> But why would we want to throw away the perfectly good C implementation, and why would we expect the C experts who have been carefully maintaining SQLite for a quarter century to be the ones to learn a new language and start over?

Because a lot of language advocacy has degraded to telling others what you want them to do instead of showing by example what to do. The idea behind this is that language adoption is some kind of zero sum game. If you're developing project 'x' in language 'y' then you are by definition not developing it in language 'z'. This reduces the stature of language 'z' and the continued existence of project 'x' in spite of not being written in language 'z' makes people wonder if language 'z' is actually as much of a necessity as its proponents claim. And never mind the fact that if the decision in what language 'x' would be written were to be revisited by the authors of 'x' that not only language 'z' would be on the menu, but also languages 'd', 'l', 'j' and 'g'.