George Mason University Antonin Scalia Law School

Ninth Circuit Confirms: Fair Use Is an Affirmative Defense to Copyright Infringement

the word "copyright" written on a typewriterThe Ninth Circuit’s recent decision clarifying transformative fair use in Dr. Seuss v. ComicMix gives much to admire (see my deep dive into the opinion here). The court held that a mash-up of plaintiff Seuss’ Oh, the Places You’ll Go! (Go!) with a Star Trek theme—entitled Oh, the Places You’ll Boldly Go (Boldly)—by defendant ComicMix was not fair use since it merely repackaged the original work without new purpose, meaning, message, expression, or character—in other words, it utterly failed to be transformative. That holding should be welcome news for those who are concerned that the rule of the derivative right is being swallowed by the exception of transformativeness.

The Ninth Circuit also addressed another issue of central importance in the fair use doctrine, namely, whether fair use is an affirmative defense such that the burden is on the defendant to demonstrate the absence of likely market harm. That the burden issue even came up is itself remarkable, given the Supreme Court pronouncements and Ninth Circuit precedents on the matter. Nevertheless, this case presented a good vehicle for the Ninth Circuit to confirm that the burden of proving fair use remains with its proponent on each of the factors, including market harm. In this post, I’ll discuss the positions of the parties and the holdings of the courts on this important procedural aspect of fair use.

District Court Holds that Burden Is on Plaintiff to Show Likely Market Harm

One of the issues before the district court was whether Seuss had to show likely market harm under the fourth fair use factor. In its memorandum supporting its summary judgment motion, ComicMix acknowledged that it had the burden of proving fair use: “The proponent of fair use has the burden to show that it applies, because it is an affirmative defense.” However, it argued that, since Boldly is transformative, there is no presumption of market harm, and the “burdens on the fourth factor shift to the plaintiff” to “demonstrate a likelihood of potential harm.” Thus, while acknowledging that it ultimately held the burden of persuasion on market harm, ComicMix argued that Seuss had to produce evidence in order to win on this factor.

The Supreme Court discussed the presumptions and burdens of proof for market harm in Sony and Campbell. In Sony, the Court stated that “every commercial use of copyrighted material is presumptively . . . unfair,” such that, “[i]f the intended use is for commercial gain,” the “likelihood may be presumed” for market harm. And when the use is “for a noncommercial purpose, the likelihood must be demonstrated” with “a showing by a preponderance of the evidence that some meaningful likelihood of future harm exists.” ComicMix latched onto this statement that likely market harm must be demonstrated, despite its use being commercial, to argue that Seuss was required to make a showing on market harm. But that position is inconsistent with what the Court held later in Campbell.

The Supreme Court in Campbell walked things back significantly, explaining that the presumption from Sony only applies in the context of “verbatim copying of the original in its entirety for commercial purposes,” not in “a case involving something beyond mere duplication for commercial purposes.” When “the second use is transformative,” the Court continued, “market substitution is at least less certain, and market harm may not be so readily inferred.” The Court in Campbell thus distinguished verbatim, commercial copies (market harm presumed) and transformative, yet commercial, uses (no such presumption). Importantly, even though there was no presumption of market harm for the transformative use at issue, the Court placed the burden on the defendant to demonstrate an absence of market harm in order to succeed on that factor because “fair use is an affirmative defense.”

This understanding of who has the burden of proof on market harm must be understood in the context of the procedural posture. Fair use is only relevant once the plaintiff has made a prima facie showing of copyright infringement. If both Seuss and ComicMix were to fail in producing evidence on market harm, the factor might be neutral. But that neutrality would only hurt ComicMix—and not Seuss—since ComicMix is trying to overcome Seuss’ prima facie showing of infringement. The Supreme Court in Campbell drove this point home, holding that, in the absence of evidence, “it is impossible to deal with the fourth factor except by recognizing that a silent record on an important factor bearing on fair use disentitled the proponent of the defense . . . to summary judgment.” That “evidentiary hole,” the Court concluded, would need to “be plugged on remand” for the defendant to prevail.

In its opposition memorandum, Seuss started with the premise that ComicMix had the burden of proof on market harm: “As with all affirmative defenses,” ComicMix “bear[s] the burden of proof.” Suess argued that ComicMix could not carry its burden because it did “not offer a shred of admissible evidence about the relevant markets for Go!, derivatives of the [Seuss] Works, and Boldly.” Thus, both parties agreed that fair use is an affirmative defense, with ComicMix bearing the ultimate burden of persuasion on whether its use was fair. The only difference was whether, and to what extent, either party could or should have to demonstrate evidence of likely market harm or its absence.

Despite arguing in its memorandum that the burden of proof was on ComicMix, counsel for Seuss told a different story at the hearing on the motion. The court asked, “What burden, if any, does [Seuss] carry in establishing a likelihood of market harm in this case?” Seuss’ counsel responded that, assuming the use is transformative, there would be no presumption and that Seuss would then have to show market harm. The district court ultimately concluded that Boldly is “highly transformative,” and with Seuss’ concession during oral argument in hand, the court held that Seuss had the burden of introducing “[e]vidence of substantial harm to it . . . by a preponderance of the evidence.” After finding that Seuss “failed to introduce evidence tending to demonstrate that the challenged work will substantially harm the market for its Copyrighted Works,” the court held that the fourth fair use factor was neutral since “the harm to Plaintiff’s market remains speculative.”

Ninth Circuit Holds that Burden Is on Defendant to Show Absence of Market Harm

On appeal to the Ninth Circuit, the parties changed their tune somewhat. In its opening brief, Seuss argued that the district court erred by placing the burden on it to show likely market harm: “fair use is an affirmative defense, and its proponent must show absence of market harm even if the challenged use is transformative.” Moreover, Seuss claimed that since Boldly is nontransformative and commercial, there should be a presumption of market harm. But even if there is no such presumption, Seuss argued that ComicMix had the burden of “offering convincing proof that the plaintiff’s markets will not be harmed by the challenged work.” And, in a footnote, Seuss argued that any concession by its counsel during oral argument before the district court “would not control this Court’s resolution of a purely legal issue: who bears the burden of proof on the fourth factor.”

In its answering brief, ComicMix argued that Seuss’ “failure to present evidence of any negative market effect weighs in favor of fair use.” It contended that there is no presumption of market harm, even though Boldly is commercial, because that presumption does not apply when a work is transformative. As to which party held the burden of proof, ComicMix argued that fair use is not in fact an affirmative defense—a remarkable claim given the holding of Campbell and ComicMix’s contrary position in the district court. To support its argument, ComicMix cited the Ninth Circuit’s decision in Lenz as well as a pair of law review articles to conclude that, “while the proponent of a true affirmative defense generally bears the burden of proof as to all elements, for a standard defense like fair use, it suffices to show that the plaintiff failed to overcome it.”

The citation to Lenz is a puzzling one. The issue there was whether a copyright owner must consider fair use before sending a takedown notice under the Digital Millennium Copyright Act (DMCA). The copyright owner argued that fair use is not “authorized by the law” under the DMCA since it is an affirmative defense excusing conduct that would otherwise infringe. The Ninth Circuit disagreed, finding that this view conflates “an affirmative defense that is labeled as such due to the procedural posture of the case, and an affirmative defense that excuses impermissible conduct.” The court ultimately decided that fair use is “authorized by the law” as a statutory matter, even if it’s an affirmative defense as a procedural one. However, the court said nothing to the contrary about who bears the burden of proof on likely market harm in a context outside of the DMCA—like in overcoming a prima facie case of infringement in litigation over whether a use is fair.

The cites to the law review articles are equally perplexing as they both explicitly acknowledge that the burden of proof rests with the party claiming fair use. The first, Proving Fair Use: Burden of Proof as Burden of Speech by Ned Snow, acknowledges that, “[a]s an affirmative defense, fair use places the burden of proof on its proponent.” The thrust of the article is that, while “the defendant bears the burden of proof” for fair use, it would be better to shift that burden to the copyright owner because of free speech concerns. The second article, Fair Use: An Affirmative Defense? by Lydia Loren, likewise concedes that Campbell “[c]learly . . . placed the burden of producing evidence to support fair use on the defendant.” The article then argues that, as a normative matter, we should put that burden on the copyright owner. Neither of these articles provides any support for the point ComicMix sought to establish, i.e., that the burden of proof is on Seuss.

In assessing the fourth fair use factor, the unanimous Ninth Circuit panel sided with Seuss: “Having found that Boldly was transformative—a conclusion with which we disagree—the district court . . . erred in shifting the burden to Seuss with respect to market harm. That shifting, which is contrary to Campbell and our precedent, led to a skewed analysis of the fourth factor.” The court declined to hold that ComicMix’s nontransformative, commercial use raised a presumption of market harm, though it did recognize that market harm could be inferred from the circumstances. The court made quick work of ComicMix’s argument that fair use is not an affirmative defense and that the burden is on Seuss to show likely market harm: “Not much about the fair use doctrine lends itself to absolute statements, but the Supreme Court and our circuit have unequivocally placed the burden of proof on the proponent of the affirmative defense of fair use.” ComicMix’s argument went nowhere because “Campbell squarely forecloses” it.

To ComicMix’s argument that Lenz compels placing the burden of proving market harm on Seuss, the Ninth Circuit pointed out that Lenz “involved fair use in a different corner of the copyright law,” namely, the safe harbors under the DMCA. The question there was of statutory interpretation under the DMCA, and the analysis was explicitly limited to “that context.” Lenz changed nothing about the procedural burdens in other contexts: “In no way did we deviate from our characterization of fair use as an affirmative defense under § 107. To the contrary, in addition to clarifying that, unlike copyright misuse and laches, fair use is not an excuse to copyright infringement, we reiterated that ‘the burden of proving fair use is always on the putative infringer.’”

Turning to the merits, the Ninth Circuit held that it was ComicMix, “as the proponent of the affirmative defense of fair use,” that “must bring forward favorable evidence about relevant markets.” Given that ComicMix chose to argue instead that it didn’t have the burden of proof, there was only “scant evidence” to consider. ComicMix’s principal evidence was an expert report, but the court found that the “entire report is premised on Boldly being transformative, which it is not, and on the expert’s misunderstanding about fair use and U.S. copyright law.” But even accepting the report’s “methodology and conclusions,” the court found that it still “fails to account for key fourth-factor considerations,” including the fact that Boldly was “intentionally targeted and aimed to capitalize on the same graduation market as Go!” and that “Boldly would curtail Go!’s potential market for derivative works.” Having found that ComicMix failed to carry its burden, the court held that ComicMix’s fair use defense failed as a matter of law.


In Dr. Seuss v. ComicMix, the Ninth Circuit addressed important aspects of the fair use doctrine—and some novel arguments that flew in the face of settled precedent. Not only did the court rein in various overly broad notions of transformative use urged by ComicMix and its amici that would have further eroded a copyright owner’s exclusive right to make derivative works, but it also confirmed the fundamental holding of Campbell that fair use is an affirmative defense, with the burden of proof to show an absence of market harm falling squarely on the party claiming fair use. ComicMix attempted to sidestep its duty by claiming that there was none; it wanted the benefit of fair use without its burden. Thankfully, the Ninth Circuit emphatically rejected this approach as one that would “‘create incentives to pirate intellectual property’ and disincentivize the creation of illustrated books,” contrary to copyright’s constitutional goals.