Fair Use Isn’t Optional: Judges Can Help Reclaim It for Creators

1. Introducing Judge Leval and Romanova v. Amilus

If you have spent much time exploring U.S. copyright law, you’d likely have encountered the writings—or at least the influence—of Judge Pierre Leval. Among many other contributions, his “Toward a Fair Use Standard” article built a solid foundation for modern fair use jurisprudence. On the bench, Judge Leval’s opinions are famed for their clarity and balance, always mindful of “copyright’s overall objective of contributing to public knowledge,” as can be seen in his landmark decision in Authors Guild v. Google. 

On May 23rd, another very important opinion was authored by Judge Leval for the Second Circuit Court in Romanova v. Amilus Inc. Although we won’t delve into it further in this blog post, his opinion provides much-needed clarification on “justification” post-Warhol—guidance that we hope many courts will follow in the coming years.

The case facts in Romanova were simple and straightforward: a photographer sued a website for republishing her photo without permission. The district court wrongly found fair use, and the Second Circuit reversed. 

The opinion authored by Judge Leval was characteristically thoughtful, not only in its substantive treatment of fair use, but also in its strong affirmation of the district court’s decision to raise fair use sua sponte (Latin for “of one’s own accord “—sua sponte is used to indicate that a court has taken notice of an issue on its own motion, without prompting or suggestion from either party). This post focuses on that procedural dimension of Romanova and goes one step further, explaining why fair use should never have been treated as an affirmative defense in the first place.

For decades, Campbell v. Acuff-Rose has been celebrated as a decision that fortified fair use. But ironically, the decision also ossified the view that fair use is strictly an affirmative defense—significantly limiting the power of fair use to safeguard free expression in practice. Even where lower courts are doubtful, they nevertheless feel bound by the way the Campbell court construed fair use as an affirmative defense. In a footnote in Suntrust Bank v. Houghton Mifflin, the Eleventh Circuit Court mused: “fair use is commonly referred to an affirmative defense, see Campbell v. Acuff-Rose Music . . . Nevertheless, the fact that the fair use right must be procedurally asserted as an affirmative defense does not detract from its constitutional significance as a guarantor to access and use for First Amendment purposes.”

Now, Judge Leval (whose influential scholarship was cited with approval in Campbell) is helping to restore the power of fair use as a fundamental right—one that cannot be waived, forfeited, or overlooked due to procedural missteps. 

2. Romanova, a course correction to misreadings of Campbell 

Although the case facts and substantive fair use ruling in Romanova are relatively routine, its procedural holding stands out as especially noteworthy. It restores to fair use its legal force and its human dimension. 

The district court in Romanova, acting on its own initiative, dismissed the plaintiff’s copyright infringement claim on the basis of the defendant’s use being a fair use, even though the defendant had not appeared in court and never raised the fair use defense. 

The Second Circuit upheld the decision of the district court to consider fair use sua sponte. While the court did not go so far as to declare that fair use is not an affirmative defense, Judge Leval’s opinion strips away some of the harmful consequences of treating fair use procedurally rigidly. According to Judge Leval, justice should not depend on one party’s ability to afford litigation, and that courts are always empowered to consider obvious, meritorious fair use defenses. Recognizing how imbalances in power play out in costly copyright litigation, Judge Leval reasoned with clairvoyance:

At times, small corporations simply cannot afford the expense of counsel needed to defend a suit. Default does not necessarily preclude the court’s consideration of affirmative defenses available to the defendant, especially when they have obvious merit and their applicability is evident from the face of the complaint. Otherwise, plaintiffs could often easily inflict unjustified harms on small corporate enemies. Intimidation tactics would threaten to strip small creators of their content, and would silence the numerous small platforms that need the protection of the fair use doctrine. (emphasis added)

He concluded:

The district court here believed that its consideration of the fair use defense would serve justice and advance the goals of copyright. [The district] court misunderstood the fair use defense, which in fact had no proper application to these facts. But we cannot fault the district court for considering a defense which it believed (albeit mistakenly) was valid and important. While district courts should indeed be cautious before sua sponte invoking affirmative defenses on behalf of defaulting defendants, they should also be cautious about not considering such defenses.

Up until this opinion, many courts have doggedly followed the mistaken belief that defendants must plead fair use in their answer or a motion to dismiss, or risk waiving it—because fair use has been characterized rigidly as an affirmative defense, a position reinforced by lower courts’ reading of Campbell. 

An affirmative defense first admits that a wrongful act has occurred, then offers a justification or excuse that negates culpability. The wrongfulness of the act is central to the concept of an affirmative defense. For example, a bribe given in a foreign country is wrongful by nature, but if the foreign country’s written law allows for such payments, it can serve as an affirmative defense; a board of directors neglecting fiduciary duty is wrongful by nature, but they can use a later shareholder rectifying vote as an affirmative defense; most famously, perhaps, killing another human is wrongful by its nature, but self defense is an affirmative defense. Because of the innate wrongfulness of the acts in such circumstances, it seems reasonable that the presumed wrongdoer must raise an affirmative defense and support it with evidence. 

However, fair use is not a wrongful act by nature, and treating it as though it were fundamentally misunderstands its role in copyright law. Unlike bribery, breach of fiduciary duty, or homicide, fair use is not ostensibly an unlawful act—fair use is a declaration that someone exercised their first amendment rights despite copyright laws, and that no wrong occurred in the first place. The idea that fair use must be affirmatively raised by a defendant or else be forfeited misunderstands its broader function. 

The Copyright Act and the First Amendment Rights were contemplated by the Framers of the U.S. Constitution around the same time. Fair use, along the idea-expression dichotomy, was specifically introduced to ensure that copyright did not overly encroach on the Freedom of Expression so central to the national identity. Fair use, in essence, is the First Amendment safety valve written into the Copyright Act. Indeed, the Copyright Act itself refers to fair use as a right, and never even remotely suggests that it is an affirmative defense. Section 106 grants copyright rights “[s]ubject to Sections 107 through 122,” and Section 107 clearly states that fair use “is not an infringement of copyright.”

Also different from defending someone who bribes, who breaches fiduciary duty, or who kills, defending someone who exercises fair use serves broader societal goals. Affirmative defenses are used to safeguard a defendant’s individualized rights or privileges. Fair use, on the other hand, aims to promote public interest as much as it protects individual’s free speech rights. Fair use ensures the public can engage with, critique, learn from, and build upon existing works. It preserves space for education, journalism, commentary, and innovation. To treat fair use as merely a tool to exculpate an individual is to fundamentally misunderstand its constitutional and statutory role.

The evidentiary burden associated with affirmative defenses is also ill-suited to be superimposed on fair use. Take the fourth fair use factor for instance, in Warhol, the court observes: “burden of proving that the secondary use does not compete in the relevant market is . . . borne by the party asserting the defense.” Yet plaintiffs are in a much better position to present evidence of market harm than defendants are to show its absence. After all, how does one prove a negative—that no market harm exists—without it sounding fragmentary and incomplete? The burden of disproving market harm amounts to demanding that defendants prove an empirical negative, where an exhaustive search is mandatory. Instead of plaintiffs bearing the burden of proving wrongful conduct, defendants are left to prove their innocence. This imposes a significant burden on lawful expression. 

So, how did it come to be that fair use is so often referred to as an affirmative defense? 

In Fair Use and Market Failure: Sony Revisited, Professor Glynn Lunney explains that fair use was first characterized as an affirmative defense in Campbell due to poor lawyering by Campbell’s counsel. Specifically, Professor Lunney notes that Campbell’s lawyer conceded affirmative defense and the burden of proof issue without challenging it or even referencing the contrary holding in Sony Corp. v. Universal City Studios where the Court placed the burden on the plaintiff to prove unfair use. 

Professor Lunney suggests that because the burden of proof issue in Campbell was uncontested, the Court’s characterization of fair use as an affirmative defense should be considered non-binding dicta rather than a definitive ruling. Judge Leval has also written about how dicta should not be treated as binding law because they lack adversarial scrutiny and can undermine the constitutional role of courts to adjudicate only live cases and controversies. Judge Leval advises his fellow judges: 

If a rule was declared only in dictum, the question remains undecided, and we have a constitutional duty to make our own determination of the answer. Unless we do, we have not done our job.

Professor Lunny considered the dictum in Campbell calling fair use an affirmative defense to have sidestepped Sony‘s balancing approach and inadvertently narrowed the scope of fair use. 

Furthermore, as Professor Lydia Loren pointed out in Fair Use: an Affirmative Defense?, the Campbell court relied on misread legislative history to reach the conclusion that fair use is an affirmative defense. The 1965 Supplementary Report and 1967 House Report explicitly rejected assigning the burden of proof to either party, calling it “unfair and undesirable.” The 1976 Act’s legislative history emphasizes codifying existing judicial doctrine without altering it, including not shifting burdens. The Supreme Court’s reliance on a 1992 House Report (post-enactment) for the “affirmative defense” label is flawed, as it does not reflect the 1976 Congress’s intent.

When legislating fair use into the 1976 Copyright Act, after weighing testimony from different stakeholders, the Copyright Office counseled in its 1965 Supplementary Report: 

We believe it would be undesirable to adopt a special rule placing the burden of proof on one side or the other. When the facts as to what use was made of the work have been presented, the issue as to whether it is a “fair use” is a question of law. Statutory presumptions or burden-of-proof provisions could work a radical change in the meaning and effect of the doctrine of fair use. The intention of section 107 is to give statutory affirmation to the present judicial doctrine, not to change it.

Similarly, the House Report counseled that “any special statutory provision placing the burden of proving fair use on one side or the other would be unfair and undesirable.”

Judge Leval’s opinion in Romanova does not expressly contradict Campbell. Yet it deliberately attacks the procedural rigidity that has developed since Campbell. Romanova reaffirms that fair use is not an exception to infringement: fair use is the built-in safety valve that ensures copyright serves—rather than stifles—freedom of expression. When courts treat fair use as another affirmative defense, to be raised by savvy defendants at just the right procedural moment, they let the tail wag the dog: public interest is sacrificed in the honor of procedural rigidity. 

One would hope, with this opinion, that more courts will follow suit—raising fair use where it matters, even sua sponte in default judgments. Romanova plants a seed we are eager to see blossom. It is an opinion that gives us hope.