I have just read this article on E. O. Wilson and I don't understand what the difference is between what he is arguing and "standard" natural selection.

I read "the extended phenotype" some years ago and from what I vaguely remember that argued that everything about selection comes down to genes: they are the units of what is inherited and drive everything else (and their reach can be quite surprising). At first glance that seems inconsistent with the article, since it mentions "group selection". But when you read the details it includes things like

Group selection begins when a colony of creatures develops a behaviour that gives it a competitive advantage over other groups. Initially, this could be down to a random genetic mutation.

which sounds like it's gene-based too (and examples seem to come from ants and the like which I thought were closely genetically-related across a colony too...).

So what is the fuss about? And how will it be resolved?

Maybe the difference is easier to explain in terms of some difference between mathematical models? If so, that's fine (I hope - it might make more sense than jargon in a field I am unfamiliar with ;o).

[Is this too "general public" for this site? Apologies if so.]


2 Answers 2


After reading the article, the fuss is about this:

In currently accepted theory Eusociality or "kin selection" explains altruistic behavior (the sacrifice of yourself or resources you control for the betterment of something else besides you) by relating the act to the amount of genetic information passed on.

The relevant equation is Hamilton's rule:

                                                              $rB > C$

C is the cost to the actor (the person giving up resources). B is the number of others involved, and r is the relationship value to the actor in terms of genetic similarity.

So if you and two full-siblings (who each share on-average 50% of your genome) and a bus comes straight at the group, and you have the capability to push them out of the way while dying yourself, kin selection suggests you shouldn't because the genetic cost is equivalent. You eliminate yourself from the gene pool, and save the equivalent of yourself for future generations (2 x 1/2 = 1).

However, if you and three siblings were in the situation, then it's evolutionarily advantageous to be altruistic. You would die (a cost of 1), but you would save the equivalent of 1.5x your genome (3 x 1/2 = 1.5) to pass on to future generations.

Thus, communal and social behaviors like those of ants basically boil down to: Everybody is so closely related that spending resources to help your kin is essentially spending resources to ensure the majority of your genome survives (which is the big point of evolution, after all).

Wilson's argument, what I can glean from the small brief of the study and the article, is that you don't need the relatedness equation to explain altruistic behavior in social situations (basically the row is he's saying the currently accepted theory is wrong). All you need to start with is a genetic mutation somewhere along the line that causes the offspring to stick close to home. Once you have a few generations that don't leave the nest, you start to develop social behavior that results in altruism because you've gone from selfish individuals to a group setting.

The article and Wilson don't give any more insight other than altruism apparently develops spontaneously from the formation of groups. Simply being social is the explanation for why some individuals will give resources to others to ensure the success of the others.

Whether or not Wilson is correct isn't really discussed. He openly admits to needing a lot of research, some of which is being performed.

  • $\begingroup$ Hamilton's rule... :) I feel that the fundamental argument for individual selection is as much ideological as it is scientific. There are quite a few edge cases which both sides explain away I feel. To me the big question is how does this explain human social behavior, which is the brass ring, still ungrabbed by biology. $\endgroup$
    – shigeta
    Aug 18, 2012 at 5:21
  • 1
    $\begingroup$ I started reading David Sloan Wilson's work on group selection a decade ago and am surprised that the field still has such a reluctance to move on. $\endgroup$
    – Abe
    Sep 5, 2012 at 3:40
  • $\begingroup$ @Abe - The duty of scientists is only to move on if better information is presented. I'm not familiar with D. S. Wilson's work, but the current textbooks feature Hamilton's Rule. If Wilson's work provides a better explanation, it will eventually get accepted. :-) $\endgroup$
    – MCM
    Sep 5, 2012 at 3:57

So what is the fuss about?

The fuss isn’t so much about biology as it is about the circumstances of the argument. in particular, I gather that there are two complaints people have with E. O. Wilson’s (and his collaborators’) arguments:

In addition, proponents of kin selection argue that group selection theory simply doesn’t hold up in theory1. I find those arguments highly compelling. At the same time, I’m not an expert and the Wilsonians argue that kin selection is equally flawed so I won’t weigh in on the debate.

1 The article is not written by an expert in the field but it has been endorsed by experts as a lucid explanation of the matter.

  • $\begingroup$ How is the fact that group selection does not hold in theory reconcile with observations / experimental tests? Is there another explanation for inherited religious, social, or political groups? $\endgroup$
    – Abe
    Sep 5, 2012 at 3:44
  • $\begingroup$ @Abe I’m not an expert but as far as I know none of those actually requires group selection. Many phenomena are simply explainable by kin selection. And while we don’t have comprehensive explanations for large-scale phenomena such as religion etc., you can construct potential explanations without invoking group selection. Indeed, just because there’s interaction within and between groups involved doesn’t mean that evolutionary selection happens on the level of groups, just like genetic selection doesn’t happen on the level of individuals, it happens on the level of alleles. $\endgroup$ Sep 5, 2012 at 7:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.