Truth and Reconciliation for Group Selection VII: If You Make A Mess, Should You Clean It Up?

One memorable Christmas morning, as our kids were gathering around the tree, I was on my way upstairs to get a sweater when I smelled something really bad. I knew that smell. Our cat had diarrhea and had deposited a wet one somewhere. I walked all over the house trying to find it before realizing that I had stepped in it the moment that I smelled it and now had tracked it all over the house.

Did I clean it up? Of course I did. Anyone would.

Here's another example of a mess: Imagine a man who has made a mess of his life. He has taken advantage of those who loved him and piled lies upon lies until he can't keep them straight anymore. Now he has been abandoned and has only himself to blame. If only he could go back to the beginning!

Should this man clean up his mess? Of course he should, and would be much better off if he did, but we wouldn't be surprised if he didn't. It would require great courage. Twelve-step programs are designed for people like that and include acknowledging one's faults and apologizing to others.

Here's a third example of a mess: Imagine that an entire scientific community has made a decision that turns out to be a mistake. The decision was momentous. It was regarded as a watershed event for the field. Its architects were celebrated as heroes. It was enshrined in textbooks. Nevertheless, subsequent events had proven it to be incorrect. If the current information was known back then, a different decision would have been made and the entire field would have taken a different path.

Should the field clean up its mess? It should, but the likelihood that it will is even less than for the individual who messed up his life. After all, "the field" is not even a corporate unit capable of making a decision like an individual. Instead of a collective decision, a cacophony of responses can be expected, including acceptance, denial, and halfway positions that attempt to acknowledge change while still clinging to the patriotic history.

That is what has happened with the group selection controversy. The categorical rejection of group selection in the 1960's was wrong, plain and simple. If they knew then what we know now, it would never have happened. Some evolutionists are perfectly comfortable with this conclusion. Others act as if nothing has changed since the 1960's. Others say that the original rejection of group selection remains valid and what passes for group selection today is different. Others claim that group selection and its alternatives are equivalent, making it a matter of preference which to employ. Others construct models and perform experiments that anyone would have identified as group selection in the 1960's, but just don't use the G-word. In short, the field as a whole is a big mess.

To see what I mean, consider a four-page news feature on group selection that appeared in the November 20, 2008 issue of Nature magazine. The author is a science writer named Marek Kohn who did a thorough job researching the subject. The article begins this way:

If biologists have learned one thing about evolution over the past 40 years, it is that natural selection does not work for the good of the group. The defining insight of modern Darwinism is that selection 'sees' individuals and acts on them through the genes they embody. To imagine otherwise, generations of students have been warned, is to fall into a naïve error definitively exposed as such in the 1960's.

So far so good. But then we learn that the two sides actually agree on a great deal. For example, everyone today supposedly agrees that group selection occurs and that group selection and kin selection are formally equivalent to each other. Evidently, the debate about group selection is largely semantic and choosing between one and the other is a matter of preference.

I will untangle some of these issues in future blogs. The point I wish to make here is that the positions reported in the Nature article bear almost no resemblance to the issues at stake in the 1960's. Back then, the center of the debate was what I have called "The Original Problem (see T&R II)", almost everyone agreed that group selection did not occur (although they agreed it was possible in principle), and kin selection was regarded as a theory that succeeded where group selection had failed. Does anyone seriously think that the "defining insight of modern Darwinism" was merely semantic? Somehow, the issues at stake in the 1960's have been permuted into a different set of issues, while everyone pretends that it is the same controversy. I do not fault the author of the article. As a science writer, the best that Kohn could do was report the views of the experts. The problem is that he had to report a mess.

As for everyone today agreeing that group selection occurs, I wish that someone would tell John Alcock, author of Animal Behavior, the most widely used textbook in the field. Alcock truly believes that nothing has changed since the 1960's. So great is his loathing of group selection that he calls it "non-Darwinian"--an irony, since Darwin clearly originated the concept (see T&R II). Regardless of what the Master thought, Alcock can't even bring himself to say that group selection is a form of natural selection that results in adaptations at the group level, when and if it occurs. Legions of college students have been taught by Alcock that "the overwhelming majority of scientists studying the evolution of animal behavior employ Darwinian theory, rather than group selection theory in any of its forms (7th edition, published in 2001)."

Or how about my colleague, Dr. X, who has been harassed throughout his career for writing about subject Y from a group selection perspective? For example, colleagues refused to publish articles with him unless he removed all references to group selection. Dr. X is a real person but--I'm not kidding--he does not want his identity revealed to avoid further harassment. He recently sent me the following message about a newly published article on subject Y:

It just gives you an idea of what people like me are always up against. The argument in this paper is completely group selectionist but neither the term nor the concept is invoked. Instead, the buzz terms become "coalition building", the formation of "alliances", etc. etc. It's fashionable among the cooperation bunch to talk about coalitions and alliances, but they never come to grips with the levels-of-selection issue... So, things are just as they have been all along.

The problem of packaging old wine in new bottles--ideas that anyone would have associated with group selection in the 1960's, but without using the G-word--pervades the modern literature, as I will show in future blogs.

I don't want to overstate the degree of censorship and persecution that people have suffered by daring to invoke group selection. My career has not suffered, for example, and believe it or not I count George C. Williams as a good friend. Moreover, the biggest tragedy is not injustices to individuals but the fog of confusion that descends over an entire field when major mistakes are not acknowledged.

At the end of T&R II, I stressed that there is an underlying simplicity to the group selection controversy. The only way to recover the simplicity is by cleaning up the mess that was made by falsely rejecting group selection in the 1960's. That is my next task, but I warn you: It smells really bad.

To be continued.

Comments

[RussAbbott]

David,

I'm afraid that I see you as contributing to the confusion. Do you know what you mean by a group as distinct from an individual? As you yourself point out in "Evolution for Everyone" everything is both a group and an individual. There are no individuals that aren't groups -- except perhaps elementary particles. And any group that one can talk about as an entity is an individual in that it coheres and persists as an entity.

It is these group-like individuals that evolve. I don't know of any examples in which arbitrary aggregations of elements evolve. I'm sure you are not suggesting that evolution applies to arbitrary aggregations. But by not distinguishing between groups that are in fact entities and arbitrary aggregations, you are, I'm afraid, adding to the confusion.

A persistent group that operates according to some established internal order/pattern evolves when a somewhat improved pattern replaces it. It's the same for both persistent groups and traditional individuals. Entities (i.e., persistent patterns of interactions among components) evolve as better patterns of interactions become more suited to their environment.

[Ajita]

The debate in evolutionary biology is not as abstract as you express. The individuals and groups concerned are well defined biologically and the problem of group selection vs individual selection is a population genetics problem to be studied by observing gene frequencies over time.

You are simply stating general rules of systems evolution. The debate is about the specific mode of evolution in well-defined biological systems.

[David Sloan Wilson - Huffpost Blogger]

I agree with Ajita's comment (below). Groups are the sets of individuals that interact with each other. Those interactions can be "for the good of the group" or not. Self-serving interactions are locally advantageous. Group selection theory attempts to explain how "for the good of the group" traits can evolve, even when they are locally disadvantageous. When we say that everything is both a group and an individual, we may be right in general systems terms but we are not addressing this particular set of issues. See my next installment (T&R VIII) for how group selection is represented in a formal theoretical model.

[royniles]

Of course it works for the good of the group. The good of the group is good for the individual or there wouldn't be an evolving group. The question you seem a bit reluctant to answer is about the how and why of the matter. You're dragging this thing out as if it's a mystery series. Is it that you;re still trying to make up your own mind on that part of the subject?

[David Sloan Wilson - Huffpost Blogger]

I assure you, I know exactly where I am going with this series of blogs! I disagree with your statement that "the good of the group is good for the individual or there wouldn't be an evolving group." This is where formal theoretical models are useful, as I recount in T&R VIII. In the haystack model, a gene for aggressiveness evolves, even though docility is for the good of the group. Group still exist--they just don't work as well as they could. The world is full of dysfunctional groups. Group selection is about how functional groups can evolve, and how selection within groups can undermine group-level functionality.

[Ajita]

I wasn't aware of the extreme position taken by Alcock in his book.

Also the statement from the Nature article: "If biologists have learned one thing about evolution over the past 40 years, it is that natural selection does not work for the good of the group." seems ridiculous.

Group selection is very active in the evolution of behaviors in groups of species that are genetically isolated. Without genetic isolation, I can see some validity to the counter arguments, but there are many species of insects, fish and plants that form genetically isolated sub-populations, get certain group-specific traits fixed in these sub-populations over time, and gain a selection advantage over other groups because of that trait. Even in the case of human groups that are genetically isolated by culture (religion) although not geographically, such forces may be at play.

From what I remember, it is the origin and spread of traits within a population that has very little possibility of group selection. The maintenance of such traits between isolated populations could very easily be due to group level selection forces.

I confess that I have also ignored group selection, primarily because I was always more interested in speciation, even if evol. bio. is only a hobby. I look forward to the "underlying simplicity" you promise to show us.

[David Sloan Wilson - Huffpost Blogger]

The "underlying simplicity" is contained in the haystack model, which elegantly describes what I called "the original problem" in T&R II. How can "for the good of the group" traits evolve, when they are often not locally advantageous? It's that simple.