Tuesday, June 19, 2012

Steve Pinker vs group selection

Steve Pinker is an evolutionary psychologist, a great thinker, a popular author, an impressive speaker (I remember one of his talks about the evolutionary explanation of the religion: it was cool), and also one of the few brave Harvard professors who have defended Larry Summers when professional women and professional blacks slammed the former Harvard president for the very suggestion that there may exist group differences.

Now, I want to discuss another scholarly issue that is related to groups: group selection. It's the idea that natural selection applies to groups and societies as well. In particular, societies composed of members who are willing to sacrifice themselves in ways that benefit the society prevail – because of analogous rules as the natural selection that is routinely applied to individuals or genes. In an Edge.org article called
The False Allure of Group Selection,
Pinker explains in quite some detail why he has a problem with the very concept and philosophy of group selection.

He tries to present the topic in as apolitical a manner as possible (although he hasn't avoided the ideological ramifications of these ideas elsewhere). That's perhaps also why he connects the idea of group selection with the promotion of conservative values. However, I personally think that Pinker realizes that the dominant advocates of group selection in the present world are leftists – people on the left side from him, to be more accurate – such as Sascha Vongehr, a Maoist and trained physicist from whom I learned about Pinker's essay.

Group selection as a theory is something that is designed to support collectivist ideologies and collectivist political regimes. They exist on both sides of the political spectrum; it just seems fair to say that the extreme left-wing ones are much more widespread than their right-wing counterparts today. Nevertheless, Pinker mentions fascism as something that should be the best organization of a society if the group selection theory were right.

Vongehr accuses Pinker of rejecting evolution. That's a very bizarre accusation as you will realize once you will swallow the totally crisp and highly conservative explanations of the evolution paradigm, as communicated by Pinker. Vongehr just likes the idea of evolution that is applied at many levels – multilevel selection – but he doesn't offer any truly rational argument for his sentiment – and he also claims that facts arising from biological research such as a variable mutation rate undermine Pinker's claims. It's not clear to me why Vongehr thinks so (the mutation rate and even its dependence on the parameters of the environment is just another quantity encoded in the genes and is subject to natural selection applied to the genes – the mutation rate in your body isn't controlled by the government, for example); his text is pretty much unintelligible so that's everything I want to say about it.

Back to Pinker.

He begins by a clear universal explanation of the mechanism of natural selection and evolution. Societies and cultures are claimed to compete in similar ways but Pinker says that this is at most a metaphor because societies and groups in general fail to obey three key conditions of the classical evolution theory:
  1. the number of copies isn't what measures the success; instead, we often talk about one society in its own category and its size, wealth, or longevity etc. matters; the statistical aspect is absent
  2. mutations aren't random but often planned; natural selection is meant to be a process that selects the preferred mutations while the process of mutation itself didn't anticipate or plan its consequences
  3. the success affects the group itself rather than its distant descendants
For those reasons, the natural selection doesn't apply to groups and cultures literally. It's just another interpretation of the single history that doesn't bring any new insights or predictive power.

Some sentences betray the postmodern intellectual atmosphere of the environment in which Pinker works (Edward O. Wilson is one of the fellow Harvard authors whose ideas Pinker argues against). For example, the group selection theory is "nice because it is non-reductionist and ecumenical", Pinker reproduces a theme heard in his environment. Needless to say, reductionism is considered by many as a major "villain" that partially drives the group selection advocates to their conclusions. Pinker makes a convincing case that it's the individual (in a species) or the individual gene that is always a player in the game of selection and the society or context may always be treated as a type of an environment.

It's surprising that Vongehr, a trained physicist whom I know from TASI-99 in Colorado (but I don't really remember him well), decided to argue against reductionism but I have already reduced this surprising attitude to Vongehr's Maoism (which brought us another piece of evidence in favor of reductionism).

The self-sacrificing behavior that benefits the group is the ultimate situation in which the classical evolution and group selection offer different predictions. Pinker analyzes some suicidal strategies whose carriers would be quite obviously driven towards extinction; time scales are important in some of them (individuals following these suicidal strategies may die out well before the group could see substantial benefits so the theory with a dominant group selection driver simply doesn't work).

He also says that the correct explanation of some suicidal behavior of bees involves the genes' desire to spread themselves (even if the genes are located in the queen, a different individual from the kamikaze bee). Human altruism may be predominantly explained by the individual desires to spread their broader kin with similar genes (regardless of more noble explanations that are sometimes being offered); so whenever the individual humans aren't the right "unit" that enters natural selection, the more correct unit is actually a smaller one (gene) and not a larger one (society).

At some level, this discussion may be just about the accuracy of effective theories we use to describe the phenomena. Pinker chooses the more microscopic, more reductionist viewpoint, as physicists would naturally do. That's the approach to describe the phenomena most accurately. (If Pinker were fully principled in this reductionist struggle, he would surely continue up to string theory instead of mentioning Lee Smolin's crackpot theory of Cosmic Natural Selection – although the latter didn't really play a role in Pinker's model of the world.) Some other people may prefer more macroscopic, emergent – and therefore less accurate, more social-scientific or humanities-oriented, and more metaphorical – descriptions working with greater, more composite building blocks.

Pinker says that even though the bees' behavior wouldn't be evidence in favor of group selection because it may be explained by the propagation of genes, the humans behave nothing like the bees, anyway. The human trait that group selection theory depends upon doesn't really seem to exist. And when some people do happen to sacrifice themselves – suicide attacks are Pinker's natural ultimate examples of self-sacrifices, of course – they lead to negative consequences for the individuals who do such things and people connected to them. The military history would look very different if suicidal attacks were benefiting the groups.

It's a clever reading which I recommend to you...


  1. If group selection is anything like herd mentality, it'll send you over a cliff and to your death:


  2. If group selection is anything like herd mentality, it'll send you over a cliff and to your death:


  3. It seems obvious, to me at least, that the processes of natural selection can't be magically acting on populations, they must act at the individual level. Are not the mutations themselves random? Is not the mechanism were by helpful mutations are preferred, first, that they help the individual survive and pass on those genes?

    Isn't this all obvious? I had no idea there were proponents of such a strange idea as "group selection" at least that anyone could possibly take seriously.

  4. I think this question hinges on what you define as a group. If you mean the whole species -- as in evolution is "for the good of the species" -- then obviously the answer is no. But if you mean a small family of closely related individuals, then the theory of kin selection would say the answer is yes: family members will sacrifice for the benefit of the survivors, with whom he shares many genes.

    Darwin thought group selection at the tribal level was probable. Tribes that displayed valor and cohesion in battle for instance would tend to triumph. What was unexplained was how you get from kin to tribe. Most tribes in prehistory were no doubt related genetically so it really boils down to a math problem.

    One problem is identifying who is and is not closely related. I've often thought that proximity might be a good proxy -- on average people who grow up in the same immediate neighborhood have been genetically related in most of past history (villages, bands, etc) so a down and dirty algorithm that programmed people to sacrifice for the people they grew up with might be good enough via the inclusive fitness/ kin selection mechanism. I've never heard a serious biologist make this argument however.

    In class societies the aristocracies were often closely related -- cousins marrying cousins and the like -- which helps explain the concept of honor on the battlefield: say what you will aristocrats were ready to die on the battlefield rather than risk dishonor. If they lost the war they and all their peers were liable to be slaughtered, enslaved, impoverished. Does an aristocracy count as a group?

    Then there are ethnically homogeneous nation states and the whole issue of nationalism. To say nothing of highly out-bred societies like the United States. You see self-sacrificing behavior on the battle field but it more difficult to imagine how these traits evolved.

    Hbd* chick talks about these issues all the time. I recommend her blog.

  5. Suicide missions aren't representative. One should think in terms of probabilities, of taking chances, not kamikaze behavior.

  6. And then of course there is the whole issue of sexual selection. Women love heroes and have lots of children by them. That would propagate genes for high risk behavior on the battlefield.

    I'm obviously not an expert but this isn't string theory.

  7. Darwin was the first, and as usual, the most acute in discussing group selection. His key recognition was that it couldn't work unless free riders could be punished. The thing is, most human societies, especially primitive societies, are pretty good at this.

  8. Events act on individuals, not groups. But groups share characteristics, even if only what the reactions will be, in part based on non-biological things as culture or shared history/memories. Groups can become extinct the same way that groups can survive: the Massada Jews didn't survive the Roman army while the Kiowa (who retreated into the hills in the face of the US Army) did. Each responded for non-biological reasons.

    Group differences can be adopted as well as in-born. What characteristic humans choose to develop is determines more about what a person (or group) will achieve and be "successful" because of, than his genetic makeup.

    Lamarkism has a real place in human development, but it is a cultural place, not (necessarily) a biological one.

    We are potentially good at a lot of things. That some groups are very good at some things may mean that they focus on it more than it does that they are genetically better or different. In some ways I see that we are already mutated and evolved for unusual circumstances: we just haven't had the opportunity for those special skills and adaptations to show up.

  9. He should study the modern greek mentality. Sacrificing for the good of the group/village/town/nation is a meme that started and is kept up with learning at school of Leonidas and his 300 men in Thermopylae. The sacrifice of Iphigenia is also on those lines ( not oneself, but one's kins).

    Modern greek history is rife with self sacrificing acts, to get out of the turkish rule, out of the WWII occupation.

    A saying learned at grade school: "better an hour of free life than forty years of slavery and prison" was a main attractive pole in this Syriza outbreak. Those who view the European contracts/memoranda as servility to outside forces reverberated to this. What EU perceived as a blackmail Syriza sold as a gamble : free versus servile. That is what grabbed the young out of work imagination. And lo, a group that got 4% of the vote grabbed 27%. Of course promising everything to everybody was also crucial, but the protagonists and strong defenders in blogs were caught be the meme.

    I prefer the meme paradigm because it rests on the brain functions, not on genes.

  10. Dear Anna, what you describe is some aspects of the Greek history and the Greek present. But what Pinker disagreed with was a particular - that this self-sacrifice etc. helps the societies of this kind to survive and multiply. Given the current state of Greece, this assertion would be very hard to make even if there were no other problems with the model.

    You may "prefer" a paradigm and there may be certain analogies. However, just because someone "prefers" the meme paradigm doesn't mean that every claim based on the assumption that memes are important is correct. The claims you prefer may very well be incorrect and one could make a rather strong case that they are.

    The truth in science isn't about "preferences". It's about the victory of the hypotheses, theories, and models that have been successfully verified against the evidence and you haven't done so with the ideas you "prefer".

  11. Group selection certainly operates, though the larger the group the smaller the effect. It's mostly kin selection in practice. Besides in pack animals which depend on group for survival there is no clear way to distinguish between group and individual selection.

    The ultimate proof of group selection is death, the organism pays the ultimate price but makes room for his and others offspring increasing their chances to succeed. The lifespan of all species is carefully tuned by evolution.

  12. You are correct that altruism is not a gene, if it is in a gene, or a meme very evident in greek society. I have said that our educational program should seriously undertake to promote altruism as a stance, the way it promotes heroism.

    "Give me liberty or give me death" is a heroic stance.

    "I will share my goods with you". is not heroic, though very important in maintaining a society.

  13. It is not hard to imagine how, among many troupes (or extended family groups) of proto humans who competed for territory, a troupe in which a single pleiotropic gene (or, less ideally imagined, two complementary genetic tweaks) had quickly become thoroughly entrenched and ‘endemic’ because it allowed the first carrier(s) and a few generations of their descendants to gain great selective advantages, and that this group’s genetic/phenotypic uniqueness was of a kind that levered the function of language to a new level - a level at which the group (troupe) became better than any other group at coordinating its attacking (and defending) and to better (or for the first time ever) hang on to some emboldening and fear blocking idea - a rudimentary religious one.
    This (not so fictitious) example of group selection would easily have come to dominate within the world of human intraspecies competition (a naturally selective competition that of course often involved cooperation).
    The above is part of a recognition of mine that I believe has either no (complete) precedent or is very rare. That is, I will believe it until - if ever - someone can point out to me that it is not. However, I know for sure that only I have conceptualized this recognition the way that I have done it (for better or for worse)! %-|

  14. Lubos,
    How do I delete the confused cross-eyed comment I just wrote??

  15. You may both delete it and edit it. Editing is below the comment, Reply Edit Share, and Delete is an option behind the downward triangle in the right upper corner which you first click.

  16. As I understand it, group selection as a historic concept places emphasis that much or mostt natural selection takes pllace at level of major populations or entire species (think end of age of dinosaurs) rather then in the form of incremental selection of individual traits, further the concept of the "Selfish Gene" after Dawkins would explain same selection from point of view of the gene itself, so really there are maybe three levels in gross terms at which natural selection might operate, I am sure that individual scientists might be influenced by non-pertinent political ideologies to put primacy at a given level.

  17. The Pinker article has a very fascinating history. Until about 15 years ago, the prevailing orthodoxy among academic and professional experts in the behavioral sciences was the so-called Blank Slate, according to which there is no such thing as human nature or, if it exists, it is insignificant. The reasons for this orthodoxy were ideological rather than scientific. Perfectly malleable human beings were required for the socialist utopias the ideologues were preparing for us. Anyone who objected was furiously attacked as a right winger and a fascist. The greatest debunker of the Blank Slate, by far, was a man named Robert Ardrey. You don't have to take my word for it. The Blank Slaters themselves admit as much very explicitly in an invaluable little book entitled "Man and Aggression," edited by arch-Blank Slater Ashley Montagu, published in 1968, and still available for a pittance on Amazon the last I looked.

    Eventually, the Blank Slate idiocy collapsed under its own weight. Some of the more respectable scientists in the field of human behavior accepted the existence and significance of evolved human nature, and developed new fields to study it, such as evolutionary psychology. Unfortunately, it was too much of an insult to their academic gravitas to admit that all the "experts" in their field had been wrong for so long, and a man like Robert Ardrey had been right. You see, Ardrey was a "mere playwright." Well, as Orwell said, "He who controls the present controls the past." A whole new revised history was created that relegated Ardrey to the role of an unperson, along with a couple of others who had written against the Blank Slate at about the same time, such as Konrad Lorenz and Irenaus Eibl-Eibesfelt. A new knight in shining armor, who had come forth and defeated the Blank Slate almost single-handedly, was anointed in the person of E. O. Wilson. His first book against the Blank Slate orthodoxy, "Sociobiology," wasn't published until 1975, long after Ardrey's most influential books, and was entirely derivative as far as the theme of human nature is concerned. No matter, Wilson possessed academic gravitas, and the "experts" recognized them as one of their own. The man who played the greatest role in creating this fantastic historical concoction was none other than Steven Pinker in his book, "The Blank Slate." He actually wrote a whole book about the Blank Slate that only mentioned the man who was most influential in defeating that orthodoxy, and then only to claim that he had been "totally and utterly wrong."

  18. ...continued,

    To "back up" this claim, he cited the following lines from Richard Dawkins "The Selfish Gene":

    These are the claims that could have been made for Lorenz’s On Aggression, Ardrey’s The Social Contract, and Eibl-Eibesfeldt’s Love and Hate. The trouble with these books is that their authors got it totally and utterly wrong. They got it wrong because they misunderstood how evolution works. They made the erroneous assumption that the important thing in evolution is the good of the species (or the group) rather than the good of the individual (or the gene).

    In other words, the sole justification for the claim that Ardrey was "totally and utterly wrong" was group selection! No matter that group selection was in no way, shape or form central to the thought of Ardrey, not to mention Lorenz or Eibl-Eibesfeldt. That single quote from Dawkins book was enough for Pinker to drop them down the memory hole.

    But wait, there's more! E. O. Wilson, that great savior of the behavioral sciences, has just published a book, "The Social Conquest of Earth," which is a defiant defense of, you guessed it, group selection!! Science is deliciously ironic sometimes, isn't it? This has to be one of the best practical jokes history has ever played on the self-anointed experts of science. That's why Pinker felt he had to put in his two cents worth about group selection. I suspect he was motivated by his guilty conscience. And now you know the rest of the story.

  19. the choice is between the "individual" and "groups of individuals" as the unit of natural selection but what about looking inside to the cells which make up individuals as the unit that benifits from the evolution of complexity.Single cells came together to form organisms because they derived benifits from doing this the individual is just a byproduct of evolution perhaps not its purpose.

  20. The "unknown" is Doug Drake. David Sloan Wilson is a major advocate of group selection, and has published papers with the other Wilson. His website, This View of Life, currently features a prominent link to Pinkers article. I'm sure he'll send a salvo back at him in the near future. Doug Drake/Helian

  21. Quite extensive study on this matter: