As a Labrador swam by me out to sea his owner said I hope he doesn't meet a seal

It always fascinated me that particular behaviors, like herding, can be so ingrained to a particular breed of dog. The dog is no longer in a setting where this is crucial to their survival, yet the urge exists. I do wonder for how many generations the behaviors would last, assuming the dominant genes were not surpressed. That is of course assuming genes are the factor that drives it. It's almost as if environment has little to do with the behavior in this case, other than having opportunity to exhibit the behavior.

I was flabbergasted when I learned that herding dogs have the instinct to bite behind the legs, even of humans. It is a pressure tactic to make the herded animal go in a certain direction.

To me this is an (unpopular) argument against the tabula rasa theory of humans.

If such a complex behaviour can be congenital, who knows what behaviours are congenital in humans.

bite behind leg doesn't sound so complex.

bite behind leg if multiple animals going towards X but animal A goes towards Y and biting will make A go towards X would be complex.

bite behind leg is simple and crude and by placing dog in right context produces complex and useful results.

It still requires solving the symbol grounding problem. How does DNA code for the brain's network weights that correspond to things like the definitions [non-prey target], [goal location], [incentive], [coerce] and [back of leg], or some other suitable set of concepts?

Birds building elaborate nests is something I can't wrap my head around. How do you encode that in DNA, and have a (comparatively tiny) brain execute such complex social behavior?!