Gould vs. Dawkins

It took me many years of reading Stephen Jay Gould and Richard Dawkins to really grasp what their disagreement was. I’ll try to summarize it here.

They both, of course, completely accept the standard story of evolution as descent with modification and natural selection. Their disagreement is about the degree to which species will change over time, and the amount of change that is possible. Gould stresses that species are constrained by the amount of genetic diversity they have available. Dawkins stresses the ability of species to adapt.

Thus Gould talks about the panda’s thumb, which is an example of evolution reusing an existing structure rather than inventing something new and more efficient. He talks about spandrels, which he describes as nonadaptive changes which are linked, genetically or structurally, to adaptive changes.

Dawkins talks about the selfish gene, and treats it as a unit of evolution without considering the other genes which occupy the same body. He discusses the extended phenotype, and argues that you can consider a beaver’s pond to be a part of the genetic inheritance of the beaver.

Who is right? Obviously, they both are. It’s just a question of emphasis and points of view. Sometimes evolution is constrained in nonobvious ways. Sometimes species can adapt freely. Which is the case in a particular scenario can be very difficult to determine.

Their positions are both sufficiently similar and sufficiently distinct to have generated a lot of strong words over the years before Gould’s death. I personally tend to favor Gould’s arguments when it comes to evolution–I think they are more helpful in understanding what we really see in the world. They helped me recover from a strong adaptationist viewpoint I held in my youth–the belief that all animals must be ideally adapted for their environment. On the other hand, Gould loses me completely in his discussion of nonoverlapping magisteria: it’s either very obvious or rather misleading, I’m not sure which.