I don't see anything about science in this essay. Philosophically, it may be a valid critique of essentialism, in the meaning that primitive conclusions Dawkins is drawing - as for rabbits or state politics - are indeed looking invalid. But nobody currently advocates such concepts - i.e. nobody really builds a scientific theories based on the fact that rabbits literally represent the "ideal rabbit" and nobody actually thinks all Florida residents are Democrats or Republicans. We may act as if they are, in order to simplify certain things, such as deciding who will be the president or which pills to give to a specific rabbit, but we know they really aren't.
OTOH, if you drop the concept entirely, then you'd need to throw a significant part of modern science out of the window. Modern science bases on the fact that there are some laws of nature, which are universal and fixed, and by doing certain actions and making certain conclusions using certain techniques we can discover these laws and thus discover how ideal Platonic objects would behave, and by reasoning about those objects we could derive the useful conclusions about real world objects. If you reject this method, then you'd have to make a scientific theory anew for each object, which would be kind of hard to make practical. Modeling is necessarily idealization, and if you reject idealization, not much is left of the scientific method. What Dawkins seems to argue is that one should realize the map is not the territory, but isn't it obvious to everybody by now?
OTOH, if you drop the concept entirely, then you'd need to throw a significant part of modern science out of the window. Modern science bases on the fact that there are some laws of nature, which are universal and fixed, and by doing certain actions and making certain conclusions using certain techniques we can discover these laws and thus discover how ideal Platonic objects would behave, and by reasoning about those objects we could derive the useful conclusions about real world objects. If you reject this method, then you'd have to make a scientific theory anew for each object, which would be kind of hard to make practical. Modeling is necessarily idealization, and if you reject idealization, not much is left of the scientific method. What Dawkins seems to argue is that one should realize the map is not the territory, but isn't it obvious to everybody by now?