“Some sheep are being treated like goats and the resulting mélange can satisfy no one except those who happen to profit from the confusion.”
Benjamin Kaplan, Harvard Law Professor, on sound recording copyright, from Publication in Copyright Law: The Question of Phonograph Records, 103 U. Penn. L. Rev. 469 (1955).
To be clear, what follows is solely my opinion, and should not be attributed to any institution, public or private. Much to my chagrin I received no support for writing this post, directly or indirectly.
There’s been a lot of talk lately about the CLASSICS Act, now incorporated into the Music Modernization Act which passed the House and is now pending before the Senate. There’s been a number of posts arguing whether it’s an extension of copyright to 144 years, or a clarification of the status quo. I’ve personally spent a while on the issue of pre-1972 sound recordings and copyright, and my article Common-Law Copyright should be coming out soon, so I figured it might be helpful to explain a bit more why I tend to think this bill is largely a clarification and federalization of the status quo, by way of providing the background for pre-72 sound recordings. This is fairly condensed, and I’d urge folks who want the whole story to read my article. Those interested in the subject should also look at the Copyright Office’s Report from 2011 on pre-1972 sound recordings, as well as its related documents.
The very first sound recordings were developed roughly contemporaneously with the player piano, and were essentially lumped together for purposes of copyright law – the first case as to sound recording and copyright, 1901’s Stern v. Rosey, which held that a sound recording did not constitute an infringement of the underlying musical composition, relied heavily on an earlier player piano case. This view was affirmed by the Supreme Court in 1908 another player piano case, White-Smith v. Apollo, which led Congress to include compulsory licensing for mechanical reproductions of musical compositions – including piano rolls and sound recordings – in the major revision of the copyright laws passed the following year. The 1909 Act included no provisions protecting the mechanical reproductions themselves, and this silence was taken by the Copyright Office and courts to be a bar to registration of sound recordings for copyright.
In the first years after the 1909 Act the record industry entered into a golden age of soaring sales, but the rise of radio combined with the Great Depression led to the worst decline of the record industry in history, with record sales down over 90% in just a few years. Different groups in the music industry organized and regrouped to meet the new threat of radio, with both the record companies and musicians’ unions taking their own tactics. A third interest group was the bandleaders whose big bands were the most popular music of the 1930s, and they organized the National Association of Performing Artists (NAPA) to protect their interests. Aside from the bandleaders the legal force behind NAPA was one Maurice J. Speiser, who translated Robert Homburg’s Legal Rights of Performing Artists from French and argued that the law should protect the performances of musicians, not just their compositions. This was critical because the big bands made most of their income off of live radio shows, and the playing of their records created a situation where the band was competing with a recorded version of itself.
These suits were generally premised on variants of the law of common-law copyright. Common-law copyright is a perpetual right held by the creator (or their assignee) in a work of authorship. It persists until publication, at which point it is supplanted by statutory copyright (assuming statutory copyright is validly obtained – if not the work enters the public domain). Publication consists of two separate elements – the investitive publication, aka publication under federal law, which is publication sufficient to register copyright in the work as a published work, and divestitive publication, or publication at state common law, which destroys the common-law copyright in the work. As discussed below, because there was no ability to protect sound recordings under federal statute, most state courts have adopted the legal fiction that sound recordings were not published for purposes of divestitive publication, because the creators of the sound recording did not intend to lose protection in the work by selling it commercially. Thus, even a sound recording from before 1972 that has sold millions of copies is “unpublished” in most states. As common-law copyright is perpetual, this protection applies equally to sound recordings made in 1891 and 1971 – both are protected until federal law preempts common-law protection. The 1976 Copyright Act abolished common-law copyright, but specifically retained it for pre-1972 sound recordings.
To establish such a performance right at common law, numerous lawsuits were filed against radio stations for playing their records. The key things plaintiffs needed to establish were that their sound recordings were unpublished (if published federal copyright would govern, and at the time they had no rights under federal law), and that the common law of the state provided that playing their records on the radio constituted infringement. The second question was probably the simpler one at the time – my research shows that in the 19th century it was widely understood that common-law copyright included performance rights regardless of whether or not federal law included such a performance right. In the first suit , Fred Waring was successful before the Pennsylvania Supreme Court, which held that he had not intended to surrender rights in his sound recordings, and thus they were unpublished, and that state law protected his sound recordings (which included the text indicating that they were not licensed for radio play on the record) from being played on the radio. Paul Whiteman was less lucky before the 2nd Circuit Court of Appeals, applying New York Law, which held that his sound recordings were published and he had no rights. Waring would bring a suit in federal court in North Carolina too, and prevailed at trial, but in response both North and South Carolina passed statutes abrogating this result and legislating that sound recordings possessed no rights at common law in those states (later clarified to only refer to radio play, not reproduction). This led to a situation where unlicensed radio play was prohibited in Pennsylvania and permitted in three other states, which would become four when Florida passed a similar law abrogating common-law copyright in 1941.
However, the wheels would soon begin to turn more favorably to the creators of sound recordings. Legal Academics began to argue that the exclusion of sound recordings from the 1909 Act’s protection was not a matter of the constitution but simply legislative judgment. In 1951 a New York trial court held that unlicensed sale of records sourced from Metropolitan Opera broadcasts constituted unfair competition and violation of common-law copyright in part. Based on this decision, in a case involving U.S. rights to resell records of classical recordings made in Nazi Germany, the 2nd Circuit reversed its previous holding from the Whiteman case, holding that sound recordings were not published for purposes of copyright law under New York Law.
Following that 1955 decision by the 2nd Circuit, Courts in a number of other states likewise found that commercial sale of a sound recording does not constitute a publication that divests the work of common-law protection. However, unlike the earlier cases, none of these cases focused on commercial radio airplay of sound recordings, likely because (a) radio airplay was needed to drive record sales and (b) out of a concern that pushing for broadcasters to pay royalties would lead to a political backlash. California became the only state to formalize this common-law doctrine into a civil statute (California never updated its dates of protection from 2047 to 2067 following that change in federal law in 1998 – it’s unclear of this is an oversight or intentional). CLIR did some studies cataloging which states have what rights for sound recordings at common law.
Despite these successes at common law, further remedies were needed. Especially as prerecorded tapes began to be widely sold, it was estimated that every third tape sold was an illicit copy. And the mafia had long been in the habit of stocking jukeboxes with their own pirate pressings of records. Accordingly states began passing statutes criminalizing record piracy in the 1960s, which would culminate in a situation today where 48/50 states have such criminal statutes (the exceptions are Indiana and Vermont)
Even so, problems of record piracy were not addressed. It had been hoped that the inclusion of sound recordings in federal copyright would be part of the omnibus revision process that had started formally in the early 1960s, but in 1971 it became clear that a more immediate fix was needed, and Congress passed a brief statute prospectively protecting sound recordings made on or after February 15, 1972 against unlawful reproduction, but specifically declining to provide sound recordings with a right of public performance as well. With the issue somewhat resolved, the 1976 statute largely incorporated the 1972 law, and the new law provided that state law remedies for pre-1972 sound recordings would survive until 2047 (changed to 2067 with the 1998 term extension). A lengthy study of copyright and sound recordings was released in 1978, but it did not result in major changes. In 2005, in a case involving classical recordings from the 1930s, the New York Court of Appeals reaffirmed that sound recordings were unpublished even if commercially sold, and made clear that unless state law was preempted, pre-1972 sound recordings enjoy indefinite protection in New York. This remains the law in states that have considered the question, at least regarding reproduction of sound recordings.
In 1995, after decades of wrangling, sound recordings finally received a public performance right, but it was limited to digital audio transmission. At first this seemed like a fairly minimal right, but recent SoundExchange payments show that sound recording copyright stakeholders had substantial foresight in pushing for such a right. The law did not per se discuss pre-1972 sound recordings, but around 2012, following a decision of the Copyright Royalty Board, SiriusXM stopped paying royalties for pre-1972 sound recordings, and Pandora followed suit. (this part is poorly documented, but is a key to understanding why the issue recently has come up) The major record labels sued, as did Flo& Eddie of the Turtles, hoping to represent all pre-1972 sound recording stakeholders in a class action to force digital streaming services to pay royalties under state law. Several other lawsuits followed, notably the ABS cases (claiming performance rights for terrestrial broadcasts of pre-1972 sound recordings) and the Sheridan cases (claiming a bit of both).
The major record labels settled with SiriusXM and Pandora fairly quickly. Flo & Eddie would ultimately reach a settlement as well, but it was not meant to stop the litigation – in fact the amount to be paid depended on the outcome of the suits. Flo & Eddie were successful in New York and California federal courts, which found that a performance right existed for sound recordings under common law / the California statute, and were unsuccessful in Florida state court. However, when the question was certified to the New York Court of Appeals, they held that no performance right existed for sound recordings, essentially stating that if such a right existed, it was not for the court to define, because of problems of administrability.
The Florida Supreme Court held similarly, but in a risk apparent from bringing these suits, also, at least suggested that the rationale for protection at common law that these recordings are unpublished is not law in Florida. In a little noticed decision in Illinois, a trial court held similarly regarding that state’s common law. It is unclear how meaningful this change will be, or what it would mean if sound recordings are considered published in some states but not others. As of right now Pennsylvania gives sound recordings effectively perpetual rights of public performance, and the most current case law in California gives them performance rights until 2047 – although the question is currently sub judice before the California Supreme Court. The Carolinas deny sound recordings performance rights by statute, and case law in New York and Florida does similarly, although as mentioned the New York decision looked to the legislature for such a right. It’s worth noting that the Pennsylvania case and statutes in the Carolinas are from the 1930s and it’s not totally clear how they apply in the modern era.
Of course, this is all in the air because the CLASSICS Act was passed by the House and is currently pending in the Senate as part of the Music Modernization Act (“MMA”). The CLASSICS Act makes clear that digital streaming services must pay royalties to pre-1972 sound recordings, but does not actually federalize protection for these sound recordings. Recently, in a move widely seen as trying to remove the CLASSICS Act from the MMA, the ACCESS Act was introduced in the Senate, which would federalize protection for sound recordings made in 1923-1972, and put earlier sound recordings into the public domain.
To supplement this post, I’ve included scans of a 1965 CRS Report (well, Legislative Reference Service Report, its predecessor), entitled The Phonograph Record Industry; An Economic Study. I actually uploaded it about a year ago, but only shared it via Twitter – it includes a lot of detail about the record industry and its history from before 1972.