Second Circuit Resoundingly Rejects Broad Application of Google v. Oracle

Leading up to the Supreme Court’s Google v. Oracle decision in April, some were hoping for a game-changing outcome that would alter the course of copyright’s fair use doctrine. But then came Justice Breyer’s opinion, and with it a good amount of head scratching. While it was no doubt a victory for Google, the extremely limited nature of the Court’s fair use analysis did little to advance copyright law or reshape the parameters of the four-factor test outside of a very specific type of software code. As a result, what might have been a very important fair use case now looks more like a footnote in the history of copyright.

Regardless of the Supreme Court’s narrow interpretation, supporters of ever-expanding fair use boundaries celebrated the decision, insisting it should be applied broadly to unauthorized uses involving all types of copyright protected works. The first test of this illusory theory came within a couple weeks of Google, when the Andy Warhol Foundation (AWF) filed a petition for rehearing of a Second Circuit decision finding that Warhol’s unauthorized use of photographs of Prince did not qualify as fair use. The petition, and supporting amicus briefs, went all in on the Supreme Court’s transformative fair use determination, arguing that the Second Circuit was now in direct conflict with the high Court. But on August 24, the Second Circuit issued an amended opinion “emphatically” rejecting petitioner’s claims and reenforcing the clearly limited applicability of Google v. Oracle.

Google v. Oracle is a Misguided but Ultimately Limited Decision

It’s probably safe to say that very few predicted how the Supreme Court would approach the questions of copyrightability and fair use at the heart of the software infringement dispute in Google v. Oracle. Justice Breyer’s opinion inexplicably addressed the fair use factors out of order, starting its analysis with the second factor—the nature of the copyrighted work. Finding right off the bat that the underlying declaring code at issue served a “functional purpose” for which copyright protection is “thin,” Breyer set the stage for an inevitable fair use holding. Added to this skewed start were equally perplexing analyses of the third and fourth fair use factors, as well as a problematic factor-one transformative use analysis that excused Google’s copying even though it merely shifted use between computing mediums or devices.

While Breyer’s opinion presented a troubling misapplication of the fair use factors and a greater misunderstanding of the goals of the copyright system, one saving grace is that Breyer made abundantly clear that the Court’s decision would have very limited applicability. As we explained just after the decision, the Court clearly states that the decision “does not overturn or modify its earlier cases involving fair use.” Numerous times throughout the opinion, Breyer says that the fair use analysis is limited to the case at hand and only applies to software, and more specifically to declaring code. So despite arriving at questionable fair use conclusions, the Court’s approach is unambiguously confined to a very specific subset of software code and should not be applied outside of those distinct works.

Warhol Foundation and Amici Try to Turn Google v. Oracle into Something it’s Not

In Warhol Foundation v. Goldsmith, the Second Circuit reversed and remanded a district court’s fair use holding, finding that courts have misapplied the standard it set forth in Cariou v. Prince and by doing so have greatly diminished the copyright owner’s right to control the creation of derivative works. At issue in Warhol was a series of 15 silkscreen prints and pencil illustrations that the late appropriation artist based, without authorization, on a photographic portrait of Prince taken by Goldsmith in 1981. Rejecting the district court’s transformative use analysis, the Second Circuit found that to be transformative enough to qualify as fair use a secondary work must be “fundamentally different and new” and embody an “entirely different artistic purpose” so that it “stands apart from the raw material”—all of which the Warhol prints failed to do. The opinion provided welcome guidance on how to analyze what constitutes transformative use as opposed to derivative use (falling under the derivate work right), a crucial clarification that corrected case law and commentary that have increasingly blurred the line between transformative use and derivative works in recent years, all but eliminating a copyright owner’s exclusive right to prepare derivative works.

Just over a week after Google v. Oracle was decided, the Andy Warhol Foundation (AWF) filed a petition for rehearing claiming that the Second Circuit’s transformative fair use determination was overruled by Google. Specifically, the petition argues the Second Circuit’s decision denies transformative status to any work in which copyright protected source material is recognizable, which it claimed cannot be squared with the Supreme Court’s holding that verbatim copying of software code qualified as a transformative use. But AWF conveniently ignores the Supreme Court’s repeated explanations that its approach in Google is confined to very distinct forms of computer code that incorporates functional elements. The “thin” protection Breyer speaks of has never and should never apply to works of pure creative expression, and the broad transformative use standard adopted in Google cannot be applied to any work outside of declaring code.

Supporting AWF’s petition for rehearing was a group of 60 law professors who filed an amicus brief that also argued the Second Circuit’s opinion conflicts with the Supreme Court’s holding of transformative purpose fair use. The brief argues that the Second Circuit erred in adopting a transformative standard that asks: “how the works may reasonably be perceived,” which the professors claim conflicts with the Supreme Court’s reliance on Google’s intent and the understanding of third parties to evaluate the purpose of the use. But the brief doesn’t acknowledge that intent and purpose of a use is much more easily evaluated in a case involving the appropriation of a semi-functional computer code to develop software and smart phone technology. When considering more traditional works of creative expression, like works of visual art, relying on the stated intent of the secondary user—or any other subjective evaluation—cannot be the determining factor in a fair use analysis. If it was, any accused infringer could escape liability by simply claiming their work was meant to change the underlying meaning or message of the source material—even when no one else recognizes said transformative purpose.    

Both AWF and the law professors also claim the Second Circuit’s fourth factor market harm analysis conflicts with Google because the Supreme Court found that any market harm to the copyright owner must be weighed against the benefit of the unauthorized use to the public. AWF and its amici attempt to apply the Supreme Court’s public benefit analysis to the Warhol silkscreen prints is completely misguided, as it ignores the drastic differences between types of copyright protected works at issue and distinctions made by the Supreme Court. Discussing the need to balance the public benefit of Google’s copying with the market harm to Oracle, Breyer’s opinion could not be clearer: “We do not say that these questions are always relevant to the application of fair use, not even in the world of computer programs.”

Second Circuit Unequivocally Rejects the Broad Application of Google v. Oracle

In its amended opinion, the Second Circuit leaves no doubt that it is entirely unpersuaded by the arguments of AWF and its amici. In an early footnote addressing AWF’s petition, the opinion states that “[a]part from its reliance on the Google opinion, the petition mostly recycles arguments already made and rejected, and requires little comment.” Before addressing Google, the opinion reiterates the court’s reliance on a transformative standard that considers how the secondary work may “reasonably be perceived.” It also makes clear that any “overly liberal” standard of transformative use risks eliminating protections for derivative works—such as film adaptations of a novel—which often add “new expression, meaning, or message.”

Turning to the effect of Google, the Second Circuit says that “we emphatically reject AWF’s assertion that Google ‘comprehensively refutes the panel’s reasoning.’” The amended opinion calls out AWF for misreading both the Second Circuit’s earlier decision and Google v. Oracle, leading to a misinterpretation that both opinions adopt “hard and fast categorical rules of fair use.” Addressing the limited nature of the Google v. Oracle, the opinion reminds the parties that the Supreme Court “took pains to emphasize that the unusual context of that case” make its conclusions “less applicable to contexts such as ours” that involve artistic works that serve no utilitarian function.

Rather than “effectively outlawing” an entire genre of art, as AWF accuses it of doing, the Second Circuit offers the following explanation:

We merely insist that, just as artists must pay for their paint, canvas, neon tubes, marble, film, or digital cameras, if they choose to incorporate the existing copyrighted expression of other artists in ways that draw their purpose and character from that work (as by using a copyrighted portrait of a person to create another portrait of the same person, recognizably derived from the copyrighted portrait, so that someone seeking a portrait of that person might interchangeably use either one), they must pay for that material as well.

Addressing AWF’s and its amici’s argument that the Second Circuit’s fourth-factor analysis fails after Google for not giving appropriate weight to the public benefit of Warhol’s copying—or the alleged chilling effect a ruling against AWF would have on the creation of art—the court explains that nothing in its opinion stifles creation “that may reasonably be perceived as conveying a new meaning or message, and embodying a new purpose.”

Importantly, the Second Circuit focuses on the harm to potential derivative markets and the effect unauthorized uses would have if they became widespread. The harm to potential markets, which is expressly laid out in factor four of the fair use doctrine, is a critical part of the factor-four analysis that the Supreme Court mistakenly disregarded in Google v. Oracle. The Second Circuit doesn’t miss the opportunity to highlight the cognizable (if only potential) harm caused by Warhol’s misappropriation, ultimately finding “the public benefit of the copying at issue in this case does not outweigh the harm identified by Goldsmith.”

Context is key in all fair use cases, and the Second Circuit concludes that “[l]ike the Supreme Court in Google,” it applied the four-factor test to the specific facts before it. The amended opinion makes clear that because “the principles enunciated in Google are fully consistent with our original opinion,” AWF’s fair use defense fails as a matter of law. 

Conclusion

From the day Breyer’s opinion was handed down, we’ve maintained that the decision is limited to specific forms of declaring code software, and thus should not be applicable to any other types of copyrighted works. Thankfully, the Second Circuit agrees and had no problem recognizing the limits Breyer’s opinion expressly imposes on its fair use analysis. Warhol v. Goldsmith is only the first of what will likely be many tests of the boundaries of Google v. Oracle, but it’s an important initial proclamation that reinforces a sensible approach to transformative fair use that protects the integrity of the derivative use right.