I see what Maud Karpeles is getting at. It's an "iffy" situation. She is both right and wrong.

It's true that having a song on the printed page or on a recording makes it possible to refer back to that source and treat it as if it were "the correct version." This is a widespread practice in classical music, and appropriately so. If you're playing with or singing to the accompaniment of a full symphony orchestra, you need to "know the score" and perform it correctly in order to avoid chaos. Popular music is a lot looser. Frank Sinatra, Barbra Streisand, Clint Black, and Sting may all sing the same song, but each will adapted it to his or her own unique style—maybe even alter the words and tune. Also, if one particular version of a song is constantly pouring out of radio and jukebox speakers wherever you go, this tends to "lock in" that version. But that's usually a short-lived phenomenon, and once that song is replaced by something else, it's free again.

Folk songs are subject to these same influences. I remember being reprimanded because I didn't sing Tom Dooley or The Sloop John B. the way the Kingston Trio sang them, even though I learned the songs before the Kingston Trio's first record came out (Tom Dooley from a Frank Warner record and The Sloop John B. from Walt Robertson in person). I ignored this and continued singing them as I learned them—more or less.

It's that "more or less" part that makes the difference. When learning a song from a recording, sometimes a word or two doesn't come through clearly no matter how many times you listen to it, so you have to bung in the word that you think it might be. Or maybe a line sings very awkwardly, so while trying to keep the original meaning, you change a word or two to make the line easier to sing or more clear. I am opposed to making changes indiscriminately and trying to justify it by saying that I'm just helping the folk process along. If I make a conscious change in a song, I have to have a damned good reason for doing so. Or sometimes you don't realize that you are making changes, and the song gets altered inadvertently. I'm frequently surprised when I pull out an old record I used for a song source years ago and see how much I've changed a particular song without realizing it. Sometimes I change it back, sometimes I just keep doing the changed version, depending. I don't slavishly return to the source if I like my way better. That is the folk process.

Then someone learns a song from me with the changes I've made and they make changes of their own. Thus it goes. The folk process is alive and healthy. It's a long, slow process, and well it should be.

So Maud Karpeles is both right and wrong. Printed and recorded versions of songs have a certain "authority." They tend to cast a song in stone. But if you ignore the "authenticity police" and don't keep referring back to the same songbook or record to make sure you're "doing it right," the folk process is inevitable.

But as for the gink who stands there at the open mike and says "This is a folk song I wrote last week" and then drones on tunelessly for thirty verses. . . . A folk song? Well . . . maybe. But not bloody likely. Time will tell.

Don Firth