Sorry for the essay, I realise that the point I'm trying to make was also said by
@flannan and
@Hrien but hopefully this is a little more convincing.
There is no such thing as improper use. There is only real language as spoken by native speakers.
Nope. Language is usable exactly and only because
rules make it interpretable.
I absolutely agree that "there is no such thing as improper use" is false, and that there are rules of English. For example, there's the rule of SVO order, which says that the subject goes before the verb and the object goes after the verb in a sentence. So "The cat bites the mouse" would be correct, whereas "Bites the mouse the cat" wouldn't.
Another example is the rule that if you're talking about an action that happened in the past, you have to put the verb in past tense. So for example, "Yesterday I baked a cake" is correct, while "Yesterday I bake a cake" isn't (despite being perfectly understandable). I imagine both of the above "rules" I listed are blindingly obvious though, to the point that you might not even think of them as rules, but they certainly are rules that every proficient speaker of English would follow.
But where do these rules of English come from? As a native English speaker, I certainly didn't learn those rules of English grammar from a textbook or a dictionary, or a kindergarten teacher. And neither did any of my other classmates, but yet we still follow the above rules all the time. What happened, of course, is that we learnt the rules of English as we learnt the language from our parents, our teachers, and our surroundings. (You could even say that learning the language is the same as learning the rules.) And similarly, our parents, and the people of the previous generation, they learnt English from the generation before as well. And so on, all the way back through the history of English.
Now, if none of us or our ancestors learnt English from textbooks, then how is it that our language lines up so well with the rules of grammar given in textbooks (e.g. ones for second language learners) and with the definitions given in dictionaries? Obviously, that's thinking about it the wrong way round. Rather than native speakers learning from books, it's the books that copy from the native speakers. The authors who wrote rules like "Whom is used to refer to the object of a verb" were observing how the word "whom" was used in native speech and writing.
While having a book as a reference to check the rules of grammar is often helpful, books suffer from a great flaw, which is that they cannot change with time. You see, English changes (and has changed) over time, since the English that children learn is never an exact copy of the English of their parents. This can be down to many factors, such as children wanting to sound different from the older generation to fit in better socially, or by non-native speakers carrying over features from their native language. Language change occurs even for scholarly English - try comparing a paper written now to one written 100 years ago. A book on grammar will be unable to adapt to changes in what is considered correct usage. (As an example, did you know that 300 years ago, "it's" (with an apostrophe) was considered the correct way to write the possessive form of "it"! See
https://www.grammarphobia.com/blog/2022/11/it-is-or-it-has.html . It even lists a few examples of grammar guides that prescribe "it's" over "its".)
Nevertheless, there are many who follow the rules that dictionaries and grammar guides give, even as the rules become older and older. Eventually, these rules find themselves restricted to fewer and fewer contexts, until they are only used by the most formal or pedantic. For example, the usage of "whom" is rapidly dwindling, and nowadays guides and dictionaries such as Merriam-Webster state that "it is perfectly standard to use 'who'" even in cases where "whom may be technically correct", in reflection of how it is actually used now.
The fact is that the process of researching and writing a dictionary or grammar guide is slow, so a book can never hope to be truly up-to-date, even at the moment it is first published. So the only way to truly know whether something is grammatically correct is to ask a number of native English speakers. Of course, most native English speakers are not consciously aware of all the thousands of rules that make up their understanding of the English language, so it becomes a question of whether something "sounds natural" to them.
"The cat bites the mouse" sounds natural. "Yesterday I bake a cake" doesn't sound natural. And to me, and evidently to many others from as far back as the 1800s all the way up to the current thread we're in, "Recommend me a book" sounds natural.