Think of this volume as a folk music equivalent of a Greek Tragedy: It's a great book with a fatal flaw.

It's a great book because it is one of the largest collections of American folk songs ever published. There is no question but that it was a seminal production.

The problem is, the songs have been Lomax-ized.

This is an old, old problem in folk music publications. Percy's Reliques started the trend: Take an old song and rewrite and refine and don't admit to it -- and even if the author does admit to it, he doesn't list where he made changes.

This is not to absolutely reject rewriting. If a song collector finds a version of a song which lacks a key verse, and wants to include it in a book for general audiences (and this book was intended for popular audiences), then the collector needs to put in that verse. But put it in [brackets] to show that it is spurious, and list the source for the interpolation.

And, while he's at it, he needs to list from whom the song was collected, and where, and when.

This volume fails on all these counts. And the Lomaxes did an incredible amount of tampering. Bottom line: There are a lot of great songs here. There is also a lot of very, very bad scholarship. If all you care about is the songs, by all means, pick up this book. But don't trust anything it says unless you can (for instance) verify it against the Lomax field recordings. Those, at least, are highly valuable and unadulterated.

I didn't read the whole book; I skimmed a lot of the text and skipped ahead to the music. There's a lot to discover in here. There are a few (now) popular songs everyone knows and loves, alongside (equally good) songs you've probably never heard of. But I find most of the songs are different versions of songs I know -- such as "Amazing Grace" with an unrecognizable melody, or "Yankee Doodle" with words about fighting the Civil War (and no mention of macaroni).

The book is sort of torn between being a songbook for general readers and being a sort of reference for people with a scholarly interest in American folk songs; whether it's the best of both worlds or the worst, I guess depends on what you're looking for. The Lomaxes combined different versions of songs, picking the bits they happened to think were best (not most representative), which makes for more of a popular songbook than anything else. But they also leave out any harmonization (ironically, for the sake of not editing), and they include songs which have no written music and songs which cannot be notated (but they approximate notating them anyway), so there's a lot of content which is useless as a songbook but potentially interesting as a reference work.

That duality has a lot to do with why I like the work of the Lomaxes. They were the sort of people who wrote arrangements for their field recordings. For people like me who are just interested in good music, not in "scholarly" accuracy, it's perfect. ( )