About the SA Blog Network

Guest Blog

Guest Blog

Commentary invited by editors of Scientific American
Guest Blog HomeAboutContact

Selfish Genes Also Must Cooperate

Email   PrintPrint

Many followers of reason think it natural and rational to be selfish. They believe that’s just how evolution works. But Richard Dawkins, the cardinal spokesperson for that oversimplified and unnaturally selective view, is guilty of logical lapses and false prophecy. His pop-science of selfishness is widely misunderstood. “Selfish” genes that don’t cooperate don’t survive. A more fitting view is that there are evolutionary limits to selfishness. Nature dooms all that damages what it depends on.

The gene centered view of evolution was popularized by Dawkins’s The Selfish Gene, which mixed the best thinking available, with great prose, logical errors and sinfully unscientific sermonizing. It remains influential, even beyond its readers, its misleading title seeming sufficient to substitute for its contents.

Dawkins promoted this gist: “a society based simply on the gene’s law of universal ruthless selfishness would be a very very nasty societyBe warned that if you wish, as I do, to build a society in which individuals cooperate generously and unselfishly towards a common good, you can expect little help from biological nature…because we are born selfish”.

But Dawkins’ doctrine of everlasting selfish doom, a kind of an evolutionary original sin, contains errors. He over-extrapolates from incomplete categories, and makes an error so common it has its own name, the fallacy of composition. Dawkins’s devil isn’t in the details, but in straying too far from them.

Dawkins defines: X as “altruistic if it behaves in such a way as to increase another such entity’s [Y] welfare at the expense of its own. Selfish behavior has exactly the opposite effect.” His scheme sees only two outcomes, selfish or altruistic, and is zero-sum: X gains by Y’s loss. This accurately describes genes competing against variations of themselves for the single slot of dominance in future populations. But does all of creation fit into that scheme? Clearly not, since it excludes: X and Y both lose; X and Y both gain cooperatively. Many biologists confuse cooperation with altruism, but by Dawkins’ definition win-win cooperation is neither selfish nor altruistic. Yet this logical space is crucial for all species with team survival strategies.

Oddly, Dawkins describes how every “selfish” gene depends on many other genes in the “intricate cooperative venture” of propagation, and notes that advantages accrue to any gene “that cooperates well with most of the other genes” it depends on (a minimum of 181), but he still makes irrationally inconsistent prophecies like we “can expect…little help from biological nature” towards cooperation. Dawkins isn’t entirely responsible for the cooperative components of his book not being as well known as his misleading title, but his unscientific sermonizing has lent the shield of science to much bad thinking. Especially the false idea of a “universal ruthless” aspect of biology, a sort of evolutionary original sin, that dooms all that lives to live in a selfish world. Cooperation abounds and all genes depend on it.

Dawkins falls for a “fallacy of composition.” He inappropriately extends properties of parts to wholes. An absurd example is: each atom in a teacup is invisible, therefore the teacup is invisible. Dawkins projects his preferred “selfish” zero-sum property of genes onto everything built by genes, and falsely concludes everything that has “evolved…should be selfish.” But all genes are also cooperative. And besides, no gene-level property can be safely projected onto all things that have genes, or everything they do. Biology isn’t that simple. It mixes competition and cooperation.

In later editions Dawkins partially recants his central selfish dogma, saying “without departing from fundamental laws of the selfish gene theory… cooperation and mutual assistance can flourish.” Game Theory had proved his prophecy of “little help from biological nature” to be false. Cooperating generously can be an evolutionarily stable strategy, with higher productivity than selfishness.

Sadly the unsimple details of Dawkins partial reversal haven’t spread as successfully as his initial sermonizing. In the 30th anniversary edition, he conceded that “born selfish is misleading” and asked readers to “Please mentally delete that rogue sentence and others like it.” Specifying those others would help. It’s time all needed corrections were preached as zealously as the prior errors.

The cooperative, interdependent team aspect of genes suggests cautious generalization. Just as no gene can survive alone, neither can members of any interdependent species. What could be called the “natural dependency principle” can be useful in mapping the evolutionary limits of selfishness: nature ultimately eliminates all that damages what it depends on.

Similar team logic is built into human social instincts, which should limit what counts as rational self-interest. But many leading practitioners of reason somehow deem it rational to damage what they depend on. Their unskilled reasoning yields poor results in many of life’s social coordination problems, like the Tragedy of the Commons.

The pop-science of selfishness needs an upgrade. Cooperation and team survival and selfishness are all natural and rational. Each is sometimes fittest for the circumstances. Dawkins says he could have called his book “The Cooperative Gene.” Evolution would be better understood if he had.

Illustration by Julia Suits, The New Yorker Cartoonist & author of The Extraordinary Catalog of Peculiar Inventions.

Previously in this series:

It Is in Our Nature to Be Self-Deficient
Inheriting Second Natures
Our Ruly Nature
It Is in Our Nature to Need Stories
Tools Are in Our Nature
We Fit Nature To Us: Evolutions two way street
Justice Is In Our Nature
Behavioral Telescope Shows How Cooperation Works

Jag Bhalla About the Author: Jag Bhalla is an entrepreneur and writer. His current project is Errors We Live By, a series of short exoteric essays exposing errors in the big ideas running our lives, details at His last book was I'm Not Hanging Noodles On Your Ears, a surreptitious science gift book from National Geographic Books, details at It explains his twitter handle @hangingnoodles Follow on Twitter @hangingnoodles.

The views expressed are those of the author and are not necessarily those of Scientific American.


Comments 8 Comments

Add Comment
  1. 1. marclevesque 5:56 pm 06/12/2013

    I enjoyed that very much, and it’s good to see such a balanced post.

    Link to this
  2. 2. Mark Sloan 11:26 pm 06/12/2013

    I am sympathetic to your proposed sources of Dawkins’ puzzling obliviousness, even still in the 30 anniversary edition, to the damage done to evolutionary science by The Selfish Gene’s characterization of genes as necessarily ‘selfish’.

    As you mention, he seeks to justify the ‘selfish’ label by falsely focusing on zero-sum rather than non-zero “benefits of cooperation” interactions. Sure, if zero sum interactions are all there is, then calling genes necessarily selfish makes sense.

    But non-zero sum “benefits of cooperation” interactions do exist. And they don’t just exist: they are responsible for all of biology, all the way from the level of mitochondria in individual cells, to all plants and animals, and the incredible evolutionary success of people. Biology only makes sense when biological agents are understood as the products of the benefits of cooperation.

    I also criticize Dawkins’ definitions of “selfish” and “altruistic”. How can it be useful to call ‘selfish’ the genes responsible for the biology underlying our emotions of empathy, loyalty, and guilt? Of course Dawkins would likely reply “Well, it is the vehicle (organism) that is acting unselfishly, not the gene”. I am unable to see why he thinks this is a useful perspective; it seems to me cobbled together merely to defend his “selfish gene” idea.

    His definition of altruistic is even worse. In his defense, he does use a common definition as you quote. But being common does not make it the most useful.

    For example, assume I believe that acting according to “Do to others as you would have them do to you” is likely to maximize my well-being over my lifetime. Then, by Dawkins’ definition, I cannot act altruistically in following this maxim because I expect it will be in my long term best interest.

    I prefer something like “An altruistic act benefits others, is done at a cost to one’s self, and is done without consideration of future benefits to the actor.” While I have used a cultural norm example that is central to morality, understanding altruistic behaviors encoded in genes is improved by using the same definition which also applies to biological altruism when benefits are only reproductive fitness.

    Maybe someone could have a try at writing a book called Cooperating Genes Created Biology?

    Link to this
  3. 3. hangingnoodles 10:10 pm 06/13/2013

    @MarkSloan – thanks for your thoughtful comment. I agree Dawkins conceptual scheme is far too simple. And a key danger is that such oversimplifications are much easier to popularize. So many in other fields believe they know how evolution works, based on a very incomplete picture. Much work remains to dig us out of the hole Dawkins has dug.

    Link to this
  4. 4. Mark Sloan 11:22 pm 06/13/2013

    Hangingnoodles, I have just today started reading a new book you might like.
    It is Evolution, Games, and God: The Principle of Cooperation (2013), edited by Sarah Coakley and Martin A. Nowak. It is an assembly of 20 or so author’s papers whose science represents what I understand to be the state of the art in the subject of the evolutionary origins of cooperation. It may be the best reply yet to Dawkins’ The Selfish Gene. There are several references to Dawkins; none are complimentary.
    For example, to show the contrast with Dawkins’ ideas, in one of the included papers,
    Nowak “hazard(s) the controversial conclusion that ‘cooperation’ is actually a third ‘principle of evolution’, next to mutation and selection. These three ‘principles’ are, however, arranged in a hierarchy: mutation is first needed to generate (genetic) diversity upon which selection can act; mutation and selection then together give rise to evolutionary processes that sometimes lead to cooperation— if one or more of the five ‘mechanisms are operative. For, whenever life discovers a new level of organization (such as the emergence of cells, multicellular organisms, insect societies, or human language), cooperation is involved in one form or another. Mutation and selection (based on selfish competition) alone, without cooperation, may not give rise to complexity.”
    That is, cooperation, not selfishness, is what is primarily responsible for all biology beyond pre-cellular replicating chemicals. (As, just by coincidence, I much less eloquently suggested in my post yesterday.)
    The only immediate negative for me is the remarkable claim, as implied by this book’s title, that useful knowledge about the nature of the supernatural can be derived from science. But so what, the rest of the material, which makes no such remarkable claims, still is very promising.

    Link to this
  5. 5. hangingnoodles 10:47 am 06/17/2013

    @Mark_Sloan: thanks for the tip. Nowak’s work on Supercooperators deserves more attention. Its lessons need to be widely popularized, to dispel Dawkins false doctrine of everlasting evolutionary selfish doom. Will look forward to reading “The Principle of Cooperation.”

    Link to this
  6. 6. karl 9:23 pm 06/20/2013

    I remember having read in SciAm about Enron collapse having to do with the Selfish Gene book, ok, Enron was a social construct, but it proves that a complex system (be it a greedy oil company or an animal) that turns it’s components against itself is destined to (Invader Zim’s tone) DOOOM!.
    Selfish gene might be a dominant trait, pun intended, but you need that your liver cooperates with you instead of hoarding all you can eat to reproduce itself, because that is cancer.

    Link to this
  7. 7. samdiener 9:27 pm 07/16/2013

    I like your emphasis on the evolution of cooperation (to cite Axelrod’s famous title), and its good to know that Dawkins has at least partially recanted.

    But I think you took the critique too far when you wrote,
    “What could be called the ‘natural dependency principle’ can be useful in mapping the evolutionary limits of selfishness: nature ultimately eliminates all that damages what it depends on.”
    You wrote above that both competition and cooperation exists. Similarly, win-win dynamics and lose-lose dynamics both exist, as do win-lose and lose-win. So, there are many parasitic life forms that damage what it depends on (the malarial parasite, to name one notorious example). As long as the host lives long enough for the parasite to complete the portion of its life cycle in that host, it can continue to propagate despite the grievous and too often deadly harm it causes to that host.

    Link to this
  8. 8. timcliffe 8:32 pm 10/8/2013

    The Dawkins-bashing is regrettable, and very often inaccurate. The book that everyone loves to hate (although more hate it than have read it) was in fact a very useful corrective to the organism-centered evolutionary thinking that predominated at the time. YES, YES, YES, I know (and Dawkins knows) that the genes have to exist in organisms and their actions are expressed in organisms, and YES, that involves high levels of inter-gene cooperation. But at that organismal level, natural selection works on the differences *between” organisms (asterisks represent italics; I wish we could use them). What are those differences based on? They are based on genes that differ. In theory and in fact, two organisms can differ in having different versions (alleles) of one gene that makes all the difference in the world between the reproductive success of those two organisms. Thus, it is the difference (and competition) between those two genes that ultimately determines the fate (or the very existence) of those organisms’ offspring.

    So natural selection on organisms ultimately functions on differences between competing versions of genes. And anyone who thinks natural selection is primarily about cooperation hasn’t been paying attention. It is often a ferociously cruel process. The genes do not care; they are after all just bits of DNA. The genes’ indifference holds true even if the organisms themselves *do* care. When a conquering male lion takes over a pride, he kills all existing infants despite violent resistance from their mothers. But shortly thereafter, these same females willingly mate with that murderous male. Why? Because their (very selfish) genes incline them that way. Think it through. If a female refuses the male, her future offspring (and future copies of her genes) simply won’t exist.

    Dawkins originally stated his case in provocative terms, because he was trying to get people to think. And it worked; many biologists started out enraged but were forced to come around by their own attempts to refute his thesis. And as I said, this was useful. Organisms are a mish-mash of traits; genes are singular. Thinking about natural selection in terms of genes is far more manageable, and therefore far more measurable, than trying to understand competition between whole organisms with all their complexities.

    And by the way, the attempt to show Dawkins up by talking about the myriad sorts of cooperation in nature is just silly. It’s like saying look, the entire sky and ocean both are blue so anyone who says trees are green is wrong.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Scientific American MIND iPad

Give a Gift & Get a Gift - Free!

Give a 1 year subscription as low as $14.99

Subscribe Now >>


Email this Article