As you rightly note, language is a shifting changing thing that evolves over time. Sheard’s The Words We Use provides a wonderful (if rather dated now) account of linguistic shift. Grammar likewise evolves over time. And the “inkhorn controversy” rules that arose out of the rediscovery of Latin and Greek were, inevitably, unhelpful and absurd because they fundamentally misunderstood not only the nature of language but also, as you correctly point out, the origins of English in Germanic (technically Geat) roots.
At the same time, if there are no rules at all, then language ceases to function as a means of communication. When we can’t differentiate between girls and girl’s, we lose the capacity for clarity. When we mistake too for to (or two) and don’t know the difference between their, there, and they’re then we risk incoherence. Sure, sometimes context enables us to determine what likely was the writer’s intention, but… sometimes not.
So there’s a balance between change and conservation, between anything goes and agreement on the basics. The fact that this balance is constantly shifting is one of the things that makes the study of language so interesting!
As for the passive voice, that was the result of a desperate attempt by some practitioners of science to appear impartial by removing the “human element” from accounts of experiments. But it was always bogus and today, with empiricism firmly established, it’s mostly been abandoned — at least in non-US countries. So the test tube no longer needs to have been placed over the Bunsen burner; I can simply put it there myself.