How group selection? As I understand it, Red Queen advises an organism that its descendants will be more successful if they are diverse, rather that being a bunch of identical copies. That argument may involve extrapolating more than a single generation into the future—something evolutionary theorists are usually reluctant to do—but it does not seem to involve group selection. As long as your own descendants are diverse, you don’t care that everyone else’s descendants constitute a monoculture—they are the ones that die in the ensuing epidemic, not your folk.
That argument may involve extrapolating more than a single generation into the future
I cringe a bit when I read evolutionary things presented this way. It makes a kind of sense from the perspective of a personified advisor but even so caries a huge risk of confusing people. It’s backwards, dammit, improved reproductive success because of traits that produced desirable outcomes multiple generations in the past.
(I do not think you are confused and I upvoted your comment.)
I cringe a bit when I read evolutionary things presented this way. … It’s backwards, dammit …
I sympathize. I have a related cringe whenever anyone mentions Omega and Timeless or acausal decision theories. At least I hope my cringe is related because I don’t want to think that the people who mention these things are actually confused.
Yes, some arguments against altruism do not apply as well to sex because there is less danger of being invaded by free-riders. But what they have in common is a short-term cost cost to the individual balanced against a long-term danger of the whole group losing. Group selection is an amorphous word and I don’t think it’s terribly important if we label any particular example with it, but I think there is a great danger of using it to mean “arguments I disagree with.”
Group selection is an amorphous word and I don’t think it’s terribly important if we label any particular example with it, but I think there is a great danger of using it to mean “arguments I disagree with.”
I hope I am not doing that thing you call dangerous. But it is not quite true that careless labeling is harmless to a rationalist. Recall that this entire discussion took off when Phil cited a paper dealing with sex (selfing). He called it a paper about group selection whereas I preferred to call it a paper about species selection. What difference does it make how you label it? Well, please recall the reason Phil cited the paper. He called it the best evidence yet that group selection works. And he did so in response to EY’s anti-group-selection interjection where the focal example had nothing at all to do with sex or speciation.
So is the paper Phil cited evidence against EY? I suppose it depends upon whether Phil’s paper is about group selection or not.
And he did so in response to EY’s anti-group-selection interjection where the focal example had nothing at all to do with sex or speciation.
The focal example was foxes (a species) adapting to control their eating of rabbits to avoid exterminating them. That’s species selection.
His key objection is mathematical: “Specifically, the requirement is [C < FB] where C is the cost of altruism to the donor, B is the benefit of altruism to the recipient, and F is the spatial structure of the population: the average relatedness between a randomly selected organism and its randomly selected neighbor.” This applies equally to clades, species, and smaller groups. So if it’s a knockdown argument against small-group selection, it’s also a knockdown argument against species selection, which exists.
And, as I pointed out in my post, the problem with that analysis is that it assumes that there is no selection of groups. It’s arguing against a strawman “group selection” theory that has no selection. The argument thus both fails analytically, and is disproven by an example that it applies to.
What if some of our cognitive biases are evolved adaptations that make human society work better? It would be pretty surprising to me if this weren’t the case!
No foxes, no sex, and no species selection in what I was talking about. Edit: Inflamatory and non-responsive comment deleted.
How group selection? As I understand it, Red Queen advises an organism that its descendants will be more successful if they are diverse, rather that being a bunch of identical copies. That argument may involve extrapolating more than a single generation into the future—something evolutionary theorists are usually reluctant to do—but it does not seem to involve group selection. As long as your own descendants are diverse, you don’t care that everyone else’s descendants constitute a monoculture—they are the ones that die in the ensuing epidemic, not your folk.
I cringe a bit when I read evolutionary things presented this way. It makes a kind of sense from the perspective of a personified advisor but even so caries a huge risk of confusing people. It’s backwards, dammit, improved reproductive success because of traits that produced desirable outcomes multiple generations in the past.
(I do not think you are confused and I upvoted your comment.)
I sympathize. I have a related cringe whenever anyone mentions Omega and Timeless or acausal decision theories. At least I hope my cringe is related because I don’t want to think that the people who mention these things are actually confused.
That’s generous of you. I know you are particularly wary of TDT, etc. :)
Yes, some arguments against altruism do not apply as well to sex because there is less danger of being invaded by free-riders. But what they have in common is a short-term cost cost to the individual balanced against a long-term danger of the whole group losing. Group selection is an amorphous word and I don’t think it’s terribly important if we label any particular example with it, but I think there is a great danger of using it to mean “arguments I disagree with.”
I hope I am not doing that thing you call dangerous. But it is not quite true that careless labeling is harmless to a rationalist. Recall that this entire discussion took off when Phil cited a paper dealing with sex (selfing). He called it a paper about group selection whereas I preferred to call it a paper about species selection. What difference does it make how you label it? Well, please recall the reason Phil cited the paper. He called it the best evidence yet that group selection works. And he did so in response to EY’s anti-group-selection interjection where the focal example had nothing at all to do with sex or speciation.
So is the paper Phil cited evidence against EY? I suppose it depends upon whether Phil’s paper is about group selection or not.
The focal example was foxes (a species) adapting to control their eating of rabbits to avoid exterminating them. That’s species selection.
His key objection is mathematical: “Specifically, the requirement is [C < FB] where C is the cost of altruism to the donor, B is the benefit of altruism to the recipient, and F is the spatial structure of the population: the average relatedness between a randomly selected organism and its randomly selected neighbor.” This applies equally to clades, species, and smaller groups. So if it’s a knockdown argument against small-group selection, it’s also a knockdown argument against species selection, which exists.
And, as I pointed out in my post, the problem with that analysis is that it assumes that there is no selection of groups. It’s arguing against a strawman “group selection” theory that has no selection. The argument thus both fails analytically, and is disproven by an example that it applies to.
You must be thinking of a different EY interjection than I was when I wrote that. I meant this EY comment:
which responded to this comment of yours:
No foxes, no sex, and no species selection in what I was talking about. Edit: Inflamatory and non-responsive comment deleted.