MYTH 20

THAT NEO-DARWINISM DEFINES EVOLUTION AS RANDOM MUTATION PLUS NATURAL SELECTION

David J. Depew

Evolution in the sense of common ancestry might be true, but evolution in the neo-Darwinian sense—an unguided, unplanned process of random variation and natural selection— is not.

—Cardinal Christoph Schönborn, “Finding Design in Nature” (2005)

Since the 1940s, the principles of what has been called neo-Darwinism (or the modern evolutionary synthesis) have guided professional evolutionary inquiry.¹ These principles stemmed from the fusion of Mendelian genetics with Darwin’s idea of natural selection. Therefore, it might seem natural enough to summarize neo-Darwinian evolution as random genetic variation plus natural selection. This is, in fact, how antievolutionists prefer to characterize it. For instance, the Discovery Institute, an organization that promotes the intelligent-design version of antievolutionism, says it is “skeptical of … neo-Darwinism’s … claims for the ability of random variation and natural selection to account for the complexity of life.”²

This way of construing neo-Darwinism might appear to have spread to the Roman Catholic Church.³ Christoph Schönborn (b. 1945), the cardinal archbishop of Vienna and a former student of retired Pope Benedict XVI (b. 1927), wrote the epigraph to this chapter in an op-ed piece in the New York Times in 2005. Some Catholic evolutionary biologists soon intervened, however, to deflect the church from supporting this way of characterizing neo-Darwinism. More factors than mutation and selection, they pointed out, affect evolutionary change, and therefore the portrayal of evolution as a process in which accidents are selectively preserved or eliminated is a myth.4 Chance variation certainly plays an essential role in the evolution of adaptations, which Darwinians as much as their creationist opponents recognize as a prominent feature of organisms. However, adapted traits and properties, which are so functional and goal directed that they seem intentionally designed, are also the outcome of so many other factors that “random genetic variation plus natural selection” is an inadequate description of how adaptive evolution works.

A good way to clear up misunderstandings about neo-Darwinism is to relive some of its history. Prior to the end of the nineteenth century, little was known about the mechanisms of heredity. However, in the 1880s, the embryologist August Weismann (1834–1914) began showing that only germ-line factors are heritable and that bodily characteristics acquired in the course of a lifetime cannot be passed on to descendants. Weismann’s account of inheritance gave rise to neo-Darwinism in the original and most proper sense of the term. Neo-Darwinians were “neo” because they explained adaptation solely in terms of natural selection’s elimination of unfit and preservation of fit variants in germ-line factors—unlike Darwin himself, who acknowledged use and disuse as an auxiliary explanation of some adaptations and assumed that acquired bodily characteristics could be passed on to descendants (see Myth 10). Weismann proposed “blastogenesis” in the place of Darwin’s “pangenesis” in order to insist that only characters in the germ line and not in the whole body are heritable.⁵

One might imagine that the (re)discovery in 1900 of what Mendelians thought they saw in Mendel’s neglected work on hybridization (see Myth 16) would have supported neo-Darwinism. At the time, however, this was far from the case. Weismann’s germ-line factors, it now appeared, came in discrete unblended but combinable units. Unless they spontaneously mutated, these heritable factors would remain unchanged in gene pools generation after generation, especially when they were recessive rather than dominant and were thus shielded from natural selection. Early neo-Darwinians held that in addition to eliminating unfit germ-line variants, natural selection could slowly spread mutations that kept lineages adapted to their environments. Unfortunately, they also assumed that the adaptive effects of a markedly superior variant in one generation would progressively wear off in later generations, leaving descendants with characteristics that were no better than average and therefore with no net gain in adaptedness.

Early Mendelians—such as William Bateson (1861–1926), Hugo de Vries (1848–1935), and Wilhelm Johannsen (1857–1927), who coined the term “gene” in 1909⁶—spotted the tensions in this early neo-Darwinian view. They were skeptical of the power of natural selection ever to beat the tendency of traits to regress to an undistinguished mean. Instead, they relied on sudden, single-leap mutations that just happened from the start to be adaptive. Accordingly, for these Mendelians, mutation—not natural selection—was evolution’s “creative factor.”⁷ In response, neo-Darwinians pointed out that single-shot mutations were overwhelmingly likely to disrupt the adaptedness on which the viability of organisms depends and so would be summarily purged by natural selection. Therefore, at the beginning of the twentieth century, it would have appeared not just dubious but incoherent to say that evolution consists of random mutation plus natural selection.⁸

Eventually, findings in genetics did support the unification of mutation and selection, but it took a few decades for this fusion to be accomplished.⁹ The breakthrough came in 1918, when the statistical genius Ronald A. Fisher (1890–1962) used a mathematical theorem first derived in 1908 to demonstrate that when Mendel’s laws are expanded to whole populations of freely interbreeding organisms, the supposed tendency of new variations to regress to an undistinguished mean is nowhere to be found.10 From this perspective, Mendelians were right to hold that unexpressed “alleles” (from Greek for “other” or “alternative”) remain in gene pools as recessives. But neo-Darwinians were right to doubt whether single-shot mutations with large effects are evolution’s innovative factor. Mutations of this sort are statistically highly improbable, and so it is even less probable that a concerted sequence of lucky mutations will drive a lineage toward greater adaptedness. In fact, it would be almost as miraculous as advocates of intelligent design hope it is. Even though unification between neo-Darwinian adaptationists and mutationists occurred a century ago, the implications of their integration are still widely misunderstood.

What, then, is the creative factor in evolution? When empirical examination of how organisms actually live is combined with statistical-probabilistic thinking about populations, we can see that mutations, each with a small effect, and natural selection can indeed slowly move populations away from an equilibrium distribution of alleles that in the absence of these factors would remain the same generation after generation.¹¹ So natural selection can be as innovative as Darwin thought, though usually in combination with other factors. In addition to mutation, there can be gene flow, when migrants spread variants by moving from one subpopulation to another, interbreeding as they go. Genetic drift, too, can occur, in which new variants can get a toehold purely by chance in the small populations in which many species live. Drift occurs in small populations for the same probabilistic reason that in roulette a ball may land six or seven times on red without violating the expectation that in the long run the number of red and black landings will even out.¹² If these variant alleles are adaptively favorable, natural selection can then spread them to and through an interbreeding population.

From the statistical-population perspective, we can also see that the genetic variation that natural selection uses to fuel the process of adaptation is not exclusively mutational. Although mutation in the genetic material is a source of variation, the recombination of genetic material during meiosis—the cell division during which sperm and egg are produced—provides so much of the proximate variation on which natural selection works that, as Theodosius Dobzhansky (1900–1975), one of the founders of the modern evolutionary synthesis, remarked, “Suppression of the mutation process … would probably have little effect on the evolutionary plasticity of a population for some time to come.”13 In recent decades, even more sources of variation have come into view, as the genetics of the developmental process has become better understood. Gene sequences that regulate the timing and rate of developmental processes are a source of heritable variation, making it clear that natural selection affects traits because more fundamentally it affects developmental trajectories. This discovery shows how large morphological changes can result from minor genomic changes.¹⁴ There are also sources of epigenetic heritable variation, such as chemical side chains that attach themselves to DNA (methylation) and more.¹⁵

Population thinking, accordingly, enables us to appreciate that natural selection does more than just eliminate unfit accidental variation. It selects against harmful inherited factors, to be sure, but working on variation of several kinds, it also selects for traits of organisms, populations, groups, and species that become adapted as natural selection, working in combination with other factors, amplifies reproductively more successful variants through interbreeding populations. In this process, natural selection acts in several modes. It acts disruptively, so that populations of the same species exploit a slightly different resource base and eventually become reproductively isolated, evolving into new species. In stable environments it acts directionally, favoring one among several genotypes if the former happens to confer an advantage to its bearers. In unstable environments it “balances” currently adaptive genotypes with others that may be useful when circumstances change by preserving the respective variants as recessive alleles. In sum, as neo-Darwinism matured it showed natural selection to be the preeminently creative factor in evolution that Darwin thought it to be.16

Based on all of the above, it becomes clear that attempts to describe neo-Darwinism as “random mutation plus natural selection” fail to do justice to a framework of inquiry that has guided the discovery of scientific knowledge for well over half a century. It is possible, however, that some still prefer this formula because this way of describing neo-Darwinism’s limited but real reliance on chance makes it look less persuasive than intelligent design. “How can all these wonderful adaptations be nothing more than preserved accidents?” ask supporters of intelligent design. They can’t, Darwinians reply—nor should the work of natural selection and related factors be looked at that way.

The confusion stems from different understandings of what is meant by the term “random.” “Random” does not mean “haphazard” or “open to all possibilities” but rather “unintentional” and “unpredictable.” When Darwin spoke of variation as “chance,” he meant only that the causes of variation are unrelated to its subsequent adaptive utility, not that variation has no causes. In fact, Darwin and his early supporters presumed that variation is caused by unknown deterministic laws of physics and chemistry, like those at the heart of the science of their day. It was only after molecular geneticists discovered that spontaneous changes in DNA sequences are a primary source of mutations that scientists could entertain the notion that mutation is random in the strong sense of being stochastic.

This created an ambiguity. Present-day Darwinians still use the term “chance” the way Darwin did. The genetic variation that is relevant to natural selection arises independently of any effect it might have on reproductive success. In this sense, “chance variation” is part of the definition of natural selection. However, in positioning “random genetic variation” as prior to and independent of the process of natural selection, the formula “random genetic variation plus natural selection” slyly leads students and others to suspect that, as one anti-Darwinian put it, “According to Darwinism our existence is a mere accident.”17 That is because the formula stresses natural selection’s weeding-out role; renders invisible its “creative” role in adapting organisms to environments; greatly shortens the long chain of causes and levels between mutations and traits; and overlooks Darwin’s appeal to natural processes with unpredictable outcomes to explain, not explain away, the functional, goal-directed, and purposive characteristics of organisms.

If the adapted traits of adapted and coadapted organisms were products of conscious design, we would expect a good designer, like a good engineer, to have a larger end in view, such as the evolution of humans. Darwinians debate how much of the history of life on earth is adapted and how much is accidental, but they can’t satisfy demands for a history that is guided toward the appearance of our species.¹⁸ This can be disheartening to those who ask biology to slake their thirst for overall evolutionary purpose. In their disappointment, they may underestimate the vast array of functional parts and goal-directed behaviors that neo-Darwinians can vouch for. Darwinians can help correct this misperception by declining to characterize adaptive natural selection as “design without a designer,” which itches for a fight it need not have.