A discussion on genetics, evolution, and information with Richard Dawkins

It’s a good conversation. It mostly focuses on River Out Of Eden and the ideas within that book. I know Richard has a new book out, but it’s refreshing here that he takes a deep dive into his past writings. While he touches upon atheist arguments, most of the conversation concerns Darwin, evolution, the genetic code, information theory, computers and function. This, I think, is where Dawkins is at his finest — talking about evolution. There’s a great moment when Dawkins is talking about the genetic code and machine code and Perry pulls out a strip of computer tape! [JAC: this happens at 12:48.] A great illustration of the ideas discussed!

Because of my past as a working biologist, I found the discussion of biology (sexual selection, brood parasitism, etc.) more interesting than the long discussion of code, the genetic code, information, and so on.

I enjoyed the section about whether animal signals evolve via genes that improve “cooperation.”  Whether you answer this “yes” or “no” depends on how you conceive of “cooperation”.  If you mean that cooperative signals evolve even though they reduce the fitness of the replicators within populations (i.e. cooperation as pure altruism), there’s no way that cooperation can evolve by individual selection (more accurately, by differential replication of genes among individuals in a population). Remember, you have to include kin and reciprocity when dealing with the evolution of cooperation within a population.

Most biologists think that the vast bulk of cooperation in animals evolves in a way that increases the fitness of the cooperators in a population. It confers an individual advantage to cooperate. You scratch my back and I’ll scratch yours, ergo you give excess food to your fellows so long as they remember to give excess food to you when you need it. Lions in a pride can gain advantages by cooperating in a hunt by being able to get more per capita food by being able to bring down larger prey or by being more successful at catching prey.

If you want a general increase in cooperation that does not enhance the fitness of individuals, you’ll have to posit forms of group selection.

I know of no examples of cooperation in animals—including any evolved cooperation in primates like ourselves—that cannot be seen as having evolved by individual (or genic) selection. Such examples, to be convincing, would have to show that while they may increase the longevity or “splittability” of a group, would have to reduce the fitness of the cooperators themselves, even when you include their kin. Some aspects of social insect behavior might conform to a group selection model, but recent work refuting such suggestions by Martin Nowak and his colleagues suggests this isn’t the case. At this point we can say that evolutionists know of know adaptations in organisms that must have evolved by group rather than “individual” selection. In the last chapter of my book on Speciation with Allen Orr, however, we describe how some evolutionary trends might be due to a form of group selection, but these are not features or behaviors of individuals.