Cato adjunct scholar David Post writes on the Volokh Conspiracy blog about the sticky copyright wicket facing some impressive jazz recordings from the 30s and 40s.

I get pretty excited … when I read that the collection also contains live performances of a Goodman-Wilson duet on “Lady Be Good” (with Wilson playing harpsichord!), Lester Young and Herschel Evans on “Tea for Two,” Charlie Christian playing electric guitar with the Goodman sextet in a 1939 performance of “Shivers,” the Count Basie and Duke Elllington bands’ performances at the 1938 “Carnival of Swing” on Randalls Island, … all previously unreleased. Oh, lordy — you’ve got to be kidding me! And listening to the excerpts from the recordings here, if anything, makes me even more delirious — this is truly great stuff by some of the greatest musicians that ever lived.

But:

[T]he potential copyright liability that could attach to redistribution of these recordings is so large — and, more importantly, so uncertain — that there may never be a public distribution of the recordings. Tracking down all the parties who may have a copyright interest in these performances, and therefore an entitlement to royalty payments (or to enjoining their distribution), is a monumental, and quite possibly an impossible, task, and it may well be that nobody steps forward with the resources to (a) undertake the efforts required and (b) take on the risk of liability.

Some of Post’s observations on copyright policy:

[I]f you give people a property interest in their creations, they’ll be able to work out market arrangements to receive compensation for them; knowing that in advance, they’ll create more works of art than they otherwise would absent that protection, and we’re all better off as a result. That’s easy enough to see. What’s harder to see is why that principle should ever be limited…

This case makes seen some costs of overbroad copyright protection through the medium of what we may not hear.