What do you think about group selection?

By Razib Khan | April 8, 2012 9:50 pm

I just received a review copy of E. O. Wilson’s The Social Conquest of Earth. One of the reasons why this book is “hot” is that Wilson has recently been revisiting the “levels of selection” debates, and significantly downgraded kin selection in the pantheon of evolutionary dynamics (at least in his mind). There has been a lot of talk on the blogs about Wilson’s ideas, in large part because of his partisan position on the Nowak vs. most other biologists debate, in favor of Nowak.

I don’t know if I’ll have time to review the book (a reality I honestly explained already to the people working at the publisher), but, it did get me thinking: what are the opinions of biologists in relation group selection? My personal experience is that opinions actually vary by discipline and by department. It’s hard to get a real sense, because people tend to be in their own “bubble.” With that in mind, I’ve put together a small survey to assess opinions. My core audience here are people who consider themselves biologists, though I can’t prevent someone with strong opinions from participating obviously!

So, a survey on group selection. You can see the results here.

CATEGORIZED UNDER: Evolutionary Genetics
MORE ABOUT: Group Selection
  • Dallas

    Why didn’t you add the field ‘evolutionary biology’? I mean, that is the field in which this debate principally occurs, right?

  • http://blogs.discovermagazine.com/gnxp Razib Khan

    #1, evolutionary biology spans a lot of disciplines. there are people studying evolution in genetics and ecology, for example.

  • Stephen

    Anymore, everyone is a geneticist of some sort. So that is usually true and a rather un-useful category. For purposes of the survey, it seems you want to peel off those folks who consider themselves “population geneticists” or “evolutionary geneticists” — that is, people who have thought about the question previously.

  • http://julianodea.blogspot.com/ Julian O’Dea

    Much as I admire Wilson and generally dislike psychological explanations, I do have a theory as to why he has downgraded kin selection. I think he regrets not having had Hamilton’s key insight into why social insects may have evolved, the haplodiploidy point.

    Wilson said somewhere that he wished he had spent more time thinking about the problem, and less time in the field gathering data. I think he feels that Hamilton beat him to a key insight, and he now seeks to downplay it.

  • http://behavecology.wordpress.com Steven

    I’ve been convinced by people like Stuart West and Steven Frank (among others) that kin selection and MLS (or ‘new group selection’, not old-style Wynne-Edwards group selection) are mathematically the same thing and that there hasn’t yet been a formulation of a problem in MLS-terms that is more useful than the corresponding kin-selection formulation. Thus, for the first question I’m torn between ‘theoretically impossible’ because MLS = kin-selection and is therefore reducible to the individual, or ‘theroetically possible and ubiquitous in nature’ because MLS = kin-selection and is everywhere. Which option would you advise I choose in terms of how you meant the question?

  • http://forwhattheywereweare.blogspot.com/ Maju

    For whatever is worth, IMO altruism is actually a trade at different times: I scratch your back now and expect to be scratched tomorrow (by you or someone else in the partnership). And, behold, it happens (or the solidarity system collapses in the mid-run). The less important element is who is the partner, although, of course, if the partner is a relative there’s some extra value in the cooperation system (but members of the same species are all close relatives anyhow, sharing most of their genes: any human and my fourth-degree cousin is nearly the same).

  • Darkseid

    so they mean that its possible with ants and bees but probably not with other stuff? The only time I’ve spoken to my biologist neighbor i asked him this question and he said he thought group selection was bunk.

  • http://timtyler.org/ Tim Tyler

    Which “group selection” are we talking about? The term “group selection” has come to mean several different things. The “new” group selection – the models of which are broadly equivalent to kin selection models – represents a reasonable perspective. I generally agree with S. West, A. Gardner & J. Marshall about its overall utility.

  • dave chamberlin

    I wish you had more time to read books and write your unique reviews. You have got a gift there Razib my boy, the gift to peer into the cutting edge of science and let us intellectual peons grok it. I know you don’t have the time to read and review the best non fiction books any more but the seething masses are out there clamoring for your brand of infotainment.

  • Matt

    I’m a first-year student in Life Sciences (finishing up first year), planning on majoring in molecular biology and genetics (McMaster University in Canada). I’m not really trying to weigh in here, but rather provide the perspective of someone’s who’s just getting into the field. We’re taught that animals do not do things for the good of the species, and that ‘good for the group’ is only a side effect of being good for the gene. A couple examples that were presented to us include foraging-vigilance trade-off, which is described as neither group selection nor altruism, and the famous lemming suicide myth. “Unless the altruistic genes produce more copies of themselves than other genes that don’t lead to altruism, the good of the group cannot explain the evolution of altruism, or any other social behaviour.” Also, my biology textbook (Biological Science by Freeman, Harrington & Sharp, 2008) maintains that “no instance of purely self-sacrificing behaviour – where the individual received no fitness benefit in return – has ever been recorded in nature.”

    We’re introduced to group selection from a very skeptical perspective, and are then taught about kin selection, inclusive fitness, and Hamilton’s rule as better explanations for altruism. So that’s what first-year students are being taught (at least at my school), if it wasn’t already known.

  • http://dispatchesfromturtleisland.blogspot.com ohwilleke

    “Groups” certainly are a slippery thing. The definition seems clear in humans until you start looking closely and trying to figure out at what level group selection operates, but few animals other than social insects have very large groups.

    Are there more than hundred hoofed animals in a typical herd? How many predators or ominvores form groups of more than a dozen? Yes, primates have moderate sized bands, but it isn’t obvious that they have social structure beyond that, unless one can consider the development of traits like respect of individual or small group hunting territories to be a “group” trait.

    The really interesting issue, one that mostly distinguishes humans from others, is selection that acts on shared cultural innovation at the group scale (cultural innovation at the individual or very small group scale may be impossible empirically to distinguish from biological evolution without actually looking at genes and bodies, it may not have the distinguishing features of group level selection). Yes, other animals have been demonstrated to learn and even diffuse innovation culturally. It doesn’t seem to be the case, for example, that even social insects engage in cultural innovation. A bee hive today looks and operates pretty much exactly as a bee hive did tens of millions of years ago.

    But, the case that selective advantage for humans producing the modern gene pool have been mostly about technological packages and cultural-political organization advantages that flow from group membership, rather than biological advantages, seems pretty strong.

    Neolithic peoples, at least initially when the gene pool shift took place, it doesn’t seem as if they came to make up a hugely increased share of the gene pool because they were stronger or smarter or prettier. They learned a way of life and that way of life allowed for more population density and food security in the long run.

    Or not. The alternative argument, and it is a touchy one, is that Neolithic peoples self-domesticated themselves and were different and that this is why food production was more of a demic and less of a culturally transmitted phenomena. In other words, perhaps there were personality traits and hereditary immune system features and intellectual capacities and abilities to handle living in close quarters with others (perhaps degraded sense of smell), that made Neolithic peoples better biologically adapted to their lifestyle than their hunter-gather neighbors beyond the well known middle to late Neolithic trait of latase persistance. Of course, the two possibilities aren’t mutually exclusive.

  • http://blogs.discovermagazine.com/gnxp Razib Khan

    #10, that sort of information is very useful. though i’m not surprised. my own experience is that there are certain groups of behavioral ecologists who are very pro-group selection.

    The really interesting issue, one that mostly distinguishes humans from others, is selection that acts on shared cultural innovation at the group scale

    i agree that this is an interesting possibility.

  • Matt

    Also, I’ve assumed that the quote “no instance of purely self-sacrificing behaviour – where the individual received no fitness benefit in return – has ever been recorded in nature” from my textbook is referring to non-human species only. It seems to me that it’s patently false when human behaviour is considered. Although it may be rare, and humans are much more likely to behave selflessly towards kin, one can certainly conjure up an example of a human sacrificing him/herself for a complete stranger. To me, the explanation for this seems to be that although one would expect genes for purely self-sacrificing behaviour to die out, we’ve also evolved the capacity to reason, for unrelated reasons. So humans are an exception to this rule, with self-sacrificing behaviour, even if rare, being a by-product of our reasoning capabilities.

  • Dm

    To what extent has a possibility been explored that altruistic behavior in our species may be a pleiotropic trait? Could it be one of the faces of a complex multifaceted trait, which is mostly subject to selection through its other faces? Say if the same genes also influenced our grandparent-grandchild bonds and were subject to selection through the latter effect? Or even things not specific to our species (kinship bonds, longer-term planning / forecasting abilities, sexual courtship, whatnot). IMVHO one does not have to postulate that altrusitic behavior is _directly_ selected to explain why it is prevalent.

  • S.J. Esposito

    Razib, in response to your suspicion in 12, I’ve had a rather lengthy discourse with a primate behavior ecologist on group selection, and she seemed to be the only one of my professors to ever give it any real credence.

    Like Matt (in 10), I’ve had mostly the experience of being taught that anything that seems like group selection is probably better explained by some other phenomena. I don’t know how much this differs between undergraduate and graduate level education (haven’t made it to the latter yet), nor do I really know how people come down on the issue based on their field of study, but I’m really interested to find out.

  • miko

    “no instance of purely self-sacrificing behaviour – where the individual received no fitness benefit in return – has ever been recorded in nature”

    This is definitely false, even for non-humans, because behavior is noisy and exploitable. Unless you want to get into a tricky discussion of intentionality in animal behavior (it’s not altruism if you didn’t “mean” to do it), but that would be pointless. I pretty much buy that humans are altruistic (to the extent that they are) because culture has parasitized our brains and exploited our kin selection instincts into doing maladaptive things. Other animals are also prey to manipulation of their instincts – not least by conspecifics – to produce objectively altruistic acts.

    Group selection as a concept is mostly crap and Nowak et al. is unadulterated effluvium from a glorified Templeton think tank. Only a Fox News style “balance” could present this as even a legitimate debate. But there is semantic confusion related to “levels of selection” discussion that is useful. That certain clades are “better” at generating variation and thus morphological evolution and speciation — this is why we see teleosts and insects of every describable morphology and sharks pretty much look like sharks — is non-controversial, though the reasons are enormously complex and interesting. Some people get confused and call this group selection, though they should not.

  • DK

    Thus, for the first question I’m torn between ‘theoretically impossible’ because MLS = kin-selection and is therefore reducible to the individual, or ‘theoretically possible and ubiquitous in nature’ because MLS = kin-selection and is everywhere”.

    My dilemma exactly. In the end, I answered “ubiquitous in nature”.

  • Tom Bri

    A counter-force against kin selection is that the more closely related two individuals are, the more they compete for the identical resources. Culturally, we see this in family fights over inheritance. On a group level, my ‘village’ has a limited amount of farmland, and every bit more one individual controls, another person from the in-group loses. Outsiders don’t even come into it. On a species level, we compete more against chimps than we do against, for example, cattle. Most of what chimps eat, we can too. If human numbers increase, chimp numbers must decrease.

    So, my genes may benefit more when an in-group member dies and I get command of his resources than if I were to act altruistically and keep him alive. Positive kin selection would have to be pretty strong to overcome this. IN-group it may be better if a close relative dies, and kin selection would only come into play against outsiders.

    I wonder if this may explain why human #s stayed low for so long, and we seem to have derived from such a limited gene pool, our near relatives competed directly for the identical resources, so it was a slow slog until they were eliminated.

  • j mct

    I’d say that that the extreme most extreme example of group selection isn’t really thought of as an example of group selection and is an example of a badly done thinking at what the ‘unit of survival’ is. The ‘unit of survival’ is usually thought of as ‘the organism’, but the real unit of life, or ‘true organism’ is a single cell.

    A multicelled organism is a ‘group’ for this purpose, and since all the cells of any multicelled organism are descended from the same cell, Hamiltons inclusive fitness stuff at the level of say a man’s brain cells versus his muscle cells… kicks in with a vengeance, and a multicelled organism is the ‘edge’ case that illustrates the cases where the effect is weaker. Since all the cells of a multicelled organism are genetically perfectly related, they can exhibit ‘perfect altruism’ where each cell will give his all for the team, or is perfectly willing to lay down his life for the tribe as a whole, even forego reproducing to help some other member of the team, who would be his ‘full brother’, i.e. is a genetic twin, to reproduce. From that the exquisite level of teamwork multicelled animals, i.e. groups, display can come about such that a human cell cannot survive outside a human body since human cells are so specialized that a specific human body, the one with it’s genetic twins, is the only place a human cell can live. If one took a human cell out of a human body and put it in a pond with where the amoebae live it would not last very long. In this case, the ‘group’ is what survives or not, the individual organism, the cell, does not. In addition, a mutation that was made the muscle cells more efficient wouldn’t be selected for if they made the liver cells much less efficient, since the group has to stand or fall together. Multicelled organisms would seem to me to be a perfect illustration of group selection in action.

    Also, it seems to this non biologist clueless bozo that the Hamilton’s inclusive fitness explanation was not, at least originally, an explanation as to why there was altruism. Prior to when he thought it up, altruism was thought to be a big problem since evolutionairy theory would predict cheaters and the like, so altruism was thought to be impossible. Hamilton’s explanation doesn’t seem to me to be an explanation as to why there was altruism, clearly lots of other stuff has to be going on, it was an explanation as to why altruism was not impossible, which isn’t necessarily it’s cause.

    Altruism is never general either, it is alway limited as to how far it goes, and for most multicelled organisms, I don’t think that there is any altruism of any kind beyond the organism itself. It would seem that kin selection wouldn’t get the job done on it’s own as far as causing altruism.

    I think that all this would a useful avenue for thinking that would could lead to some fruitful results.

  • BDoyle

    It always boils down to a desire to explain altruism as an adaptation, doesn’t it? If it weren’t for that, I think the whole debate over group selection would shrink to a few esoteric specialists in the field. For me, it is easier to think “If it is so hard to explain altruism as an adaptation, then maybe it *isn’t* one.” I am kinda with miko on this, I think. It is clear that it is advantageous to the individual for everybody else to be altruistic, maybe what is responding to selection pressure is the ability to manipulate others into behaving this way.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at http://www.razib.com


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar