TheSelfishGene

Context : SocioBiology

RichardDawkins's famous pop-science books on genetics.

Mainly advancing the case for individual rather than group selection (see UnitsOfSelection) in evolutionary theory.

Often held up and vilified as the worst offender in promoting genetic determinism and SocioBiology

But I'm amazed by this. I'm not sure many people who talk about The Selfish Gene have actually read it. Or at least not the second edition from the 80s.

Here are three things I remember Dawkins saying in the book that undermine this public perception.

1) He explicitly says genes are less like a "blueprint" than like a "recipe". By which he explicitly and fully acknowledges that translating genetic information into the phenotype is messy / stochastic process that depends on the development scaffolding.

I've lost count of the number of times people say things like "the Dawkins Blueprint model of genes is wrong because of the openness of the development process". Dawkins says exactly this in the book.

2) He advances a "functional" demarcation of "genes". He doesn't say that a gene is a particular section of DNA. It's all the things that give rise to the phenotype. Take, for example, a "gene for blue eyes". If this depends on dozens of scattered bits of DNA at various locations, and if some of these are dependent on DNA elsewhere. Or there are multiple alternative places the information is encoded so that if one fails another succeeds, then taking Dawkins's notion of a functional definition seriously, ALL of this is "the gene for blue eyes". The gene is superpositional across all of it.

Now you can push back and say this makes "genes" pretty tautological. And that might be valid. But if you take Dawkins seriously, many of the stories you read that allege that "this new discovery of the extra complexity that codes for X undermine the selfish gene theory", don't actually do any such thing. In fact, while I don't know if Dawkins himself would go this far, a "functional" definition of genes could easily expand to including "epigenetic" (ie non-DNA) components of the transmission of phenotypes.

The tautology question might become even more of an issue at this point. Because as you relax claims about what "genes" are, their predictive value diminishes. At some point you only have features of the phenotype and some indefinite notion of "inheritance". I'd guess Dawkins doesn't go all the way to that end of the spectrum. But the idea that he's hard at the other extreme is just wrong.

3) His entire notion of memes and memetics is proposed as a theory of culture which is autonomous of biology. Memes have their own survival criteria which do not reduce to the survival of the individual holding them. This is a huge rupture with SocioBiology which did take culture as a kind of behaviour which was derivative of, and determined by, the the survival of the animal.

In other words, it would be reasonably fair to say that EdwardOsborneWilson implied that cultural beliefs and practices are either justified because they helped the species survive. Or faulty (if they don't). Despite many people lumping Dawkins and Wilson together, Dawkins completely breaks with this idea. For Dawkins of course beliefs and cultural practices can survive simply because their memes are "fit". Which is entirely separate from any claim about their contribution to human survival or well-being. There is no justification for beliefs in terms of their holders. SurvivabilityDoesNotEqualTruth

I'll make one caveat to the above. Possibly some of the ideas I'm emphasizing were more explicit in The Blind Watchmaker than The Selfish Gene. I read both this and the 2nd edition Selfish Gene around the same time in the late 80s. However, even if they weren't in TSG, Dawkins of this period is very different from the image of him I continually see held up in much popular discourse and criticism.