This site is not maintained. Click here for the new website of Richard Dawkins.


← Dear E. O. Wilson: Please retire or stick to ants

kriton's Avatar Jump to comment 31 by kriton

It seems to me that group selection, or selection at the gene pool level, could happen, but only under certain circumstances. It might be that these circumstances only can be found in humans, but perhaps also in some other animals that live in flocks.

First of all, it must be a situation where relevant competing genes do not completely out-compete each other, but instead exist in some sort of dynamic equilibium. For example, genes that promote competition and genes that promote cooperation can exist in an equilibrium. This is because if everybody else competes, those who cooperate have a great advantage, but if everybody else cooperates, it can be a great advantage to be more competitive. Dominant and submissive, extrovert and introvert could be other examples. One hasn't out-competed the other. This could be because there are direct advantages of being submissive or introvert, or because those traits are connected to to other traits that are advantageous. It actually doesn't matter in this case, the important thing is that there is an equilibrium where no traits, or their corresponding genes, are winning a complete victory.

If some gene variants are more advantageous than others in all situations, those will dominate completely in the end. But if it depends on the situation which ones are more adaptive, and the situation changes often enough, there might not be enough time to win total victory.

Second, if there is selection at the gene pool level, it must mean that the mix of genes in one gene pool competes with the mix of genes in another gene pool. But if all the individual genomes in the gene pool have pretty much the same genes, this would really just be competition between individual genomes.

I would argue then that selection at the gene pool level could only happen if there is what we could call a diversity advantage. There must an advantage of several different types of individuals being present in the group simultaneously, compared to all individuals present being similar. And one diversity mix can be more adaptive than another diversity mix.

So group selection can only happen when individuals take on different specialized roles, such as happens in human societies. It would probably have to be quite large groups, maybe like a city-state.

A city-state with some soldiers, some leaders, some traders, some engineers, some skilled craftsmen and so on could be more sucessful than one with only agressive soldier types. If competition only happened on the genome level, the soldier-types would probably dominate, because the other types would be overrun without protection from soldiers. But with that protection, those other types can produce the wealth, the technology, the organization and the motivation that will make their soldiers more powerful and effective than others. This would be a diversity advantage, and that would make group selection possible.

Sun, 22 Apr 2012 14:29:16 UTC | #936462