Page 1 of 2

Haidt vs. lots of people on group selection (Ch. 9)

Posted: Fri Jul 13, 2012 1:20 am
by Dexter
(I thought it'd be better to have a descriptive subject rather than putting it in the chapter thread)

In this chapter, Haidt makes a case for group selection in explaining the evolution of some human morality. As he says, the concept of group selection has been all but banished from evolutionary biology.

Just recently Steven Pinker wrote an essay criticizing group selection. There are also comments from Haidt, Dawkins, and others. It's quite a long collection, I've only read some of it. (Recently Dawkins went off on E.O. Wilson in a review of his book, as Wilson is also trying to revive group selection.)

http://edge.org/conversation/the-false- ... -selection

I thought Haidt was pretty persuasive in arguing it as a common-sense idea, but Pinker explains why it doesn't really add anything to the gene/individual selection theory.

After reading some of this, I sort of lost the reason why the debate is that crucial, it seems they are arguing semantics at some points (this is part of the problem, says Pinker). On the other hand, nailing down the mechanisms probably is important.

Re: Haidt vs. lots of people on group selection (Ch. 9)

Posted: Fri Jul 13, 2012 10:14 am
by Robert Tulip
I find this group selection debate fascinating. The question is to what extent is the unit of natural selection the gene, the organism or the group?

Wilson gets excoriated, for example in a recent New York Review of Books article, for his advocacy of group selection.

But group selection is obvious. England conquered Ireland because as a group the English were more adaptive than the Irish. This is unpalatable but true, and applies to all examples of group conflict. When members of a group work as a team and a unit, they are more adaptive than an uncoordinated opposition.

I'm probably being unfair, but I get the impression that opponents of group selection have a mythic attachment to the modern capitalist cult of the individual, and just dislike the idea of subordination to the collective.

Re: Haidt vs. lots of people on group selection (Ch. 9)

Posted: Fri Jul 13, 2012 10:48 am
by Dexter
Robert Tulip wrote: But group selection is obvious. England conquered Ireland because as a group the English were more adaptive than the Irish. This is unpalatable but true, and applies to all examples of group conflict. When members of a group work as a team and a unit, they are more adaptive than an uncoordinated opposition.
It is obvious that groups matter, but Pinker's and Dawkins' argument is precisely that this does not amount to group selection in any meaningful evolutionary sense.

As Pinker says in his reply to critics in the above link, "Group selection has become a scientific dust bunny, a hairy blob in which anything having to do with "groups" clings to anything having to do with "selection.""

From Pinker:
In the real world, success in group-against-group competition is not primarily determined by differences in within-group empathy and kindness but by differences in ideology, technology, military strategy, and organization coerced by brutal discipline, none of which need be costly to the individuals who implement them.

Re: Haidt vs. lots of people on group selection (Ch. 9)

Posted: Fri Jul 13, 2012 3:40 pm
by geo
Robert Tulip wrote:I find this group selection debate fascinating. The question is to what extent is the unit of natural selection the gene, the organism or the group?

Wilson gets excoriated, for example in a recent New York Review of Books article, for his advocacy of group selection.

But group selection is obvious. England conquered Ireland because as a group the English were more adaptive than the Irish. This is unpalatable but true, and applies to all examples of group conflict. When members of a group work as a team and a unit, they are more adaptive than an uncoordinated opposition.

I'm probably being unfair, but I get the impression that opponents of group selection have a mythic attachment to the modern capitalist cult of the individual, and just dislike the idea of subordination to the collective.
I'm not familiar with the controversy per se, but in The Selfish Gene, Dawkins himself says that selection pressures work on different levels. Remember, the Necker cube metaphor? Probably the controversy is over some of the more esoteric details.

But it occurs to me after reading Robert's post that group selection may work almost entirely on a cultural level. England's flexibility was surely not genetic-based, was it? It's the cultural differences that meant the difference. However, just as many genetic mutations are random, so are many cultural mutations, so England's more adaptive attitude may have been a completely random development. Oops, I think I just brought up "memes."

Re: Haidt vs. lots of people on group selection (Ch. 9)

Posted: Fri Jul 13, 2012 9:50 pm
by Robert Tulip
Dexter wrote:Steven Pinker wrote an essay criticizing group selection. There are also comments from Haidt, Dawkins, and others. It's quite a long collection, I've only read some of it. (Recently Dawkins went off on E.O. Wilson in a review of his book, as Wilson is also trying to revive group selection.) Pinker: http://edge.org/conversation/the-false- ... -selection
I read Pinker's essay, and while I would normally wish to defer to his authority as a brilliant scientific intellectual, in this case we have to note that EO Wilson, also one of the most brilliant modern scientists, stands up for the opposing view, so there is no simple scientific consensus on group selection.

Knowing that fools rush in where angels fear to tread, I nonetheless think this question of whether group selection is a meaningful evolutionary concept is worth debate within the generalist philosophical capacity we have here at Booktalk.

As a successful superorganism evolves, it provides the niche that governs the evolution of the successful organisms within it. Part of this niche involves the advantage conferred by conformity to the united needs of the group. Evolution will tend to smooth off any 'rough edges' within a successful group, just as a river smooths the cobblestones in its bed. Diversity and conflict within the group will mainly emerge when the group is not growing and is preparing for a phase change.

In politics, this means that when growing empires clash, they will each tend to favour individuals who embody the ideals of the empire, and repress those who do not. Part of the subtext in the scientific debate on group selection is the atheist critique of how the empires of the west have demanded conformity to Christian belief as part of their unity strategy in the conquest of the world. We see this pressure for group conformity at work in the American military with the widespread doubts about separation of church and state.

Dawkins and Pinker see Christian dogma as an example of group conformism that is an evolutionary negative, simply because Christianity promotes so many factual errors and is a brake on knowledge and reason. They extrapolate from this critique (quite invalidly) to the view that because current religion is wrong, therefore all religion is wrong. They dispute the whole idea of group selection, in large part because as atheists they see group selection as justifying religion, but as atheists they are dissidents against the group to which they belong, the 'Christian West', rejecting the group identity conferred by religion.

Scientific atheists would not like to admit that their own actions and views might be maladaptive in some way, but I think the issue here turns on whether religion as such is adaptive in a group sense. Specific religious views are often maladaptive. Atheism as such is not maladaptive, simply because conventional Christianity is headed to the big cliff called human extinction, and the group evolution of dogmatic conformity in this case is a recipe for destruction. But the atheist idea that religion is obsolete seems to be an effort to deny group selection, and therefore to be maladaptive, a blind alley. This is at the root of why I have argued over the years that atheism should promote the mutation and evolution of religion, not its extinction.

Group selection works where the environment favours group growth. When the environment changes, group conformity produces severe inertia that prevents adaptation to a new situation. Human group selection is now needed more than ever given the new planetary context requiring global ecological regulation. Unless humanity starts acting as a group fast, civilization will be destroyed by climate change. We have the resources for successful group adaptation using rational morality. So it does not help to see Haidt pressing wrong ideas about the relation between reason and ethics. Unless we base morals on logic as a basis of human group identity we are doomed.

Re: Haidt vs. lots of people on group selection (Ch. 9)

Posted: Fri Jul 13, 2012 10:47 pm
by DWill
Do successful football teams win because of group selection? Did Roger Federer win Wimbledon under individual selection, whereas whoever won the doubles titles won under group selection? These may seem to be absurd questions, but they might indicate that saying a temporarily successful group, whether a nation or a sports team, was selected in the same way that an animal with a particular mutation was selected by the environment to breed and pass on offspring, could be meaningless. I think I side with Pinker.

Re: Haidt vs. lots of people on group selection (Ch. 9)

Posted: Sat Jul 14, 2012 11:44 pm
by Dexter
Now that I've read some critiques of group selection, some of the material looks a bit sloppy scientifically, but I'm trying to keep an open mind, and I still think there are a lot of valuable insights there.

By the way, here is Dawkins' review of Wilson, criticizing group selection. Wilson replied below, unfortunately didn't really address the controversy.

http://www.prospectmagazine.co.uk/magaz ... n-species/

I wouldn't want to suggest that Dawkins represents the last word, but I think proponents of group selection are often taking a plausible sounding theory, and not really thinking it through in terms of evolution.

Re: Haidt vs. lots of people on group selection (Ch. 9)

Posted: Sun Jul 15, 2012 4:21 am
by Robert Tulip
Thanks Dexter, always a pleasure to read Dawkins.

Let me explain further how I see this, recognising that I am somewhat ignorant about biology and eager to be instructed. I have read several of Wilson’s books, and like him a lot. He introduces something of a metaphysical quality to the science of the superorganism with his theories of sociobiology and group selection in a way I can well imagine gets Dawkins’ goat.

Dawkins explained, iirc in The Selfish Gene, that the three factors of a successful evolver are that it is fecund, stable and durable. This framework helps to show why evolution provides a successful explanatory tool that goes beyond genetics, into technology and culture. It is about the natural causal laws of complex living systems. Technology and culture are living systems that evolve, with the more fecund, stable and durable examples out-competing the less, due to the material operation of physical causality.

The evolution of a group may not be simply genetic in the Hamiltonian kin sense, as are neither technology nor culture, but that does not mean a group does not evolve by a process that we can call group selection. Is Dawkins trying to unduly restrict the concept of evolution, even though as he himself argued with the theory of the meme it applies more broadly than to simple genetics?

A successful group, whether in culture or nature, has the evolutionary qualities of being fecund, stable and durable. These qualities in the group are not the same as in the individuals that comprise it: the whole is more than the sum of the parts. Adapting to the group need provides a real factor for individual success, such that it makes little sense to say that the existence of kin selection exhausts natural evolution and means that group selection does not exist.

Dawkins demands the elegance of reduction, where the hypothesis of group selection indicates that evolution operates at irreducible levels of entities comprising many genetic individuals. Irreducible is a hot button word, but the associations commonly drawn with supernatural arguments are out of place - here it just means that groups evolve in a way characteristic of groups, and which cannot be explained just as the sum of the individual genetic evolution of its members. The identity of the group as an entity is a main factor influencing which genes within it will succeed.

That is just how I see it, perhaps intuitively. I would welcome being informed as to why this is wrong according to the 140 scientists, or if I am missing something essential. Dawkins notes the dangers of invoking authority in such a debate, but still I’m reminded of what Einstein said in a similar situation - "if I were wrong, then one would have been enough".

Re: Haidt vs. lots of people on group selection (Ch. 9)

Posted: Sun Jul 15, 2012 6:54 am
by DWill
Dexter wrote: I wouldn't want to suggest that Dawkins represents the last word, but I think proponents of group selection are often taking a plausible sounding theory, and not really thinking it through in terms of evolution.
You have a good way of cutting to the chase, and that brings the whole thing into clarity for me.

Re: Haidt vs. lots of people on group selection (Ch. 9)

Posted: Sun Jul 15, 2012 7:30 am
by DWill
Robert Tulip wrote:
The evolution of a group may not be simply genetic in the Hamiltonian kin sense, as are neither technology nor culture, but that does not mean a group does not evolve by a process that we can call group selection. Is Dawkins trying to unduly restrict the concept of evolution, even though as he himself argued with the theory of the meme it applies more broadly than to simple genetics?

A successful group, whether in culture or nature, has the evolutionary qualities of being fecund, stable and durable. These qualities in the group are not the same as in the individuals that comprise it: the whole is more than the sum of the parts. Adapting to the group need provides a real factor for individual success, such that it makes little sense to say that the existence of kin selection exhausts natural evolution and means that group selection does not exist.
I'm having a here-we-go-again moment, Robert. I think Dexter put it best and most succinctly. For me, too, the evolution of social groups is analogous to that of individuals in certain ways. But once you note the differences between the two cases, it becomes very difficult to maintain that there is anything here besides analogy. We see similarities due to the fact that both individual creatures and human groups are things that change. As far as the criteria we would use to peg success for each case, and an identical mechanism underlying each, we seem to be at sea. Dawkins' three criteria, applied in isolation to social groups, don't prove that our social groups undergo anything like natural selection. It's important to keep in mind that evolution has scientific meaning only in that it has to be identical to 'natural selection.' Otherwise, we simply have a fancy word designating things that change.

This is not to say that there couldn't have been some selection for groups in our early evolution. I don't know enough about that evidence, and I haven't read the chapter where Haidt defends group selection. But it seems fantastic to say that Britain's success at such and such a time is natural selection at work. Both Britain and its supposed success are arbitrary human constructs.