I have said here that the theory of evolution by natural selection makes the simplifying assumption that the variation which underpins selection is un-directed.
I forgot to add that this assumption is very probably untrue - and at any rate unfounded.
And if the underlying assumption is arbitrary, then the theory as a whole cannot be the whole truth.
Does the untruth of the assumption invalidate the theory of natural selection?
I think not. Because this is precisely the kind of simplifying assumption which is made in every specialized branch of human knowledge.
Indeed, that is pretty much what science is: grossly simplified models of reality. Humans (being simple souls) can only use simple theories.
In being a simplified and ultimately untrue theory, natural selection is therefore merely conforming to the nature of all scientific theories.
The problem with natural selection, as with many other specialized forms of knowledge, is knowing when it is sufficiently true to be useful; and when using it is likely to be use-less, misleading or dangerous.
And for this there is no algorithm.
Which is why scientists must be honest, why they must be truth-seekers - because it is only this transcendental impulse which stands between natural selection and infinite error - and the same applies to any science - or philosophy, law, medicine etc.
Humans cannot decide when a theory is applicable and when it is not, experience and investigation cannot decide (because they are framed by theory) - only that bottom-line assumption which stands outside of theory for that person, that culture, can decide and will decide.
That natural selection theory is widely misunderstood and misapplied is true; but the fault does not lie in the specific theory nor in its specific limitations; the fault lies in the way our culture uses theories as such, in the basic assumptions about the nature of things - in the metaphysics of modernity.
Having an incoherent metaphysical basis (nihilism, relativism, secularism, materialism) modernity cannot help but misuse theories.