19-2420-cv

The Andy Warhol Foundation for The Visual Arts, Inc. v. Lynn Goldsmith, et al.

UNITED STATES COURT OF APPEALS

FOR THE SECOND CIRCUIT

August Term, 2020

Argued: September 15, 2020 Decided: March 26, 2021

Docket No. 19-2420-cv

THE ANDY WARHOL FOUNDATION FOR THE VISUAL ARTS, INC.,

Plaintiff-Counter-
Defendant-Appellee,
— v. —

LYNN GOLDSMITH, LYNN GOLDSMITH, LTD.,

Defendants-Counter-
Plaintiffs-Appellants.

B e f o r e:

JACOBS, LYNCH, and SULLIVAN, Circuit Judges.

Defendants-Appellants Lynn Goldsmith and Lynn Goldsmith, Ltd., appeal

from a judgment of the United States District Court for Southern District of New
York (Koeltl, J.) granting summary judgment to Plaintiff-Appellee The Andy
Warhol Foundation for the Visual Arts, Inc. on its complaint for a declaratory
judgment of fair use and dismissing Defendants-Appellants’ counterclaim for
copyright infringement. We conclude that the district court erred in its
assessment and application of the fair-use factors and that the works in question
do not qualify as fair use as a matter of law. We likewise conclude that the Prince
Series works are substantially similar to the Goldsmith Photograph as a matter of
law. We therefore REVERSE the judgment of the district court and REMAND
the case for further proceedings consistent with this opinion.
JUDGE SULLIVAN concurs in the Court’s opinion, and files a concurring
opinion in which JUDGE JACOBS joins.
JUDGE JACOBS concurs in the Court’s opinion, and files a concurring
opinion.

THOMAS G. HENTOFF (Lisa S. Blatt, Katherine Moran Meeks, on

the brief), Williams & Connolly LLP, Washington, D.C.,
for Defendants-Appellants.

LUKE NIKAS (Maaren A. Shah, Kathryn Bonacorsi, on the brief),

Quinn Emanuel Urquhart & Sullivan, LLP, New York,
NY, for Plaintiff-Appellee.

Christopher T. Bavitz, Harvard Law School Cyberlaw Clinic,

Cambridge, MA, for Amici Curiae Law Professors.

Jason Schultz, Christopher Morten, New York University

Technology Law and Policy Clinic, New York, NY, for
Amici Curiae Latipa (née Michelle Dizon) and Viêt Lê.

Ira J. Levy, Goodwin Procter LLP, New York, NY; Jaime A.

Santos, Goodwin Procter LLP, Washington, D.C., for
Amicus Curiae The Robert Rauschenberg Foundation.

Gregory J. Dubinsky, Evan H. Stein, Holwell Shuster &

Goldberg LLP, New York, NY, for Amicus Curiae
Professor Terry S. Kogan.

2
 Thomas B. Maddrey, Maddrey PLLC, Dallas, TX; Russell J.
Frackman, UCLA School of Law Copyright Amicus
Brief Clinic, Los Angeles, CA, for Amici Curiae The
American Society of Media Photographers, Inc., National
Press Photographers Association, Professional Photographers
of America, Graphics Artist Guild, and North American
Nature Photography Association.

Benjamin S. Akley, Pryor Cashman LLP, New York, NY, for

Amicus Curiae Recording Industry Association of America.

GERARD E. LYNCH, Circuit Judge:

This case concerns a series of silkscreen prints and pencil illustrations

created by the visual artist Andy Warhol based on a 1981 photograph of the

musical artist Prince that was taken by Defendant-Appellant Lynn Goldsmith in

her studio, and in which she holds copyright. In 1984, Goldsmith’s agency,

Defendant-Appellant Lynn Goldsmith, Ltd. (“LGL”), then known as Lynn

Goldsmith, Inc., licensed the photograph to Vanity Fair magazine for use as an

artist reference. Unbeknownst to Goldsmith, that artist was Warhol. Also

unbeknownst to Goldsmith (and remaining unknown to her until 2016), Warhol

did not stop with the image that Vanity Fair had commissioned him to create, but

created an additional fifteen works, which together became known as the Prince

Series.

3
 Goldsmith first became aware of the Prince Series after Prince’s death in

2016. Soon thereafter, she notified Plaintiff-Appellee The Andy Warhol

Foundation for the Visual Arts, Inc. (“AWF”), successor to Warhol’s copyright in

the Prince Series, of the perceived violation of her copyright in the photo. In 2017,

AWF sued Goldsmith and LGL for a declaratory judgment that the Prince Series

works were non-infringing or, in the alternative, that they made fair use of

Goldsmith’s photograph. Goldsmith and LGL countersued for infringement. The

United States District Court for the Southern District of New York (John G.

Koeltl, J.) granted summary judgment to AWF on its assertion of fair use and

dismissed Goldsmith and LGL’s counterclaim with prejudice.

Goldsmith and LGL contend that the district court erred in its assessment

and application of the four fair-use factors. In particular, they argue that the

district court’s conclusion that the Prince Series works are transformative was

grounded in a subjective evaluation of the underlying artistic message of the

works rather than an objective assessment of their purpose and character. We

agree. We further agree that the district court’s error in analyzing the first factor

was compounded in its analysis of the remaining three factors. We conclude

upon our own assessment of the record that all four factors favor Goldsmith and

4
that the Prince Series works are not fair use as a matter of law. We further

conclude that the Prince Series works are substantially similar to the Goldsmith

Photograph as a matter of law.

BACKGROUND

The relevant facts, which we draw primarily from the parties’ submissions

below in support of their respective cross-motions for summary judgment, are

undisputed.

Goldsmith is a professional photographer primarily focusing on celebrity

photography, including portrait and concert photography of rock-and-roll

musicians. Goldsmith has been active since the 1960s, and her work has been

featured widely, including on over 100 record album covers. Goldsmith also

founded LGL, the first photo agency focused on celebrity portraiture. LGL

represents the work of over two hundred photographers worldwide, including

Goldsmith herself.

Andy Warhol, né Andrew Warhola, was an artist recognized for his

significant contributions to contemporary art in a variety of media. Warhol is

particularly known for his silkscreen portraits of contemporary celebrities. Much

of his work is broadly understood as “comment[ing] on consumer culture and

5
explor[ing] the relationship between celebrity culture and advertising.” Cariou v.

Prince, 714 F.3d 694, 706 (2d Cir. 2013). AWF is a New York not-for-profit

corporation established in 1987 after Warhol’s death. AWF holds title to and

copyright in much of Warhol’s work, which it licenses to generate revenue to

further its mission of advancing the visual arts, “particularly work that is

experimental, under-recognized, or challenging in nature.” J. App’x at 305.

On December 3, 1981, while on assignment from Newsweek magazine,

Goldsmith took a series of portrait photographs of (then) up-and-coming

musician Prince Rogers Nelson (known through most of his career simply as

“Prince”) in her studio. Goldsmith testified that, prior to Prince’s arrival at her

studio, she arranged the lighting in a way to showcase his “chiseled bone

structure.” Id. at 706. Goldsmith also applied additional makeup to Prince,

including eyeshadow and lip gloss, which she testified was intended both to

build a rapport with Prince and to accentuate his sensuality. Goldsmith further

testified that she was trying to capture Prince’s “willing[ness] to bust through

what must be [his] immense fears to make the work that [he] wanted to [make].”

Id. at 1557. Goldsmith took black-and-white and color photographs using a

6
Nikon 35-mm camera and a mixture of 85- and 105-mm lenses, which she chose

to best capture the shape of Prince’s face.

Prince, who according to Goldsmith appeared nervous and uncomfortable,

retired to the green room shortly after the session began and ultimately left

without allowing Goldsmith to take any additional photographs. During the

truncated session, Goldsmith took 23 photographs, 12 in black and white and 11

in color. Goldsmith retained copyright in each of the photographs that she took.

Most relevant to this litigation is the following photograph, hereinafter referred

to as the “Goldsmith Photograph”:

In 1984, Goldsmith, through LGL, licensed the Goldsmith Photograph to

Vanity Fair magazine for use as an artist reference. Esin Goknar, who was photo

7
editor at Vanity Fair in 1984, testified that the term “artist reference” meant that

an artist “would create a work of art based on [the] image reference.” Id. at 783.

The license permitted Vanity Fair to publish an illustration based on the

Goldsmith Photograph in its November 1984 issue, once as a full page and once

as a quarter page. The license further required that the illustration be

accompanied by an attribution to Goldsmith. Goldsmith was unaware of the

license at the time and played no role in selecting the Goldsmith Photograph for

submission to Vanity Fair.

Vanity Fair, in turn, commissioned Warhol to create an image of Prince for

its November 1984 issue. Warhol’s illustration, together with an attribution to

Goldsmith, was published accompanying an article about Prince by Tristan Vox

and appeared as follows:

In addition to the credit that ran alongside the image, a separate attribution to

Goldsmith was included elsewhere in the issue, crediting her with the “source

8
photograph” for the Warhol illustration. Vanity Fair did not advise Goldsmith

that Warhol was the artist for whom her work would serve as a reference, and

she did not see the article when it was initially published.

Unbeknownst to Goldsmith and LGL, Warhol created 15 additional works

based on the Goldsmith Photograph, known collectively, and together with the

Vanity Fair image, as the “Prince Series.”1 The Prince Series comprises fourteen

silkscreen prints (twelve on canvas, two on paper) and two pencil illustrations,

and includes the following images:

1
Though it acknowledged that the depiction of Prince in the Prince Series is
similar to that in the Goldsmith Photograph, AWF did not concede below that the
Goldsmith Photograph was the source image for the Prince Series, arguing
instead that “somehow, Warhol created” it. Dist. Ct. Dkt. 55 at 18. In its brief
before this Court, however, AWF describes the Goldsmith Photograph as the
“source image” for the Prince Series. Appellee’s Br. at 6-7.

9
Although the specific means that Warhol used to create the images is unknown

(and, perhaps, at this point, unknowable), Neil Printz, the editor of the Andy

Warhol Catalogue Raisonné, testified that it was Warhol’s usual practice to

reproduce a photograph as a high-contrast two-tone image on acetate that, after

any alterations Warhol chose to make, would be used to create a silkscreen. For

the canvas prints, Warhol’s general practice was to paint the background and

local colors prior to the silkscreen transfer of the image. Paper prints, meanwhile,

were generally created entirely by the silkscreen process without any painted

embellishments. Finally, Warhol’s typical practice for pencil sketches was to

project an image onto paper and create a contoured pencil drawing around the

projected image.

At some point after Warhol’s death, AWF acquired title to and copyright in

the Prince Series. Between 1993 and 2004, AWF sold or otherwise transferred

custody of 12 of the original Prince Series works to third parties, and, in 1998,

transferred custody of the other four works to The Andy Warhol Museum. AWF

retains copyright in the Prince Series images and, through The Artist Rights

Society (a third-party organization that serves as AWF’s agent), continues to

license the images for editorial, commercial, and museum usage.

10
 On April 22, 2016, the day after Prince died, Condé Nast, Vanity Fair’s

parent company, contacted AWF. Its initial intent in doing so was to determine

whether AWF still had the 1984 image, which Condé Nast hoped to use in

connection with a planned magazine commemorating Prince’s life. After learning

that AWF had additional images from the Prince Series, Condé Nast ultimately

obtained a commercial license, to be exclusive for three months, for a different

Prince Series image for the cover of the planned tribute magazine. Condé Nast

published the tribute magazine in May 2016 with a Prince Series image on the

cover. Goldsmith was not given any credit or attribution for the image, which

was instead attributed solely to AWF.

It was at this point that Goldsmith first became aware of the Prince Series.

In late July 2016, Goldsmith contacted AWF to advise it of the perceived

infringement of her copyright. That November, Goldsmith registered the

Goldsmith Photograph with the U.S. Copyright Office as an unpublished work.

On April 7, 2017, AWF sued Goldsmith and LGL for a declaratory judgment of

non-infringement or, in the alternative, fair use. Goldsmith countersued for

copyright infringement under 17 U.S.C. §§ 106, 501.

11
 On July 1, 2019, the district court granted summary judgment for AWF on

its fair-use claim. See Andy Warhol Found. for the Visual Arts, Inc. v. Goldsmith, 382

F. Supp. 3d 312, 316 (S.D.N.Y. 2019). Upon evaluating the four statutory fair-use

factors set forth in 17 U.S.C. § 107, the court concluded that: (1) the Prince Series

was “transformative” because, while the Goldsmith Photograph portrays Prince

as “not a comfortable person” and a “vulnerable human being,” the Prince Series

portrays Prince as an “iconic, larger-than-life figure,” id. at 326; (2) although the

Goldsmith Photograph is both creative and unpublished, which would

traditionally weigh in Goldsmith’s favor, this was “of limited importance because

the Prince Series works are transformative works,” id. at 327; (3) in creating the

Prince Series, Warhol “removed nearly all [of] the [Goldsmith] [P]hotograph’s

protectible elements,” id. at 330; and (4) the Prince Series works “are not market

substitutes that have harmed – or have the potential to harm – Goldsmith,” id. at

331. This appeal followed.

12
 DISCUSSION

I. Standard of Review

“We review a grant of summary judgment de novo,” applying the

standards set forth in Federal Rule of Civil Procedure 56(c). Cariou, 714 F.3d at

704. While fair use presents a mixed question of law and fact, it may be resolved

on summary judgment where, as here, the material facts are not in dispute. See,

e.g., Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 560 (1985).

II. Copyright, Derivative Works, and Fair Use

The Constitution empowers Congress to enact copyright laws “[t]o

promote the Progress of Science and useful Arts.” U.S. Const. art. I, § 8, cl. 8.

Congress has exercised this delegated authority continuously since the earliest

days of the nation, beginning with the Copyright Act of 1790 and, more recently,

through the Copyright Act of 1976. Under the 1976 Act, copyright protection

extends both to the original creative work itself and to derivative works, which it

defines as, in relevant part, “a work based upon one or more preexisting works,

such as a[n] . . . art reproduction, abridgement, condensation, or any other form

in which a work may be recast, transformed, or adapted.” 17 U.S.C. § 101.

13
 The doctrine of fair use has developed along with the law of copyright.

“[A]s Justice Story explained, ‘in truth, in literature, in science and in art, there

are, and can be, few, if any, things, which in an abstract sense, are strictly new

and original throughout. Every book in literature, science and art, borrows, and

must necessarily borrow, and use much which was well known and used

before.’” Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 575 (1994), quoting

Emerson v. Davies, 8 F. Cas. 615, 619 (No. 4,436) (C.C.D. Mass. 1845) (alterations

adopted). The fair use doctrine seeks to strike a balance between an artist’s

intellectual property rights to the fruits of her own creative labor, including the

right to license and develop (or refrain from licensing or developing) derivative

works based on that fruit, and “the ability of [other] authors, artists, and the rest

of us to express them- or ourselves by reference to the works of others.” Blanch v.

Koons, 467 F.3d 244, 250 (2d Cir. 2006).

Though it developed as a creature of common law, the fair-use defense was

formally codified with the passage of the 1976 Act. The statute provides a non-

exclusive list of four factors that courts are to consider when evaluating whether

the use of a copyrighted work is “fair.” These factors are:

14
 (1) the purpose and character of the use, including
whether such use is of a commercial nature or is for
nonprofit educational purposes;
(2) the nature of the copyrighted work;
(3) the amount and substantiality of the portion used in
relation to the copyrighted work as a whole; and
(4) the effect of the use upon the potential market for or
value of the copyrighted work.

17 U.S.C. § 107.

As the Supreme Court has held, fair use presents a holistic context-

sensitive inquiry “not to be simplified with bright-line rules[.] . . . All [four

statutory factors] are to be explored, and the results weighed together, in light of

the purposes of copyright.” Campbell, 510 U.S. at 577-78; see also, e.g., Cariou, 714

F.3d at 705 (“[T]he fair use determination is an open-ended and context-sensitive

inquiry.”). We consider each factor in turn.

A. The Purpose and Character of The Use

This factor requires courts to consider the extent to which the secondary

work is “transformative,” as well as whether it is commercial. We address these

considerations separately below.

15
 1. Transformative Works and Derivative Works

Following the Supreme Court’s decision in Campbell, our assessment of this

first factor has focused chiefly on the degree to which the use is “transformative,”

i.e., “whether the new work merely supersedes the objects of the original creation,

or instead adds something new, with a further purpose or different character,

altering the first with new expression, meaning, or message.” 510 U.S. at 579

(internal quotations marks and citations omitted) (alterations adopted). We

evaluate whether a work is transformative by examining how it may “reasonably

be perceived.” Cariou, 714 F.3d at 707, quoting Campbell, 510 U.S. at 582; see also,

e.g., Leibovitz v. Paramount Pictures Corp., 137 F.3d 109, 113-15 (2d Cir. 1998).

Paradigmatic examples of transformative uses are those Congress itself

enumerated in the preamble to § 107: “criticism, comment, news reporting,

teaching . . . , scholarship, or research.” And, as the Supreme Court recognized in

Campbell, parody, which “needs to mimic an original to make its point,” 510 U.S.

at 580-81, is routinely held transformative. See, e.g., Brownmark Films, LLC v.

Comedy Partners, 682 F.3d 687, 693 (7th Cir. 2012). These examples are easily

understood: the book review excerpting a passage of a novel in order to comment

upon it serves a manifestly different purpose from the novel itself. See Authors

16
Guild v. Google, Inc., 804 F.3d 202, 215-16 (2d Cir. 2015) (“[C]opying from an

original for the purpose of criticism or commentary on the original . . . tends most

clearly to satisfy Campbell’s notion of the ‘transformative’ purpose involved in the

analysis of Factor One.”).

Although the most straightforward cases of fair use thus involve a

secondary work that comments on the original in some fashion, in Cariou v.

Prince, we rejected the proposition that a secondary work must comment on the

original in order to qualify as fair use. See 714 F.3d at 706. In that case, we

considered works of appropriation artist Richard Prince that incorporated,

among other materials, various black-and-white photographs of Rastafarians

taken by Patrick Cariou. See id. at 699. After concluding that the district court had

imposed a requirement unsupported by the Copyright Act, we conducted our

own examination of Prince’s works and concluded that twenty-five of the thirty

at issue were transformative of Cariou’s photographs as a matter of law. See id. at

706. In reaching this conclusion, we observed that Prince had incorporated

Cariou’s “serene and deliberately composed portraits and landscape

photographs” into his own “crude and jarring works . . . [that] incorporate[d]

color, feature[d] distorted human and other forms and settings, and measure[d]

17
between ten and nearly a hundred times the size of the photographs.” Id. Thus,

we concluded that these works “used [Cariou’s photographs] as raw material,

transformed in the creation of new information, new aesthetics, new insights and

understanding,” and were transformative within the meaning of this first factor.

Id., quoting Castle Rock Ent. v. Carol Publ’g Grp., 150 F.2d 132, 142 (2d Cir. 1998).

In adjudging the Prince Series transformative, the district court relied

chiefly on our decision in Cariou, which we have previously described as the

“high-water mark of our court’s recognition of transformative works.” TCA

Television Corp. v. McCollum, 839 F.3d 168, 181 (2d Cir. 2016). And, as we have

previously observed, that decision has not been immune from criticism. See id.

(collecting critical authorities). While we remain bound by Cariou, and have no

occasion or desire to question its correctness on its own facts, our review of the

decision below persuades us that some clarification is in order.

As discussed supra, both this Court and the Supreme Court have

emphasized that fair use is a context-sensitive inquiry that does not lend itself to

simple bright-line rules. E.g., Campbell, 510 U.S. at 577-78; Cariou, 714 F.3d at 705.

Notwithstanding, the district court appears to have read Cariou as having

announced such a rule, to wit, that any secondary work is necessarily

18
transformative as a matter of law “[i]f looking at the works side-by-side, the

secondary work has a different character, a new expression, and employs new

aesthetics with [distinct] creative and communicative results.” Warhol, 382 F.

Supp. 3d at 325-26 (internal quotation marks omitted) (alterations adopted).

Although a literal construction of certain passages of Cariou may support that

proposition, such a reading stretches the decision too far.

Of course, the alteration of an original work “with ‘new expression,

meaning, or message,’” Cariou, 714 F.3d at 706, quoting Campbell, 510 U.S. at 579,

whether by the use of “new aesthetics,” id., quoting Blanch, 467 F.3d at 253, by

placing the work “in a different context,” Perfect 10, Inc. v. Amazon.com, Inc., 508

F.3d 1146, 1165 (9th Cir. 2007), or by any other means is the sine qua non of

transformativeness. It does not follow, however, that any secondary work that

adds a new aesthetic or new expression to its source material is necessarily

transformative.

Consider the five works at issue in Cariou that we did not conclude were

transformative as a matter of law. Though varying in degree both amongst

themselves and as compared to the works that we did adjudge transformative,

each undoubtedly imbued Cariou’s work with a “new aesthetic” as that phrase

19
might be colloquially understood. Prince’s Canal Zone (2007) is a collage of thirty-

six of Cariou’s photographs, most of which Prince altered by, for example,

painting over the faces and bodies of Cariou’s subjects, in some instances altering

them significantly. See Cariou, 714 F.3d at 711. In Graduation, Prince added blue

“lozenges” over the eyes and mouth of Cariou’s subject and pasted an image of

hands playing a blue guitar over his hands. Id. Both of these works certainly

imbued the originals from which they derive with a “new aesthetic;”

notwithstanding, we could not “confidently . . . make a determination about

their transformative nature as a matter of law.” Id.

Moreover, as we have repeatedly observed, there exists an entire class of

secondary works that add “new expression, meaning, or message” to their source

material but are nonetheless specifically excluded from the scope of fair use:

derivative works. As one of our sister circuits has observed, an overly liberal

standard of transformativeness, such as that embraced by the district court in this

case, risks crowding out statutory protections for derivative works. See Kienitz v.

Sconnie Nation LLC, 766 F.3d 756, 758 (7th Cir. 2014) (“To say that a new use

transforms the work is precisely to say that it is derivative and thus, one might

suppose, protected under [17 U.S.C.] § 106(2).”).

20
 We addressed derivative works in Cariou, characterizing them as

secondary works that merely present “the same material but in a new form”

without “add[ing] something new.” 714 F.3d at 708 (citation omitted); see also

Google, 804 F.3d at 215-16 (“[D]erivative works generally involve transformations

in the nature of changes of form.”) (emphasis in original). While that description

may be a useful shorthand, it is likewise susceptible to misapplication if

interpreted too broadly. Indeed, many derivative works “add something new” to

their source material.

Consider, for example, a film adaptation of a novel. Such adaptations

frequently add quite a bit to their source material: characters are combined,

eliminated, or created out of thin air; plot elements are simplified or eliminated;

new scenes are added; the moral or political implications of the original work

may be eliminated or even reversed, or plot and character elements altered to

create such implications where the original text eschewed such matters. And all

of these editorial modifications are filtered through the creative contributions of

the screenwriter, director, cast, camera crew, set designers, cinematographers,

editors, sound engineers, and myriad other individuals integral to the creation of

a film. It is for this reason that we have recognized that “[w]hen a novel is

21
converted to a film . . . [t]he invention of the original author combines with the

cinematographic interpretive skills of the filmmaker to produce something that

neither could have produced independently.” Google, 804 F.3d at 216 n.18.

Despite the extent to which the resulting movie may transform the aesthetic and

message of the underlying literary work, film adaptations are identified as a

paradigmatic example of derivative works. See, e.g., Authors Guild, Inc. v.

HathiTrust, 755 F.3d 87, 95 (2d Cir. 2014) (“Paradigmatic examples of derivative

works include . . . the adaptation of a novel into a movie or a play.”).

In evaluating the extent to which a work is transformative or derivative (or

neither), we typically consider the purpose of the primary and secondary works.

In Bill Graham Archives v. Dorling Kindersley Ltd., for example, we held that the

reproduction in a book about the Grateful Dead of images of posters originally

created to advertise Grateful Dead concerts was transformative because that use

was “plainly different from the original purpose for which they were created.”

448 F.3d 605, 609-10 (2d Cir. 2006). Likewise, in HathiTrust we held that the

defendants’ creation of a searchable “digital corpus” comprising scanned copies

of tens of millions of books that enabled researchers, scholars, and others to

pinpoint the exact page of any book in the catalogue on which the searched term

22
was used was a “quintessentially transformative use.” 755 F.3d at 97. In Google,

we reached the same conclusion when faced with a larger digital corpus

complete with tools that enabled researchers to track how a specific word or

phrase has been used throughout the development of the English language,

despite the fact that, unlike the database in Hathitrust, Google’s database also

permitted the searcher to view a “snippet” from the original text showing the

context in which the word or phrase had appeared. 804 F.3d at 216-17.

But purpose is perhaps a less useful metric where, as here, our task is to

assess the transformative nature of works of visual art that, at least at a high level

of generality, share the same overarching purpose (i.e., to serve as works of visual

art). While this is not the first time we have had to conduct this inquiry, our cases

on such works are considerably fewer in number, and a brief review of them

yields conflicting guidance. In Blanch v. Koons, for example, we adjudged

transformative a Jeff Koons painting that incorporated a copyrighted photograph

drawn from a fashion magazine where Koons had testified that he intended to

“us[e] Blanch’s image as fodder for his commentary on the social and aesthetic

consequences of mass media.” 467 F.3d at 253. Some time earlier, however, in

Rogers v. Koons, we denied Koons’s fair-use defense as applied to a three-

23
dimensional sculpture recreating a photograph, notwithstanding his claim that

he intended his sculpture to serve as a commentary on modern society. 960 F.2d

301, 309-11 (2d Cir. 1992).2 And, in Cariou, we held twenty-five of Richard

Prince’s works transformative as a matter of law even though Prince had testified

that he “was not ‘trying to create anything with a new meaning or a new

message.’” 714 F.3d at 707.

Matters become simpler, however, when we compare the works at issue in

each case against their respective source materials. The sculpture at issue in

Rogers was a three-dimensional colorized version of the photograph on which it

was based. See 960 F.2d at 305. In Blanch, however, Koons used Blanch’s

photograph, depicting a woman’s legs in high-heeled shoes, as part of a larger

work in which he set it alongside several other similar photographs with

“changes of its colors, the background against which it is portrayed, the medium,

the size of the objects pictured, [and] the objects’ details.” 467 F.3d at 253. In so

doing, Koons used Blanch’s photograph “as raw material for an entirely different

2
We note that Rogers predates the Supreme Court’s formal adoption of the
“transformative use” test and thus does not phrase its inquiry in precisely the
same manner as the cases that have followed. However, it remains a precedential
decision of this Court, and we believe it particularly relevant in this case.

24
type of art . . . that comment[ed] on existing images by juxtaposing them against

others.” Id. at 262 (Katzmann, J., concurring). And in Cariou, the copyrighted

works found to have been fairly used were, in most cases, juxtaposed with other

photographs and “obscured and altered to the point that Cariou’s original [was]

barely recognizable.” 714 F.3d at 710. The works that were found potentially

infringing in Cariou, however, were ones in which the original was altered in

ways that did not incorporate other images and that superimposed other

elements that did not obscure the original image and in which the original image

remained, as in the Koons sculpture at issue in Rogers, a major if not dominant

component of the impression created by the allegedly infringing work. See id. at

710-11.

A common thread running through these cases is that, where a secondary

work does not obviously comment on or relate back to the original or use the

original for a purpose other than that for which it was created, the bare assertion

of a “higher or different artistic use,” Rogers, 960 F.2d at 310, is insufficient to

render a work transformative. Rather, the secondary work itself must reasonably

be perceived as embodying an entirely distinct artistic purpose, one that conveys

a “new meaning or message” entirely separate from its source material. While we

25
cannot, nor do we attempt to, catalog all of the ways in which an artist may

achieve that end, we note that the works that have done so thus far have

themselves been distinct works of art that draw from numerous sources, rather

than works that simply alter or recast a single work with a new aesthetic.

Which brings us back to the Prince Series. The district court held that the

Prince Series works are transformative because they “can reasonably be

perceived to have transformed Prince from a vulnerable, uncomfortable person to

an iconic, larger-than-life figure.” Warhol, 382 F. Supp. 3d at 326. That was error.

Though it may well have been Goldsmith’s subjective intent to portray

Prince as a “vulnerable human being” and Warhol’s to strip Prince of that

humanity and instead display him as a popular icon, whether a work is

transformative cannot turn merely on the stated or perceived intent of the artist

or the meaning or impression that a critic – or for that matter, a judge – draws

from the work. Were it otherwise, the law may well “recogniz[e] any alteration as

transformative.” 4 Melville B. Nimmer & David Nimmer, Nimmer on Copyright

§ 13.05(B)(6); see also Google, 804 F.3d at 216 n.18 (“[T]he word ‘transformative,’ if

interpreted too broadly, can also seem to authorize copying that should fall

26
within the scope of an author’s derivative rights.”). Rather, as we have discussed,

the court must examine how the works may reasonably be perceived.

In conducting this inquiry, however, 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. That is so both because judges are typically unsuited to make

aesthetic judgments and because such perceptions are inherently subjective.3 As

Goldsmith argues, her own stated intent notwithstanding, “an audience viewing

the [Goldsmith] [P]hotograph today, across the vista of the singer’s long career,

might well see him in a different light than Goldsmith saw him that day in 1981.”

Appellants’ Br. at 40. We agree; it is easy to imagine that a whole generation of

Prince’s fans might have trouble seeing the Goldsmith Photograph as depicting

anything other than the iconic songwriter and performer whose musical works

they enjoy and admire.

Instead, the judge must examine whether the secondary work’s use of its

source material is in service of a “fundamentally different and new” artistic

3
As the Supreme Court observed over a century ago, “[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, outside of the narrowest and most
obvious limits.” Bleistein v. Donaldson Lithographing Co., 188 U.S. 239, 251 (1903).

27
purpose and character, such that the secondary work stands apart from the “raw

material” used to create it. Although we do not hold that the primary work must

be “barely recognizable” within the secondary work, as was the case with the

works held transformative in Cariou, the secondary work’s transformative

purpose and character must, at a bare minimum, comprise something more than

the imposition of another artist’s style on the primary work such that the

secondary work remains both recognizably deriving from, and retaining the

essential elements of, its source material.

With this clarification, viewing the works side-by-side, we conclude that

the Prince Series is not “transformative” within the meaning of the first factor.

That is not to deny that the Warhol works display the distinct aesthetic sensibility

that many would immediately associate with Warhol’s signature style – the

elements of which are absent from the Goldsmith photo. But the same can be

said, for example, of the Ken Russell film, from a screenplay by Larry Kramer,

derived from D.H. Lawrence’s novel, Women in Love: the film is as recognizable a

“Ken Russell” as the Prince Series are recognizably “Warhols.” But the film, for

all the ways in which it transforms (that is, in the ordinary meaning of the word,

28
which indeed is used in the very definition of derivative works, see 17 U.S.C.

§ 101) its source material, is also plainly an adaptation of the Lawrence novel.

As in the case of such paradigmatically derivative works, there can be no

meaningful dispute that the overarching purpose and function of the two works

at issue here is identical, not merely in the broad sense that they are created as

works of visual art, but also in the narrow but essential sense that they are

portraits of the same person.4 See Gaylord v. United States, 595 F.3d 1364, 1372-73

(Fed. Cir. 2010) (photograph of Korean War Memorial used on stamp not

transformative despite “different expressive character” brought about by

subdued lighting and snow since sculpture and stamp shared purpose of

“honor[ing] veterans of the Korean War”). Although this observation does not per

se preclude a conclusion that the Prince Series makes fair use of the Goldsmith

Photograph, the district court’s conclusion rests significantly on the

4
As much as art critics might distinguish Warhol’s aesthetic intentions from
those of portrait photographers, Warhol’s celebrity prints are invariably
identifiable likenesses of their subjects. The district court’s description of the
Prince Series works as transformative because they “can reasonably be perceived
to have transformed Prince from a vulnerable, uncomfortable person to an iconic,
larger-than-life figure,” 382 F. Supp. 3d at 326, rests implicitly on the Warhol
depiction being perceived as a recognizable depiction of Prince.

29
transformative character of Warhol’s work. But the Prince Series works can’t bear

that weight.

Warhol created the series chiefly by removing certain elements from the

Goldsmith Photograph, such as depth and contrast, and embellishing the

flattened images with “loud, unnatural colors.” Warhol, 382 F. Supp. 3d at 326.

Nonetheless, although we do not conclude that the Prince Series works are

necessarily derivative works as a matter of law, they are much closer to presenting

the same work in a different form, that form being a high-contrast screenprint,

than they are to being works that make a transformative use of the original.

Crucially, the Prince Series retains the essential elements of the Goldsmith

Photograph without significantly adding to or altering those elements.

Indeed, the differences between the Goldsmith Photograph and the Prince

Series here are in many respects less substantial than those made to the five

works that we could not find transformative as a matter of law in Cariou. Unlike

the Prince Series, those works unmistakably deviated from Cariou’s original

portraiture in a manner that suggested an entirely distinct artistic end; rather

than recasting those photographs in a new medium, Richard Prince added

material that pulled them in new directions. See, e.g., Cariou, 714 F.3d at 711

30
(“Where [Cariou’s] photograph presents someone comfortably at home in nature,

[Prince’s] Graduation combines divergent elements to present a sense of

discomfort.”). Nevertheless, we could not confidently determine whether those

modest alterations “amount[ed] to a substantial transformation of the original

work[s] of art such that the new work[s] were transformative,” and remanded the

case to the district court to make that determination in the first instance. Id.

In contrast, the Prince Series retains the essential elements of its source

material, and Warhol’s modifications serve chiefly to magnify some elements of

that material and minimize others. While the cumulative effect of those

alterations may change the Goldsmith Photograph in ways that give a different

impression of its subject, the Goldsmith Photograph remains the recognizable

foundation upon which the Prince Series is built.

Finally, we feel compelled to clarify that it is entirely irrelevant to this

analysis that “each Prince Series work is immediately recognizable as a

‘Warhol.’” Warhol, 382 F. Supp. 3d at 326. Entertaining that logic would

inevitably create a celebrity-plagiarist privilege; the more established the artist

and the more distinct that artist’s style, the greater leeway that artist would have

to pilfer the creative labors of others. But the law draws no such distinctions;

31
whether the Prince Series images exhibit the style and characteristics typical of

Warhol’s work (which they do) does not bear on whether they qualify as fair use

under the Copyright Act. As Goldsmith notes, the fact that Martin Scorsese’s

recent film The Irishman is recognizably “a Scorsese” “do[es] not absolve [him] of

the obligation to license the original book” on which it is based. Appellants’ Br. at

37.

In reaching this conclusion, we do not mean to discount the artistic value

of the Prince Series itself. As used in copyright law, the words “transformative”

and “derivative” are legal terms of art that do not express the simple ideas that

they carry in ordinary usage. We do not disagree with AWF’s contention that the

cumulative effect of Warhol’s changes to the Goldsmith Photograph is to produce

a number of striking and memorable images. And our conclusion that those

images are closer to what the law deems “derivative” than “transformative” does

not imply that the Prince Series (or Warhol’s art more broadly) is “derivative,” in

the pejorative artistic sense, of Goldsmith’s work or of anyone else’s. As

Goldsmith succinctly puts it, “[t]here is little doubt . . . that the Prince Series

reflects Andy Warhol’s talent, creativity, and distinctive aesthetic.” Appellants’

Br. at 36. But the task before us is not to assess the artistic worth of the Prince

32
Series nor its place within Warhol’s oeuvre; that is the domain of art historians,

critics, collectors, and the museum-going public. Rather, the question we must

answer is simply whether the law permits Warhol to claim it as his own, and

AWF to exploit it, without Goldsmith’s permission. And, at least as far as this

aspect of the first factor is concerned, we conclude that the answer to that

question is “no.”

2. Commercial Use

The statutory language of the first factor also specifically directs courts to

consider “whether [the] use is of a commercial nature or is for nonprofit

educational purposes.” 17 U.S.C. § 107(1). Although finding that a secondary use

is commercial “tends to weigh against” finding that it is fair, we apply the test

with caution since “nearly all of the illustrative uses listed in the preamble

paragraph of § 107 . . . are generally conducted for profit in this country.”

Campbell, 510 U.S. at 584-85 (citation and internal quotation marks omitted).5

5
To recognize this is not to read the commercial/non-profit factor out of the
statute. There are other situations in which the absence or presence of a
commercial motive may be highly significant. Producing a small number of
copies of a short story to be distributed for free to a high school English class may
be quite different from producing a similar number of copies for a lavishly bound
and illustrated “limited edition” of the work to be sold in the marketplace at a
high price.

33
And, since “[t]he crux of the profit/nonprofit distinction is . . . whether the user

stands to profit from exploitation of the copyrighted material without paying the

customary price,” Harper & Row, 471 U.S. at 562, the commercial nature of a

secondary use is of decreased importance when the use is sufficiently

transformative such that the primary author should not reasonably expect to be

compensated. See, e.g., Blanch, 467 F.3d at 254.

We agree with the district court that the Prince Series works are

commercial in nature, but that they produce an artistic value that serves the

greater public interest. See Warhol, 382 F. Supp. 3d at 325. We also agree that,

although more relevant to the character of the user than of the use, the fact that

AWF’s mission is to advance the visual arts, a mission that is doubtless in the

public interest, may militate against the simplistic assertion that AWF’s sale and

licensing of the Prince Series works necessarily derogates from a finding of fair

use. Nevertheless, just as we cannot hold that the Prince Series is transformative

as a matter of law, neither can we conclude that Warhol and AWF are entitled to

monetize it without paying Goldsmith the “customary price” for the rights to her

work, even if that monetization is used for the benefit of the public.

34
 Of course, even where the secondary use is not transformative, the extent

to which it serves the public interest, either in and of itself or by generating funds

that enable the secondary user to further a public-facing mission, may be highly

relevant when assessing equitable remedies, including whether to enjoin the

distribution or order the destruction of infringing works.6 But just as the

commercial nature of a transformative secondary use does not itself preclude a

finding that the use is fair, the fact that a commercial non-transformative work

may also serve the public interest or that the profits from its commercial use are

turned to the promotion of non-commercial ends does not factor significantly in

favor of finding fair use under the circumstances present here.

B. The Nature of the Copyrighted Work

The second factor directs courts to consider the nature of the copyrighted

work, including (1) whether it is “expressive or creative . . . or more factual, with

a greater leeway being allowed to a claim of fair use where the work is factual or

informational, and (2) whether the work is published or unpublished, with the

6
Goldsmith does not seek such remedies, and it is highly unlikely that any court
would deem them appropriate in this case. See Campbell, 510 U.S. at 578 n.10
(“[T]he goals of the copyright law . . . are not always best served by automatically
granting injunctive relief when parodists are found to have gone beyond the
bounds of fair use.”).

35
scope of fair use involving unpublished works being considerably narrower.”

Blanch, 467 F.3d at 256 (citation omitted). Although courts are required to

consider and weigh this factor, it “has rarely played a significant role in the

determination of a fair use dispute.” Google, 804 F.3d at 220.

The district court correctly held that the Goldsmith Photograph is both

unpublished and creative but nonetheless concluded that the second factor

should favor neither party because LGL had licensed the Goldsmith Photograph

to Vanity Fair and because the Prince Series was highly transformative. See

Warhol, 382 F. Supp. 3d at 327. That was error. That Goldsmith, through LGL,

made the Goldsmith Photograph available for a single use on limited terms does

not change its status as an unpublished work nor diminish the law’s protection of

her choice of “when to make a work public and whether to withhold a work to

shore up demand.” Id., citing 4 Nimmer on Copyright § 13.05(A)(2)(b). Further,

though we have previously held that this factor “may be of limited usefulness

where the creative work is being used for a transformative purpose,” Bill Graham

Archives, 448 F.3d at 612, this relates only to the weight assigned to it, not whom

it favors. See also Blanch, 467 F.3d at 257 (“[T]he second fair-use factor has limited

36
weight in our analysis because Koons used Blanch’s work in a transformative

manner.”).

Having recognized the Goldsmith Photograph as both creative and

unpublished, the district court should have found this factor to favor Goldsmith

irrespective of whether it adjudged the Prince Series works transformative within

the meaning of the first factor. And, because we disagree that the Prince Series

works are transformative, we would accord this factor correspondingly greater

weight.

C. The Amount and Substantiality of the Use

The third factor considers “the amount and substantiality of the portion

used in relation to the copyrighted work as a whole.” 17 U.S.C. § 107(3). “In

assessing this factor, we consider not only ‘the quantity of the materials used’ but

also ‘their quality and importance’” in relation to the original work. TCA

Television, 839 F.3d at 185, quoting Campbell, 510 U.S. at 587. The ultimate

question under this factor is whether “the quantity and value of the materials

used are reasonable in relation to the purpose of the copying.” Campbell, 510 U.S.

at 586 (citation and internal quotation marks omitted). To that end, there is no

bright line separating a permissible amount of borrowing from an impermissible

37
one; indeed, we have rejected the proposition that this factor necessarily favors

the copyright holder even where the secondary user has copied the primary work

in toto in service of a legitimate secondary purpose. See Swatch Grp. Mgmt. Servs.

Ltd. v. Bloomberg L.P., 756 F.3d 73, 89-90 (2d Cir. 2014); see also Rogers, 960 F.2d at

310-11 (“Sometimes wholesale copying may be permitted, while in other cases

taking even a small percentage of the original work has been held unfair use.”).

In this case, AWF argues, and the district court concluded, that this factor

weighs in its favor because, by cropping and flattening the Goldsmith

Photograph, thereby removing or minimizing its use of light, contrast, shading,

and other expressive qualities, Warhol removed nearly all of its copyrightable

elements. We do not agree.

We begin with the uncontroversial proposition that copyright does not

protect ideas, but only “the original or unique way that an author expresses those

ideas, concepts, principles, or processes.” Rogers, 960 F.2d at 308. As applied to

photographs, this protection encompasses the photographer’s “posing the

subjects, lighting, angle, selection of film and camera, evoking the desired

expression, and almost any other variant involved.” Id. at 307. The cumulative

manifestation of these artistic choices – and what the law ultimately protects – is

38
the image produced in the interval between the shutter opening and closing, i.e.,

the photograph itself. This is, as we have previously observed, the

photographer’s “particular expression” of the idea underlying her photograph.

Leibovitz, 137 F.3d at 115-16.

It is thus easy to understand why AWF’s contention misses the mark. The

premise of its argument is that Goldsmith cannot copyright Prince’s face. True

enough. Were it otherwise, nobody else could have taken the man’s picture

without either seeking Goldsmith’s permission or risking a suit for infringement.

But while Goldsmith has no monopoly on Prince’s face, the law grants her a

broad monopoly on its image as it appears in her photographs of him, including

the Goldsmith Photograph.7 Cf. Mattel, Inc. v. Goldberger Doll Mfg. Co., 365 F.3d

133, 136-37 (2d Cir. 2004) (vacating summary judgment where district court had

7
It is for this reason that the cases that AWF cites in support of its position (and
on which the district court relied) are not particularly instructive; each involves a
claim in which a second, distinct work was alleged to infringe the protected
expression of the original work, and each such claim was rejected on the basis
that the second work copied only the unprotected idea of the original. See, e.g.,
Bill Diodato Photography, LLC v. Kate Spade, LLC, 388 F. Supp. 2d 382, 393 (S.D.N.Y.
2005) (involving separate photographs of women in bathroom stalls with jauntily
placed handbags); see also infra Section III. Had Warhol used a different
photograph that Goldsmith alleged was similar enough to her own to render the
Prince Series an infringement of her work, these cases might be more instructive.
But he did not, so they are not.

39
concluded that “defendant could freely copy the central facial features of the

Barbie dolls” and holding that Mattel could not monopolize the idea of a doll

with “upturned nose, bow lips, and wide eyes,” but the law protected its specific

rendition thereof). And where, as here, the secondary user has used the

photograph itself, rather than, for example, a similar photograph, the

photograph’s specific depiction of its subject cannot be neatly reduced to discrete

qualities such as contrast, shading, and depth of field that can be stripped away,

taking the image’s entitlement to copyright protection along with it.

With that in mind, we readily conclude that the Prince Series borrows

significantly from the Goldsmith Photograph, both quantitatively and

qualitatively. While Warhol did indeed crop and flatten the Goldsmith

Photograph, the end product is not merely a screenprint identifiably based on a

photograph of Prince. Rather it is a screenprint readily identifiable as deriving

from a specific photograph of Prince, the Goldsmith Photograph. A comparison of

the images in the Prince Series makes plain that Warhol did not use the

Goldsmith Photograph simply as a reference or aide-mémoire in order to

accurately document the physical features of its subject. Instead, the Warhol

40
images are instantly recognizable as depictions or images of the Goldsmith

Photograph itself.

To confirm this, one need look no further than the other photographs of

Prince that AWF submitted in support of its motion below to evidence its

contention that Prince’s pose was not unique to the Goldsmith Photograph. [JA

1707-09] Though any of them may have been suitable as a base photograph for

Warhol’s process, we have little doubt that the Prince Series would be quite

different had Warhol used one of them instead of the Goldsmith Photograph to

create it. But the resemblance between the Prince Series works and the Goldsmith

Photograph goes even further; for example, many of the aspects of Prince’s

appearance in the Prince Series works, such as the way in which his hair appears

shorter on the left side of his face, are present in the Goldsmith Photograph yet

absent even from some other photographs that Goldsmith took of Prince during

the same photo session. In other words, whatever the effect of Warhol’s

alterations, the “essence of [Goldsmith’s] photograph was copied” and persists in

the Prince Series. Rogers, 960 F.2d at 311. Indeed, Warhol’s process had the effect

41
of amplifying, rather than minimizing, certain aspects of the Goldsmith

Photograph.8

Nor can Warhol’s appropriation of the Goldsmith Photograph be deemed

reasonable in relation to his purpose. While Warhol presumably required a

photograph of Prince to create the Prince Series, AWF proffers no reason why he

required Goldsmith’s photograph. See TCA Television, 839 F.3d at 181-82, 185

(wholesale borrowing of copyrighted comedy routine not reasonable where

“defendants offer[ed] no persuasive justification” for its use). To the contrary, the

evidence in the record suggests that Warhol had no particular interest in the

Goldsmith Photograph or Goldsmith herself; Vanity Fair licensed a photograph

of Prince, and there is no evidence that Warhol (or, for that matter, Vanity Fair)

was involved in identifying or selecting the particular photograph that LGL

provided.

8
For example, the fact that Prince’s mustache appears to be lighter on the right
side of his face than the left is barely noticeable in the grayscale Goldsmith
Photograph but is quite pronounced in the black-and-white Prince Series
screenprints. Moreover, this feature of the Goldsmith Photograph is, again, not
common to all other photographs of Prince even from that brief session. The
similarity is not simply an artefact of what Prince’s facial hair was like on that
date, but of the particular effects of light and angle at which Goldsmith captured
that aspect of his appearance.

42
 To be clear, we do not hold that this factor will always favor the copyright

holder where the work at issue is a photograph and the photograph remains

identifiable in the secondary work. But this case is not Kienitz v. Sconnie Nation

LLC, in which a panel of the Seventh Circuit held that a t-shirt design that

incorporated a photograph in a manner that stripped away nearly every

expressive element such that, “as with the Cheshire Cat, only the [subject’s] smile

remain[ed]” was fair use. 766 F.3d at 759. As discussed, Warhol’s rendition of the

Goldsmith Photograph leaves quite a bit more detail, down to the glint in

Prince’s eyes where the umbrellas in Goldsmith’s studio reflected off his pupils.

Thus, though AWF urges this court to follow the Seventh Circuit’s lead, its

decision in Kienitz would not compel a different result here, even if it were

binding on us – which, of course, it is not.

The district court, reasoning that Warhol had taken only the unprotected

elements of the Goldsmith Photograph in service of a transformative purpose,

held that this factor strongly favored AWF. Because we disagree on both counts,

we conclude that this factor strongly favors Goldsmith.

43
 D. The Effect of the Use on the Market for the Original

The fourth factor asks “whether, if the challenged use becomes

widespread, it will adversely affect the potential market for the copyrighted

work.” Bill Graham Archives, 448 F.3d at 613. “Analysis of this factor requires us to

balance the benefit the public will derive if the use is permitted and the personal

gain the copyright owner will receive if the use is denied.” Wright v. Warner

Books, Inc., 953 F.2d 731, 739 (2d Cir. 1991) (internal quotation marks omitted). In

assessing market harm, we ask not whether the second work would damage the

market for the first (by, for example, devaluing it through parody or criticism),

but whether it usurps the market for the first by offering a competing substitute.

See, e.g., Bill Graham Archives, 448 F.3d at 614. This analysis embraces both the

primary market for the work and any derivative markets that exist or that its

author might reasonably license others to develop, regardless of whether the

particular author claiming infringement has elected to develop such markets. See

Salinger v. Colting, 607 F.3d 68, 74, 83 (2d Cir. 2010) (affirming that fourth factor

favored J.D. Salinger in suit over unauthorized sequel to Catcher in the Rye despite

the fact that Salinger had publicly disclaimed any intent to author or authorize a

sequel, but vacating preliminary injunction on other grounds). As we have

44
previously observed, the first and fourth factors are closely linked, as “the more

the copying is done to achieve a purpose that differs from the purpose of the

original, the less likely it is that the copy will serve as a satisfactory substitute for

the original.” Google, 804 F.3d at 223, citing Campbell, 510 U.S. at 591.

We agree with the district court that the primary market for the Warhol

Prince Series (that is, the market for the original works) and the Goldsmith

Photograph do not meaningfully overlap, and Goldsmith does not seriously

challenge that determination on appeal. We cannot, however, endorse the district

court’s implicit rationale that the market for Warhol’s works is the market for

“Warhols,” as doing so would permit this aspect of the fourth factor always to

weigh in favor of the alleged infringer so long as he is sufficiently successful to

have generated an active market for his own work. Notwithstanding, we see no

reason to disturb the district court’s overall conclusion that the two works occupy

distinct markets, at least as far as direct sales are concerned.

We are unpersuaded, however, by the district court’s conclusion that the

Prince Series poses no threat to Goldsmith’s licensing markets. While Goldsmith

does not contend that she has sought to license the Goldsmith Photograph itself,

the question under this factor is not solely whether the secondary work harms an

45
existing market for the specific work alleged to have been infringed. Cf. Castle

Rock, 150 F.3d at 145-46 (“Although Castle Rock has evidenced little if any

interest in exploiting this market for derivative works . . . the copyright law must

respect that creative and economic choice.”). Rather, we must also consider

whether “unrestricted and widespread conduct of the sort engaged in by [AWF]

would result in a substantially adverse impact on the potential market” for the

Goldsmith Photograph. Campbell, 510 U.S. at 590 (internal quotation marks

omitted) (alterations adopted)); see also Fox News Network, LLC v. TVEyes, Inc., 883

F.3d 169, 179 (2d Cir. 2018).

As an initial matter, we note that the district court erred in apparently

placing the burden of proof as to this factor on Goldsmith. See, e.g., Warhol, 382 F.

Supp. 3d at 330. While our prior cases have suggested that the rightsholder bears

some initial burden of identifying relevant markets,9 we have never held that the

rightsholder bears the burden of showing actual market harm. Nor would we so

9
See HathiTrust, 755 F.3d at 96 (“To defeat a claim of fair use, the copyright holder
must point to the market harm that results because the secondary use serves as a
substitute for the original work.”); Leibovitz, 137 F.3d at 116 n.6 (“Leibovitz has
not identified any market for a derivative work that might be harmed by the
Paramount ad. In these circumstances, the defendant had no obligation to present
evidence showing lack of harm in a market for derivative works.”).

46
hold. Fair use is an affirmative defense; as such, the ultimate burden of proving

that the secondary use does not compete in the relevant market is appropriately

borne by the party asserting the defense: the secondary user. See Campbell, 510

U.S. at 590 (“Since fair use is an affirmative defense, its proponent would have

difficulty carrying the burden of demonstrating fair use without favorable

evidence about relevant markets.”); Infinity Broadcast Corp. v. Kirkwood, 150 F.3d

104, 110 (2d Cir. 1998) (“As always, [the secondary user] bears the burden of

showing that his use does not” usurp the market for the primary work); Dr. Seuss

Enters., L.P. v. ComicMix LLC, 983 F.3d 443, 459 (9th Cir. 2020) (“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.”).

In any case, whatever the scope of Goldsmith’s initial burden, she satisfied

it here. Setting aside AWF’s licensing of Prince Series works for use in museum

exhibits and publications about Warhol, which is not particularly relevant for the

reasons set out in our discussion of the primary market for the works, there is no

material dispute that both Goldsmith and AWF have sought to license (and

47
indeed have successfully licensed) their respective depictions of Prince10 to

popular print magazines to accompany articles about him. As Goldsmith

succinctly states: “both [works] are illustrations of the same famous musician

with the same overlapping customer base.” Appellants’ Br. at 50. Contrary to

AWF’s assertions, that is more than enough. See Cariou, 714 F.3d at 709 (“[A]n

accused infringer has usurped the market for copyrighted works . . . where the

infringer’s target audience and the nature of the infringing content is the same as

the original.”). And, since Goldsmith has identified a relevant market, AWF’s

failure to put forth any evidence that the availability of the Prince Series works

poses no threat to Goldsmith’s actual or potential revenue in that market tilts the

scales toward Goldsmith.

Finally, the district court entirely overlooked the potential harm to

Goldsmith’s derivative market, which is likewise substantial. Most directly,

AWF’s licensing of the Prince Series works to Condé Nast without crediting or

paying Goldsmith deprived her of royalty payments to which she would have

otherwise been entitled. Although we do not always consider lost royalties from

10
In Goldsmith’s case, photographs other than the Goldsmith Photograph, which
she has withheld from the market.

48
the challenged use itself under the fourth factor (as any fair use necessarily

involves the secondary user using the primary work without paying for the right

to do so), we do consider them where the secondary use occurs within a

traditional or reasonable market for the primary work. See Fox News, 883 F.3d at

180; On Davis v. Gap, Inc., 246 F.3d 152, 176 (2d Cir. 2001). And here, that market

is established both by Goldsmith’s uncontroverted expert testimony that

photographers generally license others to create stylized derivatives of their work

in the vein of the Prince Series, see J. App’x 584-99, and by the genesis of the

Prince Series: a licensing agreement between LGL and Vanity Fair to use the

Goldsmith Photograph as an artist reference.11

Further, we also must consider the impact on this market if the sort of

copying in which Warhol engaged were to become a widespread practice. That

harm is also self-evident. There currently exists a market to license photographs

of musicians, such as the Goldsmith Photograph, to serve as the basis of a

stylized derivative image; permitting this use would effectively destroy that

broader market, as, if artists “could use such images for free, there would be little

11
Of course, if a secondary work is sufficiently transformative, the fact that its
“raw material” was acquired by means of a limited license will not necessarily
defeat a defense of fair use. As discussed supra, however, that is not the case here.

49
or no reason to pay for [them].” Barcroft Media, Ltd. v. Coed Media Grp., LLC, 297 F.

Supp. 3d 339, 355 (S.D.N.Y. 2017); see also Seuss, 983 F.3d at 461 (“[T]he

unrestricted and widespread conduct of the sort ComicMix is engaged in could

result in anyone being able to produce” their own similar derivative works based

on Oh, the Places You’ll Go!). This, in turn, risks disincentivizing artists from

producing new work by decreasing its value – the precise evil against which

copyright law is designed to guard.

Thus, although the primary market for the Goldsmith Photograph and the

Prince Series may differ, the Prince Series works pose cognizable harm to

Goldsmith’s market to license the Goldsmith Photograph to publications for

editorial purposes and to other artists to create derivative works based on the

Goldsmith Photograph and similar works. Accordingly, the fourth factor favors

Goldsmith.

E. Weighing the Factors

“[T]his court has on numerous occasions resolved fair use determinations

at the summary judgment stage where there are no genuine issues of material

fact.” Cariou, 714 F.3d at 704 (internal quotation marks omitted) (alteration

50
adopted) (collecting cases). As no party contends that there exist any issues of

material fact in this case, we believe it appropriate to exercise that discretion here.

Having considered each of the four factors, we find that each favors

Goldsmith. Further, although the factors are not exclusive, AWF has not

identified any additional relevant considerations unique to this case that we

should take into account. Accordingly, we hold that AWF’s defense of fair use

fails as a matter of law.

III. Substantial Similarity

AWF asks this Court to affirm the district court’s decision on the alternate

grounds that the Prince Series works are not substantially similar to the

Goldsmith Photograph. We decline that invitation, because we conclude that the

works are substantially similar as a matter of law.

The district court did not analyze the issue of substantial similarity

because, in its view, “it [was] plain that the Prince Series works are protected by

fair use.” Warhol, 382 F. Supp. 3d at 324. While “it is our distinctly preferred

practice to remand such issues for consideration by the district court in the first

instance,” Schonfeld v. Hilliard, 218 F.3d 164, 184 (2d Cir. 2000), we are not

required to do so. In this case, because the question of substantial similarity is

51
logically antecedent to that of fair use – since there would be no need to invoke

the fair use defense in the absence of actionable infringement – and because the

factors we have already discussed with respect to fair use go a considerable way

toward resolving the substantial similarity issue, we do not believe a remand to

address that issue is necessary in this case.12

In general, and as applicable here, two works are substantially similar

when “an average lay observer would recognize the alleged copy as having been

appropriated from the copyrighted work.” Knitwaves, Inc. v. Lollytogs, Ltd., 71

F.3d 996, 1003 (2d Cir. 1995), quoting Malden Mills, Inc. v. Regency Mills, Inc., 626

F.2d 1112, 1113 (2d Cir. 1980). “On occasion, . . . we have noted that when faced

with works that have both protectable and unprotectable elements, our analysis

must be more discerning and that we instead must attempt to extract the

unprotectable elements from our consideration and ask whether the protectable

elements, standing alone, are substantially similar.” Peter F. Gaito Architecture,

LLC v. Simone Dev. Corp., 602 F.3d 57, 66 (2d Cir. 2010) (internal citations and

quotation marks omitted). AWF and its amici contend that this “more discerning

12
We express no view on the viability of AWF’s remaining defenses, which are
appropriately considered by the district court in the first instance.

52
observer” test should apply here because photographs contain both protectable

and unprotectable elements. See Appellee’s Br. at 65; Law Professors’ Br. at 8. The

same could be said, however, of any copyrighted work: even the most

quintessentially “expressive” works, such as books or paintings, contain non-

copyrightable ideas or concepts. See 4 Nimmer on Copyright § 13.03(B)(2).

Moreover, the cases in which we have applied the “more discerning

observer” test involved types of works with much “thinner” copyright protection

– i.e., works that are more likely to contain a larger share of non-copyrightable

elements. See, e.g., Zalewski v. Cicero Builder Dev., Inc., 754 F.3d 95, 102 (2d Cir.

2014) (architectural designs); Tufenkian Import/Export Ventures, Inc. v. Einstein

Moomjy, Inc., 338 F.3d 127, 136 n.13 (2d Cir. 2003) (Tibetan-style carpets); Boisson

v. Banian, Ltd., 273 F.3d 262, 272 (2d Cir. 2001) (quilts). By contrast, “photographs

are ‘generally viewed as creative aesthetic expressions of a scene or image’ and

have long received thick copyright protection[,] . . . even though photographs

capture images of reality.” Brammer v. Violent Hues Prods., LLC, 922 F.3d 255, 267

(4th Cir. 2019), quoting Monge v. Maya Magazines, Inc., 688 F.3d 1164, 1177 (9th

Cir. 2012). We therefore reject AWF’s contention that we should be “more

discerning” in considering whether the Prince Series is substantially similar to

53
the Goldsmith Photograph and apply the standard “ordinary observer” test. See

Knitwaves, 71 F.3d at 1002-03.

Though substantial similarity often presents a jury question, it may be

resolved as a matter of law where “access to the copyrighted work is conceded,

and the accused work is so substantially similar to the copyrighted work that

reasonable jurors could not differ on this issue.” Rogers, 960 F.2d at 307 (citation

omitted); see also Gaito, 602 F.3d at 63 (“The question of substantial similarity is by

no means exclusively reserved for resolution by a jury.”).

Here, AWF has conceded that the Goldsmith Photograph served as the

“raw material” for the Prince Series works. See Appellee’s Br. at 6-7. AWF

nevertheless attempts to compare this case to several decisions from our sister

circuits concluding that the secondary works in question were not substantially

similar to the original photographs on which they were based. See, e.g.,

Rentmeester v. Nike, Inc., 883 F.3d 1111, 1121-23 (9th Cir. 2018) (Nike’s iconic

“Jumpman” logo and the photograph used to create it were not substantially

similar to a photograph of Michael Jordan dunking a basketball); Harney v. Sony

Pictures Television, Inc., 704 F.3d 173, 188 (1st Cir. 2013) (recreated image in made-

for-TV movie was not substantially similar to the photograph that inspired it).

54
But the secondary users in those cases did not merely copy the original

photographs at issue; they instead replicated those photographs using their own

subjects in similar poses. By contrast, Warhol did not create the Prince Series by

taking his own photograph of Prince in a similar pose as in the Goldsmith

Photograph. Nor did he attempt to copy merely the “idea” conveyed in the

Goldsmith Photograph. Rather, he produced the Prince Series works by copying

the Goldsmith Photograph itself – i.e., Goldsmith’s particular expression of that

idea. This case therefore stands in sharp contrast to the situation presented by

Rentmeester, for example, in which the court explained that “[w]hat [the original]

photo and the [allegedly infringing] photo share are similarities in general ideas

or concepts: Michael Jordan attempting to dunk in a pose inspired by ballet’s

grand jeté; an outdoor setting stripped of most of the traditional trappings of

basketball; a camera angle that captures the subject silhouetted against the sky.”

883 F.3d at 1122-23.

This is not to say that every use of an exact reproduction constitutes a work

that is substantially similar to the original. But here, given the degree to which

Goldsmith’s work remains recognizable within Warhol’s, there can be no

reasonable debate that the works are substantially similar. See Rogers, 960 F.3d

55
307-08. As we have noted above, Prince, like other celebrity performing and

creative artists, was much photographed. But any reasonable viewer with access

to a range of such photographs including the Goldsmith Photograph would have

no difficulty identifying the latter as the source material for Warhol’s Prince

Series.

CONCLUSION

For the foregoing reasons, we REVERSE the grant of AWF’s motion for

summary judgment, VACATE the judgment entered below dismissing Lynn

Goldsmith and LGL’s amended counterclaim, and REMAND this case for further

proceedings consistent with this opinion.

56
Richard J. Sullivan, Circuit Judge, joined by Dennis Jacobs, Circuit Judge,
concurring:

I fully join the majority’s thoughtful opinion and its conclusion that the

Prince Series works are substantially similar to the Goldsmith Photograph and are

not protected by fair use. I write separately only to highlight what I see as an

overreliance on “transformative use” in our fair use jurisprudence, generally, and

to suggest that a renewed focus on the fourth fair use factor, “the effect of the use

upon the potential market for or value of the copyrighted work,” 17 U.S.C. § 107(4),

would bring greater coherence and predictability to this area of the law.

In the wake of the Supreme Court’s decision in Campbell v. Acuff-Rose Music,

Inc., 510 U.S. 569 (1994), the “transformative” nature of a secondary work has

become the dominant focus in determining whether that work is protected by fair

use. Courts and commentators have recognized this trend and have observed that

it threatens to collapse the four statutory fair use factors into a single, dispositive

factor. See, e.g., Kienitz v. Sconnie Nation LLC, 766 F.3d 756, 758 (7th Cir. 2014)

(“[A]sking exclusively whether something is ‘transformative’ not only replaces the

list in § 107 but also could override 17 U.S.C. § 106(2), which protects derivative

works.”); 4 Melville B. Nimmer & David Nimmer, Nimmer on Copyright

§ 13.05[A][1][b] (noting that many courts’ applications of the transformative use

1
test “are conclusory – they appear to label a use ‘not transformative’ as a shorthand

for ‘not fair,’ and correlatively ‘transformative’ for ‘fair’”). Indeed, one recent

empirical study found that, among a sample of 238 district and circuit court

decisions, whether a secondary work was transformative correlated with the

ultimate fair use outcome 94% of the time. See Jiarui Liu, An Empirical Study of

Transformative Use in Copyright Law, 22 Stan. Tech. L. Rev. 163, 180 (2019).

This pattern is perhaps best illustrated in the district court’s opinion below.

Having concluded that the Prince Series works were transformative, the district

court found that the second fair use factor was neutral (despite noting that the

Goldsmith Photograph was creative and unpublished, which “would ordinarily

weigh in Goldsmith’s favor”) and found that the third factor weighed heavily in

AWF’s favor because “Warhol transformed Goldsmith’s work into something new

and different.” Andy Warhol Found. for the Visual Arts, Inc. v. Goldsmith, 382 F. Supp.

3d 312, 327, 330 (S.D.N.Y. 2019) (internal quotation marks omitted). But perhaps

most notably, as the majority’s opinion recognizes, the district court completely

dismissed evidence of harm to Goldsmith’s potential licensing and derivative

markets after concluding that the Prince Series was transformative. See Majority

Op. at 45–50.

2
 Placing dispositive weight on transformative use while reducing evidence

of market harm to an afterthought is difficult to square with the Supreme Court’s

guidance that the fourth factor “is undoubtedly the single most important element

of fair use.” Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 566 (1985).

Indeed, we have previously explained that focusing on the importance of the

fourth factor “is consistent with the fact that the copyright is a commercial right,

intended to protect the ability of authors to profit from the exclusive right to

merchandise their own work.” Authors Guild v. Google, Inc., 804 F.3d 202, 214 (2d

Cir. 2015) (“Google Books”).

To be sure, some of this Court’s earlier decisions suggest that the Supreme

Court “retreated” from its emphasis on the fourth factor when it explained in

Campbell that “[a]ll [four fair use factors] are to be explored, and the results

weighed together, in light of the purposes of copyright,” 510 U.S. at 578. See, e.g.,

Blanch v. Koons, 467 F.3d 244, 258 n.8 (2d Cir. 2006); Castle Rock Ent., Inc. v. Carol

Publ’g Grp., Inc., 150 F.3d 132, 145 (2d Cir. 1998). But the statements in Campbell

and Harper & Row are not necessarily at odds with one another: courts can

consider all four factors while still recognizing that evidence of harm to the

3
potential market for the original work (or derivative works based on the original,

see Campbell, 510 U.S. at 593) should be given substantial weight.

The Supreme Court’s decision in Campbell does not suggest otherwise. To

the contrary, even though Campbell is recognized for crystallizing the concept of

transformative use, the opinion “characterizes the first factor inquiry as

subservient to the fourth.” Pierre N. Leval, Campbell as Fair Use Blueprint?, 90

Wash. L. Rev. 597, 605 (2015). Campbell explained that transformative works are

more likely to be fair uses because they are less likely to “act[] as a substitute” for

or “‘supersede[] the objects’” of the original work, and are therefore less likely to

“affect the market for the original in a way cognizable under [the fourth] factor.”

510 U.S. at 591 (quoting Folsom v. Marsh, 9 F. Cas. 342, 348 (No. 4,901) (C.C.D. Mass.

1841)); see also Leval, supra at 605 n.38. Moreover, Campbell explicitly

acknowledged that the defendants in that case “left themselves at . . . a

disadvantage when they failed to address the effect [of their work] on the market

for rap derivatives” and remanded for further fact-finding on the fourth factor

despite concluding that the defendants’ secondary work was transformative. 510

U.S. at 590, 594. Campbell therefore does not stand for the proposition that

transformative use should be the dispositive factor in the fair use inquiry; rather,

4
evidence of harm to the potential market for the original work (and its derivatives)

is still integral to the analysis.

By returning focus to the fourth fair use factor and being particularly

attentive to “whether unrestricted and widespread conduct of the sort engaged

in” by an alleged infringer would adversely affect the potential market for the

original work, id. at 590 (internal quotation marks omitted), courts can escape the

post-Campbell overreliance on transformative use. Fortunately, several of our

more recent fair use decisions have placed greater emphasis on the fourth factor.

See, e.g., Capitol Records, LLC v. ReDigi Inc., 910 F.3d 649, 662 (2d Cir. 2018)

(describing the fourth factor as “undoubtedly the single most important element

of fair use” (quoting Harper & Row, 471 U.S. at 566)); Fox News Network, LLC v.

TVEyes, Inc., 883 F.3d 169, 174, 180 (2d Cir. 2018) (finding no fair use despite

concluding that the defendants’ technology “serve[d] a transformative purpose,”

in part because the technology “usurped a function for which [the plaintiff was]

entitled to demand compensation under a licensing agreement”); TCA Television

Corp. v. McCollum, 839 F.3d 168, 186 (2d Cir. 2016) (recognizing that the district

court improperly “disregarded the possibility of defendants’ use adversely

affecting the licensing market for the [original work]”). And our sister circuits

5
have followed suit. See, e.g., Dr. Seuss Enters., L.P. v. ComicMix LLC, 983 F.3d 443,

459–61 (9th Cir. 2020) (emphasizing that the defendant did not “address a crucial

right for a copyright holder – the derivative works market”); Kienitz, 766 F.3d at

758 (“We think it best to stick with the statutory list, of which the most important

usually is the fourth (market effect).”).

This is not to suggest that the majority’s opinion runs counter to this trend.

To the contrary, it properly recognizes the harm to the potential licensing markets

for the Goldsmith Photograph and its derivatives, and reaffirms that the burden

of proving a lack of market harm rests with the alleged infringer. See Majority Op.

at 45–50. I write simply to stress that this renewed attention to the fourth fair use

factor will ultimately better serve the purposes of copyright, which remains at its

core “a commercial doctrine whose objective is to stimulate creativity among

potential authors by enabling them to earn money from their creations.” Google

Books, 804 F.3d at 223.

6
DENNIS JACOBS, Circuit Judge, concurring:

I concur in the opinion of the Court as well as Judge Sullivan’s

concurrence. I write briefly to make a single point: the holding does not

consider, let alone decide, whether the infringement encumbers the original

Prince Series works that are in the hands of collectors or museums.

It is very easy for opinions in this area (however expertly crafted) to have

undirected ramifications. A sound holding may suggest an unsound result in

related contexts.

The sixteen original works have been acquired by various galleries, art

dealers, and the Andy Warhol Museum. This case does not decide their rights to

use and dispose of those works because Goldsmith does not seek relief as to

them. She seeks only damages and royalties for licensed reproductions of the

Prince Series.

Although the Andy Warhol Foundation initiated this suit with a request

for broader declaratory relief that would cover the original works, Goldsmith did

not expressly join issue. The Declaratory Judgment Act is reserved for disputes

that are percolating over parties’ rights and obligations while harm threatens to

accrue. See United States v. Doherty, 786 F.2d 491, 498–99 (2d Cir. 1986)
(Friendly, J.); see also Broadview Chem. Corp. v. Loctite Corp., 417 F.2d 998, 1001

(2d Cir. 1969) (articulating the criteria for deciding whether to entertain a

declaratory judgment action). But Goldsmith does not claim that the original

works infringe and expresses no intention to encumber them; the opinion of the

Court does not necessarily decide that issue.

The issue, however, still looms, and our holding may alarm or alert

possessors of other artistic works. Warhol’s works are among many pieces that

incorporate, appropriate, or borrow from protected material. Risk of a copyright

suit or uncertainty about an artwork’s status can inhibit the creativity that is a

goal of copyright.

A key consideration in this case is the effect of the Prince Series on the

market for Goldsmith’s photograph. Our decision depends heavily on the

commercial competition between the photograph and the reproduced versions of

the Prince Series.

As the opinion observes, the market for the photograph and the market for

the original Prince Series works are distinct. See Majority Op. at 45. An original

work of art is marked by the hand or signature of the artist, which is a

preponderating factor in its value. When the work is reproduced, it loses that

2
mystique, as anyone who has browsed a gift shop can appreciate. In a word, the

original works and Goldsmith’s photograph are not “substitutes.” Castle Rock

Ent., Inc. v. Carol Publ’g Grp., 150 F.3d 132, 145 (2d Cir. 1998).

But when represented on a magazine cover, the Prince Series functions as a

portrait of the musician Prince--as does Goldsmith’s photograph. The Prince

Series retains the photograph’s expressive capacity for Prince portraiture and is

sought for that purpose. It may well compete for magazine covers, posters,

coffee mugs, and other media featuring the late musician. If the Foundation had

refuted the evidence of such market displacement, the weight of the analytical

considerations would have changed.