This site is not maintained. Click here for the new website of Richard Dawkins.

← Dear E. O. Wilson: Please retire or stick to ants

Dear E. O. Wilson: Please retire or stick to ants - Comments

Rich Wilson's Avatar Comment 1 by Rich Wilson

Read More is missing the link http://freethoughtblogs.com/blaghag/2012/04/dear-e-o-wilson-please-retire-or-stick-to-ants/

Sat, 21 Apr 2012 16:40:54 UTC | #936296

Peter Grant's Avatar Comment 2 by Peter Grant

http://freethoughtblogs.com/blaghag/2012/04/dear-e-o-wilson-please-retire-or-stick-to-ants/ Dear E. O. Wilson: Please retire or stick to ants.

Yes! Just saying "group selection" doesn't actually explain anything, that's simply restating the problem.

Sat, 21 Apr 2012 17:09:28 UTC | #936302

drumdaddy's Avatar Comment 3 by drumdaddy

Excellent article. Sad about Wilson.

Sat, 21 Apr 2012 17:31:34 UTC | #936306

Metamag's Avatar Comment 4 by Metamag

Is this similar to what real quantum physicists go through?

Sat, 21 Apr 2012 17:31:42 UTC | #936307

Neodarwinian's Avatar Comment 5 by Neodarwinian

Yes, very sad about E. O. Wilson and that is why I am blaming David Sloan Wilson and his running dog philosopher friend Eliot Sober for this debacle. Evidence? I am using about as much evidence for my assertion as that evidence supporting group selection's motivating force for human eusociality.

Sat, 21 Apr 2012 17:51:21 UTC | #936311

Rawhard Dickins's Avatar Comment 6 by Rawhard Dickins

Is there not a grey area where group selection can work, when segregated groups can be considered in the same light as an "individual". If a group's shared genes promote help between individuals within the group, the group as a whole would have better reproductive success. A less altruistic genome would have less success. To some degree there is always some segregation, this segregation controls run-away success of the actual individual.

Fire away, I've got my tin hat on!

Sat, 21 Apr 2012 18:26:52 UTC | #936320

rationalmind's Avatar Comment 7 by rationalmind

As someone who has been fascinated with the natural world since early childhood. I have three great heroes that I would like to meet. One is David Attenborough, who I have been lucky enough to meet and talk to for a while.Another is Richard Dawkins whom I hope to meet someday. The third is E. O. Wilson. I meet him briefly when he gave a talk in London some years ago. A friend managed to get me a ticket and it fitted in with a trip to London.

I think he is wrong I think he is very wrong, but I would also like to look at the nature papers myself. I have access and I will probably dig them out at some point.

The value that such disagreements have though is that they ensure that minds are focused on ensuring that we know the truth.

Sat, 21 Apr 2012 19:03:57 UTC | #936325

Schrodinger's Cat's Avatar Comment 8 by Schrodinger's Cat

Comment 6 by Rawhard Dickins

If a group's shared genes promote help between individuals within the group, the group as a whole would have better reproductive success. A less altruistic genome would have less success. To some degree there is always some segregation, this segregation controls run-away success of the actual individual.

If 10% of a group have an atruistic gene that benefits the entire population, for example if they have a gene that makes them more likely to fight to the death against predators, then the group benefits at the cost of the spread of the altuistic gene. But nothing the rest of the population is doing is actually 'selecting' this gene.......they are merely benefiting by default. So at what tipping point in the spread and rise of this gene, if there is one, does the benefit to the group become the same thing as the benefit to the gene itself ? Isn't group selection merely confusing the two ?

Sat, 21 Apr 2012 19:13:49 UTC | #936328

Stafford Gordon's Avatar Comment 9 by Stafford Gordon

Dear oh dear oh dear, I suspect that Wilson is being misrepresented; at least I hope so.

My dim understanding of the matter is that ants behave as single organism. Now, whether that's a cross over from gene to group selection I have no idea, but I'm certain that I'll be enlightened by someone on this website.

Sat, 21 Apr 2012 20:07:38 UTC | #936338

Tord M's Avatar Comment 10 by Tord M

Before I make up my mind, I would like to see what E.O. Wilson actually said. (I hope he's not suffering from antony flu.)

Sat, 21 Apr 2012 20:13:01 UTC | #936339

Anaximander's Avatar Comment 11 by Anaximander

Does group selection (in this case) mean that the group is selected or that the group is selecting?

Sat, 21 Apr 2012 21:00:45 UTC | #936351

Fouad Boussetta's Avatar Comment 12 by Fouad Boussetta

To Schrodinger's Cat and by Rawhard Dickins:

I just finished reading Jonathan Haidt's new book "The Righteous Mind". Really an excellent book. I just loved it. You can read the 4-page conclusion/recapitulation/summary towards the end of the Amazon.com preview, if you want to have an idea of the contents.

I found nothing to dispute there EXCEPT MAYBE (but I am no expert) the group selection stuff in chapter 9, which the author graciously offers to send you free of charge if you request it at: haidt at virginia dot edu

I'm still waiting for Richard Dawkins, Jerry Coyne, and other specialists to read that chapter (and its numerous notes) and express their criticisms!!!

[Note: Whether or not group selection actually occurs does not change anything to the main ideas of the book. But then again, I'm curious!]

Sat, 21 Apr 2012 21:33:29 UTC | #936355

Steven Mading's Avatar Comment 13 by Steven Mading

Doesn't an ant colony all share the same DNA? The workers, the soldiers, the queen - they're all the same genes, and differentiate only based on what they are fed - is that true?

If so, then I don't see what the difference is between group selection and kin selection in an ant colony. They're the same thing. If you're an ant, then your whole group you interact with is the tightest sort of kin you can have - they're your clones.

Could it be that someone who studies ants primarily would mistake kin selection for group selection because of this?

Sat, 21 Apr 2012 21:36:41 UTC | #936356

Alan4discussion's Avatar Comment 14 by Alan4discussion

Comment 8 by Schrodinger's Cat

But nothing the rest of the population is doing is actually 'selecting' this gene.......they are merely benefiting by default. So at what tipping point in the spread and rise of this gene, if there is one, does the benefit to the group become the same thing as the benefit to the gene itself ? Isn't group selection merely confusing the two ?

I would think competition between groups with different levels of the altruistic gene that benefits the entire population within those populations, would be relevant. A group with lower levels of the altruistic gene in its population would be less likely to survive or multiply.

Sat, 21 Apr 2012 21:37:13 UTC | #936357

Sample's Avatar Comment 15 by Sample

Jump to comment 11 by Anaximander

Does group selection (in this case) mean that the group is selected or that the group is selecting?

Maybe I misunderstand your question, but wouldn't a group theorist be required to say both?

Mike

Sat, 21 Apr 2012 21:49:35 UTC | #936360

Peter Grant's Avatar Comment 16 by Peter Grant

Comment 14 by Alan4discussion

A group with lower levels of the altruistic gene in its population would be less likely to survive or multiply.

Alan, I usually approve of your posts, but this is just nonsensical. Groups don't survive or multiply.

Sat, 21 Apr 2012 22:03:49 UTC | #936364

Anaximander's Avatar Comment 17 by Anaximander

Maybe I misunderstand your question, but wouldn't a group theorist be required to say both?

I don't know. In principle, I think, a group can select something that is not a group. And a group can be selected by something that is not a group. Though I don't know how a group can be selected, as it seems not to replicate.

Sat, 21 Apr 2012 23:16:04 UTC | #936374

Richard Dawkins's Avatar Comment 18 by Richard Dawkins

No they are not a clone. and no the colony doesn't behave as a single organism. It's much more interesting than that. It's the middle of the night and I must try to get back to sleep, but it isn't hard to look up. It's all in The Selfish Gene and lots of other places.

Good night.

Comment 13 by Steven Mading :

Doesn't an ant colony all share the same DNA? The workers, the soldiers, the queen - they're all the same genes, and differentiate only based on what they are fed - is that true?

If so, then I don't see what the difference is between group selection and kin selection in an ant colony. They're the same thing. If you're an ant, then your whole group you interact with is the tightest sort of kin you can have - they're your clones.

Could it be that someone who studies ants primarily would mistake kin selection for group selection because of this?

Sun, 22 Apr 2012 01:10:23 UTC | #936388

xmaseveeve's Avatar Comment 19 by xmaseveeve

Comment 18, Richard Dawkins,

Get some sleep! Sweet dreams, nice man. xx

Sun, 22 Apr 2012 01:29:55 UTC | #936393

Helga Vieirch's Avatar Comment 20 by Helga Vieirch

Comment 14 by Alan4discussion

I am leaning toward agreeing with Alan on this. I was just speaking of heroes on another thread here and it just so happens that E.O. Wilson is one of mine. I need to review exactly what Wilson said this time. I keep hoping that he will take some time one day to really tackle how we can fit cultural systems into models of evolution.

To me, as an anthropologist, there has always been this difficulty of meshing cultural systems with gene pools. Both are at the mercy of natural selection, but at different levels. Trying to find a model where selection works among cultural systems (meme-complexes and associated technology?) to the benefit of some gene pools and the the detriment of other gene pools (with less advantageous cultures). Insofar as cultural systems begin to compete with one another, the successful ones might be expected to increase in population, and any genes that furthered that success would therefore increase as well.

Of course cultures are not quite synonymous with genetic populations; moreover they are both permeable to both gnu flow in and out and meme flow in and out.

And yet, in humans at least, the organism is so specialized in its adaptation to learn culture, that it cannot survive without becoming uncultured. So cultural systems have contributed enormously to the election pressures acting on individuals.

The question then needs must be asked - what degree of a selective pressure, acting on the cultural system as a whole, might drive some genes to high frequency?

Genes that regulate the individual's ability to assimilate language and other aspects of culture, mutations which contribute to a cognitive system which produces a surge of dopamine in response to fair division of food, or just punishment for cheaters, or compassionate actions towards others, even strangers… might be the kinds of genes so targeted.

In a culture where sharing and cooperation and generalized reciprocity was a more successful strategy for long term survival, such mutations, moderating the behaviour of individuals, would still be selected for at the level of the individual. But the culture system itself would be selected for also at the level of the whole culture. Within an environment where food sources were sometimes precarious and varied from season to season and from year to year, the nutritional outcome for each individual was more likely optimal when food sharing was as generalized as possible, it also follows that infant survival was optimized by the same practices.

Now, the older alleles of those genes, which might have produced in each individual more dopamine in response to hoarding as much as their individual status would let then get away with, and/or produced very little dopamine when sharing food or showing tolerance toward an outsider - well, these older alleles might work better in another kind of culture. These organisms might be able to survive in the same environment as the the ones in the above example, but there would be fewer individuals in good condition and fewer females able to rear offspring optimally. This group would eventually leave fewer offspring behind and over time this could lead to extinction, possibly even competitive exclusion by the other, more cooperative and "altruistic" group.

Well, I've got to give this some more thought. I've still got evening stables to do and already my head is feeling like I have mould growing out of my ears. Too tired to make sense, maybe.

Sun, 22 Apr 2012 03:11:48 UTC | #936402

QuestioningKat's Avatar Comment 21 by QuestioningKat

It's the middle of the night and I must try to get back to sleep,... Good night.

Just a little advice...turning on the computer when you can't sleep is rarely ever a good idea. Pick something less serious to read in a paper book form to avoid the bright light. A cup of hot decaf tea also helps. Good night for me. Good morning to you.

Sun, 22 Apr 2012 03:46:00 UTC | #936405

Rawhard Dickins's Avatar Comment 22 by Rawhard Dickins

Does this not depend on the definition of an organism? In Smith's haystacks, individuals have to cooperate to form a successful group, although the real world may look like a farmyard after a hurricane with a carpet of shallow peaks and troughs representing a degree of segregation.

Some of the actual individuals will have survived the successful pressure-cooker environments of the haystacks (excuse metaphors) and temporarily become the mobile germ-line for the cluster and altruism is hence selected for because haystacks with largely uncooperative individuals don't succeed. Individuals benefit from clustering and don't spread in a run away fashion.

Sun, 22 Apr 2012 06:10:32 UTC | #936414

Arthur Noll's Avatar Comment 23 by Arthur Noll

I agree with Alan and Helga- and EO Wilson. I don't see a problem with group selection, and find that there is any argument about it, a bit strange. If people cannot live independently of each other, then the only way they survive at all is if their group survives, and group survival cannot be taken for granted. Are human societies beyond destruction? That seems like it should be a silly notion to anyone scientifically minded.

There are things that can destroy groups no differently than there are things that can destroy individuals. If groups are inefficiently organized, if individuals are constantly quarreling with each other, they are weaker. They are wasting energy on inefficiency, on internal quarreling, that might have been spent on dealing with external threats. Decisions of what is sustainable behavior, that were not based on evidence, can lead to situations where there isn't enough for the group, which can be a really major threat to groups. Overshoot of resources with carrying capacity damage is a well known problem in biology. I can't see any reason why people are different from other organisms on the issue, and see that people have in fact been making decisions on this issue without evidence for a long time. When overshoot starts collapsing, the shortages combined with previous quarreling, could turn into really serious fighting, something we are a lot more lethal at doing than most animals. Some countries are already doing this. If seriously bad decisions about sustainability have been made, the fighting could be incidental, most could die whether they fight or not, though it will happen faster with fighting. Fighting can be like negative feedback, making a bad situation worse and the result is even more fighting making things worse again.

Social groups can do something that individuals can't do- they can split. People who see that the group they are depending on is seriously unhealthy and the majority of the people in it are beyond rational discussion about it, might form small new groups that avoid the health problems of their old groups. The people who do this must have the right stuff, the right genes, that enable them to see more clearly what the problems are, what the potential problems are, and what to do to avoid them, survive those problems with the highest odds. I've written specifics about this in other places, can certainly talk more about it. The leaders of the old groups may not want this split to happen, but as things get worse and worse for them, they have less and less energy to prevent it or to care about it. They can easily feel like they have much bigger problems than a few people non violently splitting off- they are likely to have a lot more people violently wanting to take over the system and run it by a different system of fairy tales, they are likely to have deep concerns about violent threats from other nations that are looking for someone to blame for their own internal problems. Being physically non violent about this split is a protection- a huge dying social system can still be really dangerous in its death spasms. Avoiding all physical provocations is important. It isn't efficient to force behavior, people are in these new groups on a voluntary basis only.

It is a selection of humanity, it is seeing some people as more capable of forming fit groups than others- but it is self selected, people volunteer to try, nobody is directly killing anyone. The old groups are merely in the last stages of a suicide march that started ages ago, with people readily believing objective nonsense of many different kinds, and unable to stop believing.

Sun, 22 Apr 2012 06:34:09 UTC | #936417

Jos Gibbons's Avatar Comment 24 by Jos Gibbons

Comment #936307 by Metamag

Is this similar to what real quantum physicists go through?

How?

Comment #936320 by Rawhard Dickins

Is there not a grey area where group selection can work, when segregated groups can be considered in the same light as an "individual". If a group's shared genes promote help between individuals within the group, the group as a whole would have better reproductive success. A less altruistic genome would have less success.

No. A new gene doesn’t get to be common in a group unless it has immediate individualistic advantage. A gene that is common in a group didn’t get that way by group selection. And if a rival mutation subsequently occurs with immediate individualistic advantage, it will stomp out the old gene, which means the old gene doesn’t enjoy group-selectionist protection. And groups don’t have genomes; they have gene pools. Gene pools aren’t penalised in the same way genomes are.

Comment #936351 by Anaximander

Does group selection (in this case) mean that the group is selected or that the group is selecting?

That a genetic property of the group is selected. Which can’t happen, for the reasons above.

Comment #936356 by Steven Mading

Doesn't an ant colony all share the same DNA? The workers, the soldiers, the queen - they're all the same genes, and differentiate only based on what they are fed - is that true?

Not quite. Males are fatherless haploid offspring of their diploid mother whereas females are diploid offspring of a diploid mother and a haploid father. Therefore, males are as genetically dissimilar (even where they have the same mother, which if they’re similar in age they likely will) as a human’s gametes, whereas females inherit identical DNA from their father (besides mutations therein) but the DNA they inherit from their mother exhibits the former level of diversity. It’s called haplodiploidy. http://en.wikipedia.org/wiki/Haplodiploid_sex-determination_system

Comment #936357 by Alan4discussion

I would think competition between groups with different levels of the altruistic gene that benefits the entire population within those populations, would be relevant. A group with lower levels of the altruistic gene in its population would be less likely to survive or multiply.

That’s assuming that groups end up with those properties that make for the best groups in a heritable manner that is stable to internal mutations that are individualistically superior to what came before. But, if you do the maths, it turns out they’re not.

Comment #936402 by Helga Vierich

find a model where selection works among cultural systems (meme-complexes and associated technology?) to the benefit of some gene pools and the the [sic] detriment of other gene pools (with less advantageous cultures)

Natural selection doesn’t operate at the level of gene pools.

Insofar as cultural systems begin to compete with one another, the successful ones might be expected to increase in population, and any genes that furthered that success would therefore increase as well

And what happens when the genes mutate inside the population? If the genes weren’t already going to thrive for individualistic reasons, they’ll disappear.

the organism is so specialized in its adaptation to learn culture, that it cannot survive without becoming uncultured. So cultural systems have contributed enormously to the election [sic] pressures acting on individuals.

That’s not the same thing as group selection.

a selective pressure, acting on the cultural system as a whole

Selective pressures don’t do that.

In a culture where sharing and cooperation and generalized reciprocity was a more successful strategy for long term survival, such mutations, moderating the behaviour of individuals, would still be selected for at the level of the individual. But the culture system itself would be selected for also at the level of the whole culture.

A genuine example of group selection would be one in which traits end up common in a group despite lacking individualistic advantage. What does it mean for a culture system to be selected at the level of the culture? Explain in terms of what’s happening to genes over time.

these older alleles might work better in another kind of culture. These organisms might be able to survive in the same environment as the the [sic] ones in the above example

By definition, the culture that you live in is part of the environmental conditions governing natural selection.

Comment #936417 by Arthur Noll

If people cannot live independently of each other, then the only way they survive at all is if their group survives

Natural selection doesn’t favour a trait because if it didn’t extinction would result; it favours whatever gives short-term individualistic advantage, even if that makes extinction inevitable. Here’s an example. There’s a gene in mice that gets into 95% of the gametes of an individual who inherited it from exactly 1 parent, rather than the usual 50 %. However, in a double dose it’s fatal at a young age. Group selectionist logic would imply it wouldn’t prosper. Instead it does, and wild mice populations are regularly brought to extinction by mutations producing this gene. If group selection is to be taken seriously empirically, someone needs to find an example where this sort of thing doesn’t happen. If group selection is to be taken seriously conceptually, someone has to show how the Price equation can be compatible with group selection actually happening.

They are wasting energy on inefficiency, on internal quarreling, that might have been spent on dealing with external threats.

Given that we already have the mechanism by which natural selection can favour reciprocal altruism without group selection (see The Selfish Gene), why think that group selection is involved here?

People who see that the group they are depending on is seriously unhealthy and the majority of the people in it are beyond rational discussion about it, might form small new groups that avoid the health problems of their old groups. The people who do this must have the right stuff, the right genes, that enable them to see more clearly

Why is that group selection rather than individuals varying in how well they withstand the environmental conditions in which they find themselves? (Bear in mind the environment of natural selection includes the properties of other individuals.)

It is a selection of humanity

What does that mean? What is being selected at what level against what things of the same type, due to what happening with genes?

Sun, 22 Apr 2012 07:16:40 UTC | #936423

Stafford Gordon's Avatar Comment 25 by Stafford Gordon

I've woken up to remember that, of course, a genetic mutation which renders an adavantage to any organism will spread throughout that comunity; in the case of ants, (Family Formicidea) it causes them to behave in the way they do. Similarly, bees, (Superfamily Apoidea) behave as they do, because of the cumulative non random survival of randomly varying self replicating information.

Phew! Better go back to bed.

Sun, 22 Apr 2012 08:11:42 UTC | #936425

Rawhard Dickins's Avatar Comment 26 by Rawhard Dickins

When a mutation promoting altruism is passed to offspring, surely those offspring would have an immediate advantage?

Sun, 22 Apr 2012 09:20:32 UTC | #936428

Anaximander's Avatar Comment 27 by Anaximander

Jos Gibbons: That a genetic property of the group is selected. Which can’t happen, for the reasons above.

I didn't mean to say it is selected. Just asking what it means. Because "natural selection" does not mean that nature is selected, but that it is selecting; similarly, "group selection" could (as a term) mean that the group is selecting (a gene). Is group selection possible in that sense? Seems like it could be, because we can define the group in any way we like.

Sun, 22 Apr 2012 10:01:22 UTC | #936433

superatheist's Avatar Comment 28 by superatheist

"selection usually takes place at the level of the individual or the gene" sure but not just at these two levels. It has to also happen at the level of the genome, in the context of genetic networks. I am sort of disappointed in this attempt to explain these natural phenomena, as I have written before on Prof.Coyne's website, natural selection doesn't have to explain everything about an organism. One of the issues that group election tries to explain is adoption. Some primates, birds and insects adopt totally unrelated individuals, that can't possibly increase the altruist's fitness. I suspect it is a by-product of being social, so this debate seems to me to ignore very good explanations and argue based on adaptationist assumptions.

Sun, 22 Apr 2012 12:08:04 UTC | #936446

Jos Gibbons's Avatar Comment 29 by Jos Gibbons

Comment #936428 by Rawhard Dickins

When a mutation promoting altruism is passed to offspring, surely those offspring would have an immediate advantage?

I’m not following you. Could you please elaborate on what you mean?

Comment #936433 by Anaximander

"group selection" could (as a term) mean that the group is selecting (a gene). Is group selection possible in that sense? Seems like it could be, because we can define the group in any way we like.

In analogy with natural selection as you propose, so that the first word refers to the source of the selection pressures, a gene could be favoured because the group provides environmental conditions which favour it over its rivals, so I suppose the answer there is yes. But that’s not what is meant by “group selection” when biologists debate whether it can happen.

Comment #936446 by superatheist

"selection usually takes place at the level of the individual or the gene" sure but not just at these two levels.

How else could it happen?

It has to also happen at the level of the genome, in the context of genetic networks.

What do you mean? Could you talk through an example of how it works?

I am sort of disappointed in this attempt to explain these natural phenomena ... natural selection doesn't have to explain everything about an organism.

Which doesn’t necessarily mean group selection explains that which other forms of selection don’t.

One of the issues that group election [sic] tries to explain is adoption. ... I suspect it is a by-product of being social, so this debate seems to me to ignore very good explanations and argue based on adaptationist assumptions.

One way adoption could be “a by-product of being social” is as what is called a spandrel, i.e. an effect of a gene that is favoured for other effects it has rather than this one. Presumably both that style of explanation and a suggestion of how the adoption itself could be beneficial would count as “adaptationist” approaches. So too, presumably, would group selectionist explanations. Let’s put aside for the moment semantic disagreements regarding which of these will “count” as adaptationist. At any rate, while it is true some things may not be worth construing as the product of adaptations in any sense$, this still doesn’t make a case for the occurrence of group selection. And for group-selectionist explanations to be “very good” requires them not just to sound plausible intuitively, but for group selection to actually be possible given the maths of population genetics, and it isn’t. I previously issued an empirical challenge for situations where group-selectionist predictions actually work out, and it’s never been achieved. And this isn’t an argument of mine based on “adaptationist assumptions”; neither data nor theory are compatible with group selection.

Sun, 22 Apr 2012 12:39:40 UTC | #936449

Arthur Noll's Avatar Comment 30 by Arthur Noll

Jos Gibbons: You aren't addressing what I've said. Mice do not live by cooperating with each other as a super organism. The fact that this gene sometimes kills wild populations of mice is interesting, but doesn't say anything about organisms that are required to work together as a super organism in order to live. Price's equation doesn't include a way to include overpopulation problems, that I can see, so it doesn't work with what I'm concerned about. It is too simple for the issues here.

You show the way to getting such a situation of interdependence, when you say that reciprocal altruism can be selected. Dependence on that can be selected. And once this situation of being a highly social creature is there, it is not difficult to see a situation where common behaviors that worked in the original environment, could become maladaptive in a changed environment. But if some had slightly different genetics that allowed them to regroup and avoid these problems, deal properly with these changes, they could survive while the first group failed. How is this not group selection? One group fails, the other does not. The genes held in common in one group are functional, the genes held in common in the other group are not functional.

You ask: "Why is that group selection rather than individuals varying in how well they withstand the environmental conditions in which they find themselves? (Bear in mind the environment of natural selection includes the properties of other individuals.)"

It is group selection because of the specific way in which they withstand the environmental conditions in which they find themselves- cooperative behavior. They are bound together by that behavior. Certainly you can also say they are individuals withstanding environmental conditions, but that gives no information about how they are doing it. Why do you want to avoid words that describe how they are managing this feat?

Sun, 22 Apr 2012 13:01:33 UTC | #936452