Its time to backtrack a bit and explain what I mean by an “Erie v. Tompkins” dilemma. I promised when I started this blog that I would not descend into legalese, but now I’m going to have to get a bit close. I promise that this is as bad as it will ever get.

Although you’ve probably never heard of it, 1938’s Erie Railroad  v. Tompkins is probably the single most important Supreme Court decision since Marbury v. Madison, the case that determined that the Supreme Court was the ultimate arbiter of what is or isn’t constitutional. The facts in Erie v. Tompkins were almost absurdly simple. Tompkins was a citizen of Pennsylvania. The Erie Railroad was a corporate citizen of New York. Tompkins was hurt in a railroad accident that occurred in Pennsylvania. State law in Pennsylvania favored railroads. State law in New York favored victims. Cases decided in federal court leaned towards victims as well. Whose law applied?

Most people think that the federal courts exist to decide matters pertaining to federal law, so-called “federal question” issues. Surprisingly, this is usually not so. Under Article II of the Constitution, the federal courts are given jurisdiction over disputes “between Citizens of different States.” So by the time of the Erie case, there was already a well-established way to prevent a plaintiff from gaining an automatic home-field advantage over a defendant.  For example, if a citizen of State A wanted to sue a citizen of State B, Citizen A couldn’t just walk into his home state’s court, file, and demand that Citizen B show up. Instead, A must go and file suit in the federal court in B’s home state. This is called “diversity”, and most cases in the federal courts are these diversity cases. Tompkins did this, suing the Erie in New York federal court.

The court, following then-standard practice, applied federal law and awarded Tompkins $30,000. The Supreme Court reversed this, determining that the federal court should have applied Pennsylvania law, because that is where the accident happened. In other words, the federal court should have “stepped into the shoes” of a Pennsylvania state court. The only time federal law should be applied, according to the Supreme Court, is when the dispute involves a federal question. Thus, to this day, you’ll see the seemingly bizarre spectacle of a North Dakota federal court applying California law.

The federal court in Erie said that Pennsylvania’s laws should apply because the law, in both New York and Pennsylvania, said “the court where the accident happened has jurisdiction.” Predictably, the next controversy arose over how to decide whose laws should apply when the two don’t agree. That is, whose law should be used to decide whose law gets used? Again, the first rule applied was that federal procedures (called “conflict of law” rules) should apply. During the period of time we have been discussing, the 1940s and 1950s, there was huge turmoil between many federal and state courts (especially in New York) about whether federal or state conflict of law rules should be applied in diversity cases.

Recall the Capitol v. Mercury case, the one dealing with the Telefunken imports. Judge Dimock, representing the majority view, held that phonorecords were not within the category of writings that were “copyrighted works”. Therefore, they were controlled by state law; and thus, state law determined if they were “published.” Judge Hand, to the contrary, argued that all things that are writings are “copyrighted works”  under the language of section 4 of the 1909 Copyright Act.  Therefore, the records were not controlled by state law, and federal law alone determined if they were published or not. This argument boiled down to whose law should apply. If federal law applied, “published” could be defined one way. If state law applied, “published” could be defined another way. In reality, this was a struggle over judicial power, a struggle that was also at the heart of the Capitol v. Naxos IV decision.

To be continued…

Bruce D. Epperson 

Capitol Records v. Mercury Records is justifiably famous, not for the decision itself, but for Judge Hand’s dissent. Judge Dimock read the majority opinion. Capitol was the American agent for Telefunken records. Mercury distributed the same records under license to a Czech alien properties administration. The records did not contain songs copyrighted in the USA.

The first question was: whose law applies? Judge Dimock ruled that under Erie v. Tompkins, New York state law applies. Under New York law, whose contract rights prevail? Through a relatively complex chain of legal reasoning that need not concern us here, Telefunken had the superior claim.

Now for the crucial issue: did Telefunken lose their common law copyright in the sound recordings through publication? Up to RCA v. Whiteman (1940), the answer would have been yes, but, according to Dimock, RCA v. Whiteman had been replaced by Metropolitan Opera v. Wagner-Nichols (1950). Therefore, Telefunken had not lost their monopoly to make or sell records.

Judge Hand’s dissent was longer than Dimock’s opinion. It was a refinement and extension of the things he had said a decade before in Fashion Originators’ Guild and RCA v. Whiteman. He opened with a flat-out statement of two principles: 1) The performance or rendition of a composition is a writing for the purposes of the Progress Clause; and 2) It is a thing separate and apart from the composition. By extension from these two principles, Congress could, if it wanted to, grant the power of statutory copyright to the performer of a sound recording if it were affixed to a tangible medium.

However, Congress had not extended such protection. Moreover, section 4 of the Copyright Act of 1909 said that “The works for which copyright may be secured under this act shall include all the writings of an author.” Therefore, all things eligible under the Constitution for inclusion in the Copyright Act already were included. That is, there was no such thing as “writings not covered by the Copyright Act.” Thus, there was no such thing as “common-law copyright.” There were simply writings covered under the Copyright Act, but for which the specified coverage mandated no regulation or restriction. (This is the inverse of the cupcake theory: 1) statutory copyright cuts off all common-law copyrights forever; 2) statutory copyright covers all writings; 3) however, for some writings, the protections extended in the statute are zero.)

However, Hand continued, here today, this is an academic discussion. The question is not whether Telefunken can stop the plagiarism of its records by Mercury. The question is whether by public sale its common-law property has been divested. The majority opinion assumed that under Erie v. Tompkins, state law determines whether “publication” has taken place. But because phonorecords, in both their composition and sound recording aspects, are writings under section 4 of the Copyright Act, the question of whether they are “published” is a federal question, to be decided solely under federal law.

If, for example in the case at bar [Mercury] is forbidden to make and sell these records in New York, that will not prevent it from making and selling them in any other state which may regard [Telefunken’s] sales as a “publication”; and it will be practically impossible to prevent their importation into New York. That is exactly the kind of evil at which the clause [section 4] is directed.

Now, move forward to 2005. Substitute “Capitol” and “Naxos” for “Mercury” and “Telefunken.” The case is exactly the same. The only difference is that in 1950, as Judge Dimock briefly noted, there were no applicable international treaties or federal law. In 2005, there was. Judge Graeffo’s opinion in Capitol v. Naxos was, in essence, a rebuttal to Judge Hand. Dimock argued that because there was no federal law, state law controlled. Judge Hand argued that there was applicable federal law, so it did apply. Judge Graeffo’s argument was that although there was federal law, New York law still prevailed.

To be continued…

Bruce D. Epperson

Following Fashion Originators and RCA v. Whiteman, the definitive case in this series occurred in 1955 in Capitol Records v. Mercury Records. But before then, there were some suggestive lead-up cases. The first was in an Illinois federal court case, Shapiro, Bernstein & Co. v. Miracle Record Co. The case was poorly explained by Judge Igoe, but apparently the Shapiro firm had issued a record containing a song written by one Lewis, that had never been copyrighted in the required sheet music format. Miracle Records produced their own record of the song. Miracle argued that their record was not a copy of the composition. Shapiro, Bernstein, on the other hand, argued that because the sheet music had not yet been submitted for copyright, the composition was in pre-publication status, and thus Miracle’s record violated their common-law copyrights.

Judge Igoe argued that “publication is a practical question and does not rest on any technical definition of the word ‘copy.’” Citing Judge Learned Hand’s opinion in RCA v. Whiteman as “very close to our case,” he concluded that when Lewis permitted his composition to be issued on a phonorecord, it was published, and thus amounted to a dedication to the public. He implied, but did not explicitly state, that if Lewis had secured federal copyright of his sheet music first, then the publication of the phonorecord would not have affected those rights. If so, this meant that issuing phonorecords amounted to publication, leading to divestment, for common-law copyright in the composition, but not for statutory copyright.

A second, far more significant case was Metropolitan Opera Assn. v. Wagner-Nichols Recorder Co., also in 1950. It was a fairly straightforward case. The Met had an exclusive contract with Columbia Records to record its opera performances. Wagner-Nichols lifted the Met’s famous Saturday afternoon radio performances and used them to make albums competing with Columbia’s. The Met sought an injunction to block Wagner-Nichols. There were several ancillary matters dealing with issues such as intentional misrepresentation of goods beyond the scope of the present discussion. But, the major contention between the opera company and Wagner-Nichols was the latter’s assertion that the Met had “no property right in the broadcast performances and that [Wagner-Nichols] are therefore free to record these performances and sell their recordings.”

Although it never cited the earlier New York case, Metropolitan Opera was a throwback to the 1909 Fonotopia case. Its distinctive feature was the idea that “unfair competition” could be broadened beyond the idea of “palming off” mislabeled goods to include their actual misappropriation: “in recent years its scope has been extended. It has been held to apply to misappropriation as well as misrepresentation . . .to misappropriation of what equitably belongs to a competitor.” It is axiomatic that for misappropriation to occur, there must be property to misappropriate, so it was no surprise that the court held that the Met had a property right “protected in the creative element in intellectual productions—that is, the form or sequence of expressions.”

The court also found that the broadcasts did not amount to publication. Here, the court pulled a bit of legal bait-and-switch. After quickly noting the old Ferris v. Frohman stage play line of cases, the court turned to a recent Supreme Court case, International News Service v. Associated Press. But the issue in International News wasn’t whether the press items in dispute had been published, but whether they had been abandoned — a rather different thing. Abandonment turns on the question of intent; publication depends on the extent of actual distribution regardless of intent. The court then proceeded to ignore the question of whether the Met had published its sound recordings, holding instead that it had not abandoned its property through the broadcasts, because it had not demonstrated the requisite intent.

To be continued…