The Waring, Metropolitan Opera and Capitol Records decisions may appear to conflict with the accepted principle that a public sale of a literary work is a "general publication" terminating a common-law copyright, and any copyright protection thereafter must be derived from statute. But the historical distinction in the treatment of literary and musical works by Congress accounts for the lack of federal statutory copyright protection for sound recordings. In the absence of protective legislation, Congress intended that the owner of rights to a sound recording should rely on the "broad and flexible" power of the common law to protect those property rights after public dissemination of the work.
Naxos, slip op. at 20. Not once is there a citation to anything indicating a positive intent of Congress to do anything; instead, the Court of Appeals infers such intent from Congress's apparent acquiescence to the "recordings can't be published" conclusion that many drew from White-Smith. Of course, by that logic Congress should not have been able to explicitly add motion pictures to copyrightable subject matter in 1912 (the Townsend Amendment)!
What troubles me more about Naxos is that it grasps one lacuna as evidence of positive intent, but refuses to grasp a directly relevant lacuna at all. The absence of a definition of "publication" in the 1909 Copyright Act—even in the 1971 amendments that explicitly allowed copyrighting of phonorecordings—indicates that whether a work has been published is a fact-specific inquiry. (Note, too, that the 1976 Act defines "publication" only for phonorecordings.) Further, the 1976 Act, by its own terms and by the statutory language in the transitional provisions, acts to wipe out state common-law copyright, notwithstanding that rather interesting analysis of the legislative history in Naxos (slip op. at 20–22). Of course, this runs right into Goldstein, La Cienega, and the amendments to § 303. The Second Circuit seems content to stop its analysis at that point.
Then, though, one hits another lacuna: Dastar. The Supreme Court noted recently that
Assuming for the sake of argument that Dastar's representation of itself as the "Producer" of its videos amounted to a representation that it originated the creative work conveyed by the videos, allowing a cause of action under [§] 43(a) for that representation would create a species of mutant copyright law that limits the public's "federal right to 'copy and to use,'" expired copyrights.
Dastar v. Twentieth Century Fox Film Corp., No. 02–428, slip op. at 10 (citation omitted) (relying on Bonita Boats's disallowance of a Florida common-law copyright in boat-hull designs).
The real problem here, I think, is that the New York Court of Appeals is straining to reach what it considers the right and fair result. I have some sympathy for it, for an opinion the other way would appear to throw every sound recording made prior to 1972 into the public domain. The unconsidered case, though, is this one (and it's far from trivial): Recordings actually made outside the US, which therefore should have been subject only to the statutory rights that existed wherever they were actually recorded. Regardless of what the contract says, if the "common-law copyright" in the recording wasn't recognized where the recording was made, a contract can't transfer such a right to a US record label. Thus, Sergeant Pepper's Lonely Hearts Club Band should fall outside of the reasoning in Naxos—meaning that, in the end, Naxos does not contribute to uniformity in the recording industry, whatever its intent or method. If this brings to mind nasty questions about the 1909 Copyright Act's manufacturing clause and its relationship to WIPO and other international trade treaties, you get a gold star for looking at the implications. That and $5 will get you a fancified cup of coffee.
I'm sorry, but the only solution is going to be a legislative one; Naxos doesn't, won't, and can't do what the New York Court of Appeals means it to do. What Congress needs to do is decide that sound recordings shall be treated like any other copyrighted work, but that registration may be presumed for "actually published" works issued before 1972 (since it was impossible to do so). That's not going to happen, though, if only because collectors of "Americana recordings" and such will raise the danger that they will deteriorate before those recordings can be preserved. Et cetera.