I may have not made myself clear. It looks like I suggested that numbers, squares, and circles exists somewhere, just not in our world. I didn't want to. The

I may have not made myself clear. It looks like I suggested that numbers, squares, and circles exists somewhere, just not in our world. I didn't want to. The problem comes from the word "exist", which does not mean the same thing when I talk about the physical world and when I talk about mathematical concepts. For instance, when I say "Do unicorns exist?", or "cars do exist", I am talking about actual existence, in our world, such that it could change my life: if unicorns existed, I might see one. If cars did not exist, I would be riding instead of driving. When I say "do odd numbers exists", or "there exists no even prime number greater than 2", I am talking about mere ideas. See, numbers "exist" by definition. In Peano arithmetic for instance, I declare that natural numbers exist because: (1) I declare that zero exist, (2) I declare that if a number n exist, then a different number, n+1, also exists. (There are other laws, but we'll leave them.) In short, numbers "exist" because I decided it. This is not the same kind of existence than unicorns. No matter how hard I decide, unicorns don't exist. Now the fact that everyone agrees that numbers "exist" just means that everyone follows the same convention. I actually agree with David Hume. He's right because of a very simple, now easy to guess, reason: mathematics doesn't even talk about existence of the unicorn/car kind.