One issue which is often not taken into account is the extent to which the published material has been editorially mediated. I understand that the Lomaxes published a good many quite heavily collated texts, much of the material included being from unidentified sources. In effect such texts are separate from those originally collected (and may at times bear little resemblance to them) which might add a further layer of complication where questions of intellectual property are concerned. Differing considerations will apply to the raw materials in the collections (whether in the form of audio recordings or paper transcriptions) and to the "processed" materials as they appeared in print.

The Lomaxes were by no means alone in this practice, of course, but editors who fail to specify their own interventions, and to identify the source(s) of introduced material, are failing in their duty both to scholars and to singers; however worthwhile the end result from an aesthetic point of view, it can in many cases be quite impossible to tell if it is genuine or, in effect, a forgery (as witness the problematic nature of John Jacob Niles' materials, and his notable inconsistencies over issues of attribution and authorship).

I'm afraid that this doesn't address the original question materially, but it does raise another aspect of the issue that doesn't seem to have been looked into in the kind of depth that it perhaps ought to be.