tradpiper:
I instigated this discussion because of a previous thread ('Bertsongs') in which we discussed deliberate alterations to traditional material by A. L. Lloyd, who was one of the chief instigators (along with Ewan MacColl) of the 1960s folk revival, and who passed his (improved) material along to many prominent artists of the day - to the extent that his versions often became definitive. In the course of that discussion, various contributors asked: "What's so wrong with that, singers have always done it." While I agreed intuitively that singers probably had 'always done it', I was interested to learn what evidence there might be.

I really don't want to get into another argument here about what constitutes "traditional", but in my original question I was thinking about singers who learned songs through their family or community, as opposed to learning them in the context of the folk revival. In the revival we've always taken a degree of meddling as the norm (although the extent of Bert Lloyd's interventions are only now becoming apparent), but in my early days of involvement with traditional song I was given the impression that all those wonderful song variants you find in the published collections and field recordings somehow occurred by accident - by mishearing or mis-remembering on the part of "naive" or "uneducated" singers. I never really bought that, but it's taken me until now to try and find out what actually happened. It's been very interesting to hear what people like Jim Carroll and Tradsinger, themselves song collectors, have to say on the matter, likewise the contributions from someone like yourself who has learned songs in your community. I hope that answers your question adequately.