Are makers of video games required to take licenses from celebrities depicted in their games, or does the First Amendment shield them from the need to obtain permission and pay royalties?
The NCAA student-athlete legal free-for-all has generated two remarkably aligned federal appeals opinions that have brought a rare measure of uniformity to the generally disjointed case law on balancing rights of publicity protections for individuals against First amendment protections for creative expression.
The NCAA case has been very much in the headlines recently, but began five years ago as an antitrust case in which former college athletes sought to get a piece of the $20 billion a year in licensing revenue (from both television and official merchandise) received by the NCAA and its licensing agent, Collegiate Licensing Company. At that time, we noted it might be years before the case was resolved through final appeal. In fact, the first case has just finished trial.
Since 2009, the case has morphed into a class action involving potentially thousands of athletes. The issue made further headlines recently when sports superlawyer Jeffrey Kessler filed a new suit on behalf of a class of current college athletes, not to get a slice of the licensing revenues, but to blow up the entire NCAA “amateur student-athlete” system on antitrust grounds, and allow athletes to bargain for compensation while they are still students.
What escaped attention on most sports pages, however, were opinions late last year by the Third Circuit Court of Appeals in Philadelphia in the case of former Rutgers University quarterback Ryan Hart (Hart v. Electronic Arts, Inc.) and in the Ninth Circuit Court of Appeals in San Francisco in the case of former Nebraska University and Arizona State University quarterback Sam Keller (Keller v. Electronic Arts, Inc.), both against video game publisher Electronic Arts—not on antitrust grounds, but for using their personal and athletic performance information without a license, and thereby allegedly misappropriating their rights of publicity.
The former athletes both won their appeals, resulting in a proposed $60 million settlement with them and other former athletes —$40 million from EA and CLC, $20 million added by NCAA on the eve of the antitrust trial— and EA’s decision to shut down its franchise NCAA Football videogame series for at least 2015.
Both courts agreed with each other on multiple points regarding how to properly balance an individual’s rights of publicity versus an author’s First Amendment free-speech rights, a field in which up until now there has rarely been agreement or clarity.
The one Supreme Court case on this issue, Zacchini v. Scripps-Howard Broadcasting Co., 433 U.S. 562 (1977), held that a news broadcast of Hugo Zacchini’s entire “human cannonball” act without his consent when he appeared at a county fair was a misappropriation of his rights of publicity under Ohio law. The Court held that there must be a balance between the rights of individuals to control the economic exploitation of their names and images, protected by rights of publicity laws (currently recognized in about 40 states, depending on who is doing the counting), and the right of free expression, protected by the First Amendment to the federal Constitution. However, while ruling in favor of Zacchini that the television station had exceeded the bounds of First Amendment protection for the news media and was liable for misappropriation, the Court did not offer guidelines on how to strike that balance on a general basis. Lower courts have been formulating their own tests ever since.
Three of the leading tests are:
1) The Predominant Use Test holds that if the predominant purpose of a work is to exploit the commercial value of an individual’s identity, instead of to make an expressive comment about that individual, then the right of publicity claim will prevail over the First Amendment defense.
2) The Rogers Test was originally formulated in a case brought by movie actress Ginger Rogers, who sued the producers and distributors of Ginger and Fred, an Italian movie about nightclub dancers who were nicknamed after Ginger Rogers and Fred Astaire, on the grounds that the title of the movie infringed both her trademark and her rights of publicity. The Second Circuit Court of Appeals disagreed, holding that the First Amendment trumped Ms. Rogers’ rights of publicity because: (a) the use of her name in the title bore some relationship to the contents of the movie; and (b) the use of her name in the title was not a disguised advertisement for the sale of goods or services.
3) The Transformative Use Test was first formulated by the California Supreme Court in Comedy III Productions, Inc. v. Gary Saderup, Inc., 21 P.3d 797 (2001), in which the rights owner of the Three Stooges movie comedy team sued an artist for selling t-shirts and lithographs with photorealistic sketches of the Stooges. The court held that the artist had not added enough of his own original content to the portraits to be “transformative,” that is, to make it’s the artist’s own expression rather than a mere celebrity likeness, therefore the Stooges’ rights of publicity defeated the artist’s First Amendment defense.
The Third Circuit case was brought as a class action by Hart under the New Jersey statutory rights of publicity provision, while the Ninth Circuit case was brought by Keller and eight other former NCAA Division I football and basketball players under both California’s statutory and common law rights of publicity. The main defendant was video game publisher Electronic Arts, which was the producer of the NCAA Football video game series. Although EA had licensed the rights to use the NCAA and individual university trademarks for its video games, it had not licensed the rights of publicity from the college athletes. (According to a footnote in the Ninth Circuit case, Keller claimed that EA had specifically agreed in its NCAA license not to use athlete likenesses.)
Both courts substantially agreed on the facts of the case—that EA’s NCAA Football series seeks to replicate each university’s entire team roster as accurately as possible. Every player on a team has a corresponding avatar with the same jersey number, class year, height, weight, build, skin tone, hair color, home state, equipment modifications, and playing style. Player names are not written on the jerseys of the avatars, but can be added via third-party applications.
In both cases, EA defended against claims of misappropriation of the players’ rights of publicity on the grounds that the video games were forms of expression protected by the First Amendment, under any of several legal theories, trumped the athletes’ rights of publicity, and therefore negated any need for a license from the athletes.
As a threshold issue, both courts agreed with EA that the video games are fully protected as expressive speech under the First Amendment.
The Third Circuit quickly rejected the Predominant Use Test advocated by EA on the grounds that it forced judges to become art critics who “analyze select elements of a work to determine how much they contribute to the entire work’s expressiveness.”
Both courts rejected employing the Rogers Test also advocated by EA because: 1) in prior case law, it had mainly been used in trademark infringement cases, but rarely been used in rights of publicity cases, and even then had primarily been applied when a celebrity’s name was used in a title of an expressive work, not in the body of the work itself; 2) application of the Rogers Test would almost always result in shielding the video game developer, because a video game in which a celebrity appears will almost always satisfy the Rogers Test criteria of being related in some degree to the celebrity; and 3) the test was derived from trademark law, where the primary concern is protecting the public from confusion about the source and quality of goods, not an appropriate focus for rights of publicity cases, where the focus should be on protecting a valuable intellectual property right of an individual— the commercial exploitation of his identity.
Both courts ultimately concluded that the Transformative Use Test was the appropriate balancing test. The Third Circuit felt it was more appropriate to rights of publicity cases because it was derived from the fair use defense in copyright law, which focuses on protection of individual intellectual property rights. In the context of the EA video games, both courts relied heavily on the case No Doubt vs. Activision Publishing, Inc., 192 Cal.App.4th 1018 (2011), in which the California Court of Appeals held that use of significant elements of identities of members of the pop music group No Doubt in the Band Hero video game was not a transformative use. Even though the game incorporated many creative elements in the performance environment for the members, such as performing in outer space or performing songs of rival bands, there was not a sufficient transformation of the avatars of the members, ruled the court. Likewise, concluded both the Third and Ninth Circuit panels, in the NCAA video games, EA’s use of player avatars with detailed literal biographical and appearance information was not transformed by the game’s ability to create different game environments. The athletes’ avatars were placed in the same context— playing football— as what they had become famous for, even though the game made it possible to change certain variables of that context. Further, the fact that the game allowed some degree of customization of the avatar did not change the equation—to rule otherwise would provide an easy out for game makers to capitalize on use of celebrity images, as long as they also built in an option to tweak the images.
In both cases, judges wrote dissenting opinions which argued that use of the Transformative Use Test as interpreted in No Doubt was not protective enough of First Amendment rights, and might result in imposing liability on authors of books, movies, and other creative works that incorporate historical figures, even if situated in a totally fictional environment.
Additionally, the Ninth Circuit addressed two non-First Amendment, California-specific defenses to rights of publicity claims. California law creates a defense against common law rights of publicity claims for the “publication of manners in the public interest,” and a defense against statutory rights of publicity claims “in connection with any news, public affairs, or sports broadcast or account, or any political account.” The court held that it was not necessary to weigh whether the video games were “in the public interest,” or were connected to news or sports broadcasts, because they failed the threshold requirement for both doctrines that they be in the nature of publishing or reporting. Instead, they were using the athletes’ personal information in the context of simulated games that had never occurred in the real world, so they were not “publications” or “reports.”
The court then addressed one of the toughest issues in a footnote at end of its opinion. It attempted to distinguish several cases from outside California in which courts ruled that the First Amendment trumped rights of publicity claims for fantasy sports leagues that publish player names and publicly available performance statistics, so that the leagues were not required to obtain licenses from the athletes. The NCAA case is different argued the Ninth Circuit, because EA’s creation of non-publicly available virtual avatars using player identifying information did derive from the athletes’ “identities,” and were not merely public domain statistical information. While this distinction may “feel” right, the “mere statistics” versus “statistics plus specially created avatars” distinction is a very thin dividing line, and will likely be litigated in the future.
Takeaway: once limited to California cases, the Transformative Use Test is gathering endorsements from an increasing number of appellate courts, when it is necessary to balance the rights of individuals to protect their rights of publicity from exploitation by others versus the First Amendment right protections of creative expression, and especially in the context of use of celebrity avatars in video games. But rights of publicity are grounded in diverse state laws, so that even when the applicable balancing test is settled, it is notoriously difficult to predict in advance whether a court will find a particular rendering sufficiently transformative, so caution remains in order for use of celebrity likenesses without a license.