Those of us who care to know already know the definition of "Folk music". If you want to look it can be found in my old posts and on MusTrad. That's what the definition is. A definition is a definition - for example the defition of an erg in physics, or a celsius heat unit.

So first do we need to define "traditional" folk music, or are we merely looking for a term to eliminate the confusion that is caused by those who don't like the definition of "folk" music and so use the term incorrectly?

Second, if we do, need to define it, has any authority previously defined it? I would need to go away and dig to be clearer on this, but my first reaction is "Probably no".

Third, if we are to define it, what do we want it to mean and are we in a position to assert a defnition. The answer to the second part of the question is probably "No", but if the (UK) Nats return they will need a defintion for they had and presumably will revive a competition annually for the best unaccompanied performance of a traditional folk song.

One obvious contender for the meaning of the word in context is "of unknown authorship", but that cannot be complete for it conveys no impression of antiquity. However, if a traditional song is of unknown authorship then its age must be unknown too, and I hesitate to insert a phrase lke "reasonably and generally believed to have been composed by an author not still living". Both however would admit at least in theory of there being accretions to the class of "traditional" folk songs and folk music. Conversely, if new research should establish the author of a piece then the piece might cease to be traditional.

The mere fact that some results of applying a definition lead to surprising results does not necessarily mean that the definition is bad. It may (without limitation) mean that a general assumption or interpretation is bad.

Examples might be the assertion (I do not know if it is correct or not, but it is often heard) that King Henry VIII wrote Greensleeves. If that is verifiable then on usual definitions Greensleeves is not traditional.

COnsider also the Cuddy Wren (or Cutty Wren). I have heard it asserted that it was published in written form in the C13th. That would (probably, depending on the view taken of its subsequent transmission) mean it was not a folk song.

Or should we abandon the word "traditional" and stick to the known definition of "folk song" and seek to educate the "horse music" theorists and fans of Bob Dylan and Paul Simon?

Or should we fix a limit? After all the dates before and after which a motor car is "veteran" "edwardian" "vintage" or "post-vintage thoroughbred" are all fixed, even if those ignorant of the truth in such respects abound. Should we say that a "traditional" song has to be older than 1900 (or maybe 1850, to lose most music-hall stuff) - and how would we know, if the author is unknown? Maybe we shold say "provable to be older than x, so taht ther ewould have to be a proof of existnece (perhpas in writing) before that date? But that would sit oddly with the idea that a folk song must be transmitted orally rather than in writing.

A vexing question!