Is It Really Possible to Know Everything Down to the Foundation?

There's stark differences between fields and areas of how strong this connection needs to be to innovate. I think the vast majority of fields are in this more like sparse graph, less hierarchical structure. It makes me wonder how many mistakes are out there that we just don't know about in mathematical papers because we don't know whether they all like actually are consistent right down to the bottom.