce the kinds of patterns of similarity that result in nested hierarchies, and yet won’t even consider the possibility that this could be happening in biology.

Above you can see an example of nonhierarchical patterns of similarity in human design, the tailfins that appeared in various brands of American cars in the 1950s. I remember the phenomena vividly, as I was a teenager then. And Chrysler and Ford did not have tailfins owing to the companies being descended from a common ancestor. They noticed that tailfins were popular, and copied from each other.

Reeves’s other argument

In her second post, here Reeves points out wrong assumptions
used in inferring evolutionary trees, predominantly assuming molecular clocks and similar morphological clocks:

A core prediction of evolutionary theory, and a foundational assumption for constructing the tree of life, is that similarity in phenotypic traits (and especially in underlying molecular characters like genes and proteins) reflects common ancestry.

This is simply nonsense, and shows that Reeves does not understand how phylogenies (evolutionary trees) are reconstructed. Most phylogenies are constructed by methods that do not assume morphological evolution is clocklike. For molecular phylogeies, a molecular clock is either not used, or used only when the species are closely related.

Rob Stadler’s supposed demolition of inferring evolutionary trees

Even before Telford’s book appeared, a medical engineer, Rob Stadler, posted at S&CT. He was citing a 2016 book of his, and his post here was provocatvely titled “When Can I Trust Scientists About Evolution?”. Never, apparently. He is trustworthy and we are not.

He lays out 6 criteria a trustworthy theory must meet, and concludes that the inference of common descent fails all 6. Here is the first one:

Is the evidence repeatable? The process that produced existing lifeforms cannot be repeated. Descent with modification can be repeatedly demonstrated and speciation can be demonstrated on rare occasions, but this is merely inferred to operate at higher taxonomic levels; it cannot be repeatedly demonstrated.

One wonders what he makes of geology, where we can observe local processes of erosion, but not the uplift of whole mountain chains. Not to mention astronomy …

Another of his criteria is:

Can the evidence be directly measured or observed? The passing of genes from parent to child can be directly measured, but the proposed passing of genes from a common ancestor to all existing life certainly cannot be directly measured.

Yup. Just like geology.

Was the evidence obtained through prospective study? Large-scale evolution cannot be studied prospectively.

This one misses entirely what biologists do to confirm results in inferring evolutionary trees. Yes, we cannot simply wait a million years. But we can take the species whose tree has been inferred from one set of characters, and look at another set of characters. We are predicting, not what further evolution will occur, but what inference we will make from a very different set of characters.

It was this repeatability, with morphological characters, that started to convince biologists in the late 1700s that the hierarchical (groups-within-groups) pattern of relationships might actually reflect common ancestry. With the birth and growth of molecular studies of evolution in the 20th century, we saw a massive confirmation of those morphological phylogenies, though with some interesting surprises.

That is prospective study of evolution. Similarly, geologists and astronomers can ask whether their inferences are confirmed by further studies, on different types of data and different places. A good example is the confirmation of plate tectonics, which did not involve waiting millions of years.

Tu quoque?

… which is Latin and means, “and you too?” Does Stadler apply the same strict criteria to inference of common design? Evolutionary biologists, when asked about “common design”, often ask for details. Who is the designer? When we look for a murderer, we ask about motive, means, and opportunity. Do we do this with a designer? For a supernatural designer who is omnipresent, omnipotent, and inscrutable, can anything be ruled out? How does common design explain why elephants are large, gray, and lumber about the savannah eating bushes? If the explanation is simply that “the designer wanted it that way”, there is a problem: it also explains why elephants are small, pink, and flit from bush to bush pollinating them. Is Stadler’s finger-wagging unidirectional? Or does he sometimes try looking in the mirror?

And what about the Discovery Institute?

I detect a gradual shift in the advocacy at S&CT from an emphasis on Intelligent Design to rejection of evolutionary change. In short, towards creationism. Am I being paranoid? I’d like to hear from our thoughtful commenters.