Top ten reasons for doubting natural selection

Shaun Johnston
Publisher, Evolved Self Publishing, Rosendale NY
This email address is being protected from spambots. You need JavaScript enabled to view it.

Ten reasons are given for doubting that natural selection is the primary engine driving evolution. The theory is criticized on logical, mathematical, historical and epistemological grounds. Supporting theories, of mutation and kin selection, are shown to provide natural selection with little of the support they promise. An appeal is made that the theory of natural selection be limited to use as a conceptual tool in professional laboratory and field work while a search is conducted for better alternatives.

10. Natural selection cannot through use maintain multiple homeostatic mechanisms
Natural selection is the separation of each generation of a reproducing population into two groups: those that survive to reproduce, and those that don’t. Like a simple sieve, it selects according to a 2-point scale—retain, and discard; it cannot discriminate between creatures more finely than that—see below, Schrodinger (1944). Because such a “fitness” sieve can’t select for more than one aspect of fitness at a time, it will select for only the single most critical factor. But maintaining living creatures far from chemical equilibrium with the environment involves managing and driving multiple processes and factors simultaneously. Complex living creatures therefore cannot evolve through the operation of natural selection alone.

9. Natural selection reduces, cannot create, variation
With repeated operation, a mechanical sieve reduces variation. That’s all it can do. It cannot create variation. For it to continue to function, there must be an accompanying source of variation. Mutation—damage to the “particles” of inheritance that natural selection discriminates between—has been proposed as the primary source of new variation for natural selection to act on. Collectively, however, mutations impair fitness, so a species burdened with mutation will be at a selective disadvantage compared to creatures not suffering mutation occupying the same niche. Given the extraordinary ability of chromosomes to repair themselves, if significant variation exists it is possible it arose and is maintained through some other mechanism than random mutation. One can of course create variation artificially by damaging genes, but that doesn’t prove that most natural variation is caused the same way.

8. Mathematical support for natural selection is lacking
In an essay titled “The Limitations of Evolutionary Theory” John Maynard Smith wrote (1989), “It is possible to measure mutation rates in very special circumstances in some microorganisms…. But in most situations, mutation rates cannot be measured…” Of mutation, selection, and migration between populations he writes:

Thus we have three processes which we believe to determine the course of evolution, and we have a mathematical theory which tells us that these processes can produce their effects at levels we cannot usually hope to measure directly. It is as if we had a theory of electromagnetism but no means of measuring current or magnetic force…. It means that we can think up a number of possible evolutionary mechanisms, but find it difficult to decide on the relative importance of the different mechanisms we have conceived…

If the mutation rate is doubled, does this double the maximum rate at which a species can evolve? [The question has no simple and agreed answer, he says.] But my reason for raising the question….was to make the point that a theory of evolution which cannot predict the effect of doubling of one of the major parameters of the process leaves something to be desired.

In summing up he says,

It is rarely possible in evolutionary theory to think of a single decisive experiment or observation that will settle a controversy.

Pigliucci & Kaplan (2006) issue similar cautions. They point out, for example, that current statistical techniques cannot distinguish between natural selection and drift.

7. Natural selection cannot select for gene clusters
In What is Life? (1944) by the physicist Erwin Schrodinger, one of the founding documents of molecular biology and credited by both James Watson and Francis Crick with inspiring their discovery of the DNA double helix, the author begins a section titled “The Necessity of Mutation being a Rare Event” with:

This brings out one very important point. In order to be suitable material for the work of natural selection, mutations must be rare events, as they usually are… It is essential that they be introduced one at a time, while all the other parts of the mechanism are kept constant.

However, this does not appear to match how creatures actually evolve. An elephant’s trunk presumably evolves slowly as a complex set of genes. Many adaptations to aquatic life took place simultaneously in the evolution of whales. The random re-sorting of genes that is thought to be natural selection’s primary instrument seems unlikely to be able to keep clusters of genes together so they can evolve in concert.

“Gene order conservation” is observed. But no mechanism is known by which it can have arisen and be maintained through natural selection. Here are extracts from the summary of a study by Tamames (2001).

Even between very distant species, remnants of gene order conservation exist in the form of highly conserved clusters of genes. This suggests the existence of selective processes that maintain the organization of these regions…. The reasons for the maintenance of gene order are still not well understood, as the organization of the prokaryote genome into operons and lateral gene transfer cannot possibly account for all the instances of conservation found.

6. Darwin may have over-rated the efficiency of natural selection
Natural selection was deduced to be the mechanism of evolution at least five times in the 19th Century, largely inspired by the idea published in 1798 by Malthus. It inspired both Darwin and Wallace in their independent proposal of natural selection. Malthus supposed that the appetite of a population increases as a function of its volume, while its food supply can increase only as a function of its bounding area. Every population is therefore always pressing against the limit set by the supply of available food. His aim was to convince his readers of the futility of trying to relieve human misery through welfare, but biologists saw in his thinking how constant competition among conspecifics for food would exert a powerful selective pressure for continuous incremental increases in fitness, since more progeny are produced than can survive.

Malthus’ idea involved a number of suppositions, most of which turn out to be false (Stove, 1995). Food supply is not limited to the bounding area of a community, but can grow within it—a population can be plate-like or sponge-like so food is available throughout the population’s range. More significant, the limit to natural populations seldom consists of competition between equally-matched conspecifics, and is seldom constant. It is more likely to be intermittent, and to be from predators, parasites, natural disasters, and the environment. It is to these conditions that living creatures most need to adapt, not to the competition from conspecifics on which Malthus based his theory. The falsity of his suppositions calls into question any theories, such as natural selection, based on them.

These conditions act much more at random than the imagined grueling contest of all against all that Darwin saw continually polishing whatever it was that specified inheritance, with minute advantages determining which would survive to reproduce. Take maple trees shedding seeds. Why do they shed so many? So young maple seedlings will compete that much more fiercely with each other? More reasonably it’s so there’ll be enough of them for one or two to fall on the occasional patch of suitable soil, wherever it is. Even then, if several of them grow up near one another, it’s the one that gets the most light that will push up into the canopy. That’s not competition, it’s being in the right place in the right time. And maybe that’s the main reason for over-production of creatures in each generation: to cope with random variations in the environment so there’ll be just enough progeny at the right place at the right time. Make that assumption, and there may be very little significant selection through competition among conspecifics going on, reducing how much opportunity natural selection gets for discriminating between the small increments in fitness it is assumed to work on.

5. The pace of evolution’s “progress” rules out natural selection
The instrument of natural selection—populations being separated in each generation into those that survive and those that don’t—cannot be made more efficient. It is nothing but that separation. It therefore can be expected to have unchanged potential over time. Its efficacy can be expected to change only as factors offer more or less opportunity for the expression of that potential. In fact, opportunities for its expression have greatly diminished over evolutionary time, yet the pace of evolutionary “progress” appears at least unchanged.

Factors that would limit the potency of natural selection would be reproductive events occurring at greater intervals, clutches becoming smaller, and animals becoming so large that natural habitats could support fewer of them. A female elephant will on average have 6 progeny in her lifetime, according to Darwin, providing very little opportunity for selection to maintain such a large creature so far from chemical equilibrium with its surroundings, let alone for further evolution. And there are probably fewer elephants in Africa than bacteria in a cup of average pond water. Yet despite such a drastic reduction in opportunities for the winnowing out of less fit genes, evolution among such large mammals seems to have proceeded apace (Dawkins, 2004).

4. Natural selection cannot account for altruism
Based as it was on competition between conspecifics, natural selection could not account for the common appearance of altruism between conspecifics. Lending weight to the theory’s reputation of not being open to disproof, supporters of natural selection drew on its questionable mathematical apparatus to create the theory of kin selection. This supposes that our genes individually calculate what in our behavior would best ensure their own survival, independent of us, and actually prescribe that as what we suppose is our own freely-chosen behavior.

In the hands of supporters the theory appears to prop up natural selection. However, its predictions often fail to match observations. Motherhood, the prime instance of altruism, is an example. Take a mother mammal having her first infant, in a species with average lifetime progeny of ten. Mortality occurs most frequently in infancy and childhood, so her first infant promises her genes around a 20% chance of being passed on. She herself, however, presents her genes with an 80% chance of being passed. According to the theory of kin selection you’d expect her genes to make her value her own survival over that of her young until she’d had about half her anticipated allowance. This is clearly not what happens. Kin selection also fails to account for inter-specific altruism, such as can occur between dogs and cats sharing a household, who surely have nothing to gain from their mutual affection. From a good source I heard of someone entering their kitchen at night to find their dog, with a mouse between his paws, growling fiercely while surrounded by a semicircle of watchful cats. Whatever the mechanism behind such altruism, once identified it can probably be extended to account for that among conspecifics.

3. Natural selection cannot account for human talents.
A few years after joining with Darwin in publishing the theory of natural selection, Alfred Wallace reasoned it could not account for the evolution of such human abilities as mathematics, reason, music and art. However, he was unable to engage Darwin’s interest in the issue and his objections were never adequately rebutted.

Wallace arrived at his conclusion from his familiarity with two kinds of communities. As much as any 19th century scientists he was familiar with both civilization and with communities in the 1840’s still largely untouched by it, far-upriver in South American and in Indonesia. He lived among such people for many years and studied their cultures. He noted that abilities elicited by civilization existed in non-civilized communities only at a much lower level of development.

Darwin’s assertion that natural selection could never improve an animal beyond its needs called into question how such abilities could be developed through natural selection in the first place. But even supposing that they had been, Wallace realized that natural selection should have eliminated them before the emergence of civilization through disuse.

This questioning is supported by the time frame of the rise of intelligence. Creatures with what is commonly thought of as intelligence appear only within the past 5% of the existence of the Earth. Only very much more recently, within the last 1% of the existence of the earth do we find a sudden burst of intelligence among mammals, and in only the last .1% of that time do humans separate from apes. Our brains expanded at a rate of many synapses per individual in the course of those 5 million years. The appearance of intelligence in living creatures has the time course of a rapidly self-catalyzing process not possible with the simple sieve-action of natural selection. Whatever this self-catalyzing process is, once identified, or even allowed for, it may logically be projected back over the rest of evolution. The reach of natural selection as the primary mechanism driving evolution then becomes uncertain.

2. What natural selection claims to do, is insufficient
When the idea of natural selection first occurred to Darwin, Queen Victoria, not yet married, was just ascending the English throne. Passenger railroads were in their first tentative decade in the English Midlands. Darwin and Wallace published their theory the same decade the Great Exhibition of London revealed to the world Britain’s mastery of steam power. This is the world in which natural selection seemed so obvious that, on hearing about it, Julian Huxley said “How extremely stupid not to have thought of that!”

70 years later, it didn’t seem so obvious to Erwin Schrodinger. What is life? is mainly a warning that only when you work with vast aggregates of molecules does quantum uncertainty average out and physics give you precise answers. At the level of the atom, quantum uncertainty rules. Here he is talking about chromosomes:

For it is simply a fact of observation that the guiding principle in every cell is embodied in a single atomic association existing in only one copy (or sometimes two)—and a fact of observation that it results in producing events which are a paragon of orderliness. [Then he draws on his experience to say about this orderliness.] …the situation is unprecedented, it is unknown anywhere else except in living matter. The physicist and the chemist, investigating inanimate matter, have never witnessed phenomena which they had to interpret in this way. The case did not arise, and so our theory does not cover it. [Brackets mine.]

The “case” he’s referring to is order being maintained for years, even centuries, even over billions of years, in what he calls “a non-periodic crystal”—long strings of atoms in a non-repeated order—in the face of quantum uncertainty. The kind of order you find in genes you find in non-living matter only at absolute zero, when atoms come to a stop. This is how Schrodinger puts it:

The living organism seems to be a macroscopic system which in part of its behavior approaches to that purely mechanical (as contrasted with thermodynamical) conduct to which all systems tend, as the temperature approaches to the absolute zero and the molecular disorder is removed.

Evolution is one of a trio of closely-related complex phenomena, the other two being development and homeostasis. On the face of it, there’s no reason to expect evolution to be easier to account for than the others. Yet while we accept them as being of immense complexity, we’ve come to suppose that the challenge of accounting for evolution can be met with a simple 19th century sieving procedure. We might be better off abandoning that supposition and searching for a mechanism able to account for all three phenomena together. It is possible the separation between them is primarily semantic.

1. Natural selection blinds us to other prospects.
Pigliucci and Kaplan (2006) start a section titled “Statistical and Conceptual Difficulties in Measuring Selection” with “The central challenge for these multivariate approaches to understanding natural selection is moving from the statistical analysis to scientific interpretations of the results, a problem that is, of course, of much wider import in the biological and social sciences than, say, in some areas of physics or chemistry.” Discussions in the social sciences do tend to refer at some point to human origins, even if only implicitly, limiting themselves to what the current theory of origins permits. A theory of origins with the authority of science behind it acts as a valve regulating what can be supposed to be true. As a result, scientific theories of human origins have become enormously influential in fashioning opinion about law, permissible behavior, human rights, etc. It would be naïve to regard theories of origins as value-neutral, of concern only to scientists.

Pigliucci and Kaplan point out that only recently have techniques been developed and attempts made to rigorously validate natural selection, and they stress how difficult it remains. They warn that “…there can be no automatic or foolproof way of identifying selection in natural populations through statistical analysis.” Given this lack of scientific evidence for natural selection being the primary mechanism driving evolution, and how profoundly it influences opinion throughout the social sciences, perhaps it should be limited to use as a conceptual tool in laboratory and fieldwork while a fresh search is made for a replacement.

Such a search might start by abandoning the key assumption underlying natural selection, that inheritance is particulate. Given that the body has something like 250 different kinds of cell, over 20 different organs, 600 muscles and 200 bones, over 1000 entities altogether, 30,000 genes seems too few to define them, especially since all these body parts have to be defended against failure not only in their final state in adulthood but through every stage in development. Perhaps the most significant unit of evolution is the genome throughout its history, and the appropriate way to regard it is as a library of resources gathered over billions of years of evolution, like a huge .dll computer library file. Such a resource promises to enrich rather than to impoverish the social sciences and our sense of self, which is the prospect that inspired the writing of this article. On this view, evolution should no more be accounted for primarily in terms of changes in gene frequencies than the development of literature should be accounted for primarily in terms of changes in word frequencies.

Berry, A. (2002). Infinite Tropics: An Alfred Wallace Anthology. London: Verso.

Dawkins, R. (2004). The Ancestor’s Tale: A Pilgrimage to the Dawn of Evolution. Boston: Houghton Mifflin Company.

Maynard Smith, John. (1989). Did Darwin Get It Right? New York: Chapman and Hall.

Pigliucci, M & Kaplan ,J. (2006). Making Sense of Evolution: The Conceptual Foundations of Evolutionary biology. Chicago: The Uniersity of Chicago Press.

Schrodinger, E. (1944). What is Life? Cambridge: Cambridge University Press.

Stove, D. (1995). Darwinian fairytales: Selfish Genes, Errors of Heredity, and Other Fables of Evolution. New York: Encounter Books.

Tamames, J. (2001). Evolution of gene order conservation in prokaryotes, Genome Biology, 2.