Let’s call the first idea philosophical genes. When one says that a race horse has inherited great speed and fragile ankles from its parents, one is not concerned primarily with the molecules that effect this inheritance. It could be (and seems to be) that inheriting a trait as complicated as running speed involves a constellation of molecules in the horse’s DNA. Be that as it may, for evolution by means of natural selection to work logically, there must be a mechanism or mechanisms by which a) traits that can be selected are generated; and b) such traits can be passed on.
This logical requirement is prior to whatever mechanisms embody it – we wouldn’t even go looking for physical genes if we didn’t require them philosophically to make sense of natural selection.
I’ve gotten my history of Darwin’s theory after its initial promulgation a bit piecemeal, but if my understanding is correct, this requirement of a mechanism for the production and inheritance of traits was seen as a weak spot, since inheritance itself is tricky: offspring do not always appear to be some 50/50 blend of parental traits, or any other simple formulation. Traits can skip generations, then reappear. Actual inheritance as expressed in traits that can be selected appears too messy to hang an entire revolutionary theory on.
Mendel fixed all that. (Eventually – the entertaining Catholic Encyclopedia article from 1917, less than 2 decades after Mendel’s work was finally brought to light, says: “T.H. Morgan does not hesitate to say that Mendel’s laws give the final coup de grace to the doctrine of Natural Selection, and others consider that his views, if finally proved to be correct, will at least demand a profound modification in the theories associated with the name of Darwin.” See the rabbit holes here? Now I want to find out what T.H. Morgan says – and who the heck is this T.H. Morgan, anyway?) Or sort of fixed all that – Mendel’s theories are tidy for very carefully selects pairs of mutually exclusive traits, which are far from all or even most traits that attract our attention. But at least a simple framework within which to investigate had been established.
Within that framework, we can start to see, however dimly, how the mechanical workings of organic chemistry result in, say, slime mold or birds or Yankees fans. But we’d hardly bother if we didn’t have the logical need, under Darwin, to look into it.
Dawkins concerns himself almost exclusively with philosophical genes. How it is, exactly, that perturbations in strands of DNA produce Darwin’s finches isn’t nearly as interesting to him as the existence of the finches in the first place. The evolutionary biologist needs for there to be a mechanism, but doesn’t necessarily need to know how it works.
In my chicken scratchings here, I likewise focus on the logical fall-out from embracing a gene-centric view of the origin of species. I tend to think it’s the correct way, scientifically and philosophically, to view it. From a gene-centric point of view, predictions can be made that are a bit counter-intuitive – which are the best kind. Key among these is the dizzying perspective of the gene’s eye view, how behaviors and traits that make no sense from the individual organism’s survival can nonetheless make perfect sense.