One of these is not like the others. A document from an archive that lends itself to a particular interpretation is just that. It's evidence that others can verify and reinterpret for themselves.

The other three things you mention don't reach that kind of standard at all.