Whether one focuses on the word’s connotation of silliness or excitement, or maybe even anger, or analogizes to the raucous and rhymingly-named team from Savannah that makes up its own baseball rules, US copyright law is currently going a little “bananas.” From ongoing debates about the human element (or requirement) of authorship to debates over what constitutes a transformative work to the gauntlet laid down in footnote 2 of Justice Kagan’s dissent in the Supreme Court’s May 18, 2023 decision in Andy Warhol Foundation for Visual Arts, Inc., v. Goldsmith, some strong opinions, and even raw nerves, mark the discussion.
In fact, lots is happening. There are moves aimed at criminalizing some creativity usually reserved solely for federal civil copyright consideration. There is also ongoing fallout at the intersection of trademark and expressive works as the Supreme Court has remanded for reconsideration of an earlier dismissal of a toy maker’s claim that a character’s name in Toy Story 3 infringed its Lots of Hugs’ trademark. And, there are still more challenges from authors and creators to the til-now-uncompensated use of their works to “train” artificial intelligence engines, as seen in the recent open letter from the Authors’ Guild to various tech companies.
Though articles about many of those, or related, issues have appeared here in the past, they present issues that will not be solved, or even too well-defined, in this August blog post. But I will describe what happened when I read a few news reports describing the outcome in Morford v. Cattelan and then followed Justice Kagan’s admonition to “go take a look at the decision” before deciding what I thought about what has been said about that case. Warhol, 143 S.Ct. at 1293, fn.2 (Kagan, J., dissenting)
Morford is an interesting case.
But the entire Morford analysis revolved around the 11th Circuit’s “abstraction-filtration-comparison” test, described as “initially developed to review claims of copyright infringement relating to novels and plays but also applied in contexts such as software programming.” Id. at 9-10. That test, described in greater detail below, “requires the Court to ‘reverse engineer’ the work at issue,” where one “’first breaks down the allegedly infringed program into its constituent structural parts…[N]ext, the court sifts out all non-protectible material,’” and “[f]inally, the ‘last step is to compare any remaining kernels of creative expression with the allegedly infringing [work] to determine if there is in fact a substantial similarity.’” Id. at 9. That test is, for judges, an instruction on how to look at art and understand what you see.
But any approach that, like the 11th Circuit’s, that parallels structurally basic art criticism methods seems at least somewhat inconsistent with the Supreme Court’s direction in Warhol to “not attempt to evaluate the artistic significance of a particular work.” Warhol, 143 S. Ct. at 1283. Despite the seemingly neat breakdown of the 11th Circuit’s three-step test into “(1) abstraction, (2) filtration, and (3) comparison,” Compulife Software Inc. v. Newman, 959 F. 3d 1288, 1303 (11th Cir. 2020), each step in the process forces a court into a role not unlike the art critic. The court is being asked to determine what the idea or meaning of the work is (even if the meaning may in fact be multiple, mysterious, or muffled, intentionally or not), separating some meaningfulness element from other elements of the work, and then comparing such elements to one another. While courts may think that is some disciplined test from the approach of the art critic, that approach in fact tracks the art critic’s approach quite closely.
So let’s look at each step.
But, first, let’s look at the works:
As the Morford court noted (at 3), “[b]oth works are three-dimensional wall sculptures depicting bananas that are duct-taped to a vertical surface… In each sculpture, a single piece of plain gray duct tape crosses a yellow banana at an angle and affixes the banana to a vertical surface. …In both sculptures, the banana is oriented with the stalk on the top, left-hand side of the work from the viewer’s perspective.” But there are differences, from differences in the color of the background to the fruit items included to the orientation of the banana. Id. at 3-4.
To address these differences, the Morford court first engaged in the abstraction step analysis. “’In order to ascertain substantial similarity under this approach, a court first breaks down the allegedly infringed program into its constituent structural parts—that’s abstraction.’” Id. at 9 (quoting Compulife Software Inc. v. Newman, 959 F.3d 1288, 1303 (11th Cir. 2020)). For a court, that process means “’dissect[ing] the allegedly copied [work’s] structure and isolate each level of abstraction contained within it.’”Id. at 10 (quoting Computer Assocs. Int’l, Inc. v. Altai, Inc., 982 F.2d 693, 707 (2d Cir. 1992)). While the Morford court does this matter-of-factly by stating (at 10) that:
Morford’s Banana & Orange features two green rectangular panels, each seemingly attached to a vertical wall by masking tape. The panels are stacked on top of each other, with a gap between each. Roughly centered on each green panel is a fruit: an orange on the top panel and a banana on the lower panel. The orange is surrounded by masking tape, and a piece of silver duct tape crosses the orange horizontally. The banana is at a slight angle, with the banana stalk on the left side pointing up. The banana appears to be fixed to the panel with a piece of silver duct tape running vertically at a slight angle, left to right[,]
abstraction of a more complex visual work, or a play or a novel, would likely take such analysis much deeper into criticism of the significance of each level of abstraction. For instance, for some literary critics such as Knapp and Michaels (in Against Theory, 8 Critical Inquiry 723, 731 (Summer 1982)), the process of abstraction of a work into different verbal components or blocks does not advance the analytical ball because they see the “possibility of language prior to and independent of intention and thus conceiving intention as something that must be added to language to make it work.” Indeed, Robert Scholes has noted (in Textual Power: Literary Theory & The Teaching Of English (1985), at 36) that, with certain writers like Hemingway, “[a]bstraction” is “left to the reader.” If that be the case, and the readers are judges, then the first prong of this test would hardly be expected to yield consistent results, as the conclusion is not a product of the creator or the standard applied, but of the viewer, which changes courtroom to courtroom and case to case, as discussed further below.
This is not an argument against employing abstraction steps in literary criticism or in other fields like the social sciences, as it has a proud history in both (as described by Mar and Oatly, in The Function of Fiction is the Abstraction and Simulation of Social Experience, 3 Perspectives On Psychological Science 173 (2008)). But it is a warning that any “decision of the level of abstraction is a subjective process,” as Ying He has noted (in A Cognitive Approach to the Subjectivity of Aesthetics at 158) in the computer programming and security context. But the same is true in aesthetics and the interpretation of art—and it can be more biological than cerebral, especially with visual arts:
Although brain systems differ in their functions, they are all engaged in abstraction and concept formation because of their involvement in the acquisition of knowledge, and because a similar neural process governs different ideals produced by the brain. We are not conscious of the neurological processes that underlie these abstractions because they are automatic, but we are aware of their results. Consider the physiological properties of visual neurons. They are specialized for specific orientation, detection of motion in specific directions, and specific colors. This makes it evident that abstraction is not a characteristic of higher areas of the brain, or limited to them, they are characteristic of early visual areas.
[A Cognitive Approach to the Subjectivity of Aesthetics, at 21]
This, again, makes “abstraction” a strange step as part of a prescribed analytical construct because the likelihood of uniform application is challenged, perhaps on an even biological basis.
Next, one would, like the Morford court, look at the filtration step, a step trying to decide each “particular element of a work [that] should be filtered out during this stage” because is part of the underlying idea of the piece or derived from the public domain rather than an example of the work’s protectible expressive elements. Morford, at 11. Though the “’idea expression dichotomy’ is the normal standard in copyright law,” there are times when “the expression of the idea necessarily ‘merges’ with the idea itself, making the expression therefore unprotectible.” Id. Again, this is a tough standard for a judge to apply while also trying “not attempt to evaluate the artistic significance of a particular work,” as noted in Warhol (at 1283). I say that for two reasons. First, as we have seen in the literary field, “textual criticism” of literature is itself “a filtering process,” and judges are supposed to adjudicate the copyright issues without becoming a critic. Second, in essence, the judge is faced with a direction from the highest court not to arbitrate the significance of the work while being simultaneously asked to judge the artistic significance and relevance of each element of the work as part of the filtration process.
That seems contradictory. But perhaps that is not surprising, as Warhol itself sets up that tension itself:
Elsewhere, however, the Court of Appeals stated that “the district judge should not assume the role of art critic and seek to ascertain the intent behind or meaning of the works at issue.” Id., at 41. That statement is correct in part. A court should not attempt to evaluate the artistic significance of a particular work. See Bleistein v. Donaldson Lithographing Co., 188 U.S. 239, 251, 23 S.Ct. 298, 47 L.Ed. 460 (1903) (Holmes, J.) (“It would be a dangerous undertaking for persons trained only to the law to constitute themselves final judges of the worth of [a work], outside of 1284*1284 the narrowest and most obvious limits”).[fn omitted] Nor does the subjective intent of the user (or the subjective interpretation of a court) determine the purpose of the use. But the meaning of a secondary work, as reasonably can be perceived, should be considered to the extent necessary to determine whether the purpose of the use is distinct from the original, for instance, because the use comments on, criticizes, or provides otherwise unavailable information about the original, see, e.g., Authors Guild [v. Google], 804 F.3d [202,] 215-216 [(2d Cir. 2015)].
This paragraph asks the judge, or the art critic, to carry out their tasks, and consider the meaning of a work. But to do so without considering the work’s significance and without considering the subjective intent of the person creating the derivative work or the subjective impact of that work upon the viewer. That seems a tough draw, and in many ways, the filtration step requires one to do just that element by element.
“[O]nce the Court has broken down the original work into its relevant parts and “filtered” out the unprotectible portions, the Court must compare the allegedly infringing work with the protectible parts that remain.” Morford, at 13. There are dos and donts for such comparisons:
This is not a simple comparison of any and all similarities. Herzog, 193 F.3d at 1257 (“Lists of similarities between the two works are ‘inherently subjective and unreliable,’ particularly where the list contains random similarities, and many such similarities could be found in very dissimilar works.”). Rather, “the court’s substantial similarity inquiry focuses on whether the defendant copied any aspect of this protected expression, as well as an assessment of the copied portion’s relative importance with respect to the plaintiff’s overall [work].” Altai, 982 F.2d at 710. This inquiry must focus on whether any similarities are significant to the copyrighted work—not whether those similarities are significant to the allegedly infringing work.
If one must do more than list all similarities, and one must judge which are random and which are not, and one must assess “relative importance” to the original work, then one is becoming an art critic and assessing the significance of a work, or more precisely of its constituent elements. Indeed, the Morford court doing the comparison step (after simply listing similarities despite the earlier note) only one common feature that “is, on its own, insignificant and insufficient to support a finding of legal copying.” Morford, at 14. Indeed, the Morford court moves away from looking at concurrent features to differences, noting that “these differences demonstrate that Comedian and Banana and Orange are different expressions of the underlying idea,” without noting at that point exactly what that underlying idea is. Morford, at 16. (Of course, we see elsewhere in the opinion mention of “the idea of duct-taping a banana to a wall.” Id. at 15.). With that, the Morford Court granted summary judgment to the defendant, finding no infringement.
But, on July 6, 2023, Morford appealed, so the case now goes to the 11th Circuit for further consideration (Morford’s opening brief is due September 20, 2023). That gives us reason to comment further on the abstraction-filtration-comparison test and related issues.
The abstraction-filtration-comparison test, like many of the related tests, is “designed to sift the protected from the unprotected and conclude whether it not the works are substantially similar.” Said, Reforming Copyright Interpretation, 28 Harv. J.L. & Tech. 469, 483 (2015)(discussing that test, as well as two-step copying and improper appropriation, extrinsic dissection/intrinsic judgment, abstractions test, and total concept and feel test). None of these tests are themselves the problem, despite what seem like criticisms above.
The problem really arises from those who apply such tests while pretending it is simply a legal test, or that they are not engaged—and engaged appropriately—in the interpretation of art just like many art critics. As Zahr Said has said (in Reforming Copyright Interpretation, 28 Harv. J. L. & Tech. 469, 473 (2015)), “[c]opyright adjudication requires judges to adopt interpretative methodologies, whether or not they address them explicitly.” (italics in original) The main issue I have with the majority in Warhol is the same one reflected in Justice Kagan’s dissent and predicted in some sense by the avoidance discussion in Farley, Judging Art, 79 Tulane Law Review 805, 836-39 (2005). There is a lack of express, or at least expressly clear, understanding that interpretative choices are being made and interpretative methodologies are being applied, and courts should not be operating in a vacuum or on some other level, or pretend that they are. Both Farley and Said seem to agree.
The applications of such support should not stay shrouded but should stand in the open because any “legal determination of the ‘what is art?’ problem is required in a range of diverse areas of law. Yet, it is precisely these determinations that judges are told not to make.” Farley, at 836. But judges should address such questions and should be expressed about how they do because “[i]nterpretive choices can offer some explanation for the greater divergence in outcomes and reasoning seen in infringement analysis more generally. Recognizing its importance can improve cogency of copyright doctrine throughout litigation,” as Said also said (at 473). Judges should not run from that role, but embrace it, recognizing that the “difficult questions …that courts encounter here have been addressed in philosophy, art history, and art criticism.” Farley, at 839. Courts should “acknowledge that these questions have already been theorized and that there are bodies of scholarship that are relevant and could be helpful.” Id. In that sense, one must credit the Morford court for expressly noting that “presented with a fulsome factual record and cross-motions for summary judgment, ‘what art is’ looms closer in mind.” Morford, at 6. Still, even that court’s momentary candid quickly receded behind the notion that “the Court still need not attempt to answer that age-old (and frankly unanswerable) question.”
Not only should courts attempt it, they must, in essence, do it in many cases. This is at the heart of Justice’s Kagan dissent in Warhol. While Justice Kagan surely was unhappy with a majority that “does not realize how much Warhol added” from an artistic perspective, she is much more chagrined “that the majority does not care” and was “adopting that posture of indifference” toward the work’s artistic significance. Wahol, at 1292. She noted that, before Warhol, courts were supposed to ask whether there was “something new, with a further purpose or different character,” and there anything of such nature added “to a significant degree.” Id. No more—she decries that majority for suggesting that Warhol’s efforts had been mere cropping and flattening when all the experts offering testimony in the case agreed that there had been a transformation of the photographer’s work. Id., compare 1292 with 1296 and 1301.
Justice Kagan ends her Warhol dissent with the words “It will make our world poorer.” I agreed with her then, and, aptly, borrow and transform her words here to “it will make our world poorer,” and our analysis less complete, if our courts continue to address the intersection of art and law only indirectly and obliquely. Judges are sorting through artistic issues as well as legal ones and should acknowledge that. Once we do, then we may better address the issues of what is fair and transformative use, and what is mere theft or infringement. And, these issues certainly remain because Warhol, “[i]nstead of providing clarity,” …deepens [the] confusion” on the “question of what is recycling and what is theft[,]… one of the oldest and knottiest in the aesthete’s arsenal” and our courts’ dockets.
And just to get back to where this blog post started, let me end with a fellow NJ IP lawyer’s pithy yet thematically pertinent Tweeted summary of Warhol v. Goldsmith:
and the admission that “I think I’m going bananas” with all these copyright issues.