Steven Pinker’s unconvincing debunking of group selection

Steven Pinker writes:

Human beings live in groups, are affected by the fortunes of their groups, and sometimes make sacrifices that benefit their groups. Does this mean that the human brain has been shaped by natural selection to promote the welfare of the group in competition with other groups, even when it damages the welfare of the person and his or her kin? . . .

Several scientists whom I [Pinker] greatly respect have said so in prominent places. And they have gone on to use the theory of group selection to make eye-opening claims about the human condition. They have claimed that human morailty, particularly our willingness to engage in acts of altruism, can be explained as an adaptation to group-against-group competition. As E. O. Wilson explains, “In a group, selfish individuals beat altruistic individuals. But, groups of altruistic individuals beat groups of selfish individuals.” . . .

I [Pinker] am often asked whether I agree with the new group selectionists, and the questioners are always surprised when I say I do not. . . . The more carefully you think about group selection, the less sense it makes, and the more poorly it fits the facts of human psychology and history.

I was not persuaded by Pinker’s arguments against group selection. Before getting to the details, let me just say that I think it’s ok that I think this. After all, Pinker himself says that E. O. Wilson thinks group selection is real. I’m not saying that Wilson is definitely correct, but when it comes to evolution it certainly doesn’t seem unreasonable to side with biologist E. O. Wilson over linguist Steven Pinker or popularizer Richard Dawkins. (Just to be clear: I’m not saying I found Pinker unpersuasive because I was weighing his qualifications against Wilson’s. What I’m saying is that (a) I found Pinker unpersuasive (for reasons I outline below) and (b) it seems that I’m joined Wilson, among others, in that point). I do, however, recognize that many prominent biologists despise the idea of group selection.

Pinker makes three arguments:

1. For a process to be called “natural selection,” it must satisfy the following criteria: “the groups made copies of themselves by budding or fissioning, the descendant groups faithfully reproduced traits of the parent group (which cannot be reduced to the traits of their individual members), except for mutations that were blind to their costs and benefits to the group; and groups competed with one another for representation in a meta-population of groups.” Without these, Pinker says, stories about group selection are “so poetical that they shed no light on the phenomenon” and are no better than silly analogies such as “Cars today are equipped with steel-belted radials because they outcompeted polyester-belted tires in a process of tire selection.”

2. Pinker makes the reasonable point that “it’s only when humans display traits that are disadvantageous to themselves while benefiting their group that group selection might have something to add.” He then continues: “this brings us to the familiar problem which led most evolutionary biologists to reject the idea of group selection in the 1960s. Except in the theoretically possible but empirically unlikely circumstance in which groups bud off new groups faster than their members have babies, any genetic tendency to risk life and limb that results in a net decrease in individual inclusive fitness will be relentlessly selected against.”

3. “Nepotistic altruism in humans consists of feelings of warmth, solidarity, and tolerance toward those who are likely to be one’s kin. . . . The cognitive twist is that the recognition of kin among humans depends on environmental cues that other humans can manipulate. Thus people are also altruistic toward their adoptive relatives, and toward a variety of fictive kin such as brothers in arms, fraternities and sororities, occupational and religious brotherhoods, crime families, fatherlands, and mother countries. These faux-families may be created by metaphors, simulacra of family experiences, myths of common descent or common flesh, and other illusions of kinship. None of this wasteful ritualizing and mythologizing would be necessary if “the group” were an elementary cognitive intuition which triggered instinctive loyalty. Instead that loyalty is instinctively triggered by those with whom we are likely to share genes, and extended to others through various manipulations.”

To which I reply . . .

One man’s “manipulations” are another’s group selection. The story I’ve always heard, and which makes sense to me, is that different groups have different social organizations, and some are more successful than others. It all depends on the environment, but in some environments an every-man-for-himself (or, every-man-for-himself-and-his-kin) society won’t be so economically successful. A famous example is the transition from hunter-gatherer to agricultural societies; it’s said that the latter dominate economically even while being worse for most of the individuals involved.

Pinker might argue at this point that this has nothing to do with natural selection and has everything to do with wasteful ritualizing, mythologizing, and coercion. But I don’t see how that shoots down the group-selection story, in which the qualities that help a social structure expand and be copied are not necessarily those that benefit each individual or kin. The story is that many of our social norms arise because we are part of societies that exist partly from the development of such norms.

To me, Pinker’s arguments seem inappropriately normative. Anything having to do with genes is “instinctive”, while more generally social attitudes are dismissed using words such as “faux,” “simulacra,” “myths,” “illusions,” and “manipulations.”

A clue to Pinker’s attitude comes from his statement, “the term ‘group selection’ adds little to what we have always called ‘history.'” I’m reminded of the notorious statement that all science is physics or stamp collecting. The study of history is hugely important, and what history is about (to me) is nailing down the facts. History is the ultimate descriptive social science. What happened when, who said what to whom. Sure, interpretations are central to historical work (consider, for example, A. J. P. Taylor’s “Origins of the Second World War”), but what historians contribute in particular is the particularity of history.

Just as there’s a lot of science in between physics and stamp collecting, there’s a lot of social science in between biology and history. As a psychologist, Pinker knows this, but he seems to forget it here. The study of group selection of societies would seem to me to fall in the category of anthropology or sociology. I do think that it can help to understand the social benefits of “ritualizing and mythologizing,” rather than simply dismissing them as “wasteful” and created by “manipulations.” Just as a biologist or psychologist might study some apparently strange individual behavior as a result of some evolutionary adaptation, it can make sense for a sociology to take rituals and myths seriously and not be so quick to dismiss attitudes such as patriotism that are central to our culture.

That said, I don’t want to overstate my case. I’m not a biologist or even a sociologist, and I’ve been careful to characterize these cultural evolution examples as “stories.” That’s just the way it is: we have a lot more examples of animals than of societies so these ideas are inherently more speculative, more like the sort of just-so stories for which evolutionists are so notorious. So I agree with Pinker that cultural evolution will never reach the scientific status of the evolution of species. But that doesn’t mean that group selection isn’t real. It just means that group selection makes Pinker uncomfortable.

Pinker has clearly thought more about all this than I have, so I won’t go so far as to say I think he’s wrong. Just call me unconvinced.

P.S. Pinker characterizes natural selection as “the best idea that anyone ever had.” I don’t know about that. Given that evolution by natural selection is (basically) true, it’s no surprise that someone would come up with it sooner or later. I think the title “the best idea that anyone ever had” should be attached to a truly unique idea that might never have arisen. I’m sure that Steven Wolfram and Nathan Myhrvold have come up with a few dozen better ideas all by themselves! And what about that religion that L. Ron Hubbard invented? The secrets of nature will eventually be discovered in any case, but coming up with a new religion all on his own—that’s what I call impressive.

P.P.S. The discussion by Richard McElreath in comments seems helpful.

40 thoughts on “Steven Pinker’s unconvincing debunking of group selection

  1. Maybe in a utilitarian sense the invention of farming made people worse off but in a what the genes want sense it certainly lead to many more copies of genes for those participating in agriculture. So I am not sure that it represents a divergence between group and genetic interests.

  2. Andrew:
    “[If] (basically) true, it’s no surprise that someone would come up with it sooner or later.”
    Sounds like ideal realism – so I googled.

    Ended up here: http://www.anselm.edu/homepage/dbanach/31-theories.htm

    From there this entry looked close;
    “Internal Realism: (Peirce, Late Hilary Putnam)
    Truth exists as an ideal towards which the scientific community progresses. Truth is not a relationship between our scientific activity and something else, but is what the ideal scientific community would progress towards under ideal scientific conditions. The basic unit of connection with the world is no longer the particular theory or paradigm, but the entire history of the scientific community.”

    This entry perhaps a bit off (or pre1900):
    Pragmatism: Charles Sanders Peirce (1839-1914), William James (1842-1910), John Dewey (1859-1952) Truth (and meaning) are always relative to a particular practical context, to a set of practices and values. Truth is what works (in that particular practical context).

    So I came up with a revision:
    PragmatiCism: Charles Sanders Peirce (1839-1914)
    Truth (and meaning) are always relative to a particular purposeful representation (modeling exercise), for a set of motivations and concerns. Truth is what is the least wrong representation (after infinite refinement for that particular purpose).

  3. @OneEyedMan I’m not a biologist (I’m a political scientist), but I don’t think that genes want anything at all. Sure, people want to survive as long as possible, and, as a consequence, people who survive are more likely to reproduce themselves, all else equal. But I don’t think we gain anything by saying that genes want something.

    • Saying genes want to reproduce is like saying hydrogen wants to combine with oxygen. I’m not sure it’s meaningful to say a person wants something, but we have to talk to each other and so we go along with the narrative that we make choices and stuff.

      • In high school chemistry we were taught not to anthropomorphize chemicals, but I always found that silly. Much easier to think about strong acids as wanting to let go of H+, etc.

  4. Let me try to break this down.
    The Original Question: Humans all over the world make sacrifices for their groups. The ubiquity of the behavior seems to suggest that it’s somehow hardwired into us. But intuitively it would appear that natural selection should reward selfish behavior. How did this happen?

    EO Wilson (as summarized by Pinker): Good groups improve the survival chances of their members.

    Pinker: It is hard to imagine how natural selection would work to promote loyalty to large groups when individuals reproduce much faster than large groups do. It is easier to imagine that we have repurposed existing loyalty-to-kin machinery (that makes sense from a natural selection perspective). It’s unnecessary to imagine extra hard wiring for loyalty to the group.

    Gelman:
    Possible interpretation 1 (nothing to do with hard wiring): The original question shouldn’t focus on hard wiring. Group loyalty evolved, but maybe not in the natural selection sense of the word. This seems like a tautology. It’s here, so of course it “evolved” somehow. And it has to do with groups, so of course it was by “group evolution”.
    Possible interpretation 2: It’s plausible that one type of group system beat out another type of group system (eg agriculture over hunting and gathering) and the dynamics of the new system selected for group loyalty.
    Maybe there are other possible interpretations of Andrew’s response…

    I think Andrew’s attack on Pinker’s dismissiveness of social science is misguided. Pinker is addressing the Original Question, which has to do with hard wiring, and this social science stuff really doesn’t apply to the hard wiring question. I’m sure Pinker appreciates the great role social science would play in explaining how and why we repurpose our loyalty to kin wiring to create loyalty to the group behavior.

  5. Andrew, as I understand it, the controversy is largely semantic. All selection ultimate operates through the channel of kin: it’s biology. What Wilson calls group selection exists as a result of altruistic behaviors increasing the likelihood of kin survival via the survival of the group. Is this group or kin? The position of the kinsters is that the name we give it has to recognize the necessity of the kin channel. The groupsters seem to take the opposite view, under the impression that kin selection ultimately reduces to mechanistic rules based on relatedness. (In some naive versions, this is an apt description.) But there can be an enormously complex set of mediations between individual behavior and kin survival, which looks very groupish. This is how Wilson’s ants must look to him.

    I think it’s a non-debate for anyone who has a sophisticated sense of how complex kin survival can be and also acknowledges the biology.

    On the other hand, to the extent that an analogous (emphasize analogous) process occurs at a cultural level among humans, there is no need for a kin channel because there’s no biological substrate to transmission, or at least not one operating through relatedness.

    • The boundary line between kin and “non-kin” is a lot hazier than people assume. If you go back 40 generations (about a 1,000 years), your family tree has a trillion open slots to fill. Clearly, some of your ancestors did double duty, genealogically speaking.

  6. ” And what about that religion that L. Ron Hubbard invented? The secrets of nature will eventually be discovered in any case, but coming up with a new religion all on his own—that’s what I call impressive. ”

    No No No To invent a new religion is very easy! Hubbard did his just as a stunt to prove that.

  7. In Jonathan Haidt’s new book he provided an example of a chicken farmer who bred chickens not based on how they performed individually, but on the total output per group of 12 chickens that was in a box together. He found that this 1) increased to total amount of eggs produced and 2) reduced the amount of infighting that occurred between the chickens. I thought that this was the most solid example of what pressure from the modern understanding of group selection looks like.

  8. Perhaps group selection is a term that will stay around for convenience sake recognizing that it comes down to individual selection in the end. As long as one recognizes one’s individual genes are what are selected and not the individual themselves one could twist any of the selection of groups that is successful eventually into individual selection. However, that individual selection may be hard to see, work with, or develop from in many cases and so group selection will be used. The genetic mechanism stays the same but the level of explanation used is for convenience within the domain of research.

    On an unrelated note, I’ve found it problematic that those who argue against group selection, and even for it, tend not to touch on what happens to the descendants of the unselfish. Are the hero or martyr’s children more successful? That seems like a good line of research to undertake for those interested in the topic.

  9. “Pinker makes the reasonable point that ‘it’s only when humans display traits that are disadvantageous to themselves while benefiting their group that group selection might have something to add.'”

    I disagree that this point is reasonable. Group selection can in fact only work when the net advantage to members of the group is positive. It’s still group selection if the individual chooses behavior which creates a disadvantage compared to other members of the group. By enhancing the advantage of the group over other groups, the behavior also enhances individual advantage (in the absolute) more than the reduction of advantage relative to other group members.

    For example, consider the behaviors of cooperatively assisting in the raising and protection of children. A mother would achieve an advantage within the group if she were to go around killing off the offspring of other women. However, this would create a disadvantage for the group compared to other groups where this behavior does not occur, and this group disadvantage makes the children of the baby-killer less likely to survive than if she had left the other women’s children alive. Intra-group advantage in this case translates into overall disadvantage.

    I also disagree with Pinker that the idea of these traits being (at least in part) genetically determined can be dismissed, and insist that the idea of groups competing for status within the meta-group population follows from the fact that groups even exist with mathematical inevitability. As to the first, our understanding of the determinants of human behavior is imperfect, we find genetic causes of many animal behaviors, and one can seldom (if ever) isolate a single cause for any behavior; it’s plausible that any given human behavior is influenced by genetics at least to the point of increasing or decreasing the likelihood of the behavior being learned. A perfect example is language. Language is obviously a learned behavior, but there are no human populations that lack it; the genetic predisposition to learn language is actually a point that Pinker raises himself in a different context. If that, why not other behaviors as well? The only other requirement for group selection is a statistically valid survival advantage for members of a given group over members of other groups given certain behaviors. The advantage to the group would then become, ipso facto, advantage to all of the group’s members (or at least their relatives).

  10. I guess I’ll be the first bioloogist to chime in, and I should as a I aslo am doing a post-doc that focuses on group selection. Now, I’m certainly no David Sloan Wilson, and if you want a great perspective on this issure check out his posts over on the huffington post. I also read Pinker’s piece and found it equally unconvincing. Perhaps the weakest part is the insistence on focusing on human groups and what Gould called “Adaptive story telling”. As Adam mentions perhaps the most cited example of group selection is the chicken experiments in the 80’s by Muir. http://ps.fass.org/content/75/4/447.short . What Pinker is also missing is that group selection does not need to be a group of closely related individuals, it can be any group that lives together. There is a great treatment of group selection (that’s rather technical) by Samir Okasha. In it he points out that there are two kinds of group seletion. One is that group fitness is simply the sum of the individual fitness of each group member. The other is that fitness of groups is explixcitly considered, e.g. groups beget groups (see this blog post for more info http://historiesofecology.blogspot.ca/2011/09/what-did-wynne-edwards-1962-actually.html). In anycase, Pinker seems a bit more concerned with how group selection impacts the evolution of Humans and uses historical anecdote as argument. That’s something no biologist would use. What’s strange about this whole thing is that there are arugments that are made by biologists that are mathematical, and then people who chime in (like Pinker) who make arguments against group selection that involve no math. That’s not to say that Pinker can’t make arguments for or against group selection, but he’s someone who is using his status as “public/popular intellectual” as a soapbox to make unconvincing arguments about a mathematical topic with no math. On top of that he still doesn’t seem to get the philosophy right. I’m glad to see Andrew call him out for making weak arguments.

    • I don’t know what David Sloan Wilson has posted recently, but he wrote a great monograph (_The Natural Selection of Populations and Communities_) in which he developed mathematical models that generated group selection. The logic was quite consistent with Richard Dawkins’ extended phenotype. I would have thought that this would have persuaded Dawkins similar to the way that Dawkins was initially unconvinced by Amotz Zahavi’s Handicap Principle until Alan Grafen developed a mathematical model that corroborated Zahavi’s proposal. But I get the impression that _Natural Selection of Populations …_ has not sustained much attention within evolutionary biology.

  11. I’ve recently read “Darwin’s Conjecture”, a book by Hodgson & Knudsen, that introduces some order in arguments for and against evolutionary processes at the level of groups, organizations, etc. Their main argument is that one can generalize the concept of the biological gene to a more general concept of replicator, that lives within an interactor (something that interacts with the environment), and formulate evolution solely in terms of these generalized concepts. traditional evolution then becomes a special case of an evolution, etc. They show that such a generalization helps explaining group selection and other seeming peculiarities of evolution that are hard to explain within the more traditional conception of evolution. An interesting book that is relevant to this discussion.

  12. I agree with Edmund. Pinker does not appear to be either advocating or criticising any explicit (i.e. quantitative) model or hypothesis at all – that makes his writing essentially contentless.

    To talk about group selection is to talk about (statistical) population genetics. Sometimes I get the impression that many biologists (let alone non-biologists) have opinions on the matter that are informed by other biological disciplines but not by population genetics. That’s just silly.

  13. Andrew,

    Your ideas about group selection is fairly similar to Elinor Ostrom’s, so you are in good company:

    http://socialevolutionforum.com/2012/03/26/elinor-ostrom-enhancing-the-evolution-of-institutions-for-collective-action/

    I’d recommend the commentary by some of the more mathematical commentators on Pinker’s article.

    Building on Edmund Hart’s reply, there is an interesting split in the commentaries with the more mathematically oriented researchers primarily coming out against Pinker (Boyd & Mathew, Richerson, Gintis, Henrich, Queller, DS Wilson – and also Peter Turchin at the Social Evolution Forum). And the more evolutionary psychology and philosophy crowd coming out for Pinker.

    Boyd & Mathew have an interesting response that is in line with your thinking. It is especially interesting because Pinker uses Sarah Mathew’s work as an example against group selection and she uses it as an argument for it.

    Queller is a pithy and short article from a well-regarded evolutionary (mathematical) theorist that, I think, hits the nail on the head about what this debate is really about.

    Many of these commentators (Boyd, Mathew, Richerson, Gintis, Turchin and Henrich) have contributed to a fairly sophisticated mathematical and increasingly empirical understanding of cultural evolution. AIfll of their commentaries are good, but if I had to pick one, I’d recommend Joe Henrich’s commentary as a comprehensive response to the cultural group selection parts of Pinker’s.

  14. As an outsider, these debates always remind me of the quote about academic arguments being vicious and bitter because the stakes are so low. A lot of this does sound like it’s getting into semantics. Is a group an extension of kin, or are kin a special case of a group? Who cares.

    Also, they’re arguing over a stochastic process (evolutionary selection) that is very heterogeneous over time and across populations. Some of the biggest determinants of the current traits may even be rare one-off events that look nothing like anything else in evolutionary history. The fact that the process of evolution is heterogeneous also means that the very process that we’re arguing reductionist explanations over is not well defined to begin with.

  15. This is a very complex debate, and it takes a long time to even get biologists to grasp it. I teach this material in a PhD evolutionary theory course, and it takes an entire two weeks just to derive the most basic theorem in this literature (Hamilton’s rule), so I have a lot of sympathy for people who are confused and/or skeptical.

    People new to this subject should really read David Queller’s response to Pinker, which appears way down on the page linked in the OP. Queller has spent his career refining and promoting the use of inclusive fitness accounting, so his rejection of Pinker carries a lot of weight here.

    I was a co-signer on the 100+ author rebuttal to E.O. Wilson’s paper, but I’m also a proponent of the theory that (cultural) group selection has been important in human evolution. How can I hold these beliefs at the same time?

    The E.O. Wilson dust up was partly over a narrow interpretation of inclusive fitness reasoning. Both “multilevel selection” and “inclusive fitness” can be strategies for solving or interpreting mathematical models. Wilson et al were criticizing a narrower view of “inclusive fitness,” which is quite a bit more limiting than their “group selection.” But most of us understand “inclusive fitness” to be much broader than they do. So there’s really no debate, aside from experts on the evolution of eusocial insects arguing about various causal hypotheses. That’s interesting, perhaps, but has no impact on debates about “group selection.”

    A related debate is over causal hypotheses for the evolution of human morality (whether genetic or culture or both). Some of these hypotheses go by the label “group selection,” because they suggest that properties of residential and symbolic groups can matter in the fates of individuals. Pinker, like many who learned their evolutionary theory from Richard Dawkins, doesn’t like those sorts of hypotheses. But whether they are right or wrong will have no impact on the philosophical coherence of “group selection.”

    An analogy to fights over Bayesian statistics is informative, perhaps for readers of this blog. All manner of odd, false beliefs about “Bayesian stats” persist, despite the successes of modern Bayesian statistical inference. These beliefs persist partly because of when people learned their stats and partly because of ignorance of the maths or new results. Most scientists don’t really understand either Bayesian nor non-Bayesian stats, so the arguments are just confusing to them. Tell them which button to push, they ask.

    A similar dynamic goes on in evolutionary biology over group selection. Back in the 1960’s, some smart people made some bad arguments that seemed to say that group selection couldn’t happen. But it turned out that group selection as a description of evolutionary dynamics works fine, and that was accepted by mathematical biologists in the 1970’s (actually before, but not commonly). But many people, like Pinker, have never caught up with the mathematical facts. And most biologists don’t really know what “fitness” is anyway, so just tell them which button to push.

    It isn’t that any hypothesis invoking “groups” can be true. It’s rather that just because a hypothesis invokes “groups” doesn’t tell us whether or not it’s true. Similarly, just because I use priors when I do statistics, it neither means that my inferences are wrong or right.

    A committed non-Bayesian might argue then that we should just keep being non-Bayesian, because it works. Well, sometimes (usually IMHO) being Bayesian is easier. Likewise, many biologists think, if the two perspectives are usually the same, we should just ignore group selection. Well, sometimes thinking in terms of group selection is easier than thinking in terms of inclusive fitness.

    • Thanks for this comment, Richard – a fascinating and highly accessible take on the debate. (I like the analogy to the Bayesian/frequentist divide).

  16. Your example of the transition from hunter-gatherer to agricultural societies is in line with one of Pinker’s points – the ad hoc attachment of the group selection label to anything involving groups. The transition to agriculture was beneficial to the individuals involved – don’t confuse smaller stature and worse teeth at the beginning of the transition with evolutionary outcomes. Agriculturalists were able to reduce the space between children, accumulate resources and generally increase their fitness (that is, fitness as biologists define it). Those within an agricultural “group” would not have been able to increase their fitness by defecting and shifting back to a hunter-gatherer lifestyle. It was in their self-interest to stick with the agricultural way of life.

  17. Jason,

    It gets tricky when you talk about a culturally transmitted practice like agriculture. The archaeological evidence I’ve read, and the historical record, all say pretty clearly that agriculture did not spread across the globe just because individuals that practiced it had more children, like Inuit genetic adaptations to cold environments. Instead, societies that had agriculture also had new means to displace, enslave, assimilate and exterminate forager groups. In those terms, it seems to be a clear case of cultural group selection, because the outcome derives from the properties and concerted actions of groups involved.

    • They had the means to displace hunter-gatherers because they could support a much larger population per unit of space.

      • Wonks,

        That’s true, but it’s not a sufficient explanation (notice I said “just because”). It would be if farmers just outbred nonfarmers, or convinced them to switch to farming. But that’s not what happened – the rise of agriculture has historically been a story of intergroup conflict and conquest.

    • A group of individuals acting in their own self interest (whether individually or in concert with others) and displacing others is not group selection. Again, that is Pinker’s point – group selection being attached to anything involving groups.

      • Jason,

        Of course, a group of cells acting together as an organism is not individual-level selection, by your argument, because they are acting in their own self interest. Playing that game gets us no further than where we started.

        Books on evolutionary theory (not pop sci books by Dawkins and Pinker) are pretty clear about what “group selection” means, and it supports my above statements. If you wish to see, I encourage you to read Sean Rice, “Evolutionary Theory”, Sober and Wilson, “Unto Others”. Both are excellent introductions to the mathematics behind group selection.

        • Reading Unto Others was one of the turning points for me towards deciding that multilevel selection was the less insightful way of looking at the issue. After all, cells in an organism are just maximally related kin.

          So let’s break down this agriculture argument. How is an agriculturalist being weakly altruistic towards his group of fellow agriculturalists when he decides to stay in agriculture rather than taking up a hunter-gatherer lifestyle?

          • Jason,

            It’s a common mistake to think group selection means altruism (this is discussed in Unto Others…). Group selection just means there’s a causal arrow coming from group properties, in this case the capacities for organized violence by agricultural groups. In fact, one of the best ways groups can elicit group-functional behavior is by making it the “selfish” best response through rewards and punishments, so its no surprise group selected institutions look this way. The canonical models of cultural group selection don’t even discuss altruism at all, as they are coordination problems. If you prefer economic idioms, Sam Bowles goes over this in his micro text.

          • @John Li

            Andrew wrote “A famous example is the transition from hunter-gatherer to agricultural societies; it’s said that the latter dominate economically even while being worse for most of the individuals involved.” That sounds like altruism to me.

          • Jason:

            No, when referring to hunter-gatherer and agricultural societies, I was not talking about altruism. I was saying that I’ve heard that latter dominate economically even while being worse for most of the individuals involved. This is a statement about the organization of society, not altruism. Here’s a quick analogy (probably flawed in some ways but maybe will get the point across): It’s more pleasant to paddle a kayak than to be a galley slave. But the galley will go faster.

          • Andrew,

            Groups in a multilevel selection framework are generally based on interaction between group members (why bother categorising them as a group if there is no interaction). Is there any transfer of fitness to other group members and what is the relative size of this transfer compared to their own fitness gain? In the galley example, this might be what is the relative gain in speed of all the other galley passengers from your rowing effort?

            Obviously, if your own gain is larger than the transferred fitness, we don’t need altruism. That is, if your own rowing effort adds more speed than it does to other group members, you gain fitness relative to the other passengers. But if you have other passengers and assume any cost to rowing, we know this can’t be the case as they will each get the additional speed you generate for no effort. Hence, we have at a minimum weak altruism. Of course, David Sloan Wilson defined “weak altruism” – the term I used in my earlier comment – based on the specific transfer within the group, even if across a broader set of the population it is seen to confer a fitness benefit. The relevant comparison for altruistic behaviour in a multilevel selection framework does not also include the kayak – weak altruism focuses on the galley.

            A cost of this approach, as Pinker pointed out, that a multilevel selection framework does not provide insight into the formation of the group in the first place (why pick the galley?). Pinker wrote: “[A] mathematical model that submerges the psychological forces that keep different “groups” together (such as genetic relatedness or mutual sensitivity to altruism), and requires theorists to dig them out from under the equations, is hardly a perspicuous way to analyze sociality.”

            If we apply this to your agricultural example, altruism is again a question of transfers between agriculturalists – if that is how we define groups. Unless we want to define “worse for most of the individuals involved” as being in dimensions that don’t matter in a biological sense (pleasure?), or want to rule out transfers between agriculturalists (in which case, why are we grouping them?) we’re going to find ourselves in the same altruistic territory as we did in the galley example above.

  18. It’s a little bit harsh to call Dawkins a popularizer – sure, he did a lot of that, but he did (and still might do, for all I know) good scientific work. Similarly Wilson has done a bunch of popularizing. Wilson’s probably done a little more science, and Dawkins a little more popularizing, but I don’t think you should dichotomize them like that.

    • Price’s Equation can be used to help think about inclusive fitness (Hamilton famously re-derived Hamilton’s rule from it). Or it can be used to help think about multilevel selection – by partitioning fitness into within-group and between-group components.

      The debate over the utility of the Price Equation is somewhat outside of this debate since memberrs of both camps use it to some extent.

  19. ” The study of history is hugely important, and what history is about (to me) is nailing down the facts. History is the ultimate descriptive social science. ”

    YES!

    I just read the book by Rodney Stark: “Cities of God: The Real Story of How Christianity Became an Urban Movement and Conquered Rome”, which should interest readers of this blog!

    http://www.amazon.com/Cities-God-Christianity-Movement-Conquered/dp/0061349887/ref=sr_1_7?s=books&ie=UTF8&qid=1342552560&sr=1-7&keywords=Rodney+stark

    The first ever history book I have read featuring a statistical appendix! Rodney Starks’s program is to collect
    statistical data throwing light on the rise of Christianity, and use it to test hypothesis about that. very convincing.
    Statistics are rather elementary and traditional (correlations, p-values, 5%-cutoff), but of course the real contribution is to actually collect data about things that happened almost 2000 years ago.

Comments are closed.