By interpretation, I mean the fact that we have to step outside of the theory to see that the Godel statement is true. The truth of the statement is about its interpretation. This is something I am never quite sure of.
When I looked this up in my book by A.N. Crossley, I was reminded (having forgotten) that the fact that any theory (with usual conditions) strong enough to include number theory has undecideable statements is the alternative version of the first undecideability theorem. This is what (7) was about, but I miswrote - sorry to have been confusing. Crossley says this version of the theorem is slightly stronger, and I am sure he understands better than I do. So (7) is not needed for the first version of the theorem (the existence of a true but unprovable statement). I have corrected my last post.
Yes, reportedly Hilbert blew his top. I haven't heard of (7) yet, so this is interesting. By prove that the converse cannot be proved do you mean to prove that if a statement is true then the original system is not necessarily undecideable? What do you mean by "interpretation" in (6)? Are you referring to the Lucas-Penrose Thesis?