That makes sense to me and I’m comfortable with redundant agreeing evidence as well (assuming we don’t ignore any contradictory evidence), but my impression is that these fields do not consistently have such a standard. Maybe my impression is wrong? To me it seems like you need to crawl through the dependency chain and verify that reasonable standards were used all the way up. Does the peer review process in these fields actually enforce this? It seems like no, but I am just a spectator.