With its generally high-quality online content, Khan Academy exercises a remarkable influence on what students and teachers alike learn about science and other subjects. That’s why I have devoted four posts to analyzing the outdated junk science in their video on “Evidence for Evolution.” After going through supposed lines of evidence from embryos, homology, and fossil horses, the video ends by looking at biochemical similarities between organisms, presented as a suite of arguments for common ancestry.
Khan claims that “How the DNA gets replicated and translated and transcribed is very similar from one life-form to another.” Yet a 2020 paper noted that the “Origin of DNA replication is an enigma because the replicative DNA polymerases (DNAPs) are not homologous among the three domains of life, Bacteria, Archaea, and Eukarya.” These differences are so great that one paper asked “Did DNA replication evolve twice independently?” The paper proposes that “the modern-type system for double-stranded DNA replication likely evolved independently in the bacterial and archaeal / eukaryotic lineages.” And while we’re discussing fundamental biomolecular similarities, another paper compared the genomes of 1,000 different prokaryotic organisms and found that “of the 1,000 genomes available, not a single protein is conserved across all genomes.”
It is of course true that all life uses DNA and proteins. The video argues that this universal similarity across life “hints at a common ancestry.” True, universal common ancestry is one possible explanation for such biochemical similarities — but are there others that go unmentioned by Khan? As we saw with homology in vertebrate limbs, it’s key to appreciate functional requirements. Last year Emily Reeves explained that many properties of the amino acids used in life appear optimal for our biochemical needs. So, there are good functional reasons why all life should use these same molecules.
Moreover, because all life-forms use DNA (which contains nucleotides) and proteins (made of amino acids), we are able to gain nutrients we need — amino acids and nucleotide bases — from the plants, animals, and other organisms that we eat. The fact that all life uses the same basic building blocks is precisely what makes the food web possible! These universally shared similarities might indicate the design of the ecosystem, not common ancestry.
Once we appreciate that there are good functional reasons for life re-using the same basic molecules (e.g., DNA, proteins), common design becomes an alternative explanation. Intelligent agents frequently re-use the same types of parts in different designs to meet functional requirements. Think of how both cars and airplanes use wheels, or how different versions of Microsoft Windows re-use key computer codes. As Paul Nelson and Jonathan Wells observe in the book Darwinism, Design, and Public Education:
An intelligent cause may reuse or redeploy the same module in different systems, without there necessarily being any material or physical connection between those systems. Even more simply, intelligent causes can generate identical patterns independently. … If we suppose that an intelligent designer constructed organisms using a common set of polyfunctional genetic modules — just as human designers, for instance, may employ the same transistor or capacitor in a car radio or a computer, devices that are not “homologous” as artifacts — then we can explain why we find the “same” genes expressed in the development of what are very different organisms.p. 316
The Khan Academy video never considers this possibility, but again, common design — the intentional re-use of a common blueprint or components — is a viable explanation for the widespread functional similarities among the biomolecules found in different types of organisms.
The video then compares humans and chimps, saying that the latter’s behaviors and facial expressions are “eerily human.” As we’ve seen, similarities don’t necessarily require common descent and could point to common design. Speaking of similar behaviors, I could say the same thing about my family cat, Bonsai — who is very intelligent and often seems able to read my mind and anticipate my behaviors. This doesn’t necessarily mean we’re genetically related — it means we were built with minds and mental outlooks that have important overlap. And the fact that our minds and behaviors have important overlap means that we can relate to one another quite well. In fact there are numerous examples of animals within species or across species being able to relate to one another. Just Google “animal friendship videos” and make sure you’ve got an hour to kill and tissues nearby to wipe your eyes! Here’s my point: It’s been said that shared assumptions are the basis of culture. If the designer is a relational being (i.e., a being that’s into relationships), then designing species with mental and emotional similarities that foster intra- or even inter-species deep emotional connections and friendships should not come as a surprise. No common ancestry needed.
But that overlap has significant limits. Just like chimpanzees, my cat doesn’t use complex language, build complex tools, use fire, wear clothing, engage in abstract reasoning, do math, compose music, write poetry, ponder the mysteries of the universe, practice religion, or engage in any number of advanced cognitive or spiritual activities. (I would say that he doesn’t engage in moral reasoning. However, I am pretty sure that he knows there are things I don’t want him to do — and he deliberately does those things, apparently for that very reason, like when he recently woke me up from a badly-needed nap by crying at my bedroom door. In any case, my cat definitely lacks an appreciation for morality as found in humans.) So, while some basic behaviors link all mammals — an intriguing fact that makes many animal relationships possible — there are numerous higher behaviors found only in humans. Somehow these points get left out of the Khan Academy video.
The video then claims that humans and chimps are 98 percent genetically similar, saying “their genes show just how close to human beings they actually are.” That statistic is false: it overstates human-chimp genetic similarity, as I explained here last year; see, “Human-Chimp Similarity: What Is It and What Does It Mean?” At that post, I further explained that any given percent genetic similarity between two species does not necessarily imply an ancestral relationship since that similarity could be present for functional reasons—reflecting their common design.
A Tree of Life?
The video closes by saying that “the fact that we can measure how far things are away allows us to create a very accurate tree of life.” This sounds like the claim I responded to from Richard Dawkins last year that genetics data allow us to create a “perfect hierarchy — a perfect family tree.” Except when you dig into the technical literature you find out this isn’t true at all. I also reviewed this evidence last year in response to Dawkins at “Phylogenetic Conflict Is Common and the ‘Hierarchy’ Is Far from ‘Perfect’.” A very nice treatment of problems with the tree of life is given in Jonathan Wells’s sequel to his book Icons of Evolution, titled Zombie Science: More Icons of Evolution. He summarizes there the many problems facing the tree-of-life hypothesis, including:
- Circularity of homology-based arguments for common ancestry
- A lack of fossil evidence for evolutionary transitions
- The difficulty of establishing ancestor-descendant relationship between fossil species
- The fact that phylogenetic trees are based upon assumptions that similarity results from common ancestry
- The frequent presence of convergence where similarity is not the result of common ancestry
- Frequent conflicts in different versions tree of life
- The presence of “orphan genes” that don’t seem to fit anywhere in the tree of life
- The inability to resolve basic domains of life into a “trunk” of the tree
Perhaps Wells’s most relevant point in rebuttal to Khan Academy comes when the video claims that we can easily determine percent genetic similarity between organisms, and then out pops a tree. Here’s what Wells writes in Zombie Science:
Since the rise of molecular biology in the mid-twentieth century, biologists have increasingly used comparisons of sequences in DNA, RNA and protein to construct phylogenetic trees. For example, a particular DNA sequence might be present in different species, though with minor variations. Comparing the sequence differences in species A, B, and C could lead to an inference that species A is more closely related (that is, more similar) to species B than it is to species C. The similarity between two sequences (often called homology) can be expressed as a percentage, representing how many subunits at corresponding positions are identical between them.
Similarity may be assumed to imply genealogy, but this is only an assumption. Any inference to genealogy based on sequence similarity is hypothetical. And since molecular sequences (with rare exceptions) are available only from living organisms, any inference about the evolutionary past of those organisms-including their ancestors-is even more hypothetical.p. 35
This is exactly right. As I have explained in the past, the basic logic behind building molecular trees is relatively simple. First, investigators choose a gene, or a suite of genes, found across multiple organisms. Next, those genes are analyzed to determine their nucleotide or amino acid sequences, so the gene sequences of various organisms can then be compared. Finally, an evolutionary tree is constructed based upon the principle that the more similar the nucleotide sequence, the more closely related the species. But the whole process is based upon the assumption that genetic similarities between different species necessarily result from common ancestry. But there’s no need for that assumption. If the similarities being compared are functional similarities — which is always the case when you are comparing gene sequences between organisms — then those functional similarities could reflect common design rather than common descent.
Moreover, biology is full of examples where similarity didn’t result from common ancestry. As Jonathan Wells points out, convergence is “ubiquitous” in biology. This calls the basic assumptions of the whole project of treebuilding into question.
Making Sense of Biology
The Khan Academy video opens by quoting Theodosius Dobzhansky who famously stated, “Nothing in biology makes sense except in the light of evolution.” Khan further says that the modern theory of evolution is “about as strong as theories get.” But when, to make the case for the theory, you have to resort to all these old, long-refuted icons of evolution, how strong is the evidence really? It’s worth recalling Jonathan Wells’s remarks at the end of Icons of Evolution:
[T]he claim that “nothing in biology makes sense except in the light of evolution” is demonstrably false. … The icons of evolution are a logical consequence of the dogma that nothing in biology makes sense except in the light of evolution. All the misleading claims we have examined in this book follow from the sort of thinking represented by Dobzhansky’s profoundly anti-scientific starting-point. … [S]cience at its best pursues the truth. Dobzhansky was dead wrong, and so are those who continue to chant his antiscientific mantra. To a true scientist, nothing in biology makes sense except in the light of evidence.pp. 247-248
Wells is correct that when you start with Dobzhansky’s dogmatic statement, you get bad science. You also get bad science education. It’s very unfortunate that Khan Academy — which is so good on so many other subjects — is misinforming students about the evidence, or lack of it, for evolution.