Ninth Circuit Refuses to Adopt “Ordinary Observer” Test for Substantial Similarity and Copyright Infringement
December 2 2021
The Ninth Circuit was recently asked to determine whether to continue to apply the Circuit’s two-part extrinsic/intrinsic test for “substantial similarity” with regard to a copyright infringement claim or to depart from this approach and apply the Second Circuit’s “ordinary observer” test instead. In Johannsongs-Publishing, Ltd. v. Lovland, an unpublished opinion issued on November 29, 2021, the Ninth Circuit declined to depart from its precedence and affirmed summary judgment in favor of defendants who were accused of copyright infringement in connection with the song You Raise Me Up.
Johannsongs-Publishing claimed that it owned the rights to a 1977 song, Soknodur, except as to the rights for that song’s lyrics. Johannsongs-Publishing claimed that You Raise Me Up, which was written by Rolf Lovland and Brendan Graham and released in 2001 by Secret Garden and later by Josh Groban in 2003 infringed on its rights in Soknodur. In 2018, Johannsongs-Publishing filed a copyright infringement claim against the defendants who were involved in publishing and/or selling You Raise Me Up. The defendants moved for summary judgment on the grounds that the elements claimed to be similar between You Raise Me Up and Soknudor were not actually “sufficiently similar” to support a finding of copyright infringement. They further claimed that any similarities that did exist were derivative of Irish folk songs such as Danny Boy (aka Londonderry Air), which was in the public domain.
In connection with the defendants’ motion for summary judgment, each side submitted reports by experts regarding the alleged similarities (or lack thereof) between the two songs. These reports were crucial for the district court in reaching its finding that there was no substantial similarity to support an infringement claim and granting summary judgment to the defendants.
Before turning to the differences in the expert witness reports, the district court examined the legal framework for a copyright infringement claim. The basic elements that a plaintiff must prove are: “(1) ownership of a valid copyright; and (2) copying of constituent elements of the work that are original.” There was no dispute among the parties that plaintiff did hold a valid copyright in Soknodur (except as to the lyrics). Rather, the defendants focused their attack by claiming there was no triable issue as to the element of copying.
The district court continued by acknowledging that the copying element of an infringement analysis “contains two separate components: `copying’ and `unlawful appropriation’.” The court noted that while these two issues are usually referred to in shorthand as “substantial similarity,” they are in fact distinct concepts. First, the district court noted that “[p]roof of copying by the defendant is necessary because independent creation is a complete defense to copyright infringement … [n]o matter how similar the plaintiff’s and the defendant’s works are ….” Furthermore, the requirement of “unlawful appropriation” takes into account that “[the Copyright Act] does not forbid all copying” and “excludes such matters as ideas, concepts or principles from its protection.”
Thus, the district court boiled down its analysis as follows: First, in order to prove copying, a plaintiff must show “the similarities between the two works need not be extensive, and they need not involve protected elements,” but rather, “[t]hey just need to be similarities one would not expect to arise if the two works had been created independently.” Second, to prove unlawful appropriation, a plaintiff must show that “the similarities between the works must be `substantial’ and they must involve protected elements of the plaintiff’s work.”
The district court then applied Ninth Circuit precedence in its analysis of whether there was unlawful appropriation by employing the two-part extrinsic/intrinsic test. It noted that the “extrinsic test compares the objective similarities of specific expressive elements in the two works” and “often requires expert analysis.” On the other hand, the intrinsic portion of the analysis “test[s] for similarity of expression from the standpoint of the ordinary reasonable observer, with no expert assistance.” To prevail on a claim of copyright infringement, a plaintiff must satisfy both tests “for the works to be deemed substantially similar.” However, for purposes of summary judgment, a court can only consider whether or not there is a triable issue as to the extrinsic test, not the intrinsic test, which is a question of fact left to a jury.
In turning its analysis to the extrinsic test, the district court recognized that a plaintiff “must first identify the sources of the alleged similarity between the two subject works.” For purposes of analyzing musical compositions, the court should consider a variety of elements, such as “melody, harmony, rhythm, timber, structure, instrumentation, meter, tempo and lyrics.” This analysis will generally require expert testimony and the court is also required to “filter out” the unprotectable elements in a plaintiff’s work before assessing whether there is “substantial similarity.”
Defendant’s expert submitted a detailed report that analyzed the five musical elements of Soknudor and You Raise Me Up: the song’s component elements individually and in combination; the prior art that existed as of the 1977 composition of Soknudor; and finally, the two songs in their entirety of their component elements and prior art. The defendant’s expert included visual illustrations showing and comparing the structures, harmony/chord progressions and melodies and comparing these two songs with prior art songs. The expert also attached multiple audio recordings to support his analysis. The court found the defendant’s expert analysis to apply “reliable principles and methods to the facts of the case.”
The defense expert found that the two songs lacked significant structural similarities and he gave multiple examples of how their structures differed. He found a similar lack of similarity between the harmony/chord progressions in that any harmonic similarities were either commonplace or present in prior art, such as the song Danny Boy. He further analyzed the differences between the rhythms and melodies to show how they had no significant similarities. Furthermore, he opined that “[a]ny melodic expression in common between [the two songs] is found in Irish folk songs that date back to the 18th and 19th century ….” The expert went on to do a detailed analysis of the individual notes in the first verse and chorus to again show there was no substantial similarity between the songs.
Johannsongs-Publishing, on the other hand, submitted a report from an expert that the court found to be “fatally flawed” and therefore declined to consider it. First, the plaintiff’s expert’s first report was titled “preliminary comparison” and was “far less comprehensive” than the defendant’s report, consisting of only seven paragraphs of preliminary findings. The plaintiff’s expert had apparently only compared the “skeletal pitches” between the two songs and opined that there was some similarity. Crucially, the plaintiff’s expert had not done any analysis with regard to the prior art but had merely speculated according to the court that “it is highly doubtful that another musical work will share the same lengthy series of skeletal pitches and musical events to the degree of similarity” between the two songs at issue. The district court found, however, that because this expert’s report did not analyze the two songs in a matter consistent with the Ninth Circuit’s extrinsic test, the opinions expressed therein were “legally deficient and irrelevant.”
Hoping to cure these deficiencies, the plaintiff submitted a rebuttal report by the expert, which the court found to be “more extensive” but that it did “little to reinforce the scanty conclusions presented” in the expert’s preliminary report. The district court noted that while the rebuttal report also contained some analysis of the prior art, specifically Danny Boy, it did “not do so in a legally relevant way: it does not filter out elements of these prior art songs from Soknudor and [You Raise Me Up] and compare the remainder as the case law requires; rather, it just tallies up the number of similarities and concludes that there are more similarities between Soknudor and [You Raise Me Up] than between either of them and [Danny Boy].”
The district court agreed with the defendant’s expert that the plaintiff’s expert report were “unreliable and reached unsupportable conclusions.” These defects included the plaintiff’s expert apparently considering “notes to be similar even when they appear in different places in the song’s melodies;” identifying “identical pitches” in both songs but then admitting that there were intervening pitches between some of these notes; and finally, failing to include “sufficient supporting evidence – like comparative transcriptions or sheet music – for the court to assess the validity and accuracy of [the] analysis.”
Given what it found to be the reliability of the defendant’s expert reports versus those of the plaintiff’s expert that it found to be less reliable, the court found that the defendant’s “extensive analysis of the melodies is conclusive: any melodic similarities between Soknudor and [You Raise Me Up] are either unprotectable because they are found in prior art songs including [Danny Boy], or they are too scattered to amount to substantial similarity.”
The district court likewise rejected the plaintiff’s claim that copyright protection may apply “under appropriate circumstances in the selection and arrangement of unprotected components.” The court found that “neither plaintiff nor its expert engaged in any analysis that could support a selection and arrangement theory.”
Unsurprisingly, the plaintiff appealed the district court’s summary judgment ruling to the Ninth Circuit. While the Ninth Circuit’s unpublished opinion is brief, it appears that Johannsongs-Publishing urged it to find that the district court had erred in applying the two-part extrinsic/intrinsic test in determining whether there was substantial similarity, and instead urged the court to apply the “ordinary observer” test that had been adopted by the Second Circuit in the case Andy Warhol Foundation for Visual Arts, Inc. v. Goldsmith, 11 F.4th 26 (2nd Cir. 2021). The Ninth Circuit declined this invitation finding that the two-part extrinsic/intrinsic test was Ninth Circuit precedence and that there had been no U.S. Supreme Court decision that had undercut the Ninth Circuit’s prior analysis. The Ninth Circuit concluded that it would not disturb the district court’s findings as to the reports of the parties’ various experts and therefore affirmed summary judgment in favor of the defendants.
The Johannsongs-Publishing case is a reminder of the burden that plaintiffs may have to satisfy in proving “copying” in support of a copyright infringement claim. Attorneys representing these plaintiffs are reminded that they must adhere to the precedence of the circuit in which they are litigating and should not hope that the circuit court may ignore its precedence and apply an approach from other circuits that may be more favorable.