Mathematical peer review historically relied on social trust in the author's reputation rather than checking every single line of an argument