Michael,

"Absence of evidence is not evidence of absence"; just because broadsides didn't survive doesn't mean they never existed.

to play devil's advocate on the Dust Bowl evidence, during fieldwork the answer one gets depends on the question one asks and the informant's expectations of one's wishes. If I ask "whom did you learn that song from?" I will get the answer. But I may not find out how it was learned. Even if I ask "where did you get that song?" the person may assume I mean "from whom"; and finally, many savvy informants know perfectly well that folklorist types prefer not to know if you learned all your songs from books!

Note that I'm not suggesting informants lie; though that, too, sometimes happens. The current generation of the Copper Family could rightly say that they learned songs from Bob Copper--but Bob had them written down in a book. So is this written transmission or oral? Dorman Ralph from Newfoundland learned songs from printed sources, even though he was completely blind; his brother read the words to him. So just because these people say they learned the song from another person does not mean they learned it "orally" or that the event represents an oral transmission.

That said, I haven't heard those tapes, so I don't know how the information was elicited.

To continue playing Devil's advocate, in the section of Laws' Native American Ballads given over to accidents and disasters, 15 of 33 items seem to have been known on broadsides, newspapers, phonograph records, or other published forms before they were widely popular. I haven't seen any of the printed forms, and in most cases neither has Laws, but there is testimony that such forms existed.

Remember that most Native American Ballads that we know of originated shortly before the era of commercial recordings. Once a record was available, broadsides lost much of their cachet, so many of these songs may have been printed only once or twice. It's not surprising that few of the actual broadsides survived. In the case of British songs, which existed for hundreds of years before recordings, and consequently went through many printings, it's not surprising that more copies would survive.

Finally, as to the African-American argument, just because a song was from an African-American did not mean it was composed and distributed purely by oral tradition. As an example, we can take Bill Dooley, composer of Frankie and Albert, Stackolee, and other "folk" songs. Frankie's lawyer stated in court that Bill Dooley sold copies of his compositions on the street for ten cents. I don't know if any of these copies survived.

So in general: yes, we agree, oral transmission was crucial. But the extent to which a song has been transmitted through oral tradition, writing or print can only be discovered through research into individual songs.