>>213627297
>aren't pronouns and numbers pretty rarely borrowed?
When languages already have them, then it is rare (with notable exceptions like Japanese). However, when we look deeper into history, languages in the past often didn't have numbers (basically until they adopted agriculture of some form) or pronouns, and in these cases it's very common to borrow them from languages that already have them, though from what I've seen, with numbers at least, it is unusual for the whole system to be taken from a singular source language.
One good example of why you cannot take pronoun similarity as evidence of a genetic relationship is Proto-Indo-European and Proto-Uralic, which have pronouns (and verb personal conjugations) that strongly resemble each other, as well as in some cases relatively similar suffixes for case and number. This has led some linguists (like the fairly renowned Frederik Kortlandt) to the conclusion that they must be related. This is however unlikely, since structurally PIE looks like it came from the Caucasus and PU came from East Siberia (which is also backed up by genetic evidence). What is more likely is that these languages (or their ancestors) were in contact at some point and borrowed certain grammatical particles and pronouns from each other.
