November 8, 2010

Group Selection, Again

The "Level of Selection" debate in evolutionary theory has been going on for decades, with the advocates of lower levels (e.g., the Selfish Gene) usually drubbing, through better math, the advocates of significant amounts of selection at the group level. 

Personally, I've always found the arguments for group selection plausible (although that probably just shows how bad at math I am). Think of two tribes in prehistoric northern Europe squaring off militarily along the river that divides their traditional territories. They each covet the others' lands, but both groups have been afraid to attempt a river crossing under archery fire. One tribe is mostly lactose tolerant, the other tribe is mostly not. The leaders of the lactose tolerant tribe decide to send their cavalry, without their grain wagons, on a multiday sweep up the river in an attempt to turn the enemies' flank. They can drink mare's milk on their expedition. The lactose intolerant tribe's cavalry attempts to follow them, but gets slowed down by their food wagons. The milk-drinking cavalry outdistances the non-milk drinking cavalry to the point where they can get a full day free to cross the river.

In the next generation, the women of the lactose intolerant former tribe have many children who are lactose tolerant. The men don't have many children at all.

Now, that seems to me like group selection with a bang. Membership in the group -- outflankers or outflankees -- determines much about the fate of your genes in future generations.

But, it's often been explained to me that that's not real group selection, which can't happen because cheaters would undermine the evolution of altruism. 

Okay, but, I've  long had the hunch that this debate isn't framed as productively as it could be.

In recent years, Edward O. Wilson, the ant expert and author of Sociobiology, has taken up the the cudgels for the concept of group selection, which has widely been seen as evidence of imminent senility. A commenter, however, offers an interesting perspective.
Many here have missed the terms of this controversy. EO Wilson & co's change of mind was forced by 1999 work by James Hunt who studied multiple origins of sociality in a wasp phylogeny in which all the species had about the same “kin selection potential” and found that sociality evolved when there were strong “ecological” incentives (Hunt JH 1999. Trait mapping and salience in the evolution of eusocial vespid wasps. Evolution 53: 225-237). But since ~2005 Wilson, Hoelldobler, etc, have hijacked the controversy.

People have indeed started realizing that crucial for the existence/persistence –and thus for the evolution– of animal societies is not (or not so much) “kin selection” but rather the very rewarding ecological niches out there in which biomachines that adopt group-approaches to foraging and interference competition are much more effective trophically than are biomachines which adopt “solitary-consumer/fighter/reproducer” strategies.

This means that the evolutionary success of the genetic programs encoding such group behaviors is fully subordinate to the existence of such ecological opportunities!
In other words: people have begun realizing that, e.g., “altruism” is also a winning ecological strategy, rather than just an example of the promotion, or not, of altruism genes and the rejection (or invasion) of cheater genes.

The social-ant colony, e.g., is an ecological machine that out-competes at the foraging- and interference-competition level most other organisms in almost any terrestrial ecological setting, i.e., a social-ant colony in the field cannot be reduced natural-historically and evolutionary-historically to just an example of an Evolutionarily Stable Strategy immune to “selfishness” mutations that may undermine the genetic encoding of its sociality.

Wilson indeed has always made a big deal of the fact that ant species monopolize nearly 70% of the insect biomass on earth, but he did not realize the implications of this until the wasp guy rubbed it in to him and his coterie while they were still happily repeating the empty syllogisms of kin-selection numerologists [who meanwhile have even almost managed to deny Darwin (sic!) the credit for explaining the existence of sterile ants, etc., when he mentioned in the Origin of Species that an individual's sacrifice can benefit the reproduction of relatives, i.e., kin selection].

This 70% means that evolution by “natural selection of individuals” delivers niche-occupancy strategies that suffice to claim only ~30 of the trophic energy monopolized by the insect Bauplan (assuming termites and other social insects are insignificant biomass-wise).

The situation among many mammals is the same. Wild-dog packs and hyenas, e.g., beat the hell out of tigers and lions, and biomass wise they dominate.

It is time for gratuitous faux-aprioristic arguments to be confronted with ultimate natural-historical facts. And it is also time that the applied-math peddlers posturing as evolutionary biologists learn that “natural selection” is not the same as “evolution by natural selection”, that differential fitness is always caused by differential ecological performance (and never by “genes”), and that evolution by natural selection is just something that “may” happen when there is differential ecological performance at some level of biological organization, which however does not “prove” that ecological performance is caused by molecular interactions of genes (or of proteins; genes may suffice but are not necessary for differential performance, but additive genetic variation in performance suffices for evolution driven by differential ecological performance).

In other words, Sober’s 1984 [1984 sic!] book “The Nature of Selection” should be required reading for every evolutionary biologist. Sober showed first that the “kin selection” oxymoron is a muddled verbal construct to refer to that special case of group-level differential ecological performance in which “selected” groups happen to also be kin groups (most of the time but not always... see mutualism, e.g.).

54 comments:

asdfasdf said...

"One tribe is mostly lactose tolerant, the other tribe is mostly not. The leaders of the lactose tolerant tribe decide to send their cavalry, without their grain wagons, on a multiday sweep up the river in an attempt to turn the enemies' flank."

Wow. From 'Blood and Soil' to 'Milk and Soil'.

PS. I'll bet the tribe with ovaltine would have kicked everyone's ass.

Anonymous said...

"The situation among many mammals is the same. Wild-dog packs and hyenas, e.g., beat the hell out of tigers and lions, and biomass wise they dominate."

Lions also move in packs, and a pride of lions will destroy a pack of hyenas. The National Geographic special "LIONS AND HYENAS" made this abundantly clear.


http://www.youtube.com/watch?v=Yr_4IGXhYec

Best nature documentary ever.

Anonymous said...

And though wolves and wild dog packs can attack and kill the slower bear--even the grizzly--, this rarely happens with tigers. Wolves will dare not mess with most tigers. And Indian dholes--wild dogs--will attack tigers ONLY IF they are very hungry. And even then, nearly half the pack will die before it kills the tiger.

jack strocchi said...

Individual selection is the default position, because it does not require organization. But group selection can overcome this providing the sanctions against cheaters are sufficiently powerful and pervasive.

Look at the way the Mob functioned to ensure group solidarity in the good old days. Whilst you remained a team-player your chances of survival and sexual reproduction were reasonably good. Lotta mobsters must have had kids with Vegas show girls.

Violate group solidarity principles - betray the family, go to the cops - and you stood a pretty good chance of being whacked. So "cheaters" would not have had great reproductive futures.

But eventually that group solidarity got broken down as the sociological niche the Mob inhabited became hostile - FBI surveillance, RICO legislation, greener pastures as legal upward mobility became possible amongst white ethnic working class groups.

So the "cheaters" strategy started to pay-off. More rats turned over to the Feds or broke with the Family and its traditional ways - look at Gotti whacking a Boss like that in broad day light without permission from the Commission. That would never have happened in the old days.

So the Mob's group selection advantages started to erode and its power began to wane. Till nowadays it is only a shadow of its former self.

More generally over the past generation the whole of the Occident has gradually weakened its group selective powers - its called "liberalism". So people no longer have traditional loyalties to family, faith or flag.

Meanwhile the Orient still retains plenty of group selective sanctions which are slowly starting to give it a competitive advantage. Not just the PRC which has converted this solidarity into high economic growth rate. But also the Muslim world which has followed the traditional group selection strategy of using religion to increase the feasible birth rate.

So the End of History was declared somewhat prematurely.

Anonymous said...

I hope you do more posts like this, as I have a lot of trouble understanding this topic, and yet it seems to underlie many of the larger arguments you make on the blog or in your columns.

Anonymous said...

with the advocates of lower levels (e.g., the Selfish Gene) usually drubbing, through better math, the advocates of significant amounts of selection at the group level

But the maths (Price's equation) shows that selection can work on all levels.

Anonymous said...

Sober showed first that the “kin selection” oxymoron is a muddled verbal construct to refer to that special case of group-level differential ecological performance in which “selected” groups happen to also be kin groups (most of the time but not always... see mutualism, e.g.).

Makes sense. "Group" in this context seems to be a more general term whereby kin selection, selection between ethnic groups, races, cultural groups, etc., are all cases of group selection.

DR said...

That's still individual/gene selection just with the distribution of alleles correlated with "tribes." This has nothing to do with group vs individual vs. gene selection.

The difference between the two is that individual selection says that traits are selected for that favor the survival of the individual, gene selection says that traits are selected for that favor the propogation of the gene and group selection says that traits are selected for that favor the survival of the group.

However we can make your example into group selection with a slight tweak. Let's say that 90% of Tribe A has an allele for lactose tolerance, but this gene also causes degenerative heart failure at a young age. No one in Tribe B has this allele. In this case Tribe A will still cross the river and outbreed Tribe B because they only need to carry 10% of the grain (for their few lactose intolerant members).

All of the members of Tribe A will now thrive (including the lactose intolerant ones), while Tribe B will die out. However over time the lactose intolerant members of Tribe A will outbreed the lactose tolerant ones because of the heart defects. Eventually Tribe A will be complete lactose intolerant (at least with this variant of the gene).

Thus even though the trait benefited the group it disappeared because it was not advantageous at an individual/gene level. In effect the free-riding lactose intolerant members of A used the lactose tolerant members to beat out B, but then outbred them anyway.

This is why group selection does not happen. Even if it produces massive advantages for the group, there's always an even bigger advantage to cheating and free-riding on the rest of your group as long as the gene is individually detrimental. This has nothing to do with complex math, it's a simple instance of the classic Prisoner's Dilemma problem.

Shawn said...

I just saw a weird picture and a thought popped into my mind: do Whites have this type of musculature? I have only seen it on Blacks.

http://www.google.com/imgres?imgurl=http://www.collegehoopsnet.com/008_206348_UCLA_v_stanford.JPG&imgrefurl=http://www.collegehoopsnet.com/picture-day-russell-westbrook-celebrates-41954&usg=__bP_k1qBAd2TKcZUeG-JVYW8V0IM=&h=600&w=433&sz=383&hl=en&start=0&sig2=4gJpMV_0_JO9YZnjckD22w&zoom=1&tbnid=N3sJweKSF2HKSM:&tbnh=170&tbnw=152&ei=58jYTMK4KJGanAf1qZDECA&prev=/images%3Fq%3DRussell%2BWestbrook%26um%3D1%26hl%3Den%26sa%3DN%26biw%3D1024%26bih%3D677%26tbs%3Disch:1&um=1&itbs=1&iact=hc&vpx=351&vpy=265&dur=823&hovh=264&hovw=191&tx=87&ty=146&oei=58jYTMK4KJGanAf1qZDECA&esq=1&page=1&ndsp=15&ved=1t:429,r:6,s:0

TGGP said...

Anonymous, it's long been known from Price's equation that higher levels of selection can happen. It's just that odds are stacked against them relative to lower levels of selection. "More variance within groups than between groups" isn't just a blank-slater shibboleth, it's an accurate description of what we observe.

Henry Harpending proposed a model of group selection with David Boxenhorn explains here.

John Hawks responds to Wilson on kin-selection here.

Anonymous said...

It is posts like this that show that Steve Sailer truly has mediocre ideas about how things work. Half truths, speculations, absurd theories that defy common sense to explain things when far more plausible explanations are possible, anecdotal narrations in place of empirical evidence and logical fallacies, are his modus operandis. This was a cute little anecdote to explain why one group would prevail over another so far out there in every way I can think of I simply cannot take this guy seriously. I can come up with a billion little anecdotes out of the top of my skull to prove exactly the opposite that Sailer is trying to prove.

Whiskey said...

Steve, there's a lot more to fighting in groups than drinking Mare's milk. What the hell are the Mare's going to eat? And are all the horses going to be mares? Producing Milk?

Victor Davis Hanson addresses some of this in Culture and Carnage. Basically, the Western way of fighting: mass/shock battle of annihilation of the enemy, proved devastating against pretty much all its enemies, over time.

And the selection it caused was the ability to stand in line, shoulder to shoulder with mates, slaughtering and being slaughtered in a disciplined, orderly fashion. Like chopping wood only doing it with men.

Horses are expensive. In Western Europe, they don't dominate as much, because there's always mud, another river, another valley, another set of hills. Composite bows fall apart in the constant rain. There's lots of forests. Over time it was easier to take your basic yeoman farmer, teach him a few basic things, keep him in line, and give him a spear, a pike or a musket and have him fight in the line. He would not take a generation to train, be hideously expensive, and likely to turn on your dynasty.

Whiskey said...

Which if you think about it, are the successful attributes of European businesses and culture today:

1. Unity, i.e. able to confront crises and unpleasant situations "in line."

2. Plodding cooperation rather than flashes of brilliant maneuver create competitive advantage.

3. Emphasis on direct confrontation and annihilation of the enemy.

4. Large musters of pretty much the entire available manpower to meet threats.

You could argue this has been the pattern of most European and US businesses when faced by threats, for the most part, over time.

gcochran said...

"that seems to me like group selection with a bang"

Wrong. If the lactose persistence allele had become common because of the benefits it conferred to the group as a whole - _that_ would be group selection.
But it became common because of benefits it gave to individual carriers. Any group benefits are coincidental.

Look, group selection would mean that significant genetic change happens in a way that favors actions that benefit the group as a whole - with that group advantage as the primary driver of the increase in gene frequency, rather than some individual advantage that occasionally happens to give advantage to the group by sheer luck. Europeans had a group advantage versus Amerindians, in that they were resistant to a host of Old World diseases: but that does not mean that the process was driven by the group advantage vs Amerindians. It started thousands of years before Columbus: it didn't have anything to do with Amerindians. We'd never even dreamed of them.

In just same way, we might someday run into a dangerous alien race that drops dead in amazement upon seeing a redhead. But MC1R mutations didn't become common because of an ongoing process in which tribes with redheads kept wiping out their competition, resulting in super-redheaded people. That's not group selection: it's called coincidence.

Selection can cause continuing refinements and, eventually, complex functionality. What you're talking about can't. A more accurate summation of the process you describe is "stuff happens".
Once in a while the result of individual-level selection confers a group advantage, but far more often, it confers group disadvantages - which we see plenty of - confirming that selection at the gene and individual level is strong, but rarely matters at higher levels.

There can be group selection, but groups have to compete a lot, extermination has to be common, and gene flow between groups must be limited. I know of cases where it's real enough, particularly with pathogens - but it unlikely to have any importance in humans.

As for Wilson, it's time to call a spade a spade. He has the basic prerequisite of every public intellectual in the US: he has no idea what he's talking about. And he never has.

Anonymous said...

Shawn, that's what guys with extremely low body fat look like. You could have all the same muscles, but add a little body fat and all the definition disappears.

jack strocchi said...

gcochran said:

There can be group selection, but groups have to compete a lot, extermination has to be common, and gene flow between groups must be limited. I know of cases where it's real enough, particularly with pathogens - but it unlikely to have any importance in humans.

What about tribal solidarity, xenophobic hostility and raids where the men are put to the sword and the women are all carried off? That sort of thing seems to have had some importance in human history.

And so many moral codes seem to glorify individual self-sacrifice to the group - Christianity is almost a caricature of this form. Could it not be possible that tribal groups played a part in the selection and sanction of an "altruistic" gene in the social organization? That would be consistent with the obsessive interest that religions have in sexual matters - who you marry and issue with.

Just askin'. I am not a public intellectual. So please don't judge me too harshly for not knowing what I am talking about.

Anonymous said...

What's the best introductory textbook on this subject?

Anonymous said...

One of the most interesting things I've read recently is with 23andme.com's relative finder, Ashkenazi often show up as third cousins even when there is no family connection, seemingly because their gene pool is so inbred.

James B. Shearer said...

Your example is silly. Individual selection readily explains how lactose tolerance genes can go from near 0% frequency to near 100% frequency. Your example at best explains a one time jump from 50% to say 80% (from which the lactose intolerant genes could recover in a few generations if they were individually advantageous). As gcochran says this is just stuff happens, a random shock to which all genes are at risk. You could just as well hypothesize a deadly disease which kills all the cows leaving the lactose intolerant tribe in the better position.

farmiddle said...

One thing I've found confusing about the 'group selection' argument is grammatical. When you say 'natural selection' you don't mean that nature is being selected, you mean that the individual is being selected in relation to the natural environment. Likewise, if 'group selection' didn't mean that the group is being selected, but that the individual is being selected in relation to the group, then I think group selection would make more sense.

I’m also curious about the level of complexity of different human societies. Complex societies appear to have outcompeted simpler societies throughout history. Is that group selection in the sense that the group is being selected or that individuals are being selected that are better adapted to complex groups? Both can be thought of as 'group' selection.

Anonymous said...

"The men don't have many children at all."

I'm surprised they had any. I must have been snoozing during HS biology class.

Anonymous said...

"The men don't have many children at all."

I'm surprised they had any. I must have been snoozing during HS biology class.

Anonymous said...

IIRC Lions push wild dogs out. The success of the lions in the park has translated to fewer wild dogs.

Baloo said...

Question to GCochran: the altruism thing, where young males sacrifice themselves in war, thereby preserving their sisters' genes — is that a valid concept, and if so, is it an example of group selection?

ben tillman said...

Individual selection is the default position, because it does not require organization.

Absolutely not. You're saying that a human body is disorganized, which is preposterous.

ben tillman said...

The difference between the two is that individual selection says that traits are selected for that favor the survival of the individual, gene selection says that traits are selected for that favor the propogation of the gene and group selection says that traits are selected for that favor the survival of the group.

All individuals are groups, and all groups are individuals. Once you realize this, as David Sloan Wilson has explained, the notion that there is no group selection becomes ridiculous.

MQ said...

There can be group selection, but groups have to compete a lot, extermination has to be common, and gene flow between groups must be limited.

Samuel Bowles at the Santa Fe institute has produced compelling evidence that exactly these conditions prevailed among human ancestral tribes. See Choi, Jung-Kyoo and Samuel Bowles, "The coevolution of parochial altruism and war."Science 319 (26 October 2007) and Bowles, Samuel, "Did Warfare among Ancestral Hunter-Gatherer Groups Affect the Evolution of Human Social Behaviors?" Science, 324, pp. 1293-98. (05 June 2009)

ben tillman said...

While the first few chapters explain the misunderstandings that have produced the "controversy", The most important stuff is laid out at pages 87 to 98 of Unto Others: The Evolution and Psychology of Unselfish Behavior by Eliot Sober & David Sloan Wilson. I hope you have read this by now, Steve.

Wilson & Sober summarize, "...the primary message of our book is that groups, too, can be functional units and that individuals sometimes behave more like organs than organisms."

"Multilevel selection theory" is a better descriptor than is "group selection theory", but the emphasis on selection seems to cloud the concept, so I've coined the term "multilevel organization theory".

M said...

Yeah that example you've given - is that really group selection? I would have thought that was individual level selection where the individuals with the lactase allele gain selection benefits due to their environment which just happens to be their group, which seems like its a trivial truth (interesting from the point of view of HBD observers, but not interesting to population geneticists).

I think group selection would be more like, using an ersatz example:

Say you're a member of a ethno-religious mercantile group whose members largely devote themselves to religious study.

Let's also say, that you all carry an allele (for the sake of a simplified model) that sporadically produces geniuses who earn huge sums of money sufficient to support your clan.

Let's assume that the allele is only active in the genius and inactive in the rest of the clan (i.e. has no effect), but that the active allele means that the subject has no offspring.

Now, if this allele has been subject to selection to get to where it is, that kind of selection - where it is only selected for in the individuals it is inactive in and has no phenotypic effects, through their group association with the individual in which it is active - is what group selection is.

(Note that I can't actually imagine the above dynamic being or having been operant even in the historical and contemporary populations who are... um... somewhat parallel to the above - it's merely for the purposes of illustration).

Anonymous said...

"And it is also time that the applied-math peddlers posturing as evolutionary biologists learn that “natural selection” is not the same as “evolution by natural selection”, that differential fitness is always caused by differential ecological performance (and never by “genes”), and that evolution by natural selection is just something that “may” happen when there is differential ecological performance at some level of biological organization, which however does not “prove” that ecological performance is caused by molecular interactions of genes (or of proteins; genes may suffice but are not necessary for differential performance, but additive genetic variation in performance suffices for evolution driven by differential ecological performance)."


Could a sentence get any longer?

Veracitor said...

1. I think there's an easy non-mathematical argument for group selection: how else can you explain multicellular organisms?

2. Even mathematical types who are supposed to know better get led astray by their emotional responses to the term "altruism," which means something quite different in the jargon of evolutionary theory than in common use. They keep thinking that "altruism" means an organism "gives" without expecting to "receive." Of course that is nonsense, true altruism is non-adaptive. So people get tangled in the verbal weeds of "mutual altruism," which is almost a contradiction in terms.

Everyone should switch to using the term "cooperation." As soon as they do, they will see that even purely individual selection can easily favor traits that offer to cooperate even if they only find reciprocators probabilistically-- this is the link between evolutionary theory and game theory.

This insight makes "group selection" more plausible. Once you have organisms selected for (some level of) cooperation, a mutation which enhances that cooperation (or the benefits obtained from it) can be favored on a group level.

For example, suppose members of species S signal willingness to cooperate by assuming a special posture (assume cooperation is useful, as with pack hunters). If no nearby S creature replies to the signal within interval T, the initial signaller proceeds alone to some individualistic behavior.

Now suppose a mutation makes some members of S willing to wait longer (T+) for a reply to a willing-to-cooperate signal. In the absence of any S creatures willing to cooperate, this T+ mutation may have a neutral or even detrimental effect on individual fitness (e.g., remaining stationary for interval T+ might increase danger from predators which would be less likely to pounce within interval T).

But if several members of S inherit a T+ mutation (e.g., from a common ancestor) and it allows them to cooperate more often or more effectively, it may improve their total fitness!

Individuals sharing a cooperation-enhancing mutation may have more inclusive fitness than less-cooperative individuals (even if the fitness of a few cooperative individuals suffers, e.g., under particular environmental conditions) because of gains from cooperation. When cooperating, such individuals comprise a group. Since the benefits (and costs) of various forms of cooperation accrue to members of (cooperating) groups, natural selection can operate on such groups.

At this point someone will complain that "cheaters" "must" out-prosper cooperators. Although that is a dogma oft-repeated, I think it is baseless. It's easy to show that many possible gains from cooperation only accrue to actual participants. Also, real organisms appear to have evolved a variety of mechanisms to guard their cooperating groups against various kinds of cheating. Since we have empirical evidence of cooperation, it seems unlikely that selection pressure against cooperation is very strong. On the evidence, I think there must be considerable fitness benefit to cooperative strategies since so many species employ them.

jack strocchi said...

Absolutely not. You're saying that a human body is disorganized, which is preposterous.

I mean social, not biotal, organization.

Mudpuppy said...

As has been said repeatedly, Steve, this is a terrible example. It would be nice if you would post an edit or correction as discussion of this issue is already so confused among the laity.

In order to be under "group selection," the trait in question must have neutral or negative effects on individual fitness, but still spread through populations due to the fitness it confers to inter-group interactions.

A frequently cited example is of the prairie dogs, meerkats, monkeys, etc that have "sentries" that watch out for predators and emit conspicuous cries to alert their cohorts when one is spotted. This behavior may have spread (even though it confers fitness disadvantages to the individual sentries) because other members of their group were better protected, on average, than other individuals in more selfish, less defended groups.

It may be though that traits like the above are actually examples of pro-social behaviors emerging through individual selection models like reciprocal altruism, competitive altruism, etc... Or gene selection kinship models.

jack strocchi said...

Plainly there are altruistic people, an phenomenon which has always been labelled a "problem" by ultra-Darwinists. The case of family altruism has been covered by Hamilton/Williams with the notion of kin selection. So family altruism is simply gene egotism applied to closely related con-specifics. So far so good.

But we also observe alruistic behaviour on behalf of more distantly related con-specifics in numersous societies in numerous social roles (priests, soldiers, nurses and so on). In most cases altruistic behaviour is praised and altruistic social roles are elevated in social status.

Plenty of times these individuals sacrifice chances of survival and sexual opportunities. Thats not supposed to happen in the pure Darwinian schema.

And its obvious that individual self-sacrifice can improve the chances of groups survival and sexual thriving. Pretty much every second movie is about the hero who risks his life to save the group.

If he makes it then he usually gets to go off with the beautiful girl. Which no doubt improves his individual fitness.

So is manifest alruism really just another form of latent egotism, albeit a more riskier version? Meaning that it is individual egotism all the way down?

jack strocchi said...

Of course gcochran is in the business of studying how altruistic religions administer individualistic kin-selection, as in the case of the Azkhenazi Jews. So probably not a good idea to pick a fight with him on this subject.

Anonymous said...

TGGP said...

Anonymous, it's long been known from Price's equation that higher levels of selection can happen.

--

Yes. That is why I used the word can.

Encino man said...

To give yourself a very nice understanding of group evolution and a history of the concept, read the entirety of David Sloan Wilson's well written and fun to read 'Truth and Reconciliation for Group Evolution' starting from October 2009 at http://scienceblogs.com/evolution/truth_and_reconciliation_in_gr/.

And then for fun read everything else on his blog, including 'Atheism as a Stealth Religion' and "Economics and Evolution as Different Paradigms'

Encino man

adsasdfasdf said...

I don't see it as individual selection vs group selection but individual-within-group selection.

Anonymous said...

Even mathematical types who are supposed to know better get led astray by their emotional responses to the term "altruism," which means something quite different in the jargon of evolutionary theory than in common use. They keep thinking that "altruism" means an organism "gives" without expecting to "receive." Of course that is nonsense, true altruism is non-adaptive. So people get tangled in the verbal weeds of "mutual altruism," which is almost a contradiction in terms.

Altruism has a pretty specific meaning in evolutionary biology. Altruistic acts are defined as those that benefit the recipient but harm the actor.

jack strocchi said...

someone said:

Altruism has a pretty specific meaning in evolutionary biology. Altruistic acts are defined as those that benefit the recipient but harm the actor.

That meaning seems to me pretty close to common usage. Does anyone have an explanation of institutionalized altruism - in families, religions and armies - that can clearly be explained by individual gene selection?

Wandrin said...

"So family altruism is simply gene egotism applied to closely related con-specifics. So far so good."

Wouldn't extended group altruism be the same as family altrusim assuming
a) The ability to understand probabilities.
b) The ability to understand reciprocity.
c) The ability to understand how culture can be used to modify behaviour.

So IQ above a certain level would inevitably lead to group altruism based on probable benefit?

The cheater problem is limited by group competition. If two tribes are at war the tribe with the most cheaters is much more likely to lose.

(Armies held together by being the most religious, ideological, nationalistic, hyper disciplined etc have a built in advantage because by definition those armies have been created through a process designed to minimize cheaters e.g a roman cohort that broke and ran would be decimated i.e one in ten would be executed.)

There's a cheater niche and cheaters may thrive as long as there's not too many of them but as soon as the number of cheaters grows too large (or powerful) the group as a whole loses fitness and succumbs to a group with less cheaters.

In that sense diversity and multi-culturalism could be seen as a kind of unilateral disarmament that would inevitably lead to the destruction of the group to groups who were more strongly "us" orientated.

Hence also the historically extreme attitude to traitors as the worst of the worst and historical antipathy to anything which made it harder to create a singular "us" thereby leading to racism, xenophobia, religious intolerance etc as a self-defense mechanism.

Wandrin said...

"Stuff happens" evolution wouldn't occur continuously but as occasional conquest events.

"When Israel had finished slaughtering all the inhabitants of Ai in the open wilderness where they pursued them, and when all of them to the very last had fallen by the edge of the sword, all Israel returned to Ai, and attacked it with the edge of the sword. The total of those who fell that day, both men and women, was twelve thousand -- all the people of Ai. For Joshua did not draw back his hand, with which he stretched out the sword, until he had utterly destroyed all the inhabitants of Ai. Only the livestock and the spoil of that city Israel took as their booty, according to the word of the Lord that he had issued to Joshua. So Joshua burned Ai, and made it for ever a heap of ruins, as it is to this day."

If the traits that gave the advantage were genetic then in most cases the result would be a dilution of those traits as some of the conquering males took wives from the conquered population and some of the conquered females were taken back to the conqueror's terriotory as captives.

(Matrilineal descent would counter that tendency however making the conquerors more inclined to kill the females as well as the males.)

If there was a group advantaged pool separated geographically then these "stuff happens" evolutionary events could be periodic i.e steppe nomads hitting a population limit then going on a rampage with large numbers conquering and settling down as a minority elite in Egypt or somewhere slightly raising the average IQ before merging with the process repeating every few hundred years or so.

Wandrin said...

"At this point someone will complain that "cheaters" "must" out-prosper cooperators."

One point on that. If it's mathematically true that cheating is rationally the optimal strategy for an individual as an individual and yet group co-operation (or everyone else group cooperating) is the optimal strategy for the individual as a member of a group then wouldn't the individual level trait selection mechanism be to select for some optimal level of rationality below 100%?

Maybe advanced groups need the masses to be irrationally swayed by ideas like patriotism, religion etc for the group cooperation mechanism to work?

Too little rationality and your group can't progress because of stupidity. Too much rationality and your group has too many cheaters to survive competition from hostile groups.

Steve Sailer said...

The battle of Valmy in 1792, when nationalist French novices overwhelmed a mercenary army of invading professionals was instantly recognized (by Goethe, who witnessed it) as an epochal event. This set off a sort of altruism arms race that crested in the Great War and has since receded.

It was widely believed in the 1920s and the 1930s by the English, who had fought the Somme using a volunteer army, that they had annihilated their finest young men.

All nurture? Some nature? Nobody knows.

Wandrin said...

I don't think the significance of WWI can possibly be over-estimated both as the result of earlier changes and a cause of future ones.

The changes in technology that led to the need for ultra-patriotic mass armies and the consequences of that kind of total war.

"It was widely believed in the 1920s and the 1930s by the English, who had fought the Somme using a volunteer army, that they had annihilated their finest young men."

There's a lot of truth in that i think. A lot of the males who survived, survived through being rejected from service on medical grounds because of rickets and other environmental diseases.

Vercingetorix to Verdun.

Another aspect is chance. The group eugenic benefits of mass altruism become extremely dysgenic if you happen to have a war in a period that is post-machinegun but pre-tank.

rob said...

As a multicellular organism, I think I have to believe in group selection. Otherwise my brain cells, liver cells, etc. are all suckers for letting the sperm cells do all the reproduction. If I were reproducing, I mean.

ben tillman said...

Rob, you are a smart, smart man.

Anonymous said...

Your general point about the lactose-tolerance example taken, I don't want to be pedantic, but military horse riding (or even very competent horse riding) didn't begin until around 1000 BC - 800 BC by which time I believe lactose tolerance had spread over Europe...not completely sure about the latter (time-frame of spread of lactose tolerance) but I actually think these kinds of things are much less important than military technology and social organization in facilitating conquest.

Wandrin said...

"but military horse riding (or even very competent horse riding) didn't begin until around 1000 BC - 800 BC"

I forget the actual date ranges but i think chariots came first and cavalry much later.


"I actually think these kinds of things are much less important than military technology and social organization in facilitating conquest."

If lactose tolerance had its root in steppe nomads then its effect would be to increase the population on the steppes, which would in turn mean larger armies. The larger armies combined with the chariot / cavalry advantage leading to periodic conquest.

If so then the progression of lactose tolerance off the steppe would be a marker of the results of earlier conquest rather than a cause.

Anonymous said...

Wandrin--
There is a disagreement between Indo-Europeanists on one hand and military historians/archaeologists on the other. The latter generally agree that competent riding (and military riding) didn't begin until after the invention of the bronze bit, which was roughly 1000 BC - 800 BC. Earlier than that, as you point out, the chariot was used, but not MUCH earlier. While wheeled vehicles existed early, and while rudimentary chariots were developed by 2000 BC, the military application of the chariot did not begin until around 1800-1700 BC. With the invention of the chariot very small peoples/armies were able to dominate much larger populations: such as the Hyksos in Egypt, the Hittites in Hatti, the Kassites in Mesopotamia, the Aryans in India, the Levant (limited), and in Mitanni...and I would argue, also the Aryans' kin in Europe. In either case, lactose tolerance I would suppose spread in Europe much earlier than 1600-1500 BC, if you want to argue that chariot conquest (rather than riding) was the cause (which I believe)...

Cochran and Harpending's theory is attractive for many reasons but I don't see how it can fit the archaeological, linguistic, etc., evidence which points to a very *late* dispersal of IE/Aryan

I know Sailer wasn't making this point, this is a sort of side convo...one argument you COULD make is that pastoralists generally are more warlike than sedentary agriculturalists and often conquer them. This happened also with Tutsi and Hutu in Rwanda, as he's pointed out before. But Tutsi lactose-tolerance did not spread among the Hutu I believe...

Wandrin said...

Anon,

Fair enough. I'd readily admit my view is biased by liking the idea of my ancestors being chariot riding lactose tolerant Conans.

another reader said...

If I have this right, group selection is the group being favored at the expense of the individual(s). With humans, more complex cooperative behavior had benefits on the individual level, which can be "scaled up" to the group level, but that doesn't make it group selection.

Looks like there is a lot of semantic confusion here... just because a behavior is group-orientated or benefits a group doesn't mean selection is occurring at a group-level. But if you change the definition to including those features, then group selection does occur, and what you have is a trivial observation.

Wandrin said...

"Looks like there is a lot of semantic confusion here"

True on my part.

My thought was that a trait that was more beneficial for the group than the individual e.g a non-cheater gene, could become dominant over time through a process of groups with more non-cheaters massacring the ones with more cheaters.

I'm not sure what that would be called.

Another name for the non-cheater gene might be faith gene.

Anonymous said...

Wandrin,
It is possible your ancestors were chariot-riding lactose-tolerant Conans...I believe the Aryans who colonized/conquered Europe were this, I just don't think their success had much to do with lactose tolerance. Again, I believe Aryanization of Europe was very late. But I also believe in model of "elite dominance," and so they were a minority...

The native European population probably was itself lactose tolerant. Aren't Basques lactose tolerant...I just don't think that these kinds of things can explain success. The racial superiority of Aryans went well beyond being able to process milk.