On June 26, 2000, President Bill Clinton announced the completion of the Human Genome Project, which had just deciphered the sequence of DNA in a human cell. “Today,” he said, “we are learning the language in which God created life.” At the president’s side was Francis Collins, director of the project, who had helped to write Clinton’s speech. “It is humbling and awe-inspiring,” Collins said, “to realize that we have caught the first glimpse of our own instruction book, previously known only to God.”
As its subtitle indicates, The Language of God presents evidence for Christian belief. Curiously, however, that evidence does not include DNA, which according to Collins provides “compelling” evidence for Darwin’s theory of evolution instead. In the course of defending Darwinian evolution as “unquestionably correct,” Collins argues that intelligent design not only “fails in a fundamental way to qualify as a scientific theory” but is also “doing considerable damage to faith.”
Francis Collins criticizes intelligent design (ID) on the grounds that it fails to suggest approaches for experimental verification, but then he cites experiments that he says prove it wrong. He also criticizes it for being a God of the gaps argument, but only after redefining ID as an argument from ignorance. Collins feels that ID poses a serious problem to Christian belief because it rejects Darwinian evolution, which he feels is supported by overwhelming evidence. But the only evidence Collins cites for Darwin’s mechanism of variation and selection is microevolution – minor changes within existing species. And the principal evidence he cites for Darwin’s claim of common ancestry is DNA sequences that he says have no function – though genome researchers are discovering that many of them do have functions. Collins’s defense of Darwinian theory turns out to be largely an argument from ignorance that must retreat as we learn more about the genome – in effect, a Darwin of the gaps.
From atheism to belief
Part of Collins’s book is devoted to his own journey from childhood atheism to belief in God. In his “most awkward moment” as a physician, a patient asked him what he believed, and he realized that although he was a scientist he had “never really seriously considered the evidence for and against belief.” From that moment on he was “determined to have a look at the facts, no matter what the outcome.”
One of the facts Collins looked at, and the one that impressed him most, was the “Moral Law” — the concept of right and wrong that “appears to be universal among all members of the human species.” Influenced by the writings of C. S. Lewis, and convinced that the Moral Law “cannot be explained away as cultural artifact or evolutionary by-product,” Collins concluded that its source must be God.
Collins continued to seek evidence for his belief. He found some in the Big Bang, which “cries out for a divine explanation.” He found more in the orderliness of natural laws and in the Anthropic Principle, “the idea that our universe is uniquely tuned to give rise to humans.” According to Collins, “the fact that the universe had a beginning, that it obeys orderly laws that can be expressed precisely with mathematics, and the existence of a remarkable series of ‘coincidences’ that allow the laws of nature to support life… point toward an intelligent mind.”
This is not the same as what has become known as “intelligent design.” In fact, Collins is worried that intelligent design (ID) is endangering faith by encouraging Christians to pin their faith on illusory evidence that evaporates in the light of new scientific discoveries.
God of the gaps?
According to Collins, ID rests upon three propositions: (1) “evolution promotes an atheistic worldview;” (2) “evolution is fundamentally flawed, since it cannot account for the intricate complexity of nature;” and (3) “if evolution cannot explain irreducible complexity, then there must have been an intelligent designer.”
Like many other critics of ID, Collins thus begins by using a definition of ID that is different from that of its proponents. According to the web site of Discovery Institute’s Center for Science and Culture (the worldwide headquarters of the intelligent design movement), “Intelligent design holds that certain features of the universe and of living things are best explained by an intelligent cause, not an undirected process such as natural selection.”1
Contrary to Collins’s first proposition, ID does not claim that “evolution promotes an atheistic worldview.” Evolution can mean simply change over time, which carries no religious or anti-religious implications. Or it can mean minor changes within existing species, an uncontroversial phenomenon that is also religiously neutral. Problems arise with respect to Darwin’s theory, though, which excludes design in the details of living things. Darwin himself wrote that he saw “no more design in the variability of organic beings, and in the action of natural selection, than in the course which the wind blows.” So he was “inclined to look at everything as resulting from designed laws, with the details, whether good or bad, left to the working out of what we may call chance.”2 Because Darwinism excludes design in the details, many people (including many Darwinists!) have argued that it “promotes an atheistic worldview.” But the conflict between Darwinism and ID concerns only the question of whether certain features of living things are better explained by an intelligent cause or by undirected natural processes.
Collins’s second and third propositions falsely imply that ID consists only of a negative argument – as though the absence of evidence for Darwinism automatically justifies an inference to design. Yet (as we shall see below) a design inference is warranted only when effects resemble those that we know from experience are due to an intelligent cause. The fact that Darwinian theory cannot explain something may be a necessary condition for inferring intelligent design, but it is not sufficient.
Collins argues that ID (as he defines it) is not science: “Intelligent design fails in a fundamental way to qualify as a scientific theory. All scientific theories represent a framework for making sense of a body of experimental observations. But the primary utility of a theory is not just to look back but to look forward. A viable scientific theory predicts other findings and suggests approaches for further experimental verification. ID falls profoundly short in this regard.”
Collins then criticizes biochemist Michael Behe’s claim that some features of cells are irreducibly complex. “By irreducibly complex,” Behe wrote in 1996, “I mean a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning.”3 Behe argued that whenever we encounter irreducible complexity in our daily lives we quite reasonably attribute it to an intelligent agent, since that is the only cause we know that can produce it.
Behe described several features of living organisms that he argued were irreducibly complex. One of these was the human blood-clotting cascade, a system of interacting proteins that ensures that clots will form only when and where they are needed. But Collins asserts that “the well-established phenomenon of gene duplication” shows that the component parts of the blood-clotting cascade “reflect ancient gene duplications that then allowed the new copy… to gradually evolve to take on new a function, driven by the force of natural selection.” Behe wrote a response to this criticism in 20004, which Collins ignores.
More surprising is the fact that Collins is here citing experimental evidence against a theory he maintains is unscientific because it is not open to experimental testing. In claiming that evidence from gene duplication disproves ID, Collins is demonstrating that ID can be tested with scientific evidence. Either ID is unscientific, in which case evidence is irrelevant; or evidence can be cited against it, in which case ID is scientific. Collins can’t have it both ways.
The most “damaging crack in the foundation of intelligent design” for Collins, however, is recent research that allegedly undercuts Behe’s argument for the irreducible complexity of the bacterial flagellum, a propulsion system driven by a microscopic high-speed motor. Collins first re-casts Behe’s view of irreducible complexity as a claim that “the individual subunits of the flagellum could have had no prior useful function of some other sort,” then he proceeds to argue against that claim.5
As evidence, Collins cites the type III secretory apparatus (TTSS), “an entirely different apparatus used by certain bacteria to inject toxins into other bacteria that they are attacking” that is composed of proteins similar to some of those in the bacterial flagellum motor. “Presumably,” Collins writes, the elements of the TTSS “were duplicated hundreds of millions of years ago, and then recruited for a new use; by combining this with other proteins that had previously been carrying out simpler functions, the entire motor was ultimately generated. Granted, the type III secretory apparatus is just one piece of the flagellum’s puzzle, and we are far from filling in the whole picture (if we ever can). But each such new puzzle piece provides a natural explanation for a step that ID had relegated to supernatural forces, and leaves its proponents with smaller and smaller territory to stand upon.”
As in the case of the blood-clotting cascade, Collins ignores Behe’s counterargument – in this case, that “there’s no reason that parts or sub-assemblies of irreducibly complex systems can’t have one or more other functions.”6 The fact that a fuel pump can be used for other purposes doesn’t mean that the automobile engine of which it is a part is undesigned. Furthermore, evidence suggests that the bacterial flagellum is older than the TTSS. If anything, the latter probably de-volved from the former!7
In any case, Collins is again citing evidence (as he did above) against a theory he says can’t be tested against the evidence. Without noticing the contradiction, he concludes: “Scientifically, ID fails to hold up, providing neither an opportunity for experimental validation nor a robust foundation for its primary claim of irreducible complexity.”
Collins then adds a new twist: ”More than that, however, ID also fails in a way that should be more of a concern to the believer than to the hard-nosed scientist. ID is a ‘God of the gaps’ theory, inserting a supposition of the need for supernatural intervention in places that its proponents claim science cannot explain. Various cultures have traditionally tried to ascribe to God various natural phenomena that the science of the day had been unable to sort out – whether a solar eclipse or the beauty of a flower. But those theories have a dismal history. Advances in science ultimately fill in those gaps, to the dismay of those who had attached their faith to them. Ultimately a ‘God of the gaps’ religion runs a huge risk of simply discrediting faith. We must not repeat this mistake in the current era. Intelligent design fits into this discouraging tradition, and faces the same ultimate demise.”
But Collins’s “God of the gaps” description of the history of science is inaccurate, except perhaps as an account of the demise of animism. As physicist David Snoke has written, “Did anyone ever argue for the existence of God because we did not understand magnets or the orbits of the planets? Perhaps some pagan shaman somewhere has argued that way, but I see no evidence for any serious Christian argument along these lines.”8 And Collins’s suggestion that scientific advances have eliminated intelligent design is exaggerated, to say the least. After all, for the blood-clotting cascade and the bacterial flagellum the only thing Collins offers is speculation about ancient gene duplications.
The most egregious flaw in Collins’s statement, though, is his misinterpretation of ID.
First, ID is not inserting supernatural intervention, unless intelligence itself is defined as supernatural. ID makes only the minimal claim that it is possible to infer from the evidence of nature that some features or patterns in nature are explained better by an intelligent cause than by undirected processes. True, one can then ask about the nature of the intelligence, and a reasonable answer would be God. But ID does not take us that far; it is not natural theology.
Second, and more importantly, design inferences are not arguments from ignorance. No sane person argues, “I don’t know what caused X, therefore it must be designed.” We infer design in our daily lives when X resembles things that we know are produced by intelligence and could not plausibly have been produced without it. Irreducible complexity is one hallmark of designed things; the “specified complexity” of William Dembski is another.9 In either case, we infer design most reliably when we have more evidence, not less.
Darwin and his contemporaries thought living cells were blobs of protoplasm; it was easy for them to assume that such blobs were undesigned. But as modern biologists learn more and more about the irreducibly complex biochemical cascades and molecular machines needed for life, it becomes less and less plausible to dismiss cells as accidental by-products of unguided natural forces.
Rather than criticizing intelligent design for what it is, Collins tries to make it something it isn’t and criticizes that instead. His real target is not ID, but God of the gaps reasoning, which he considers a grave danger to religious faith because of what he believes to be overwhelming evidence for Darwinian evolution.
As director of the National Human Genome Research Institute (NHGRI), Collins argues that data from DNA sequencing provide “powerful support for Darwin’s theory of evolution, that is, descent from a common ancestor with natural selection operating on randomly occurring variations.” Regarding the second part of the theory (natural selection acting on random variations), Collins writes: “Darwin could hardly have imagined a more compelling digital demonstration of his theory than what we find by studying the DNA of multiple organisms. In the mid-nineteenth century, Darwin had no way of knowing what the mechanism of evolution by natural selection might be. We can now see that the variation he postulated is supported by naturally occurring mutations in DNA.” Although most mutations are neutral or harmful, “on rare occasions, a mutation will arise by chance that offers a slight degree of selective advantage. That new DNA ‘spelling’ will have a slightly higher likelihood of being passed on to future offspring. Over the course of a very long period of time, such favorable rare events can become widespread in all members of the species, ultimately resulting in major changes in biological function.”
Collins continues: “Some critics of Darwinism like to argue that there is no evidence of ‘macroevolution’ (that is, major change in species) in the fossil record, only of ‘microevolution’ (incremental change within a species).” But he writes that “this distinction is increasingly seen to be artificial.” To prove his point, Collins cites a Stanford University study of stickleback fish10. Marine sticklebacks typically have armor plates extending from head to tail, but many freshwater sticklebacks lack such plates, and biologists have found a correlation between this difference and variations in the gene for Ectodysplasin (EDA), a molecule involved in the formation of the plates. Collins concludes: “It is not hard to see how the difference between freshwater and saltwater sticklebacks could be extended to generate all kinds of fish. The distinction between macroevolution and microevolution is therefore seen to be rather arbitrary; larger changes that result in new species are a result of a succession of smaller incremental steps.”
But marine and freshwater sticklebacks are merely varieties of the same species, Gasterosteus aculeatus. There is no evidence here that variations in their gene for EDA would (or even could) lead to the origin of a new species, much less to the new organs or major changes in body plans needed for macroevolution. The same can be said for the only other example cited by Collins, variations in disease-causing viruses, bacteria, and parasites.
In 1937, evolutionary biologist Theodosius Dobzhansky noted that “there is no way toward an understanding of the mechanisms of macroevolutionary changes, which require time on a geological scale, other than through a full comprehension of the microevolutionary processes observable within the span of a human lifetime.” He concluded: “For this reason we are compelled at the present level of knowledge reluctantly to put a sign of equality between the mechanisms of macro- and microevolution, and proceeding on this assumption, to push our investigations as far ahead as this working hypothesis will permit.”11
Like Dobzhansky, Collins merely assumes that microevolution can be extrapolated to macroevolution. Despite seventy years of genetic research the extrapolation remains an assumption, and the distinction between micro- and macroevolution is no more “arbitrary” now than it was then.
Regarding the first part of Darwin’s theory (descent from a common ancestor), Collins writes that “the study of multiple genomes” enables evolutionary biologists “to do detailed comparisons of our own DNA sequence with that of other organisms.” Since DNA mutations accumulate over time, organisms with a recent common ancestor would be expected to show fewer differences in their DNA than organisms that diverged much earlier. “At the level of the genome as a whole,” Collins writes, “a computer can construct a tree of life based solely on the similarities of DNA sequences,” and he includes an evolutionary tree (“phylogeny”) of mammals constructed in this way. Collins concludes: “This analysis does not utilize any information from the fossil record, or from anatomical observations of current life forms. Yet its similarity to conclusions drawn from studies of comparative anatomy, both of existent organisms and of fossilized remains, is striking.”
What Collins doesn’t mention is that the DNA data often lead to conflicting phylogenies. For example, his evolutionary tree in The Language of God shows flying lemurs related to tree shrews, and rabbits and monkeys on more distant branches. But a phylogeny published in the Proceedings of the National Academy of Sciences USA in 2002 shows flying lemurs related to monkeys and tree shrews related to rabbits.12 Conflicts among different DNA-based trees are a major headache for evolutionary biologists, some of whom spend their entire careers attempting to resolve them.
Not only can phylogenies constructed with DNA conflict with each other, but they can also conflict with phylogenies based on morphology. Take whales, for example – fossils of which Collins asserts are “consistent with the concept of a tree of life of related organisms.” On morphological grounds, evolutionary biologist Leigh Van Valen proposed in the 1960s that modern whales are descended from an extinct group of hyena-like animals.13 Then, in the 1990s, molecular comparisons suggested that whales are more closely related to hippopotamuses14. In 2001, however, evolutionary biologist Kenneth D. Rose reported that “substantial discrepancies remain” between the morphological and molecular evidence15. And in 2007, J. G. M. Thewissen and his colleagues pointed out that since whales appear in the fossil record 35 million years before hippopotamuses “it is unlikely that the two groups are closely related.” Thewissen and his colleagues concluded from morphological comparisons that whales are descended from a raccoon-like animal instead.16
The conflict between morphological and molecular phylogenies continues, and the problem is bigger than whales. In 2007, British scientists analyzed 181 molecular and 49 morphological trees and observed that “molecular and morphological phylogenies often seem to be at odds with each other.”17Clearly, as a general statement, Collins’s claim that DNA trees are strikingly similar to trees drawn from comparative anatomy is false.
According to Darwinian theory, natural selection would tend to eliminate DNA changes that are harmful, but not changes that have no effect on function. When molecular biologists discovered in the 1970s that the vast majority of the mammalian genome consists of DNA that does not code for proteins, some assumed that this was simply mutational garbage that had accumulated over the course of evolutionary history. Collins cites certain segments of this “junk DNA” known as “ancient repetitive elements (AREs),” which he argues originated from transposable elements (“jumping genes”). Collins writes that almost half of the human genome is “made up of such genetic flotsam and jetsam.” Remarkably, when one compares AREs in mice and humans “many of them remain in a position that is most consistent with their having arrived in the genome of a common mammalian ancestor, and having been carried along ever since.”
The key assumption underlying Collins’s argument is that the AREs are functionless. In a revealing passage Collins writes: “Some might argue that these are actually functional elements placed there by a Creator for a good reason, and our discounting of them as ‘junk DNA’ just betrays our current level of ignorance. And indeed, some small fraction of them may play important regulatory roles. But certain examples severely strain the credulity of that explanation. The process of transposition often damages the jumping gene. There are AREs throughout the human and mouse genomes that were truncated when they landed, removing any possibility of their functioning. In many instances, one can identify a decapitated and utterly defunct ARE in parallel positions in the human and the mouse genome. Unless one is willing to take the position that God has placed these decapitated AREs in these precise positions to confuse and mislead us, the conclusion of a common ancestor for humans and mice is virtually inescapable.”
Here (and elsewhere in his book) Collins is using a peculiarly Darwinian form of argument. In The Origin of Species, Darwin repeatedly argued that his theory must be true because divine creation is false. This is an odd way to defend a scientific theory, yet it is common in the Darwinian literature. For example, in a section on “Evidence for Evolution” in the 2005 college textbook Evolution, Douglas J. Futuyma wrote: “There are many examples, such as the eyes of vertebrates and cephalopod molluscs, in which functionally similar features actually differ profoundly in structure. Such differences are expected if structures are modified from features that differ in different ancestors, but are inconsistent with the notion that an omnipotent Creator, who should be able to adhere to an optimal design, provided them.”18
How do Futuyma and Collins know what a Creator would do? Where else in science are statements about a Creator used to support a theory? Obviously, there’s something very strange about Darwinian “science.”
Stripped of its dubious theological content, Collins’s argument reduces to this: Darwin’s theory predicts the accumulation of functionless DNA differences, and that is what we find. But recent genome research provides growing evidence that much “junk DNA” is not functionless at all. For example, in 2006 Japanese and American researchers discovered that “a large number of nonprotein-coding genomic regions are under strong selective constraint” – meaning that they have functions, otherwise selection would not affect them. The researchers wrote: “Transposable elements are usually regarded as genomic parasites, with their fixed, often inactivated copies considered to be ‘junk DNA’… [but many such] sequences have been under purifying selection and have a significant function that contributes to host viability.”19 In other words, the very “decapitated and utterly defunct” transposable elements that Collins considers his best evidence are turning out not to be functionless after all.
A similar result was reported by California scientists in 2007, who surveyed 10,402 noncoding elements in the human genome and found that a surprisingly high percentage functioned in gene regulation. They concluded that “mobile elements may have played a larger role than previously recognized.”20 The same year, Australian molecular biologists reported: “While less than 1.5% of the mammalian genome encodes proteins, it is now evident that the vast majority is transcribed, mainly into non-protein-coding RNAs… [of which] increasing numbers are being shown to be functional.” The Australians concluded that the percentage of the genome that encodes functional information “may be considerably higher than previously thought.”21 And in 2008, American researchers demonstrated an important function for noncoding RNAs transcribed from segments of repetitive DNA that had previously been considered junk.22 It seems that with each scientific advance Collins’s “inescapable” evidence for common ancestry shrinks.
In addition to “junk DNA,” Collins also cites “silent mutations” in protein-coding segments of DNA. Since three letters of the DNA code are needed to specify one amino acid, and since there are sixty-four possible arrangements of such letters but only twenty amino acids, most amino acids can be specified by more than one three-letter “word.” This means that even in the protein-coding segments of DNA some mutations do not change the resulting amino acid sequence. These are sometimes called “silent mutations.”
In the course of evolution, natural selection would tend to eliminate proteins that have been damaged by changes in their amino acid sequences, so according to Darwinian theory organisms are more likely to carry DNA mutations that do not produce such changes than those that do. According to Collins, when we compare DNA sequences of related species, “silent differences are much more common in the coding regions than those that alter an amino acid. That is exactly what Darwin’s theory would predict. If, as some might argue, these genomes were created by individual acts of special creation, why would this particular feature appear?”
There he goes again, invoking a theological argument to support what is supposedly a scientific theory. Theology aside, his argument (as above) relies on the assumption that “silent” mutations are functionless. In 2002, however, Uruguayan scientists discovered that a “silent” mutation in a gene in bacteria decreased the solubility of the resulting protein, even though it did not change the amino acid sequence. 23 And in 2007, scientists at the U. S. National Cancer Institute found that a “silent” mutation in mammalian cells significantly altered the functional properties of a multi-drug resistance protein while leaving its amino acid sequence unchanged.24 If “silent” mutations are not silent after all, then Collins’s argument falls apart.
So Collins defends Darwin’s theory by assuming that microevolution can be extrapolated to macroevolution, and by assuming that certain segments of DNA have no function despite growing evidence that they do.
Darwin of the gaps
Recall Collins’s principal objection to ID: “ID is a ‘God of the gaps’ theory, inserting a supposition of the need for supernatural intervention in places that its proponents claim science cannot explain… But those theories have a dismal history. Advances in science ultimately fill in those gaps, to the dismay of those who had attached their faith to them. Ultimately a ‘God of the gaps’ religion runs a huge risk of simply discrediting faith. We must not repeat this mistake in the current era. Intelligent design fits into this discouraging tradition, and faces the same ultimate demise.”
Except for the “supernatural” part, this actually sounds like a description of Collins’s own strategy of defending Darwinism by relying on supposedly functionless segments of DNA. He repeatedly assumes that if we are ignorant of the function of a stretch of DNA then it has no function; it is simply a relic fortuitously inherited from a common ancestor. But the more molecular biologists learn about DNA, the more they discover functions in what were previously thought to be functionless segments. Collins’s defense of Darwinian evolution becomes less tenable with every new advance.
How ironic. Collins claims he’s basing his case for Darwinism on new knowledge from genome sequencing, but he’s actually basing it on gaps in that knowledge. Collins himself argues that such an approach has “a dismal history.” Advances in science ultimately fill in those gaps, to the dismay of those who had attached their faith to them. Ultimately a “Darwin of the gaps” approach runs a huge risk of simply discrediting science. We must not repeat this mistake in the current era. Darwinism fits into this discouraging tradition, and faces the same ultimate demise.