Question: When is a Pre-1972 Recording Not a Pre-1972 Recording?

Answer: Unclear, but it could be more often that you might think.

In the ongoing litigation over whether recording artists are entitled to performance right royalties for the public performance of pre-February 15, 1972 recordings, we have a new wrinkle. A federal judge in the U.S. District Court for the Central District of California has determined that some recordings you might have thought to be pre-1972 recordings could actually lose that important status. This opens a whole new defense for broadcasters, satellite radio operators and webcasters who might otherwise have been sweating about potential liability galore. (How much "potential"? As I reported last June, Sirius XM figured it was the smart thing to do to settle out with four performance rights claimants for $210 million, rather than risk further litigation; and that amount didn't cover any potential royalties beyond next year.)

If you're fuzzy on just what the pre-1972 recording performance rights controversy is all about, take a minute to check out my earlier posts on the topic - this one and this one should help you out (or you can take your pick from my collection at this link). For today's lesson, here's what you need to know: in 1971, Congress created a "performance right" in recordings, but it applied that right only to recordings made on or after February 15, 1972. In the last several years, a number of recording artists (or their representatives) have relied on state laws (mainly in California and New York) to argue that transmissions of their pre-2/15/72 recordings by, most notably, Sirius XM and Pandora trigger performance rights royalty obligations. Their arguments have gotten traction in a number of rulings against Sirius XM and Pandora.

In the most recent decision - which involves broadcast and online performances - the tide may have turned.

The case pitted CBS against four plaintiffs holding the rights to a bunch of pre-2/15/72 records. (The artists' whose recordings were at issue included the Everly Brothers, Ray Stevens, Andy Williams, the Chi-Lites and another couple dozen or so ranging from reasonably familiar to pretty obscure.) There apparently was little question that CBS stations had transmitted most of the songs in California (either over-the-air or through Internet streams), so under the earlier line of cases it looked like CBS might be on the hook for royalties.

But CBS came up with a new defense and Judge Percy Anderson was persuaded.

CBS's argument is that it had broadcast only "digitally re-mastered or re-issued recordings which were created after February 15, 1972". In CBS's view, the original analog recordings had been modified sufficiently through the digitization/remastering process that, by the time they got to CBS, they had become "derivative works" separate and distinct (for copyright purposes, at least) from the originals.

Many people scoffed at CBS's argument, by the way. Who's laughing now?

The argument is best explained in a couple of publications from the Copyright Office: "Circular 56", in which the term "derivative work" is defined, and Section 803.9(F)(3) of the Office's Compendium of U.S. Copyright Office Practices, in which that definition is fleshed out a bit. According to the former, a derivative work "incorporates some preexisting sounds that were previously registered or published, or sounds that were fixed, before February 15, 1972", but those preexisting sounds must have been "rearranged, remixed, or otherwise altered in sequence or character, or there must be some additional new sounds".

According to the Office, re-mastering old recordings can produce a "derivative work" if the re-mastering involves "multiple kinds of creative authorship, such as adjustments of equalization, sound editing, and channel assignment". But if the re-mastering involves only "[m]echanical changes or processes", such as "a change in format, declicking, and noise reduction", that does not create a "derivative work" separate from the original recording. In its Compendium the Office elaborates, somewhat unhelpfully, that post-recording contributions involving "'equalization', 'reverberation,' 'reprocessing,' and 're-engineering'" may be enough to give rise to a derivative work, or not - depending on whether the changes are merely "mechanical in nature" or "too minimal". Importantly, a derivative work represents a separate copyrightable work, distinct from the original on which it is based.

You can probably see where this is going.

All recordings made prior to February 15, 1972 were made in analog format, because that's the only technology that was available back then. But nowadays, virtually all recordings used by broadcasters, satellite radio folks and webcasters are digital. So somewhere along the line the original recordings were transformed from analog to digital. If that process involved sufficient changes to make the digital version its own "derivative work", the fact that that separate work became "fixed" as a recording after February 15, 1972 - as would be the case for any digital recording - would mean that it would be subject to federal, not state, copyright law. That in turn would mean that broadcasters would be exempt from any performance royalties, and other performances (by, e.g., satellite distributors and webcasters) would be subject to statutorily-set royalties.

So CBS argued that the versions of the recordings which it broadcast were sufficiently different from the original versions of those recordings to constitute derivative works. To prove that, they brought in their experts to compare the two different versions - to the extent that the two different versions could be reliably identified. (As it turns out, there are lots of different versions of oldies available, and figuring out precisely which is the original and which happened to have been aired in the relatively recent past can be tricky.)

CBS's experts - sound engineers and producers, including some who had remastered some of the recordings at issue - emphasized that the remastering process involves the "personal aesthetic" of the sound engineer performing the process. In their view, it's not a mere "drag and drop" process to convert a sound recording from analog to digital format: adjustments to various characteristics such as the bass, treble, midrange and other frequencies can and possibly must be made to, e.g., emphasize certain instruments and de-emphasize others. One expert said that the remastering process involves "subjectivity, originality and ultimately produces works of art".

The plaintiffs, of course, countered with their own experts, who characterized the remastering process as a purely mechanical practice without any creativity involved at all - essentially the 21st Century equivalent of dubbing a cut from an album onto a cassette so you could listen to it on your Walkman. (We hope that our younger readers will appreciate the links we have provided to pictures of each of these, in case you are unfamiliar with these prehistoric music delivery media and products).

In Judge Anderson's view, CBS got the better of the argument. He was satisfied that remastering can leave a work technically unedited even though it subjectively and artistically alters the work's timbre, spatial imagery, sound balance and loudness range; when such alterations occur, the remastered work is entitled to federal copyright protection separate from the original work. And he concluded that there were "perceptible changes" between the original versions of the songs at issue and the versions aired by CBS, although on that point he seems to have taken the experts' word. It does not appear that Anderson in fact personally listened to any of the recordings to see just how "perceptible" the changes might have been to his own ear.

And with that, Judge Anderson dismissed the plaintiffs' complaint.

As radical as this decision is, it's important to recognize that it does not get broadcasters and digital performers (e.g., webcasters) completely off the hook for potential performance royalties for the transmission of pre-2/15/72 recordings. Whether such liability exists in the first place is the $640,000,000 question for the broadcast industry in some sense. As I've discussed time and time again, even though broadcasters do not have to pay a royalty for performance of sound recordings under federal law, a finding that there is a state-law-based performance right in pre-1972 sound recordings could require payment of royalties for performance of those older sound recordings in certain states - royalties that could run into the hundreds of millions of dollars. And as I've reported in earlier posts, at least a couple of courts have concluded that such a state-law-based performance right does indeed exist. Judge Anderson seems to have tacitly accepted the premise of those earlier decisions.

But his decision does afford broadcasters and others a very useful workaround to avoid liability - or even a defense against any such claims in the event they are sued. Under Judge Anderson's view, if performance rights royalties for pre-2/15/72 recordings are claimed by artists or labels, the targeted defendant (be it a broadcaster, webcaster, etc.) can counter by arguing that any recordings that happen to been transmitted were remastered versions perceptibly different from the original pre-2/15/72 recordings. If CBS's experts are correct, it appears likely that pretty much any remastered recording can be shown to differ in enough respects to be deemed a derivative work. And, since digital remastering was unavailable prior to 2/15/72, any such remastered recordings would by definition be subject to federal, not state, copyright law. The broadcast of such remastered "derivative works" would be exempt from any performance right royalties (and performance royalties for the digital transmission of such works would be determined not by state law, but by the federal Copyright Act).

So it would be a good idea to check your libraries to confirm that, for any pre-2/15/72 music you're playing, you're using digitally remastered versions.

There is an interesting further implication, involving a perhaps unintended consequence of Judge Anderson's ruling: it could result in endless extension of copyright protections as owners of sound recordings periodically remaster their works, effectively renewing their original date of copyright. In a footnote, he dismisses this concern as "unwarranted because the Court's finding of copyrightable originality is based not on a mere conversion between formats, but on the original expression added by a sound engineer during the remastering process." True enough, but since the experts certainly made it sound like remastering is in and of itself almost invariably a matter of personal aesthetics, presumably it would not be hard for the copyright holder to insure that each and every remastering could be legitimately deemed a distinct - and copyrightable - "derivative work".

At the very least this poses an interesting question - one of several interesting questions - and the reason that I think this case of all of them is the one the Supreme Court might be most interested in taking. Stay tuned.