nothing has been proved

But by your own admission, nothing can be proved! So we go on the weight of the evidence - what seems more likely?

Several songs collected by Bert Lloyd seem to follow this pattern:

- Scruffy, reasonably good song appears in sources dated 18-something
- Lloyd produces a really good song which no one has seen before, but which is recognisably related to the earlier song
- Lloyd claims that his song had actually been around for ages, and the earlier song was really a variant of it

What seems more likely? (And again, this has nothing to do with whether they're good songs or not.)