This past week, the Supreme Court in Andy Warhol Foundation for the Visual Arts v. Lynn Goldsmith grappled with the difficult relationship between “derivative” and “transformative” uses of a copyrighted work in a high-stakes dispute between two industry heavyweights, each looking to land a knockout blow. The extended judicial dialogue did not reveal a clear winner—nor should it have. In such cases, where both parties contribute to a masterpiece, some division of the spoils is the voice of principled moderation.

In 1984, the noted photographer Lynn Goldsmith granted an “artist’s reference license” to Vanity Fair for the preparation of one rendition of the famed entertainer Prince to appear twice in its November 1984 issue. That license did not specify who that artist would be, and it only allowed the chosen artist to get inspiration from the photograph, not build off it. Vanity Fair picked the famed artist Andy Warhol, a.k.a. “Andy the Appropriator,” who is now best known for his imaginative recreations of photographs of Marilyn Monroe and Campbell’s soup cans. True to form, Warhol used the Goldsmith photograph as direct source and made not one but fifteen silkscreens, each with its own distinctive feature. Two pictures tell the tale:

(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes; …

(4) the effect of the use upon the potential market for or value of the copyrighted work.

Goldsmith vigorously attacked this fair-use defense by claiming that Warhol’s works are “derivative work[s]” from her initial photograph, i.e., “work based upon one or more preexisting works,” including an “art reproduction.” Based on this view, Goldsmith sought to collect hefty royalties from the foundation and the right to all Warhol’s silkscreens under Section 106(2) of the Copyright Act: “to prepare derivative works based on the copyrighted work[.]” The District Court stressed the many transformative differences between the two works—for instance, the cropping of the picture and the treatment of the hair—to show how the two works were different in tone and world view, and concluded that the Goldsmith photograph was entitled to no creative credit for the Warhol works. The Court of Appeals for the Second Circuit reversed that decision, noting that the Warhol silkscreens were clearly derivative works that did not differ more from the Goldsmith original than a movie adapted from a book. But it did not say anything about the appropriate remedy for the breach it identified.

So, it was either feast or famine. In their briefs and oral argument, all the participants proceeded on the assumption that the Warhol silkscreens were either a derivative or a transformative use. Both sides are half-right, for the silkscreens are, even to this unpracticed eye, plainly both derivative and transformative: Warhol first used and then embellished Goldsmith’s work. At this point, the first fair-use factor seems largely irrelevant because the Warhol silkscreens were clearly intended for a commercial use, and not deserving of the broader fair-use interpretation given to charitable organizations. The transformative-use angle was strictly judge-made, but it is today so entrenched that it requires a close comparison between the two works, which to the practiced eye reveals many artistic differences in mood and technique.

The fourth factor dealing with potential competition effects is also of little use. Matters of competition are always questions of degree whenever the two works are not identical in all relevant respects. The extent of the overlap in the target markets of the two works is not clear. Indeed, even the direction of the causality should be up for grabs because the popularity of the Warhol silkscreens has boosted the value of Goldsmith portrait, and perhaps her broader work portfolio.

The relevant fair-use tests do not therefore provide much guidance. Instead, as I have argued elsewhere, a fresh start is needed to avoid the all-or-nothing solution in this case, which, as Justice Elena Kagan noted, could upend the universal custom that requires Hollywood studios to pay for the books they adapt. But Kagan offered no way out of the impasse because neither she (nor anyone else) developed a contract solution to the problem. Namely, the justices and lower court judges failed to explore a possible method for setting a price or royalty term (as with the 1984 license) to divide the gains between the two parties, roughly in proportion to their respective contributions. In this case, Warhol clearly ignored the Vanity Fair license in treating it not just as a simple reference point but as the physical base for his own transformative silkscreens.

Sadly, that cryptic license did not specify any remedy for his actions. If it had, there would be no reason to ignore the division of spoils among parties with considerable experience working in this market. The best guess is that a full contractual treatment of the matter covering Goldsmith, Vanity Fair, and Warhol would address damages and the ability to make and sell future reproductions. In trying to construct remedies under the Copyright Act, none of these contractual guardrails are in place, so it becomes necessary to return to the fundamental objective of copyright law, which is to give each creator the exclusive possession of an artistic work to spur its production.

Implementing this joint solution will simultaneously create the desired incentives for both parties. To award hefty damages to Goldsmith and to give her the copyrights to the Warhol works at issue would lead to a huge reduction in Warhol’s incentive to exercise his creative powers, were he alive today. Hence, any all-or-nothing proposal involving a joint work necessarily slights the contribution of at least one party. This is not to say, however, that the Goldsmith parties stand in perfect parity to each other, because Warhol well knew (or at least had to have known) that the Vanity Fair artist’s reference license did not authorize his multiple productions. Thus, before starting work, he should have procured a broader license or perhaps found a different photograph of Prince from another artist or in the public domain. Warhol did not do that because he prized Goldsmith’s excellence. In bypassing the contractual rules, Warhol yielded an edge to Goldsmith in the damage or royalty calculations.

Nonetheless, it would be most unwise to follow the Second Circuit case of Cariou v. Prince (2013) brought by the photographer Patrick Cariou against the artist Richard Prince over the artist’s use of Cariou’s photographs as the basis for his collages and paintings. The District Court found infringement and the remedy was unrelenting: Prince had “to deliver to Cariou all infringing works that had not yet been sold, for him to destroy, sell, or otherwise dispose of.” This would be classic overkill if applied in Goldsmith as it would allow Goldsmith to control the sale of the Warhols for her exclusive benefit.

This extreme statutory remedy for copyright infringement through creation of derivative works may make sense for simple knockoffs, but not for sophisticated and labor-intensive creative works, like those by Warhol. Even though injunctive relief is often the best way to prevent the circumvention of individual property rights, it hardly makes sense in the context of artistic masterpieces. By comparison, in patent infringement cases, the Supreme Court frequently and with far less justification has conditioned the use of injunctive relief under a multipart balancing test associated with eBay v. MercExchange (2006). But that makes no sense in cases steeped in First Amendment overtones, wherein the court would fallaciously strip from an artist his own creations—especially when it would be possible to design an effective remedy allowing both the Warhol Foundation and Goldsmith to exercise complete control over their respective artistic works.

It is a stock legal problem to separate the contributions of two parties, each of whom supplies some input toward the creation of an artistic work—in this case, the Warhol silkscreens. The best way to do this is to set a royalty that approximates her contribution to the final product. One way to set a fair royalty rate for the sale of the silkscreens is to use the relative sales prices of the Goldsmith photograph and Warhol’s paintings, as is done in, for example, a musical licensing deal. Goldsmith’s royalty should then be bumped up to penalize Warhol for his illegal actions. A basic 10 percent royalty might be a sensible first approximation, which I would increase by 50 percent for a 15 percent royalty. That does quite well by Goldsmith, and it does relatively little to inhibit the Warhol Foundation’s far-flung activities.

In the long run, the basic copyright law should be redrafted to take into account the multiple instances where single works are both derivative and transformative in order to give a firmer foundation to a much-overlooked problem. But until that is done, the justices who crafted the transformative-use doctrine should adjust the remedy to fit their new creation.

© 2022 by the Board of Trustees of Leland Stanford Junior University.