Natural selection is just a term for a class of feedback systems. As such, it does not explain complexity per se. What we do know are the various mechanisms for gene evolution, namely duplication, translocation, inversion, deletion, point mutation, etc. The feedback loop of natural selection acts on the results of the above. Together, they give rise to complexity. We can simulate this sort of activity just fine.
So far, it remains the case that "understanding" the process is pretty much the same as observing it or simulating it. This is not very satisfying compared to elegant equations which we are used to in physics and chemistry.
The same is largely true for climate science. While certain aspects like CO2 trapping heat do have nice simple equations, the dynamic effects of oceans, winds, etc, need to be simulated. We don't "understand" them with a simple set of equations. Never will.
I guess this is not really that different from lots of other fields any more. I remember taking solid state physics in college and grad school, with all those nice detailed transistor models. We "understood" things with a nice ideal view. Today, designing leading edge IC circuits requires all sorts of statistical modeling because the properties are not well controlled. So modelling and simulation are at the core of design, not some nice tidy "understanding". |
