The Law Review Forum

Who Is the Bad Art Judge? Aesthetic Nondiscrimination in Andy Warhol Foundation v. Goldsmith

Dave Fagundes

December 28, 2023

Dave Fagundes is the Baker Botts LLP Professor of Law and Research Dean at the University of Houston Law Center.


Most commentary about Andy Warhol Foundation for the Visual Arts, Inc. v. Goldsmith focuses on fair use. This Essay highlights the significance of a different feature of the recent Supreme Court decision: The Warhol Court’s treatment of the doctrine of aesthetic nondiscrimination. Famously articulated in Bleistein v. Donaldson, this doctrine holds that judges’ opinions of artistic merit should not affect their evaluations of copyright matters. But in the century since it was decided, many courts—such as the Second Circuit in its Warhol opinion—expanded this holding to a broad caution against judges analyzing works of authorship at all. This Essay makes three claims about this brief but consequential aspect of Warhol. First, it explains why the Court’s critique of the Second Circuit’s use of Bleistein correctly reads the century-old opinion by Justice Holmes. Second, it shows that the Court is right not only as a matter of reading precedent but also as a matter of necessity, because numerous provisions of the Copyright Act require judicial assessment of the content (though not the quality) of works of authorship. Finally, this Essay expresses hope that the Warhol Court’s reading of Bleistein will correct the harmful exceptionalist notion that judges are intrinsically bad at artistic analysis, thereby creating more space for courts to engage with the works that form the subject matter of copyright law.


In the months leading up to its decision, Andy Warhol Foundation v. Goldsmith (2023) was the most anticipated Supreme Court copyright opinion in years, perhaps decades. Observers wondered whether the Court would reject or embrace the Second Circuit’s narrowing construction of the first factor of the Copyright Act’s fair use defense. When the opinion was finally issued in May 2023, though, the reaction was one of muted puzzlement. On the one hand, the Court affirmed the Second Circuit’s rejection of the Foundation’s fair use defense. On the other, it did so on narrow, highly case-specific grounds, and in the process reaffirmed its leading fair use precedents. The outcome left both owners and users more confused than vindicated.

This short Essay is about an aspect of Warhol that is, by contrast, both unappreciated and highly consequential: its implications for the doctrine of aesthetic nondiscrimination. This well-settled principle holds that judges’ opinions of a work’s merit have no place in the adjudication of copyright matters. But in the century-plus since its establishment in Bleistein v. Donaldson Lithographing Co. (1903), some courts—including the lower court in Warhol—expanded this principle into a broad admonition against judges evaluating not just the quality but the content of works of authorship. The Supreme Court, in a brief but crucial passage, identified and corrected the Second Circuit’s misreading of Bleistein. In so doing, the Court situated judges in copyright cases as analyzers but not critics of art, a role that is not only permissible but necessary for resolution of numerous features of the Copyright Act.

This Essay expounds and defends the Warhol Court’s treatment of the aesthetic nondiscrimination principle in three parts. First, it contrasts the Second Circuit’s invocation of the principle with the Supreme Court’s and explains why the latter is true to Bleistein. Second, it shows that the Court’s narrower reading is necessary as a matter of statutory interpretation because the Copyright Act contains numerous provisions that not only permit but require analysis of the content (though not the merit) of works of authorship. Finally, it shows how the Court’s reading of Bleistein may push back against the pervasive notion that judges are intrinsically bad at dissecting works of authorship, which hamstrings their ability to engage in the analytical work that copyright cases often require.

I. (Over-)Reading Bleistein

When the artist pseudonymously known as Prince died in 2016, numerous publications commemorated his life and work. The cover of Vanity Fair’s retrospective included an image of Prince that Andy Warhol created in 1984. Warhol’s painting, “Orange Prince,” was a silkscreen on canvas that was in turn based on a 1981 photograph by Lynn Goldsmith. But while Warhol had acquired the right to make one painting based on Goldsmith’s photograph of Prince, he also created fourteen additional unlicensed paintings based on that photograph. Goldsmith notified the Andy Warhol Foundation for the Visual Arts (AWF) that the image on the 2016 Vanity Fair cover was unauthorized, and thus an infringement of her rights in the 1981 photograph. AWF responded by seeking a judgment declaring that Warhol’s adaptation of Goldsmith’s photograph was non-infringing fair use.

Unauthorized uses do not infringe an owner’s work if they are amount to fair use. The Copyright Act’s fair use test includes four mandatory factors, but recent Second Circuit precedent has held that the predominant consideration was whether the unlicensed use was transformative of the work. The District Court concluded that Warhol’s use of Goldsmith’s photograph was transformative and ruled in favor of AWF. On appeal, the Second Circuit reversed. The appeals court did not depart from earlier cases’ holding that transformativeness is a dominant fair use consideration, but instead defined transformativeness more narrowly than had previous courts. The court rejected the notion that any unauthorized use with a different character, expression, or aesthetic than the original was necessarily transformative. Instead, the Second Circuit imposed a higher standard, holding that only unauthorized uses that were of a “‘fundamentally different and new’ artistic purpose and character” were transformative of the owner’s work.

The appellate court’s defense of this new framework relied in part on the principle of aesthetic nondiscrimination. The Second Circuit expressed concern that the more capacious transformativeness inquiry improperly led judges to “assume the role of art critic and seek to ascertain the intent behind or meaning of the works at issue.” As authority for this proposition, the court invoked one of the most famous passages in U.S. copyright law, Justice Holmes’s admonition in Bleistein that “[i]t would be a dangerous undertaking for persons trained only in the law to constitute themselves final judges of the worth of pictorial illustrations.”

The Second Circuit read Bleistein’s aesthetic nondiscrimination principle broadly to bar any judicial analysis of works of authorship on the theory that judges are poorly qualified to make assessments of art or literature. Other courts have similarly read Bleistein as requiring judges to “limit[ the] need to engage in artistic analysis” in their evaluation of copyright cases, shying away from evaluating issues like fair use factor two, the idea/expression dichotomy, or originality because they understand aesthetic nondiscrimination to mean that judges should assiduously avoid the analysis of works of authorship.

The Supreme Court affirmed the Second Circuit’s judgment in favor of Goldsmith, but in a brief but crucial passage offered a different, narrower reading of Bleistein. It tactfully declared the lower court’s reading of that case “correct in part.” The correct part was that judges should not seek “to evaluate the artistic significance of a particular work.” The incorrect part was that judges should not seek to determine the message or meaning of a work. On the contrary, the Court stated, “[T]he meaning of a secondary work, as reasonably as can be perceived, should be considered to the extent necessary . . . .

The Court read Bleistein’s aesthetic nondiscrimination principle to bar judges from considering only the artistic merit of a work (i.e., is it of high or low quality, does it exhibit good or poor taste, is it better or worse than similar works) while still permitting them to analyze, objectively and as necessary, the content of the work itself (i.e., does it parody another work, what are its structural features, is it distinct from preexisting similar works). That courts should not consider the quality of a work, the Court’s reading goes, does not foreclose them from engaging in analyses of that work as necessary to resolve copyright matters.

So, whose reading of Bleistein is right? Short answer: the Supreme Court’s, and it’s not close. A careful reading of the language quoted by the Second Circuit makes this abundantly clear. Justice Holmes’s skepticism about judicial capacity extends only to their ability to evaluate the “worth” (that is, the quality rather than the content) of works of authorship.  This passage does not caution courts against analyzing the content of works of authorship, but rather against judging works’ merit or quality.

The context of Justice Holmes’s observation furthers the point. His “dangerous undertaking” phrasing referred to the argument of the dissent and the lower court that the works at issue in that case—lithographs used to advertise a circus—were not copyrightable because they were “low art” that did not promote the progress of science as required by the Constitution. Justice Holmes rejected this claim because the status of art as low or high, good or bad, was an inherently subjective undertaking that was “dangerous” as it would lead to inconsistent results.  But nowhere did Bleistein say or even suggest that rigorous analysis of the meaning or message of works cannot be undertaken as a general matter. On the contrary, as the following Part shows, this kind of objective analysis is necessary for resolving central issues required by the Copyright Act in infringement matters.

II. The Necessity of Interpretation

The previous Part showed that the Supreme Court was right to correct the Second Circuit’s reading of Bleistein as a matter of precedent. This Part shows that the Court’s reading of Bleistein is right also as a matter of necessity because the Copyright Act regularly requires judges to analyze the works at issue in infringement litigation. To establish copyright infringement, plaintiffs must show that they have a valid copyright in a work of authorship and that the defendant has copied constituent elements of that work without permission. Defendants can also raise statutory defenses to this prima facie case, such as fair use or first sale. At each stage, reaching some understanding of the work(s) at issue is inevitable. Three illustrations follow.

A. Copyrightability

Start with originality, a statutory prerequisite for vesting a copyright. In Feist Publications, Inc. v. Rural Telephone Service Co. (1991), the Supreme Court disaggregated the statutory originality requirement into two separate elements: whether the plaintiff’s work contains a “modicum of creativity” and whether it is an independent creation of the author. Both inquiries require analysis of the content of the plaintiff’s work on its own terms and in comparison to others.

Feist exemplifies the necessity of dissecting works of authorship in originality matters. The case tasked the Court with assessing the copyrightability of a telephone directory. To answer this question, the Court had to identify the scheme used by the plaintiff to select and arrange the directory’s data. The selection, it found, was dictated not by the plaintiff’s reflection on who should be in its phone book but by a Kansas regulation requiring a certain scope of coverage. The Court concluded that the selection of data did not originate with the author but with the Kansas Corporation Commission. In terms of the data’s arrangement, the Court found that it was similarly mechanistic, due not to the company’s creative choices but to a preexisting organizational scheme (i.e., alphabetical order) that did not exhibit originality’s requisite spark of creativity. The Court’s well-regarded decision would have been impossible without a careful parsing of the organizational choices in the plaintiff’s directory and whether those choices exhibited creativity and originated with their author, all of which the Court managed to do without opining on the relative aesthetic worth of the works in suit.

B. Impermissible Copying

A second example arises in the context of adjudicating infringement. Where a plaintiff alleges that a defendant’s work shares many features with, but is not identical to, her work, courts must undertake the difficult task of determining whether this partial overlap amounts to copyright infringement. Enter substantial similarity. This doctrine demands that judges scrutinize the plaintiff’s work and the defendant’s use to determine the degree of similarity they share and whether that similarity is due to impermissible copying. Close analysis is inextricably part of a judge’s application of this doctrine. The leading technique, “analytic dissection,” requires the judge to disaggregate each work to identify areas of overlap, ask whether those overlaps are permissible (i.e., whether they are too minor to merit copyright protection or are gleaned from the public domain), and then consider whether the remaining quantum of resemblance is “substantial.”

And while locating the line where similarity becomes substantial is notoriously difficult, the analytical techniques used to identify commonalities between works and assess their legality can be undertaken without making any assessment of merit. As an illustration, consider Christian hip-hop artist Marcus “Flame” Gray’s claim that Katy Perry’s musical work “Dark Horse” infringed the copyright in his work “Joyful Noise.” A unanimous Ninth Circuit ultimately rejected Gray’s claim. While it agreed that there was an eight-measure section that was identical between the two works, the court pointed out that this progression was an “ostinato,” a common musical phrase that had been used for centuries preceding Gray’s composition of “Joyful Noise.”

This is classic analytical dissection: The court scrutinized the two works and found a notable commonality, but found that the commonality was a public domain note sequence, and so rightly concluded that it could not be the basis for an infringement allegation rooted in substantial similarity. Such a task would plainly be impossible if, as the Second Circuit suggested in Warhol, judges were barred from assessing the content of works of authorship.

C. Affirmative Defenses

Finally, consider the fair use defense, which by statute demands that courts analyze the defendant’s work (“the purpose and character of the use”); the plaintiff’s work (“the nature of the copyrighted work”); and how they relate to each other quantitatively (“the amount and substantiality of the portion [of the plaintiff’s work] used in relation to the . . . whole”). Moreover, deciding fair use often requires courts to make qualitative comparisons of the plaintiff’s work and defendant’s use, which in turn entails analysis of the respective works in suit.

Applying the fair use defense may require judges to determine what the unauthorized use communicates about the owner’s work. In rebuffing the Second Circuit’s suggestion that judges should not “ascertain the . . . meaning of the works at issue,” for example, the Warhol Court pointed out that having some sense of what both the work and the use mean is prerequisite to the factor one fair use analysis (“purpose and character of the use”). Such meaning may be necessary to determine whether the unauthorized use “comments on, criticizes, or provides otherwise unavailable information about the original.”  It further suggested that such analyses could be undertaken “objective[ly],” so that they did not run afoul of Bleistein’s distinction between considerations of a work’s quality (categorically impermissible) from analysis of its content (permissible if objective).

Another famed fair use decision, Campbell v. Acuff-Rose (1994), exemplifies both the inevitability of analyzing works of authorship in fair use cases as well as the possibility of doing so without treading on impermissible assessment of aesthetic merit. In that case, the Court considered whether 2 Live Crew’s unlicensed hip-hop version of “Pretty Woman” was fair use of the Roy Orbison original because the former was a parody of the latter. In so doing, it led off with Bleistein to stress that “[w]hether . . . parody is in good taste or bad does not and should not matter to fair use.”

Because, however, parody is “a composition in which the characteristic style and themes of a particular author or genre are satirized by being applied to inappropriate or unlikely subjects,” the Court proceeded to analyze the content of both the Orbison original and the 2 Live Crew version, and then considered what (if anything) the defendant’s use said about the plaintiff’s. The Court’s holding that 2 Live Crew’s use of Orbison’s work was fair relied in large part on the factual premise the former’s bawdy reimagining of the latter’s lyrics “can be taken as a comment on the naivete of the original of an earlier day,” reflecting the “joinder of reference and ridicule” that is characteristic of parody. Here, too, the Court engaged in a particular kind of analysis (assessment of the meaning of works in suit) without treading on the impermissible question of their relative quality.

The examples listed in this Part illustrate that numerous features of copyright law require some degree of judicial dissection of works of authorship, rendering the Court’s reading of Bleistein correct not only as a matter of precedent but also as a matter of necessity. The idea that judges must analyze authors’ works may trigger concerns that this task is one that courts will undertake poorly because it lies outside their expertise. But as the final Part shows, this belief is rooted in a false exceptionalism that itself derives from an overly broad understanding of the aesthetic nondiscrimination principle. A final virtue of the Warhol Court’s take on Bleistein is that it may help to dispel this misconception as well.

III. Good (Enough) Art Judge

This Essay concludes with a third and final cheer for Warhol’s treatment of Bleistein: It may help to push back against the pervasive belief that judges are uniquely bad at interpreting works of authorship. This belief can be traced in part to Bleistein. Justice Holmes’s assertion that it was a “dangerous undertaking” to allow “persons trained only in the law” to evaluate the merits of art has been expanded to mean that it is similarly “dangerous” to allow judges to analyze works of authorship at all. This take on Bleistein regularly leads courts to shy away from assessing the content of works on the theory that, as the Ninth Circuit recently proclaimed, “[J]udges make terrible art critics.”

What is strange about this is that we don’t find judges proclaiming themselves “terrible economists” or “terrible historians.” On the contrary, judges often embrace other non-legal specialties with gusto, hence the emergence of law-and-economics and originalism as leading modalities of adjudication and interpretation. This is not to say that judges are necessarily great at analyzing works of authorship in copyright cases. Quite often, they’re not. Rebecca Tushnet has shown that, at least when called on to interpret visual art, judges tend to oscillate between viewing works’ meanings as either obvious (rendering interpretation unnecessary) or opaque (rendering interpretation impossible). But for that matter, judges aren’t necessarily great economists or historians either. Setting aside questions of validity, both originalism and law-and-economics have received their share of scholarly heat for being selectively and clumsily deployed by judges.

The idea that judges are distinctively bad at analyzing creative work represents an exceptionalism that is not only false, but harmful. Reading Bleistein’s “dangerous undertaking” as extending to judges’ ability to analyze works of authorship unduly clouds their adjudication of copyright matters. It leads judges to abdicate rather than engage with works of authorship, or to use proxies to deny that they are engaging in analysis of the work at all. This may not be the sole or even primary reason that judges tend to oscillate between assessing the meaning of visual art as either obvious or opaque, but it certainly doesn’t help.

Here is where Warhol may supply a partial corrective. In rebuking the Second Circuit’s take on Bleistein, the Court not only invited but required that courts analyze works when deciding copyright cases. “The meaning of a … work,” the Court stressed, should be considered to the extent necessary.” Moreover, the Court’s treatment of the works in suit in Warhol (wisely undertaken with the help of experts) provided an object lesson in how judges can make plausible (though by no means indisputable) arguments about creative work without falling into a well of conjecture and subjectivity. And in so doing, the Court pushed back against the exceptionalism of the bad-art-judge view, situating the analysis of works of authorship as just another instance of generalist judges having to engage with subject matter outside their native expertise.


It is too early to say what implications the Supreme Court’s opinion in Andy Warhol Foundation v. Goldsmith will have for fair use. This brief Essay has sought instead to highlight the case’s implications for aesthetic nondiscrimination. The Court correctly held that the doctrine disfavors judicial evaluation of the merits, but not the content, of works of authorship. This is a good thing too: Resolution of numerous core copyright issues requires some analysis of the works in suit. Finally, emphasizing the accurate scope of Bleistein is important because it may provide a corrective against the notion that judges are uniquely incompetent at analyzing works of authorship. The false exceptionalism that judges are inherently terrible at this task leads judges to avoid careful analysis of works of authorship when it is most necessary.

Of course, to say that judicial engagement with works of authorship is appropriate in copyright cases when necessary is not to say that it is easy to do well. On the contrary, analyzing the fruits of creative expression, and in particular understanding its meaning(s), can be nuanced and even elusive. There is no shortage of commentary illustrating judges fumbling when trying to analyze works of authorship. But the fact that an inquiry is difficult is no reason to avoid addressing it. Otherwise, we would have to bid farewell to the Equal Protection Clause and criminal law’s reasonable doubt standard, among many others. As with other areas outside their native expertise, judges can use familiar tools (e.g., relying on experts to supplement knowledge, issuing detailed written opinions to force analysis and give some shared understanding of their reasoning) to generate credible, nuanced analyses of art. In so doing, judges can responsibly turn toward rather than away from the artifacts regulated by copyright law.

The title of this Essay was inspired by Robert Keller, Who Is the Bad Art Friend?, N.Y. Times Magazine (Oct. 5, 2021).