3 ms·

I love Ed Wilson. But I think it's so odd that his group selection theories get such a fawning treatment by journalists. I mean, he publishes in Nature, and the

I love Ed Wilson. But I think it's so odd that his group selection theories get such a fawning treatment by journalists. I mean, he publishes in Nature, and the next month 140 other evolutionary biologists sign a letter telling him that he's wrong. He claims it's because his new idea is threatening everybody else's livelihoods. But surely it's more parsimonious that when it's 4 against 140 that the 4 are just plain wrong.

jly 7y ago
I'm also a huge fan of Wilson's books, but his most recent two evolution-focused works ('The Social Conquest of Earth' and the new 'Genesis') really double-down on the ideas of group selection. I guess this definitely qualifies as provocative, but it is certainly an extreme minority opinion, as you've pointed out. The writing is as eloquent as ever, but they're not books I would point someone to as a starting point on these ideas. His newfound doubts and critiques of inclusive fitness theory are not shared by the majority of biologists.
User23 7y ago
Isn’t that true of many scientific advances? A claim that everyone else is getting it wrong is going to be met with skepticism initially and sometimes for quite a while. Of course it could just be wrong too.
torstenvl 7y ago
Except that E. O. Wilson was an immense group selection skeptic until his 80s, when he decided he was wrong and threw in with D. S. Wilson's ideas (sometimes called "neo-group selection"). See D. S. Wilson and E. O. Wilson, Rethinking the Theoretical Foundation of Sociobiology, 82 Q. Rev. Bio. 327 (2008). It's not like this is his own pet theory and he's trumpeting it out of ego. If anything, it demonstrates humility about his earlier dismissiveness.
pierrebai 7y ago
Clearly I'm no expert in the field, but I've no understood the objections to group selection. From what I've read, the objector claim that genes get selected and this slection necessarily passes by reproduction which is done on an individual basis. Yet, especially inthe example of ants, two neighbour colonies of ants of the same specie might have a queen with different genes and if the difference grant and advantages the one with the better gene pool with spawn other similar colonies which woudl supplant the inferior one. In each colonies, non-reproductive ants with the better gene don't reproduce themselves but yet it's their better genes that supplant the neighbouring colonies. The queen does no work other than lay eggs, so it does not not contribute the the higher success rate ofthe colonies. Clearly, the behaviour of non-reproducing ants has a direct influence on the gene pool. For example, the better gene might make better warrior ants. They don't reproduce themselves, yet the gene helps the colonies. The same argument could be made of two wolf packs, or any two competing group. Of course, in those case, you can look at it at an individual level, but the better gene will sprad at the group level. (Ignoring if a wolf could switch pack to spread the gene to the other pack. I'm here assuming this doesn't happen for the sake of the argument.) The more social the species, the more the group-level genes can influence the chances of reproductions of those not harboring the gene but part of the group, which make the group more successful, which eventually will make the gene spread in the group through gene exchange and selection during reproduction. For example, if the gene leads to better caring between individuals, it benefits the entire group harboring individual with that traits and eventually the group with the highest percentage of that gene will win out.
- onychomys 7y ago
  It's a little tough to explain without resorting to a level of math nerd that is hard to type out here, but the basic problem with it is that there's too much gene flow between demes (partially isolated intra-breeding populations), and that swamps the relatively weak effects of natural selection. It's part of the reason that almost all speciation is allopatric, because without that isolation, the effects of selection get wiped out by cross-breeding.

jly 7y ago

I'm also a huge fan of Wilson's books, but his most recent two evolution-focused works ('The Social Conquest of Earth' and the new 'Genesis') really double-down on the ideas of group selection. I guess this definitely qualifies as provocative, but it is certainly an extreme minority opinion, as you've pointed out. The writing is as eloquent as ever, but they're not books I would point someone to as a starting point on these ideas. His newfound doubts and critiques of inclusive fitness theory are not shared by the majority of biologists.

User23 7y ago

Isn’t that true of many scientific advances? A claim that everyone else is getting it wrong is going to be met with skepticism initially and sometimes for quite a while. Of course it could just be wrong too.

torstenvl 7y ago

Except that E. O. Wilson was an immense group selection skeptic until his 80s, when he decided he was wrong and threw in with D. S. Wilson's ideas (sometimes called "neo-group selection"). See D. S. Wilson and E. O. Wilson, Rethinking the Theoretical Foundation of Sociobiology, 82 Q. Rev. Bio. 327 (2008). It's not like this is his own pet theory and he's trumpeting it out of ego. If anything, it demonstrates humility about his earlier dismissiveness.

pierrebai 7y ago

Clearly I'm no expert in the field, but I've no understood the objections to group selection. From what I've read, the objector claim that genes get selected and this slection necessarily passes by reproduction which is done on an individual basis. Yet, especially inthe example of ants, two neighbour colonies of ants of the same specie might have a queen with different genes and if the difference grant and advantages the one with the better gene pool with spawn other similar colonies which woudl supplant the inferior one. In each colonies, non-reproductive ants with the better gene don't reproduce themselves but yet it's their better genes that supplant the neighbouring colonies. The queen does no work other than lay eggs, so it does not not contribute the the higher success rate ofthe colonies. Clearly, the behaviour of non-reproducing ants has a direct influence on the gene pool. For example, the better gene might make better warrior ants. They don't reproduce themselves, yet the gene helps the colonies. The same argument could be made of two wolf packs, or any two competing group. Of course, in those case, you can look at it at an individual level, but the better gene will sprad at the group level. (Ignoring if a wolf could switch pack to spread the gene to the other pack. I'm here assuming this doesn't happen for the sake of the argument.) The more social the species, the more the group-level genes can influence the chances of reproductions of those not harboring the gene but part of the group, which make the group more successful, which eventually will make the gene spread in the group through gene exchange and selection during reproduction. For example, if the gene leads to better caring between individuals, it benefits the entire group harboring individual with that traits and eventually the group with the highest percentage of that gene will win out.

onychomys 7y ago
It's a little tough to explain without resorting to a level of math nerd that is hard to type out here, but the basic problem with it is that there's too much gene flow between demes (partially isolated intra-breeding populations), and that swamps the relatively weak effects of natural selection. It's part of the reason that almost all speciation is allopatric, because without that isolation, the effects of selection get wiped out by cross-breeding.