Wonderful post as usual. The arguments apply to any departure from opportunism of course, not just altruism. Consider, for instance, Schelling in The Strategy of Conflict:
‘How can one commit himself in advance to an act that he would in fact prefer not to carry out in the event, in order that his commitment may deter the other party? ... perhaps he can pretend a revenge motivation so strong as to overcome the
prospect of self-damage; but this option is probably most readily available to the
truly vengeful’ (Schelling, 1960, p. 36).
In bargaining situations, ‘the sophisticated negotiator may find it difficult to seem as obstinate as a truly obstinate man’ (Schelling, 1960, p. 22).
When faced with a threat, self-interested behavior may be far less materially rewarding than ‘genuine ignorance, obstinacy or simple disbelief, since it may be more convincing to the prospective threatener’ (Schelling, 1960, p. 38)
Schelling, Hirschleifer and Frank (Passions within Reason) motivated some of my own work when I was a graduate student, the above quotes may all be found here:
Your focus here is on individual adaptations to changes in costs and benefits but I think population level adaptations are more likely. People don't discard altruism so much as those genuinely exhibiting it becomes less prevalent when cost rise.
So interesting Rajiv - thanks for sharing. I'm a big fan of Schelling and Frank. Just glanced through your articles and they seem extremely relevant to things I am thinking about and working on at the moment.
There's an amazing paper by Robert Trivers on the evolution of a sense of fairness, hardly ever cited, I will try to find a reference. In general I think reciprocity and fairness norms are much more robust than pure altruism because less exploitable and hence less costly.
I totally agree. I did not know your papers Rajiv, otherwise I would have quoted them in "Optimally Irrational". About the genuine feeling that moral norms are objective and unconditional, I wrote: "the feeling of objectivity creates constraints on moral behaviour, and, as pointed out by Schelling (1960/1980), the restriction of possible actions may give an advantage in social interactions by allowing one to commit to some course of action. For example, you may trust people more in business transactions if you know that they believe that unfair deals are forbidden by some objective rule (therefore removing for this person the option to engage in unfair behaviour)."
A thought-provoking and extremely well written essay.
I have never been able to make sense of kin selection as a “process in which organisms spread their genes by aiding relatives who share copies of those genes”. In order to make sense of that, you need a definition of the word “share” which excludes the fact that all humans share 99% of their genome with each other. Do you know of one?
A more precise definition would focus on relative genetic similarity rather than absolute sharing. Kin selection operates on genes that are identical by descent - meaning copies inherited from a recent common ancestor. This concept accounts for the higher probability of shared rare alleles between close relatives, beyond the baseline similarity all humans share. So rather than "sharing genes," it's about the increased likelihood of having the same variants of specific genes compared to the general population. (I agree it can be confusing, though!)
Nicely explained! I have another theory that explains why it is completely rational for us to be helpful to strangers, at least in our highly connected world. It is based on some simple ideas borrowed from proven investing strategies.
Don’t we just enjoy helping other people? When we provide help, it makes the other person happy which then creates a positive emotional response in us.
You say you are skeptical of group selection, but Pinker’s post that you link to isn’t a great critique – there are major problems with many common versions of group selection, but that doesn’t imply that there is no correct version.
The strongest critique of the genetic group selection is (which Pinker adverts to) is that you can’t sustain enough variation between groups in the face of even modest gene flow. That is a very strong argument, but it is ultimately an empirical question as to whether the ancestral environment satisfied the low-flow requirement. Bowles & Gintis, A Cooperative Species Ch 8 give a model tuned to late Pleistocene that suggests genetic group selection may have played a role in the evolution of altruism. I don’t know enough to have a strong opinion about their model and I do think the gene flow argument is strong enough that it is worth being skeptical, but still, the argument can’t be dismissed without addressing specific models.
Also, while Pinker cites some cultural group selection sources, his discussion doesn’t address the point that cultural group selection models overcome the low trait flow requirement by the premise that people are predisposed to imitate, so entrants (at least in low numbers) will adopt the dominant norms.
He also has some not very persuasive general arguments ie that “so-called group selection. . . is not a precise implementation of the theory of natural selection [but] Instead it is a loose metaphor.” That is no doubt true of some models, but not of the best models. I am not an expert in this area, but Boyd & Richerson’s models strike me as being as rigorous as any gene centered model. Pinker dismisses them simply by conflating multilevel selection with group selection arguments generally. I haven’t read their work in a while, but my impression was that they had indeed developed a rigorous correct general model and Pinker’s general comments did not at all persuade me otherwise. Of course, there is still the question of whether the real world variables are such as to allow higher level selection to play a significant role, but that is different from saying that group selection theories are just a bunch of hand waving.
He also argues group selection is wrong because evolution requires random variation and “The mutations are not random. Conquerors, leaders, elites, visionaries, social entrepreneurs, and other innovators use their highly nonrandom brains to figure out tactics and institutions and norms and beliefs that are intelligently designed in response to a felt need.” This is semantics - ie by definition natural selection requires totally random variation (which is not even true genetically), or a fallacy. Pinker seems to be implying that if individual learning plays any role in generating variation, then selection plays no role in determining which persist. That is clearly false. The cultural group selection argument, as I understand it, is that individual learning results in choices that are generally better than random, with the optimality decreasing as the feedback becomes more indirect. So, in figuring out how to make an arrow, the feedback is fairly immediate and even then pure individual learning does not result in optimality. The feedback re the best social norms for a stable and prosperous society is even more indirect. It is not plausible that leaders like Donald Trump and Joe Biden will figure out the best response to a felt need, notwithstanding that their brains are “highly nonrandom.” Variation thrown up by individual brains is at best quasi-random, which is enough to allow for selection to play a significant role. In the case of high level social norms, I expect the product of individual brains is effectively random.
Your second link is to The Pleistocene Social Contract. Not sure if you are recommending it, or giving it as an example of bad group selection thinking.
Thanks Norman. Excellent points. I'm more sympathetic to Pinker's critique than you are. Because a critique of group selection will be central in my forthcoming book "Why it's OK to be cynical", I will be blogging about this specific topic a lot more in a few months. Your comments here give me a lot to think about.
- Re. whether the conditions for group selection were met: the Sterelny book, which I highly recommend, gives some reasons for thinking they weren't met, and more generally critiques the Bowles and Gintis in what is (in my view) a very persuasive way.
- For me, a big problem for group selectionist theories is that there are more plausible alternative explanations for the phenomena they set out to explain (e.g. human prosociality, costly punishment, groupishness, etc.). And I think Pinker is right - and he is not the first person to make this point - that group selectionist theories of human altruism generate incorrect predictions about how human moral and social psychology actually work.
- My feelings about cultural group selection are complicated. Basically I think that whole literature greatly understates the importance of human agency in shaping and driving cultural change, and that it doesn't provide mechanistic understanding of the actual causal processes underlying cultural evolution. I also think almost the whole cultural evolution literature greatly underestimates the sophistication and vigilance underlying human social learning and communication, which has very little to do with "imitation" or "copying". (Here I am very influenced by the work of, e.g, Dan Sperber, Olivier Morin, and Hugo Mercier).
I realise these remarks will be too brief to be unsatisfying but I will look forward to your comments - and critiques - when I write up my thoughts in more depth.
I’m afraid you caught me on a day when I don’t feel like working at my actual job.
At one level it seems to me that cultural group selection is clearly true. The Yamnaya swept the world extremely rapidly as a result of what was very probably cultural innovation, such as a mobile pastoral lifestyle with wheeled wagons. I don’t see an individual learning argument for that spread. I suppose genetic can’t be ruled out, but that would be a helluva lucky mutation. The more difficult question is whether cultural group selection contributed to ultrasociality.
1) I agree that the gene flow point is strong and at the end of the day I probably don't buy Bowles & Gintis. But I can't dismiss it. Same w Henrich v HBD Chick on why western societies are WEIRD. Anyway, I'll read Sterelny – I'd seen it before but wasn't sure whether it was worth $80.
2a) A more plausible theory would be great - one reason I like Wrangham's self-domestication theory is that it is an alternative explanation for ultrasociality. I'll look forward to your post.
2b) I thought Pinker's arguments were strawmannish. He was positing results from a very strong form of group selection that dominates everything else. No one is arguing for that. Everyone acknowledges that there are strong individual level selection forces that go against group selection and so no doubt we will see a complex outcome trading off these factors. That will be hard to distinguish from a situation where putatively pro-social behaviour is actually strictly self-interested. But that's not surprising – if ultrasociality gave huge group level advantages, we'd probably see it in more species. The question is whether it helped push us over the edge into ultrasociality.
The bigger problem for me is that cultural group selection theories seem to require a fairly strong propensity to imitate in order to overcome the individual selection pressures. It’s the equivalent of the low-flow requirement in gene models. In some spheres, eg making arrows, baking, food choices, that is arguably true. But at the norm level, it doesn't seem true. If kids strongly imitated their parents, a lot of modern politics would be different. Maybe the answer is along Pinker’s lines after all – there is strong persistence of norms that are directly individually beneficial eg pastoral steppe lifestyle, but weaker persistence otherwise. (But how can we account for intergenerational political difference, and persistence of observationally opaque yet adaptive food preparation customs, like ash in taro, or taboo on pregnant women eating certain fish.) Now, you can do with lower propensity to imitate if there is strong selection pressure eg non-stop tribal warfare with extermination as an outcome. But then you would expect the norms to break down fairly quickly when the pressure relaxed. Maybe that is descriptively plausible.
3) Lack of understanding of mechanisms is different from rejecting cultural group selection. Mechanisms provide a refinement. Traditional natural selection is so powerful precisely because it allows strong predictions without understanding mechanisms, but better understanding of mechanisms does allow better predictions eg linkage disequilibrium. Anyway, I think Cecilia Heyes, Cognitive Gadgets, is an extremely thought provoking take on possible mechanisms.
Similarly, I do think human agency matters, but I think it is refinement. Human ingenuity throws up the cultural variants that are then selected on. In general, my guess is that there is plenty of room for individual agency to play a long term role even within an cultural evolutionary model, as I suspect that some important norms are effectively coordination games so there is a lot of room for path dependence.
You say “the whole cultural evolution literature greatly underestimates the sophistication and vigilance underlying human social learning and communication, which has very little to do with "imitation" or "copying".” Can you elaborate a bit? Are you saying that imitation plays little role in culture? To me, it seems clearly true that there are important norms, eg respect for property, or how to prepare toxic foods, that are followed because of cultural imitation, and not because anyone has figured out that they are right.
Thanks Norman. Good points. Very briefly, I think:
1. The sense in which cultural group selection is "clearly true" is the sense in which CGS is not doing any interesting explanatory work. This is one of Pinker's points, and I agree. For example, in theories of state formation, it is clear that once states first emerged, state formation accelerated because states' far superior military capabilities meant that pre-state forms of social organisation were easily outcompeted, and many people therefore had strong incentives to build states. One can call that "cultural group selection" but describing it that quasi-evolutionary way has no explanatory advantage I can see and in fact misdiagnoses the causal factors actually driving the process.
2. I don't dismiss group selection with certainty. I just don't find it plausible.
3. I see your point that Pinker's arguments seem strawmannish, but I don't agree. People make big claims for group selection and its connection to altruism, claims which (in my view) are at odds with facts about how we know human beings behave.
4. Re. your points (2) and (3), the authors I cited - Mercier, Sperber, and I would also add Manvir Singh's stuff - make the points much better than I could about the problems with cultural evolution/dual inheritance theory.
Anyway, I realise all of that will seem very unsatisfying - but just some brief responses for now. I will be interested in hearing your thoughts when I publish more developed outlines of my views on these topics in the future.
Looking forward to the book and excellent essay. Your points about the differing levels of altruism in individualistic vs communitarian societies were especially interesting. I don't know if you are familiar with Bruce Schneier, he is a cryptographer, and he's written a lot about the need for security systems and tiered privacy as group size exceeds certain thresholds, specifically because of something like your zone-of-interest concept. However, the difference between your point and his point is that he sees it as a small-but-active group of highly cynical people. In his view, the trust in any entity is two fold one part fairness (safe value), and one part security (Limit and Penalize Defection). But they are the same coin. Not sure if you differ, but I think you might find some of his POVs interesting. Here is a good flavor of how he thinks about these issues. https://www.schneier.com/blog/archives/2023/12/ai-and-trust.html
Great, insightful piece as usual. However, reputation management in other social animals including primates should not be underestimated. Also, I don’t see such a stark contrast between ingroup/outgroup and coalition formation. In chimpanzees between group competition is more lethal than within and this is the same for human small scale societies. What changed, recently, is that the groups became bigger and therefore safer (larger average distance to the borders of territories) and thus ingroup became more prominent including its more peaceful social dynamics.
Is the argument that selective reputation management (partially) explains the lack of ubiquity in self-sacrifice (i.e. niceness)? That we don't act (or give) on virtue consistently and the variability in action is defined by an individual seeking evolutionary advantage?
Reciprocity and status definitely have theit place as a social incentives to self-sacrifice, but there are systems and machines that severely infere with baseline cognition. Can the actions, sincere or otherwise, be trusted as free from influence and precise in their assessment of gain? Could that variability in self-sacrifice stem from an amalgam of faulty assessments of virtue and influence that disguises self-harm as self-interest? Do deficits in available resources make self-sacrifice more detrimental than self-interest?
Are we trying to figure out whether someone is nice if the slaughterhouse still spills blood for a polite exchange, some cup holder change, a few minutes of time, and an fried glob of calories? And then asking if we're cynical for judging the person?
Does respect (in house or outsourced) change why people pay for beheading? Who would we ask for why the blood is spilt? Would Niccolo know? Or would the nugget lords?
Focussing in from your excellent timeless, universal observations - about human nature and evolution - to my more specific ones about their socio-psychological manifestations in our current Western condition.....
"...there is Love in the Abstract: Love of complete strangers. Western lefty liberals love almost everyone – all 8 billion of them. With exceptions of course. Exceptions might be white people or males or both. You cannot, of course, be of any actual, realistic use to those abstract objects of your Love but along the way you can – by unfavourable comparison - really diss the more proximate objects of your displeasure. For sentimentalising one party often entails demonising another. Loving strangers is a great way of working through your resentments against people of your own kind but a bit more successful than you. Which brings us to Hate in the Abstract and in particular Theological Hate: Some people (terrorists for example) hate whole great categories such as infidels or women who don’t obey them slavishly behind their burkhas. Or Jews. And they not only hate them but think that they deserve to be killed. But the bleeding heart liberals among us love those terrorists too. Other people (mostly white ones bizarrely) hate everyone who is deemed to have been the beneficiary of White Privilege."
Wonderful post as usual. The arguments apply to any departure from opportunism of course, not just altruism. Consider, for instance, Schelling in The Strategy of Conflict:
‘How can one commit himself in advance to an act that he would in fact prefer not to carry out in the event, in order that his commitment may deter the other party? ... perhaps he can pretend a revenge motivation so strong as to overcome the
prospect of self-damage; but this option is probably most readily available to the
truly vengeful’ (Schelling, 1960, p. 36).
In bargaining situations, ‘the sophisticated negotiator may find it difficult to seem as obstinate as a truly obstinate man’ (Schelling, 1960, p. 22).
When faced with a threat, self-interested behavior may be far less materially rewarding than ‘genuine ignorance, obstinacy or simple disbelief, since it may be more convincing to the prospective threatener’ (Schelling, 1960, p. 38)
Schelling, Hirschleifer and Frank (Passions within Reason) motivated some of my own work when I was a graduate student, the above quotes may all be found here:
www.sciencedirect.com/science/article/pii/0167268195000534
Much later, returned to this theme here:
www.sciencedirect.com/science/article/abs/pii/S0899825603000290
Your focus here is on individual adaptations to changes in costs and benefits but I think population level adaptations are more likely. People don't discard altruism so much as those genuinely exhibiting it becomes less prevalent when cost rise.
Thanks again!
So interesting Rajiv - thanks for sharing. I'm a big fan of Schelling and Frank. Just glanced through your articles and they seem extremely relevant to things I am thinking about and working on at the moment.
There's an amazing paper by Robert Trivers on the evolution of a sense of fairness, hardly ever cited, I will try to find a reference. In general I think reciprocity and fairness norms are much more robust than pure altruism because less exploitable and hence less costly.
I totally agree. I did not know your papers Rajiv, otherwise I would have quoted them in "Optimally Irrational". About the genuine feeling that moral norms are objective and unconditional, I wrote: "the feeling of objectivity creates constraints on moral behaviour, and, as pointed out by Schelling (1960/1980), the restriction of possible actions may give an advantage in social interactions by allowing one to commit to some course of action. For example, you may trust people more in business transactions if you know that they believe that unfair deals are forbidden by some objective rule (therefore removing for this person the option to engage in unfair behaviour)."
Excellent post Dan, I look forward to your book.
A thought-provoking and extremely well written essay.
I have never been able to make sense of kin selection as a “process in which organisms spread their genes by aiding relatives who share copies of those genes”. In order to make sense of that, you need a definition of the word “share” which excludes the fact that all humans share 99% of their genome with each other. Do you know of one?
A more precise definition would focus on relative genetic similarity rather than absolute sharing. Kin selection operates on genes that are identical by descent - meaning copies inherited from a recent common ancestor. This concept accounts for the higher probability of shared rare alleles between close relatives, beyond the baseline similarity all humans share. So rather than "sharing genes," it's about the increased likelihood of having the same variants of specific genes compared to the general population. (I agree it can be confusing, though!)
Nicely explained! I have another theory that explains why it is completely rational for us to be helpful to strangers, at least in our highly connected world. It is based on some simple ideas borrowed from proven investing strategies.
https://medium.com/@vinbhalerao/why-diversifying-your-life-leads-to-me-and-we-becoming-aligned-f0787419e061
Don’t we just enjoy helping other people? When we provide help, it makes the other person happy which then creates a positive emotional response in us.
Of course. (But the question is why we enjoy it, and under what circumstances).
“If your enemy is hungry, feed him; If he is thirsty, give him a drink; For in so doing you will heap coals of fire on his head.”
You say you are skeptical of group selection, but Pinker’s post that you link to isn’t a great critique – there are major problems with many common versions of group selection, but that doesn’t imply that there is no correct version.
The strongest critique of the genetic group selection is (which Pinker adverts to) is that you can’t sustain enough variation between groups in the face of even modest gene flow. That is a very strong argument, but it is ultimately an empirical question as to whether the ancestral environment satisfied the low-flow requirement. Bowles & Gintis, A Cooperative Species Ch 8 give a model tuned to late Pleistocene that suggests genetic group selection may have played a role in the evolution of altruism. I don’t know enough to have a strong opinion about their model and I do think the gene flow argument is strong enough that it is worth being skeptical, but still, the argument can’t be dismissed without addressing specific models.
Also, while Pinker cites some cultural group selection sources, his discussion doesn’t address the point that cultural group selection models overcome the low trait flow requirement by the premise that people are predisposed to imitate, so entrants (at least in low numbers) will adopt the dominant norms.
He also has some not very persuasive general arguments ie that “so-called group selection. . . is not a precise implementation of the theory of natural selection [but] Instead it is a loose metaphor.” That is no doubt true of some models, but not of the best models. I am not an expert in this area, but Boyd & Richerson’s models strike me as being as rigorous as any gene centered model. Pinker dismisses them simply by conflating multilevel selection with group selection arguments generally. I haven’t read their work in a while, but my impression was that they had indeed developed a rigorous correct general model and Pinker’s general comments did not at all persuade me otherwise. Of course, there is still the question of whether the real world variables are such as to allow higher level selection to play a significant role, but that is different from saying that group selection theories are just a bunch of hand waving.
He also argues group selection is wrong because evolution requires random variation and “The mutations are not random. Conquerors, leaders, elites, visionaries, social entrepreneurs, and other innovators use their highly nonrandom brains to figure out tactics and institutions and norms and beliefs that are intelligently designed in response to a felt need.” This is semantics - ie by definition natural selection requires totally random variation (which is not even true genetically), or a fallacy. Pinker seems to be implying that if individual learning plays any role in generating variation, then selection plays no role in determining which persist. That is clearly false. The cultural group selection argument, as I understand it, is that individual learning results in choices that are generally better than random, with the optimality decreasing as the feedback becomes more indirect. So, in figuring out how to make an arrow, the feedback is fairly immediate and even then pure individual learning does not result in optimality. The feedback re the best social norms for a stable and prosperous society is even more indirect. It is not plausible that leaders like Donald Trump and Joe Biden will figure out the best response to a felt need, notwithstanding that their brains are “highly nonrandom.” Variation thrown up by individual brains is at best quasi-random, which is enough to allow for selection to play a significant role. In the case of high level social norms, I expect the product of individual brains is effectively random.
Your second link is to The Pleistocene Social Contract. Not sure if you are recommending it, or giving it as an example of bad group selection thinking.
Thanks Norman. Excellent points. I'm more sympathetic to Pinker's critique than you are. Because a critique of group selection will be central in my forthcoming book "Why it's OK to be cynical", I will be blogging about this specific topic a lot more in a few months. Your comments here give me a lot to think about.
- Re. whether the conditions for group selection were met: the Sterelny book, which I highly recommend, gives some reasons for thinking they weren't met, and more generally critiques the Bowles and Gintis in what is (in my view) a very persuasive way.
- For me, a big problem for group selectionist theories is that there are more plausible alternative explanations for the phenomena they set out to explain (e.g. human prosociality, costly punishment, groupishness, etc.). And I think Pinker is right - and he is not the first person to make this point - that group selectionist theories of human altruism generate incorrect predictions about how human moral and social psychology actually work.
- My feelings about cultural group selection are complicated. Basically I think that whole literature greatly understates the importance of human agency in shaping and driving cultural change, and that it doesn't provide mechanistic understanding of the actual causal processes underlying cultural evolution. I also think almost the whole cultural evolution literature greatly underestimates the sophistication and vigilance underlying human social learning and communication, which has very little to do with "imitation" or "copying". (Here I am very influenced by the work of, e.g, Dan Sperber, Olivier Morin, and Hugo Mercier).
I realise these remarks will be too brief to be unsatisfying but I will look forward to your comments - and critiques - when I write up my thoughts in more depth.
Just read Sterenly, The Pleistocene Social Contract. Absolutely magnificent.
Yeah it's a great read! Very dense but worth it.
I’m afraid you caught me on a day when I don’t feel like working at my actual job.
At one level it seems to me that cultural group selection is clearly true. The Yamnaya swept the world extremely rapidly as a result of what was very probably cultural innovation, such as a mobile pastoral lifestyle with wheeled wagons. I don’t see an individual learning argument for that spread. I suppose genetic can’t be ruled out, but that would be a helluva lucky mutation. The more difficult question is whether cultural group selection contributed to ultrasociality.
1) I agree that the gene flow point is strong and at the end of the day I probably don't buy Bowles & Gintis. But I can't dismiss it. Same w Henrich v HBD Chick on why western societies are WEIRD. Anyway, I'll read Sterelny – I'd seen it before but wasn't sure whether it was worth $80.
2a) A more plausible theory would be great - one reason I like Wrangham's self-domestication theory is that it is an alternative explanation for ultrasociality. I'll look forward to your post.
2b) I thought Pinker's arguments were strawmannish. He was positing results from a very strong form of group selection that dominates everything else. No one is arguing for that. Everyone acknowledges that there are strong individual level selection forces that go against group selection and so no doubt we will see a complex outcome trading off these factors. That will be hard to distinguish from a situation where putatively pro-social behaviour is actually strictly self-interested. But that's not surprising – if ultrasociality gave huge group level advantages, we'd probably see it in more species. The question is whether it helped push us over the edge into ultrasociality.
The bigger problem for me is that cultural group selection theories seem to require a fairly strong propensity to imitate in order to overcome the individual selection pressures. It’s the equivalent of the low-flow requirement in gene models. In some spheres, eg making arrows, baking, food choices, that is arguably true. But at the norm level, it doesn't seem true. If kids strongly imitated their parents, a lot of modern politics would be different. Maybe the answer is along Pinker’s lines after all – there is strong persistence of norms that are directly individually beneficial eg pastoral steppe lifestyle, but weaker persistence otherwise. (But how can we account for intergenerational political difference, and persistence of observationally opaque yet adaptive food preparation customs, like ash in taro, or taboo on pregnant women eating certain fish.) Now, you can do with lower propensity to imitate if there is strong selection pressure eg non-stop tribal warfare with extermination as an outcome. But then you would expect the norms to break down fairly quickly when the pressure relaxed. Maybe that is descriptively plausible.
3) Lack of understanding of mechanisms is different from rejecting cultural group selection. Mechanisms provide a refinement. Traditional natural selection is so powerful precisely because it allows strong predictions without understanding mechanisms, but better understanding of mechanisms does allow better predictions eg linkage disequilibrium. Anyway, I think Cecilia Heyes, Cognitive Gadgets, is an extremely thought provoking take on possible mechanisms.
Similarly, I do think human agency matters, but I think it is refinement. Human ingenuity throws up the cultural variants that are then selected on. In general, my guess is that there is plenty of room for individual agency to play a long term role even within an cultural evolutionary model, as I suspect that some important norms are effectively coordination games so there is a lot of room for path dependence.
You say “the whole cultural evolution literature greatly underestimates the sophistication and vigilance underlying human social learning and communication, which has very little to do with "imitation" or "copying".” Can you elaborate a bit? Are you saying that imitation plays little role in culture? To me, it seems clearly true that there are important norms, eg respect for property, or how to prepare toxic foods, that are followed because of cultural imitation, and not because anyone has figured out that they are right.
Thanks Norman. Good points. Very briefly, I think:
1. The sense in which cultural group selection is "clearly true" is the sense in which CGS is not doing any interesting explanatory work. This is one of Pinker's points, and I agree. For example, in theories of state formation, it is clear that once states first emerged, state formation accelerated because states' far superior military capabilities meant that pre-state forms of social organisation were easily outcompeted, and many people therefore had strong incentives to build states. One can call that "cultural group selection" but describing it that quasi-evolutionary way has no explanatory advantage I can see and in fact misdiagnoses the causal factors actually driving the process.
2. I don't dismiss group selection with certainty. I just don't find it plausible.
3. I see your point that Pinker's arguments seem strawmannish, but I don't agree. People make big claims for group selection and its connection to altruism, claims which (in my view) are at odds with facts about how we know human beings behave.
4. Re. your points (2) and (3), the authors I cited - Mercier, Sperber, and I would also add Manvir Singh's stuff - make the points much better than I could about the problems with cultural evolution/dual inheritance theory.
Anyway, I realise all of that will seem very unsatisfying - but just some brief responses for now. I will be interested in hearing your thoughts when I publish more developed outlines of my views on these topics in the future.
Looking forward to the book and excellent essay. Your points about the differing levels of altruism in individualistic vs communitarian societies were especially interesting. I don't know if you are familiar with Bruce Schneier, he is a cryptographer, and he's written a lot about the need for security systems and tiered privacy as group size exceeds certain thresholds, specifically because of something like your zone-of-interest concept. However, the difference between your point and his point is that he sees it as a small-but-active group of highly cynical people. In his view, the trust in any entity is two fold one part fairness (safe value), and one part security (Limit and Penalize Defection). But they are the same coin. Not sure if you differ, but I think you might find some of his POVs interesting. Here is a good flavor of how he thinks about these issues. https://www.schneier.com/blog/archives/2023/12/ai-and-trust.html
Thanks for sharing. That sounds really interesting - I'll check it out.
Great, insightful piece as usual. However, reputation management in other social animals including primates should not be underestimated. Also, I don’t see such a stark contrast between ingroup/outgroup and coalition formation. In chimpanzees between group competition is more lethal than within and this is the same for human small scale societies. What changed, recently, is that the groups became bigger and therefore safer (larger average distance to the borders of territories) and thus ingroup became more prominent including its more peaceful social dynamics.
Thanks, Jan - good points. You know anything good to read on reputation management in other social animals?
Reputation in nonhuman animals:
https://link.springer.com/referenceworkentry/10.1007/978-3-319-47829-6_1884-2
Kindness as a result of group size increase:
https://peterturchin.com/book/ultrasociety/
Cheers
Fascinating essay on this topic. Led to some fresh insights after pondering such matters for years. Thank you.
Thanks John!
Quick question. Do you feel Evolution is defined by just Darwin?
No but I think evolution by natural selection is the only process responsible for the emergence of functional design.
Is the argument that selective reputation management (partially) explains the lack of ubiquity in self-sacrifice (i.e. niceness)? That we don't act (or give) on virtue consistently and the variability in action is defined by an individual seeking evolutionary advantage?
Reciprocity and status definitely have theit place as a social incentives to self-sacrifice, but there are systems and machines that severely infere with baseline cognition. Can the actions, sincere or otherwise, be trusted as free from influence and precise in their assessment of gain? Could that variability in self-sacrifice stem from an amalgam of faulty assessments of virtue and influence that disguises self-harm as self-interest? Do deficits in available resources make self-sacrifice more detrimental than self-interest?
Are we trying to figure out whether someone is nice if the slaughterhouse still spills blood for a polite exchange, some cup holder change, a few minutes of time, and an fried glob of calories? And then asking if we're cynical for judging the person?
Does respect (in house or outsourced) change why people pay for beheading? Who would we ask for why the blood is spilt? Would Niccolo know? Or would the nugget lords?
Focussing in from your excellent timeless, universal observations - about human nature and evolution - to my more specific ones about their socio-psychological manifestations in our current Western condition.....
"...there is Love in the Abstract: Love of complete strangers. Western lefty liberals love almost everyone – all 8 billion of them. With exceptions of course. Exceptions might be white people or males or both. You cannot, of course, be of any actual, realistic use to those abstract objects of your Love but along the way you can – by unfavourable comparison - really diss the more proximate objects of your displeasure. For sentimentalising one party often entails demonising another. Loving strangers is a great way of working through your resentments against people of your own kind but a bit more successful than you. Which brings us to Hate in the Abstract and in particular Theological Hate: Some people (terrorists for example) hate whole great categories such as infidels or women who don’t obey them slavishly behind their burkhas. Or Jews. And they not only hate them but think that they deserve to be killed. But the bleeding heart liberals among us love those terrorists too. Other people (mostly white ones bizarrely) hate everyone who is deemed to have been the beneficiary of White Privilege."
https://grahamcunningham.substack.com/p/love-of-the-people