It’s been too long since I made a proper post – I’ve been busy putting together a piece I’m calling An Empirical Study of 225 Years of American Copyright Registrations (title subject to change, natch), which I fully expect to be the bees knees, all while being no bigger than a breadbox. But there’s still a lot I want to share, and this seems as good a time as any to share a report the US Copyright Office issued in 1947 which has been almost completely forgotten.
Until 1947, the US Copyright Office would not register editions of public-domain music for copyright. Obviously the music could not be reinserted to copyright protection, but the Office held that merely editing public-domain music was insufficient for copyright protection. However, in that year the Office issued a report (link to my scans) that gave the history of Office rules for registration of music, and an argument for how they should be changed to include edited versions of public-domain works. The Office changed its position based on this report and allowed such editions of music to be registered for copyright. The Office had distributed the report to interested outside parties, and copies are available at a number of law libraries, but I haven’t seen it online.
The issue of whether editions of public domain music are nonetheless protected was litigated in England in the Hyperion Records case, involving a small classical music label in a suit by a scholar over his editions of the music of an early 18th century composer. However, I think the discussions in the report are illuminating for discussing the scope of music copyright protection and infringement more generally, something Joe Fischman also discusses in his upcoming article Music as a Matter of Law, and I of course discuss in my upcoming article Common-Law Copyright (never miss an opportunity to plug one’s own work).
As I’ve mentioned a few times on this blog, and will mention many more times, I currently have an article going through the editing process at the University of Cincinnati Law Review, which attempts to be a systematic study of common-law copyright, both generally and as applied to sound recordings, with a specific focus on performance rights. This is part of a series of posts where I focus on specific cases/examples from that paper, and share some of the primary source documents.
In this post I will discuss two suits against the owners of the Tivoli Opera House in San Francisco for infringement of their common-law copyright by performing unpublished light opera (aka operetta) without a license. The cases are from the mid-1880s, which might seem like deep esoterica, but in fact I think they offer unique insight into the contemporary common-law copyright cases currently ongoing, in this case specifically the Flo & Eddie v. Pandora case currently being briefed before the California Supreme Court. This is because, as I’ll explain more below, the 1880s and early 1890s offer a unique window into the question of whether a performance right exists at common law. This is because federal law did not include performance rights for music at the time, but courts essentially universally found that common law copyright, which only protects unpublished works, did offer an exclusive performance right. As a result, (a) questions of common-law copyright and its scope became especially important at this time, and (b) it strongly suggests that common-law copyright does not obey the limitations on performance rights that federal law does.
No American Theater did so much to popularize opera as the Tivoli, best remembered of all of San Francisco’s theaters, which Joe Kreling opened as a Beer Garden in 1875, with a 10-piece Orchestra and Tyrolean singers. Rebuilt in 1879, it became the Tivoli Opera House. Its career began happily with Gilbert & Sullivan’s Pinafore, which ran for 84 nights. For 26 years thereafter it gave 12 months of Opera each year, never closing its doors, except when it was being rebuilt in 1904: a record in the history of American Theater. For eight months of the year light opera – Gilbert and Sullivan, Offenbach, Van Suppe, Lecoq- was performed, and for 4 months, grand opera, principally French and Italian occasionally Wagner. From the Tivoli chorus rose Alice Nielsen, the celebrated prima donna.
Unmentioned in this excerpt is whether the Tivoli had the rights to perform any of these works. The situation of performance rights as of the 1880s was a particularly complicated one under the federal statute. There was a performance right for drama under a 1856 Act, codified into the 1870 revision of the copyright law, but it did not expressly include music (music had been added to the copyright act in 1831, but that act included no performance rights whatsoever). The question of whether the act creating a performance right for dramatic works included music remained open until 1885, when the Circuit Court in New York conclusively rejected such an argument in The Mikado Case, about the Gilbert & Sullivan operetta.
In truth, though, whether or not the federal act included a performance right for music mattered little, since almost all composers were based out of Europe, not the United States, and foreigners were not protected by federal law at all until 1891 anyway (Gilbert & Sullivan had concocted an extensive scheme to get an American copyright, which I discuss much more in an article). Accordingly, given that federal law did not protect them – both because they would not have a performance right, and because even if they did, they could not get an American copyright, European opera composers hoping for recompense for their labor in the American market were forced to rely on common-law copyright.
Into this gap stepped Leo Goldmark, who in the 1880s took on a role somewhat similar to that which Flo & Eddie have taken on in this decade – advocating for performance rights at common law. Leo Goldmark is mostly best known today for his family members than his own legacy,1 but his own career was focused on the law, which he took up in the 1870s after emigrating to America. Around 1880, with both musical training himself and connections to major composers, Goldmark began a national campaign of bringing lawsuits against various opera companies, accusing them of infringing the common-law rights of the unpublished parts of operettas.
The unpublished aspect was always a bit complicated, since for most of these operas the actual songs had been published in piano/songbook reduction. However, Goldmark premised his arguments on the unpublished parts of the opera – the unsung dialogue, the stage direction/choreography, and the orchestration. Courts across the country showed no resistance to this argument, liberally handing out injunctions. By the time Leo Goldmark brought suit against the Krelings in September of 1885, the case file reveals that he had already filed 26 other such lawsuits all across the country. In this case he accused them of performing the operetta Nanon by Richard Genée without permission of the American rightsholders.
The US Circuit Court in San Francisco (the Circuit Court was a trial court until the 1890s) quickly found that Goldmark had a right to an injunction against performance of the operetta at common law. The Court did not treat the matter lightly though, and indeed the presiding judge had initially denied the application for an injunction, but on a petition for rehearing invited the other federal judges from California to sit with him, and at that rehearing reversed his initial opinion and granted the injunction.
The matter continued to fester for a number of years, and in 1888 the Court,2found that Nanon was protected by common law in the United States, because even assuming a authorized publication had occurred of some of the songs from the show, because other aspects of the operetta like the orchestration and interstitial dialogue had not been published. The Court thus enjoined the Krelings from performing any part of Nanon that had not been published, but held that they could use their orchestration of the music for the opera, which they had created independently from the published piano score, and not copied from the unpublished original orchestration
The issue of orchestration and publication would be explored more in a second case brought against the Krelings by Thomas Henry French, manager of American operations for the theatrical licensing company (to this day) named after his father, Samuel French. The lawsuit was commenced in early 1886 and was likely inspired by Goldmark’s similar suit, and the Courts in San Francisco seemed to treat it as being of a piece with the Goldmark suit, enjoining the Tivoli from performing the operetta Falka due to French’s common-law rights in the opera.
This litigation would run for many years, until the Circuit Court in 1894 decisively ruled that although a book of the songs from the original French version of the opera had been published, the complete orchestration and other aspects of the operetta had not been published, and had been infringed under common law. Critical to that finding was one of the unpublished manuscript scores being found as part of the discovery process, and in turn included in the case file.
The decisions themselves are important reading for those seeking to understand common law copyright in California. The case files, which I scanned from the National Archives Regional Division near San Francisco, are likewise fascinating. For instance, the case file for French v. Kreling includes the transcript of a testimony of the in-house orchestrator of the Tivoli opera house, where he goes into great length about the then-common practice of having an individual such as himself on staff, tasked with orchestrating published piano scores into orchestral scores, to avoid paying royalties due to common law. There are also copies of many of the opera libretti at issue in these cases included in the case files – I didn’t scan the entire volumes, but included scans of the cover pages should they be of interest to musical historians.
At least in the 1880s, it was pretty clear that California common law included performance rights for musical compositions, regardless of whether federal law did, a situation somewhat analagous to the situation today for terrestrial radio.. As the question of whether the codification of California common law includes a performance right is currently before the California Supreme Court, this seems like an important time to analyze.3
His brother Carl Goldmark was a major composer at the time, best known today for his violin concerto and other orchestral pieces, but best known at the time for his opera The Queen of Sheba. Leo Goldmark’s son Rubin Goldmark was himself a major composer during his life, although his compositions are mostly forgotten now and he is best remembered for teaching Aaron Copland and George Gershwin. ↩
in the interim California had been split into two Districts, with the court in San Francisco becoming the Northern District ↩
Of course, the California Supreme Court is looking specifically at digital services that are required to pay performance royalties through SoundExchange, in this case Pandora, but the cases brought by ABS and others arguing that there is a performance right for terrestrial radio loom in the background. ↩
I’m not pushing a normative argument with this post, I just think it’s useful to understand how some of the debates over copyright have evolved – or not – over 200 years. Some parts of this article are expanded upon much further in my article The Twilight of the Opera Pirates, which tells the history of the origins of performance rights for music, but I wanted to focus on a different aspect of the Ingersoll Bill than that article looked at.
A question people love to hash out in the public discussions of copyright is whether copyright is a property right. Of course the question is a little misleading, because copyright is obviously at least a form of property, but what people are really debating is whether copyright is substantially equal to real property in the deference given by the law and society. These debates are typically accompanied by calls to history, but these are typically given in a general sense without reference to specific incidents. In the nineteenth century the vast majority of references to property rights in copyright in the records of Congress are in respect to the lengthy fight for international copyright protection, which did not culminate until 1891. However, the Ingersoll Copyright Bill, introduced in 1844 and quickly disappearing, is the exception, and provides I think an interesting window into what people thought about copyright at the time.
At the beginning of 1844, Charles Jared Ingersoll, a Democratic Congressman from Philiadelphia, who also had a career as a lawyer, orator, and playwright, introduced a copyright bill.[His plays included Edwy and Elgiva and Julian, and his widely circulated Discourse Concerning the Influence of America on the Mind included a discussion of copyright law as well.] This bill was intended as a complete revision of copyright law, and was generally far ahead of its time. Amendments Ingersoll added two weeks later further modernized his bill, adding performance rights as well as a system of design protection. However, Section 15 of the bill contained a clause that I haven’t seen in other copyright legislation, at least not phrased this way:
[A]ll copyright shall be deemed personal property, and shall be transmissible by bequest, or, in case of intestacy, shall be subject of the same law of distribution as other personal property.
This clause is obviously aimed partially at insuring that copyrights can be transmitted to heirs, but there’s also a clear intent to assert that copyright is a form of property. Indeed, reading over the bill and amendments, assertions that copyright is a property right are all over the bill – in stark contrast to the then-in-force 1831 Act, which does not use the word “property” once.
After being introduced the bill was referred to a Select Committee, where it failed to go anywhere. However, luckily, one of the members of this Select Committee was an elderly John Quincy Adams, still keeping his daily diary. Ingersoll and Adams were fierce opponents on issues of the gag rule and slavery, so it is perhaps unsurprising that Adams was less than enamored of Ingersoll’s proposal. When the subject was first broached, Adams noted in his diary that he “offered some suggestions as to the natural right of literary property, to the principles of which, as entertained by me, Ingersoll immediately declared his dissent. His principles are radically depraved, and never can harmonize with mine.” A month later the bill was discussed again, leading Adams to describe the bill as “an entire but most incongruous system of copyright property, fit for nothing but to multiply litigation.”
It’s of course impossible to know exactly what Adams was referring to, but the property aspects of the Ingersoll bill seem like the obvious answer (although in fairness the bill is sprawling and pointing to any one aspect is difficult). That said, the very fact that the Ingersoll bill needed to state that copyright is a property right is suggestive that this was not uniformly agreed on.1
An illustrated of the debate as to whether copyright is on par with real property was illustrated in a Senate Report from six years earlier on international copyright, where the Senate Committee on Patents wrote that
The committee do not deem it necessary to argue the question, which has been so much discussed, whether an author has a property in his written and published productions, by natural right, which society is bound to protect. It will, perhaps, be sufficient to admit, that in most cases there will be found equitable considerations, constituting strong claims for aid and protection in some form, to those from whose intellectual labors mankind derive a benefit. Partly with reference to such considerations of justice to authors, and partly with it view to the advancement of literature and science, all civilized nations have established copyright laws.
In the end the Select Committee on the Ingersoll Bill focused on issues of international copyright instead, including a memorial from Nahum Capen, Miscellaneous Memorials on the issue of international copyright from around the country (my scans from the National Archives, 2 MB), as well as a Memorial from John Jay, presumably a son of the former Chief Justice (my scan from the National Archives, 2 MB). These all concerned the recently failed effort for international copyright from the late 1830s, not the question of property in copyright.
One of these days I want to write a proper account of the history of copyright in America from 1789 to 1909, and one of the strands I want to show is that the triumph of international copyright in 1891 also represented a triumph of a property model of copyright, which was then codified by the 1909 Act. In the 1840s the issue was still much more equivocal. As Justin Hughes has demonstrated, the term “literary property” was already in wide use by authors and advocates of international copyright. However, the term “property” was assiduously avoided by others, including the pirate press and opponents of international copyright. In a way, it seems like little about the debate has changed in some 170 years, but the law itself clearly has.
Of course, the counterargument is that it went without saying in previous bills. ↩
In response to the success of this project I sent the BPL and Archives a list of 19th century patent cases from New England, and they’ve scanned the case files for these as well. The length of these files varies dramatically – some are a few dozen pages long, some are hundreds of pages long. The link on the case name goes to the case file, while the link of the citation goes to the reported decision. It’s really cool to see these case files, both to see how patent litigation was conducted in the nineteenth century, and to see how famous early patent cases came to be decided.
Lowell v. Lewis, 1 Robb, Pat. Cas. 131, Circuit Court, D. Massachusetts. May Term.1817 (“usefulness” in the IP clause of the Constitution is satisfied so long as the invention is not “frivolous or injurious to the well-being, good policy, or sound morals of society”)
Last week the Supreme Court decided the case about the Constitutionality of the clause of federal trademark law excluding from registration disparaging, scandalous, or immoral marks, and found this clause to be unconstitutional. Part of the reasoning at both the Circuit Court level and the Supreme Court level was that such a rule would be obviously unconstitutional vis a vis copyrights, and there is no good way to develop a rule that distinguishes copyrights from trademarks in this context. I’ve been telling people informally that the history of copyright registration is actually a lot more complicated, and that I should write an article explaining this. But in the interim, in the interest of getting the information out there, I’ve put the basic information in a blog post.
In rejecting the argument that trademark registration is government speech, and thus excluded from the protection of the First Amendment, Justice Alito’s opinion states the following (Slip Op. at 21-22)
For if the registration of trademarks constituted government speech, other systems of government registration could easily be characterized in the same way.
Perhaps the most worrisome implication of the Government’s argument concerns the system of copyright registration. If federal registration makes a trademark government speech and thus eliminates all First Amendment protection, would the registration of the copyright for a book produce a similar transformation?
The Federal Circuit’s opinion is much lengthier, and explores this issue in greater detail. This attitude is unsurprising, as nowadays the Copyright Office registers essentially anything. Indeed, the most frequent copyright litigant, by far, is a pornographer. However, copyright’s content neutrality has only been well established since fairly recently, and in the past the Copyright Office has denied registration – and Courts have found works uncopyrightable – driven by both legal concerns such as obscenity as well as more amorphous concerns like immorality.
As William Patry notes in a post about this question in 2005, the trial court in Bleistein held that a “picture which represents a dozen or more figures of women in tights, with bare arms, and with much of the shoulders displayed, and by means of which it is designed to lure men to a circus” was not within the protection of the statute, finding it “merely frivolous, and to some extent immoral in tendency.”1 The Supreme Court overruled this decision, but did not address the discussion of morality, focusing on other issues. Subsequent cases would hold that works that were illegal or immoral were not the subjects of copyright.
In the first part of the twentieth century, federal prosecutions for obscenity through the mail generally made most questions of copyright in works that might even have a shade of obscenity superfluous for domestic authors. For foreign authors the combination of obscenity prosecutions and the manufacturing clause of the copyright law at that time (requiring that English-language books be printed and typeset in America in order to receive copyright protection) was essentially a bar to copyright protection in America for books that might be within European sensibilities, but on the edge of American ones. Robert Spoo has written about the travails of as James Joyce in this arena, and it’s a fascinating story all its own.
In the annual report of the Register of Copyrights for 1941, the Copyright Office made clear for the first time as a written policy that it would not register “seditious, blasphemous, immoral or libelous” works, as the Office did not consider them to be copyrightable at all. 1941 Annual Report at 29-30. Although this policy was anchored in a belief that such works were generally illegal, such a policy goes far beyond the scope of the first amendment today – I can’t say for certain if it did in 1941 though.
It is unclear whether many works were rejected in these standards, but in September of 1954, the Copyright Office issued Ruling Number 32 relating to Obscene Works, reading as follows:
a. The Office will not ordinarily attempt to examine a work to determine whether it contains material that might be considered obscene.
b. If the examiner believes, upon an ordinary examination, that a work is obscene, the Chief of the Examining Division will determine whether the work should be referred to the Department of Justice for possible prosecution. If the work is not so referred, the copyright claim may be registered.
c. When a work is referred to the Department of Justice, registration will be held up pending action by that Department.
The unraveling of this system came in September 1958, with the arrest of Lawrence E. Gichner, owner of a sheet metal company and amateur sexologist, who had attempted to register his book “Erotic Aspects of Chinese Culture.” Apparently the Copyright Office had referred it on examination to the Department of Justice, who had contacted the Washington, DC police. The DC police arrested him, and took for evidence fifty crates of material. The ACLU rushed to Mr. Gichner’s defense, and the Librarian asked the President to ask his Attorney General for an opinion as to whether the Copyright Office was permitted to register works that are illegal or immoral. On December 18, 1958, the Attorney General answered that the Office was not required to turn away works on the grounds of immorality or illegality. 41 Op. Att’y Gen. 395 (1958).
However, an eagle-eyed reader may note that the Attorney General did not actually hold that immoral works were protected by copyright – in fact he strongly implied the opposite, that both immoral and obscene works are not protected by copyright, but that it was not the job of the Copyright Office to make that determination. In 1979 and 1982 the Fifth and 9th Circuits held that immoral and obscene works were protected by copyright. Jartech v. Clancy, 666 F.2d 403 (9th Cir. 1982); Mitchell Bros. Film Group v. Cinema Adult Theater, 604 F.2d 852 (5th Cir. 1979) However, in 1998, the Southern District of New York noted that it was “far from clear that the Second Circuit will follow the Fifth and Ninth Circuits in rejecting the argument that obscene material is entitled to copyright protection.” Devils Films, Inc. v. Nectar Video, 29 F. Supp. 2d 174, 176 (S.D.N.Y. 1998). This would be a minority position, but Courts have been noting a split on the issue ever since (the Second Circuits and most other Circuits have never addressed the issue head-on).
As obscenity has steadily shrank into the darkest corners and immorality has essentially disappeared as a legal standard, it’s easy to miss that there was surprisingly little of substance stating that copyright law protected works which fall into the vague standard of immorality. In fact, Tam can reasonably be stated to be a precedent for copyright’s neutrality, albeit perhaps an unwitting one. It seems that with the decision to no longer examine for immorality or obscenity, the law came to recognize that such works were protected by copyright, even in spite of a long tradition to the contrary, only partially told here. Whether there was a causality to this or it was simply a consequence of the expanding scope of the first amendment is an interesting one.
the discussion of whether the posters at issue were works of the “fine arts” may have been more relevant to the question of whether the 1874 Print and Label Act controlled than to whether they were copyrightable generally, as I discuss in my article Reimagining Bleistein↩
It’s been too long since an update…and I had a couple of interesting old things to share. However, then the Northern District of Illinois granted a motion to dismiss in Sheridan v. iHeartmedia, and I needed to say a bit about it. Some posts have been focusing on the fact that the Court found that performance rights do not exist at common law, but I think the Court actually went substantially further, and substantially abrogated civil protections for sound recordings made before 1972 in Illinois (a state criminal statute still exists for commercial record piracy). In this post I will (hopefully) explain the doctrine, and why this case is a much bigger deal than is being commonly stated. Finally, as always, I’ll include a tidbit from the archives.
As I’ve mentioned a number of times before, I have an article forthcoming on common-law copyright (edits are still being done by the Cincinnati Law Review). In that piece I largely take as a given that common-law copyright protects sound recordings made before February 15, 1972, and then spend close to 100 pages exploring what exactly that means, focusing on the question of performance rights particularly and differences between statutory federal copyright law and common-law state copyright law. And regarding New York and California, at least, that is a fairly safe presumption – given the (cryptic and outdated) California statute, coupled with caselaw from the New York Court of Appeals. However, in other states the applicability of common-law copyright to pre-1972 sound recordings is suddenly appearing questionable, and I wanted to explore a bit what that means.
My article tells the story in much more detail, but as a result of a series of cases culminating in the Supreme Court’s 1908 decision in White-Smith v. Apollo, Courts and the Copyright Office came to understand that the 1909 Copyright Act did not protect mechanical reproductions of musical compositions (i.e. discs or piano rolls) as being protected by copyright law. This began to be controversial in the 1930s, where big bands began to worry about radio stations using their recordings to compete with their (extremely lucrative) live radio programs. Fred Waring and Paul Whiteman were the two primary litigants in these suits, and Waring found success, both at the Pennsylvania Supreme Court and at a Federal District Court in North Carolina (although North Carolina and several other states would abrogate these decisions by statute). Whiteman on the other hand had his position rejected by the Second Circuit Court of Appeals, which would partially reverse itself in 1955 in the case of Capitol v. Mercury.
It is axiomatic that common law protected a creative work (under the pre-1978 laws) until it was published, but publication destroyed common-law copyright, leaving only the possibility of statutory copyright. The arguments at issue in all these cases involved a fundamental question – is a sound recording that is commercially sold “published” within the meaning of copyright law?1 If publication occurred then no rights exist at common law, but if publication did not occur despite the public sale of the phonorecord, then common-law rights remain.
The argument that the sale of phonorecords does not constitute a publication of the sound recording they contain is bipartite. The first part is objective, and argues that selling a record is more akin to public performance (which is not publication) than public sale (which is), because the record is just a captured performance. Because the American rule is that performance is not publication, therefore commercial sale of a record does not constitute a publication. The other half is subjective, and notes that a publication is a dedication to the public, and that the bandleaders in question did not intend to dedicate their recordings to the public without any intellectual property protections – accordingly the records were not “published” and common-law protection remains.
In the 1960s piracy became a serious problem with the advent of prerecorded tapes, and state common law would prove insufficient as a remedy. In 1971 sound recordings were brought within federal law, but this amendment applied only to sound recordings made February 15, or thereafter, of 1972 leaving state common-law copyright as the main source of legal protection for sound recordings made before that point against unauthorized exploitation . Shortly thereafter the US Supreme Court confirmed that state-law protections for sound recordings older than 1972 were both Constitutional and not preempted by the federal statute. A few decades passed, and the New York Court of Appeals once again affirmed that sound recordings from before 1972 remained unpublished in that state.
And that has remained the bedrock of a rich panoply of litigation in recent years over whether the common law right includes a performance rights. And at least in New York state that has resulted in a decision I took issue with that pre-1972 sound recordings lack a performance right. However, as the issue comes before other states, Courts are showing skepticism of the underlying argument made by Waring & Whiteman, both because it’s a transparent (if possibly necessary) legal fiction and because it offers a convenient way to resolve these cases neatly. And as mentioned whether there has been a publication that destroys common law copyright is a question of state law, so we possibly have fifty different standards for whether sale of a phonorecord constitutes publication of the sound recording it embodies – and most states have no caselaw on this specific question.
This leads us to the Sheridan decision which is the spur for this posting. In concluding that the motion to dismiss on common-law grounds should be granted, the Court held that “There is no dispute that the Sheridans voluntarily sold their recordings. When they did so, the Sheridans lost their common law right to control the public performance of those recordings in Illinois (and pretty much everywhere else).” This doesn’t leave the Sheridans without legal protection (48 out of 50 states, including Illinois, have criminal statutes aimed at commercial record piracy. However, these statutes do not go beyond commercial copying and resale, and leave the putative owner of the recording at common law without any civil remedies in most cases.
What happens when a state’s highest court holds that sound recordings are not published? Is federalization finally going to get the boost it needs from stakeholders? Surely leaving such a situation in place would be unacceptable to the owners of the sound recording? I’m curious to find out.
As promised, a tidbit from the archives. The Whiteman v. RCA case has a fascinating file, at the New York location of the National Archives. It includes extensive deposition testimony and more. One thing that I’m not aware of having been circulated is the District Court’s findings of fact and conclusions of law, which are substantially more detailed than its published opinion. Sadly I only have cell phone pictures of the Findings of Fact and Conclusions of Law (PDF 10 MB), but here it is.
I planned on having a whole separate section on whether the protection for pre-1972 sound recordings is really common-law at all – many commentators in the mid-20s century said no, including Nimmer. But that’s really a matter of semantics, fun as those are, and this post is more than long enough already.
One interesting (to me at least) question is whether this question of publication is a matter of state or federal law. I go into this more in the paper, but I’m inclined to think that the question of whether a publication occurred to destroy common-law copyright (a “divestative” publication) is a question of state law, except a successful federal registration is one of the ways common-law copyright is destroyed. On the other hand, whether sufficient publication occurred to have a valid registration under the terms of the 1909 copyright act (a so-called “investative publication”) is a question of federal law. One nice benefit of this understanding is that it clarifies that the differentiation between investative and divestative publication is not an artificial legal construct – they are actually two different things. ↩
A foundational story about the development of copyright in America is the question of whether product labels could – or should – be protected by copyright. As far as I can tell this question marked the beginning of denying copyright registration on the grounds of subject matter, and I think is an important story to tell for understanding a number of current debates regarding copyrightability, and also serves as an antecedent to the modern copyright examination system. I’ve written about this story some in the past, in my 2012 article Reimagining Bleistein: Copyright for Advertisements in Historical Perspective, but it’s worth highlighting this story separately from the broader narrative in that piece, and I’ve found some additional parts to this story since I published that piece.
Although the 1790 Copyright Act only prescribed a limited species of works for protection, for instance omitting music from the list of protected works, Courts would quickly hold that a musical work or even a single sheet of paper could be registered for copyright as a “book” (or after 1802, a “print”). This matches a general early tendency to be liberal in what could be registered. However, in 1848 Justice McLean, riding Circuit, held in Scoville v. Toland that a label that reads “Doctor Rodgers’ Compound Syrup of Liverwort and Tar. A safe and certain cure for consumption of the lungs, spitting of blood, coughs, colds, asthma, pain in the side, bronchitis, whooping-cough, and all pulmonary affections. The genuine is signed Andrew Rodgers” is not subject to copyright. Justice McLean reasoned that the purpose of the label was only to identify the product, and it communicated nothing beyond its association with the product – and accordingly copyright law would not protect it. In the 1850s, the copyright records show a clear upwards trend in registrations of product labels for copyright protection, and coupled with the Scoville decision, this led some to question whether product labels were protected by copyright – especially regarding patent medicines.
A letter from the Clerk of the US District Court for the Northern District of New York (Aurelian Conkling at the time), published in the New York Journal of Medicine and Collateral Sciences in 1856, highlights the problems these so-called “patent medicines” posed for Clerks who were asked to register their labels for copyright (recall that the District Court Clerks were responsible for copyright registration until 1870). The letter explains that the usage of “patent” was sardonic, since “these notrums not truly patented, the secret of their preparation being studiously withheld in defiance of…our Patent Laws.” Rather, these patent medicines would attempt protection by registering their labels for copyright and then bringing suit against competitors for infringement of the copyright. Mr. Conkling had been refusing to grant copyrights for patent medicine labels for several years at this point.1 Conkling further opined that:
Congress did not intend to prevent the imitations of the stamps and labels of any manufactured article, or goods, or merchandise [under copyright law]. That is a subject of such extensive interest and importance, that, if it had been the intention of Congress, to embrace it in the provisions of the law, that intention would have been distinctly and unequivocally manifested.
In response to a letter from Mr. Conkling, the State Department (responsible for the administration of copyright until 1859) issued a circular stating that “inasmuch as mere labels are not comprehended within the meaning” of the copyright law, clerks were instructed to “refuse, in all cases, to record or issue a certificate for the same” under the copyright law.
In 1859, copyright responsibilities were moved to the Interior Department, which delegated them to the Patent Office (although registrations themselves remained the responsibility of the clerks of the district courts). Shortly after this act was put into effect the Patent Office put out a circular to the same effect as the State Department circular a few years earlier, specifically directing the district courts not to register “stamps, labels, and other trade-marks of any manufactured articles, goods, or merchandise” because “the acts of Congress relating to copyright are designed to promote the acquisition and diffusion of knowledge, and to encourage the production and publication of works of art,” and thus labels were not “embraced within the meaning of the [copyright] acts.” Thus a label, no matter how artful, could not be registered as copyright, but might be registered as a design patent.2 It’s likely that the list of labels, trademarks, etc that the Patent Office produced in 1859 and supplemented in 1861 stemmed out of the release of this circular.
In spite of these clear pronouncements, compliance with these circulars varied. In a Series of Letters from 1864-1866 (PDF, 4 MB, from the miscellaneous files of copyright records in the Rare Book Room of the Library of Congress), a successor as clerk of the District Court for the Northern District of New York complained that he was getting many requests to register labels, and that a common complaint was that the Southern District of New York had no such compunctions about issuing copyright registrations for product labels. Given that the Southern District had a much higher volume of copyright registrations that is perhaps understandable that less attention would be paid at the Southern District (there are 4,484 pages of copyright records microfilmed from the Northern District from 1820-1870, compared with 54,822 from the Southern District for the same period). Perhaps more notable is that aside from the Northern District of New York, there’s no real record of other courts refusing to register labels (that I’ve found – the copyright records of many Courts are overflowing with labels by the end of the 1860s).
Nonetheless, the Copyright Office issued a Circular in response in 1866 (PDF, 1 MB, this copy from the NJ 1846-1870 Copyright Files at the Library of Congress) making clear that their views on whether labels were protected by copyright had not changed. The language the Commissioner of Patents used is a development from Justice McLean’s opinion in Scoville v. Toland 18 years earlier, and it has a whiff of the concept of aesthetic separability that was argued in Mazer v. Stein 88 years later.
As all the acts of Congress related to copyrights are designed to promote the acquisition and diffusion of knowledge, and to encourage the productions and publication of works of art, it has always been held that stamps, labels, and other trade-marks of any manufactured article, goods,or merchandise are not embraced within the meaning of the acts. When any production is issued as an object of art, having value in itself, and intended for sale as such, it properly comes within the provisions of the copyright law; but when, however artistically executed, it is not produced for sale as a work of art, but is designed to be affixed in the manner of a label to a manufactured article, it then plainly falls under the act relating to patents for designs, and consequently cannot be protected by copyright.
As noted, despite repeated encouragement from Washington, it does not seem that other Districts followed the lead of the Northern District of New York. As I tell in much greater detail in my article, when copyright was centralized at the Library of Congress in 1870, the Library was quickly overwhelmed with applications to register product labels for copyright, and in 1874 the Librarian convinced Congress to pass a law transferring responsibility for registering labels to the Patent Office, where they would be a sui generis form of intellectual property registered along the same rules as those for registration of copyright. The statute attempted to create this entire new form of intellectual property in one paragraph of statutory language, and hijnks predictably ensued, including a case where the Patent Office repeatedly ignored mandamus from what is now the D.C. Circuit Court of Appeals to register labels without examination.
The decision of the Supreme Court in the Trade-Mark Cases in 1879 (holding the 1870 Trademark Act unconstitutional) did not stop the process of label registration, but the Supreme Court’s 1891 holding in Higgins v. Keuffel that a label for india ink was not protected by copyright did pause the system of label registrations with the Patent Office for a number of years.
However, in 1893 the Comissioner of Patents ruled that labels could be registered, provided they were of sufficient originality to meet the authorship requirements of the copyright clause (which the label from Higgins did not). However, the Comissioner held that the advertisement at issue in Ex parte H.J. Heinz Co., 62 O.G. 1064 (Comm’r Patents, 1893) was of sufficient originality to qualify for registration as a label.
Somewhat surprisingly, commercial labels and advertisements would not be returned to the jurisdiction of the Copyright Office until 1940. The story – especially after 1874 – is told in greater detail in my article, but I’ve brought it up a few times lately, and it seemed to be worth highlighting.
Apparently the denied copyright registrations included a “elixir of life” as well as a “diarrhoea cordial.” Letter at 423. Surely the latter would test the limits of aesthetic non-discrimination even today. ↩
This paragraph is copied essentially verbatim from pg. 352 of my 2012 article. What’s next is new, though. ↩
Even more than most, this post is my opinion only, and reflects views and opinions that should only be imputed to me. The gist: The 9th Circuit should reverse in ABS v. CBS, digital remasterings of sound recordings absolutely can be protected by copyright, but aren’t always and it requires factfinding, and a digital remastering does not de facto federalize a sound recording from before 1972 protected by state law.
As some of you may know, I have a fairly lengthy piece on the scope of common-law copyright for sound recordings It’s available here, although it’s still a work in progress (the staff at the Cincinnati Law Review are currently doing edits, so pagination will change some). If you want an audio version, you can also hear a version of my talk on the issue from the Library of Congress.
One nice thing about having this blog is that I don’t need to wait for that piece to come out, though, to explain how the decision of the US District Court for the Central District of California in ABS v. CBS is incorrect and demands reversal. It misunderstands the basic ordering of authorship and derivative works, and I’d like to briefly explain why. That decision is current being briefed before the 9th Circuit, and is being briefed (the Appellants’ brief is here; Appellees’ Answering Brief is here)
ABS and a host of other parties assert that they own the common-law copyrights in sound recordings made prior to February 15, 1972. Under the current law, such older sound recordings are not protected by federal law, but may be protected by state law protections. The preemption provisions of the copyright law contain a specific exception until 2067, so protection for all pre-1972 sound recordings will terminate then. In practice, most states tend to follow the model of indefinite protection offered by New York under the doctrine of common-law copyright (the protection given at common law to unpublished works, under the legal fiction that sound recordings that have been commercially released are nonetheless unpublished), but the modern case law regarding whether common-law copyright includes a performance right (aka a broadcast right) is surprisingly sparse. One of the things I point out in my article is that if you cast a broader net there’s plenty of caselaw from the late nineteenth century, all of which agrees on a performance right at common law, but Courts thus far haven’t embraced this argument completely. In response to this confusion the Copyright Office recommended federalizing protection for pre-1972 sound recordings, but that has not yet happened.
Earlier litigation brought by Flo & Eddie of the Turtles argued that digital radio services that would be required to pay a royalty under federal law were not doing so under state law, leading to the surprising determination by the New York Court of Appeals that no performance right exists for pre-1972 sound recordings, a decision whose ramifications have yet to fully play out. However, that decision only applied to New York state in theory (how much deference other states will accord it remains unclear), so the lawsuits brought by ABS et al are going forward in a number of other states. However, ABS made a broader claim than Flo & Eddie had, claiming that all terrestrial stations (ie regular radio) had to pay a royalty for playing their sound recordings, a requirement that does not exist at federal law. CBS Radio, one of the largest station operators in the nation, was sued by ABS in the Central District of California.
The issue for the Court was that on one hand, the weight of precedent was that common-law seems to require a performance right in California (indeed, the same Court had just ruled that), and there’s no real principled way to create a separate rule for terrestrial radio as opposed to satellite radio under existing doctrine.1 However, there’s currently no requirement for paying royalties for plays of sound recordings on terrestrial radio, and there are concerns of administrability and uniformity from finding a common-law performance right on a state-by-state basis.
Facing long odds given the Court’s recent decision on performance rights at state law, CBS tried a clever move – they argued that they weren’t playing pre-1972 sound recordings at all, but were rather playing post-1972 sound recordings, which were protected by federal law – but had no performance rights for terrestrial radio. The argument is that because the radio station was not playing vinyl records made before 1972, but were rather playing subsequent digital transfers, the relevant sound recordings were the digital transfers and not the originals. As such, the case should be dismissed because all parties understood no cause of action existed under federal copyright law.
Firstly, the Court looked at a number of sources, including the Compendium of U.S. Copyright Office Practices, Third Edition (the Copyright Office’s examining manual), and held that for every sound recording at issue (over 100), a derivative work had been created when it was remastered into a digital version.2 However, section 803.9 of that manual indicates that the following words are indicative of de minimis creativity where no sufficient authorship exists for an independent copyight: “Declicking, New format, Noise reduction, Reissue.” In addition, “the term ‘remastering’ may refer to authorship that is mechanical or too minimal to be copyrightable.” Put another way, a remastering absolutely can be independently copyrightable, but this is a fact-intensive question that cannot be easily resolved. The District Court’s failure to engage in factfinding on the issue for each an every sound recording at issue is an easy ground for reversal. However, there is a deeper problem with the District Court’s decision’s, one which moots the need for such a factfinding – even if a derivative work was created in the transfer to digital and radio stations are playing that derivative work, radio stations are still playing the underlying pre-1972 sound recording as well.
The RIAA’s Amicus Brief makes this point especially well – a derivative work only includes what is new – anything in the original work is protected by the copyright in the original work only. This is the core holding the Supreme Court reached in its 1990 decision in Stewart v. Abend regarding the copyright in the film Rear Window. Creating a derivative work like a digital remaster of an analog recording does not replace the protection for the underlying work. This isn’t a statement of opinion, it’s black letter law, at 17 U.S.C. 103(b):
The copyright in a compilation or derivative work extends only to the material contributed by the author of such work, as distinguished from the preexisting material employed in the work, and does not imply any exclusive right in the preexisting material. The copyright in such work is independent of, and does not affect or enlarge the scope, duration, ownership, or subsistence of, any copyright protection in the preexisting material.
The District Court ignored a core tenet of copyright law in concluding that performance of a derivative work in a pre-1972 sound recording is only performing that derivative work, and not the original as well.
Having spent much of the recent year thinking about common-law copyright, I have devised an alternative argument along the lines of the CBS argument, one which I think avoids this pitfall. I’ve included it below the jump, but be warned that I don’t think it actually works (especially in California) – and I’ll explain why too. It’s more a conceptual exploration of how common-law copyright plays out differently from statutory copyright.
California actually has a statute for common-law copyright, so it’s not actually common-law, but it’s generally understood to just be a codification, and it’s short on detail ↩
In theory the Compendium is only given Skidmore deference by the Courts, meaning it is advisory but not binding. However, in practice the Compendium is more than a series of arbitrary examining rules – the 3rd Edition of the Compendium was one of the triumphs of Maria Pallante’s tenure as Register, and represents a thorough attempt to reckon with the body of statutes and caselaw that govern copyrightable works to determine what the Copyright Office should and should not be registering for copyright. In practice it is now the leading treatise on what copyright does and does not protect. in the United States ↩
Following that came the task of modernizing the business of copyright registration – although Ainsworth Spofford (Librarian of Congress 1865-1897) had effected the centralization of copyright registration, in contrast to the earlier practice of registering copyright in the local federal court, the actual record-keeping done by Spofford was no better than what was done by the District Courts, with a general ledger of copyright registrations being kept, containing a few fields of basic information and then individually signed by the Librarian.1 At the beginning of the 20th Century Solberg introduced the “XXC” classification system and system of application forms, which would be used in modified fashion until the effective date of the 1976 Act. In 1903 the Copyright Office improved the Catalog of Copyright Entries as it evolved from an obscure circular into a proper catalog, including statistical information for the first time.
Also at the beginning of his tenure, Solberg effectively engineered the appointment of Herbert Putnam to be librarian, and would serve alongside him from 1899-1930 (when Solberg retired). With the Copyright Office reaching a certain operational stability, attention then turned to the question of reforming the copyright laws. It’s fairly well-known that the original 1790 Copyright Act was largely an Americanized version of the 1710 Statute of Anne. The 1831 Act was a major advance, expanding the scope of copyright (music was not technically protected by the 1790 Act, for instance, although in practice music was registered as a “book” or print”), and making many other reforms. For most of the nineteenth century the main question of American copyright law was whether to extend copyright to foreigners, a measure that did not become law until 1891. In the interim there had been an omnibus revision to all intellectual property laws in 1870 – including a revised patent code and the first instance of federal trademark law – but aside from the change of venue to the Library of Congress, the 1870 Act was not a major shift from the 1831 Act. By the early 20th century motion pictures and sound recordings were already being widely consumed, and a heavily modified copyright law from an entirely different era was clearly inadequate. Given that international copyright had withstood a number of challenges in its first decade and now felt secure, it was time for a 20th century copyright law.
Beginning in his 1901 Annual Report, Solberg would annually advocate for a new copyright law. The ball began rolling in the Senate Committee on Patents, which requested that a commission be formed to study the copyright laws in January of 1905, after which the Librarian delegated to Solberg to plan a series of conferences to meet with copyright industry representatives in the Spring of 1905 to the Spring of 1906. Transcripts of these conferences were created and were much later compiled as part of the printed Legislative History of the 1909 Copyright Act, currently available on Heinonline. By this time, Solberg was also deep into a period of study of the copyright laws that would leave him the undisputed master of them. His first major work was a study of every copyright enactment in the United States up to 1900, published that same year. In February of 1905 he followed that with his study Copyright in Congress: 1789-1904, which chronicles every action regarding copyright in Congress, including failed bills and long-forgotten petitions and memorials. By the time the revision process began, Solberg had quite literally read every proposed copyright law in this Country’s history.
Following the Spring 1905 conference, Solberg and his small staff at the Copyright Office began the project of drafting a copyright bill out of the suggestions and comments made. By the Spring of 1906 a bill was ready, and was published along with comparisons to the current statute. The bill was introduced in May of 1906, and underwent extensive revision – with Solberg in consultation every step of the way.2 The Librarian was also involved in the discussion as the bill moved forward, but the Copyright Office had become the experts on copyright law.
As mentioned, Solberg would stay in office until 1930, serving alongside the Librarian he had helped appoint all along. Even after his retirement the Copyright Office maintained its position as the governmental experts on copyright, and further existed as a largely independent agency ingrown into the Library of Congress, left alone by Librarians who had other priorities – and who also were not prepared to bring the requisite subject-matter expertise in copyright to bear.
In addition, Spofford had also been alarmingly lax in his handling of funds deposited by repeat copyright registrants ↩
Many bills were introduced during this period, but they were mostly revisions of proposed revisions of the Solberg bill ↩
Those who looked closely at the guide to these records hosted by GW Law may have already noticed these records. They fall into two categories of records – letter books of copyright correspondence with the State Department, and what I’ve termed “ephemeral” copyright records, to indicate small lists of copyright registrations, usually compiled by the clerk of the court.
The letter books are correspondence related to copyright with the Secretary of State, who was mostly responsible for copyright deposits until 1870. There are three continuous letter books for 1818-1824, 1824-1828, and 1828-1832. The bulk of the letters in these books are cover letters that accompanied individual deposits, although there are other documents as well. For instance, there is a letter in the 1828-1832 book from counsel for Henry Wheaton, attempting to establish the proper deposit of Wheaton’s Reports as part of the litigation that would become the Supreme Court case of Wheaton v. Peters. There is also a book that appears to be deposits with the State Department for 1853-1854, but it is on onion paper and was too fragile and voluminous to scan easily. According to the Roberts survey of the copyright records, there is also a letter book of the Patent Office regarding copyright matters for 1859-1870 as well in the Rare Book Room, which would presumably be digitized along with the rest of the copyright-related material held there.