Hasn't there been significant discussion on this topic that noted that Eliezer's definitions and conclusions were rather out of date on this topic?
Not really outdated.
Eliezer is trying to cure people of defaulting to a group selection that seems to be, in the imagination of the regular educated audience, stronger than individual organism (let alone gene!) level selection.
I'm kind of tired of this discussion since I've read a lot of stuff on the subject and while in my opinion Eliezer may be wrong on a minor point or two or has perhaps miscommunicated a few more, people are spending a lot of time nitpicking something that is to the first approximation accurate and works great as a cure. A few of them are just being good LWers and do so in good faith but a few others seem driven by a desperate desire to rescue as much of the good fuzzy feeling group selection can bring.
I apologize if this is too harsh, but I felt it needs to be said, especially now that we have a large influx of new LWers (I wouldn't go nearly as far but a few have even spoken of an Eternal September ), some of which are not yet even near to being as less wrong on this as Eliezer's original article is.
Note: I say this as someone who had a quite good (and not totally uninformed) opinion of group selection at one point.
Not really outdated
After his posting in September, I can't agree with that. He flat out stated that "group selection cannot exist". For a group of people who are supposed to demand that their beliefs pay rent, and that we not attribute weight to beliefs based on who they come from, defending Eliezer in this case is just not justified. I mean, the whole point of the SC/SI selection example was that it was a group-selection effect happening with groups of species. The same mechanism happening there would also operate with tribal/colony units.
Side note: It is always September on the internet.
but I don't think Eliezer has changed his position.
He hadn't as of a month ago. Quoth Eliezer: "Thus demonstrating that a sort of species selection can exist, because species can't breed with each other and therefore can't be infected by SC genes from other species, while group selection can't exist, SC always infects."
(For those who didn't follow the link: the topic at hand was that of how it is that self-incompatible Nightshade plants -- that is, Nightshades that have a very complex genetic mechanism preventing them from fertilizing themselves -- continue to exist when self-compatibility/fertility allows for rapid growth of a population and diversification. The answer is that the various SI species are less likely to fixate on 'bad' genes, and thus be less vulnerable to disease. So while the individually selfish SC strategy yields short-term gains, the long-term gains of the SI-cooperating species causes them to collectively persist. The cheaters lose and the cooperators win. As a group. This gets more interesting when you consider that with species this close, cross-fertilization / hybridization has a relatively high fertile-offspring rate, meaning that there's actually gene flow across species.)
The "species cannot breed with each other" thing seems like an argument from definitions on Eliezer's part, because while there are often genetic, fertility, life-strategy and longevity issues with hybrids, they're not all sterile, and they're not all maladaptive. Indeed, there are some situations where hybridizing at the expense of the "pure" populations turns out to be an adaptive strategy. This can even be done in the lab with fruit flies.
Wild hybrids of fin and blue whales aren't uncommon at all (we don't know much about their lives as yet, though we've recorded them for quite a while now). Coyotes and wolves interbreed so frequently that some genetic analyses indicate that red wolves may all constitute coyote hybrids, implying the entire species "canis rufus" originated as a cross. Wolf-dog hybrids are quite common in the US. Beefalo and Zubron hybridize domestic cattle and wild bison, and are perfectly viable (and common in agriculture). Wild populations of grizzly and polar bears seem to produce viable hybrids; a few are known both as wild encounters and captive-bred. At least some camelid and feline hybrids are also reproductively viable. There are many, many interspecific bird hybrids.
Today's post, The Tragedy of Group Selectionism was originally published on 07 November 2007. A summary (taken from the LW wiki):
Describes a key case where some pre-1960s evolutionary biologists went wrong by anthropomorphizing evolution - in particular, Wynne-Edwards, Allee, and Brereton among others believed that predators would voluntarily restrain their breeding to avoid overpopulating their habitat. Since evolution does not usually do this sort of thing, their rationale was group selection - populations that did this would survive better. But group selection is extremely difficult to make work mathematically, and an experiment under sufficiently extreme conditions to permit group selection, had rather different results.
Discuss the post here (rather than in the comments to the original post).
This post is part of the Rerunning the Sequences series, where we'll be going through Eliezer Yudkowsky's old posts in order so that people who are interested can (re-)read and discuss them. The previous post was Beware Stephen J. Gould, and you can use the sequence_reruns tag or rss feed to follow the rest of the series.
Sequence reruns are a community-driven effort. You can participate by re-reading the sequence post, discussing it here, posting the next day's sequence reruns post, or summarizing forthcoming articles on the wiki. Go here for more details, or to have meta discussions about the Rerunning the Sequences series.