Last month, we saw the story of “Kind of Bloop”. Tech entrepreneur Andy Baio had put together a Chiptune tribute album of the classic Miles Davis work Kind of Blue. He secured licenses for the songs but failed to license the Chiptune version of the original album’s cover art that was used as the cover art for Kind of Bloop. Jay Maisel, the photographer and copyright holder of the iconic image of Davis, objected to Baio’s use, and the two settled the dispute without resorting to litigation.
Baio recounts his side of the story on his blog in a post titled “Kind of Screwed.” The story struck a chord in some areas of the blogosphere, spun as greedy-copyright-owner-shakes-down-poor-hobby-artist. The 80-year old Maisel was viciously attacked online in numerous places.
Jeremy Nicholl wrote two posts about this tale (first post and second post), describing the harassment and examining the facts more closely. The real story is a far cry from the version embraced by the online rabble and the bloggers who roused them.
Setting that aside, I want to take a look at the copyright issue underlying this story — the copyright issue that got so many so furious because they felt that, while Baio’s view was correct, the law gave Maisel an unfair upper hand.
See, the premise of Baio’s belief that he was “kind of screwed”, the premise that led some to push this story as “yet another in a long list of examples of how copyright is used to censor, rather than to promote progress”, and the premise that the online mob used to justify their harassment of an 80-year old photographer both online and off, was that the new cover art is clearly a fair use of the original work.
But it’s not.
If you record an 8-bit cover version of a Miles Davis song, you clearly need a license from the publisher, just as you need a license to record, say, a country version of a Beatles song. Baio knew this — he secured licenses for all the songs on the album before releasing it. The rules don’t suddenly change when you make an 8-bit version of a Miles Davis album cover.
To understand why, let’s take a look at Baio’s argument in more detail and see how well it holds up in reality. Baio says that “The crux of our disagreement hinges on the first factor — whether the Kind of Bloop illustration is ‘transformative'” so I’ll also talk primarily about that, but I will add a few comments on the other factors.
Transformative Purposes and Derivative Works
Baio’s argument is as follows:
In his influential paper on fair use, Judge Pierre N. Leval wrote, “Factor One is the soul of fair use.” Stanford’s Fair Use Center asks, “Has the material you have taken from the original work been transformed by adding new expression or meaning? Was value added to the original by creating new information, new aesthetics, new insights and understandings?”
From the beginning, Kind of Bloop was a creative experiment. I was drawn to the contradiction between the textured, subdued emotion in Kind of Blue and the cold, mechanical tones of retro videogame music. The challenge was to see whether chiptune artists could create something highly improvisational, warm, and beautiful from the limited palette of 1980s game consoles. (I think we succeeded.)
Similarly, the purpose of the album art was to engage both artist and viewer in the same exercise — can NES-style pixel art capture the artistic essence of the original album cover, with a fraction of the resolution and color depth of an analog photograph?
It reinforced the artistic themes of the project, to convey the feel of an entire album reimagined through an 8-bit lens. Far from being a copy, the cover art comments on it and uses the photo in new ways to send a new message.
This kind of transformation is the foundation of fair use. In a 2006 verdict, the court found artist Jeff Koons’ use of a fashion photo “adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message.”
I don’t think there’s any question that Kind of Bloop’s cover illustration does the same thing. Maisel disagreed.
Baio is incorrect for one simple reason:
“[C]opying that is complementary to the copyrighted work (in the sense that nails are complements of hammers) is fair use, but copying that is a substitute for the copyrighted work (in the sense that nails are substitutes for pegs or screws), or for derivative works from the copyrighted work, is not fair use.” (Emphasis added.) 1Ty, Inc. v. Publications Intern., 292 F.3d 512, 517 (7th Cir. 2002).
Copyright owners have, along with the exclusive right to copy and distribute their works, the exclusive right to prepare derivative works based on their own works. Most commonly, we think of derivative works such as the movie version of a book, or a foreign language translation of a book, but they can include any new work derived from an existing work.
So what distinguishes derivative works from fair uses of a work? Both can, after all, be considered “transformative” in some sense of the word.
The Second Circuit has noted that there exists “a potential source of confusion in our copyright jurisprudence over the use of the term ‘transformative.'” It explains:
A “derivative work,” over which a copyright owner has exclusive control, is defined as
a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted.
17 U.S.C. §§ 101, 106(2) (emphasis added). Although derivative works that are subject to the author’s copyright transform an original work into a new mode of presentation, such works — unlike works of fair use — take expression for purposes that are not “transformative.” 2Castle Rock Entertainment v. Carol Publishing Group, 150 F. 3d 132, 143 (1998).
If you want to understand what courts mean by “transformative purposes”, reading through the case law on the subject might not prove terribly enlightening. But in the last few years, three empirical studies of the fair use doctrine in case law have provided a greater depth to understanding its application — Barton Beebe’s An Empirical Study of US Copyright Fair Use Opinions, 1978-2005, 3156 University of Pennsylvania Law Review 549 (2008). Pamela Samuelson’s Unbundling Fair Use, 477 Fordham Law Review 2537 (2009). and Matthew Sag’s Predicting Fair Use. 5Unpublished, July 6, 2011.
In his new paper, Making Sense of Fair Use, Neil Netanel examines these studies and adds his own research to this area. He writes:
[C]ourts have made clear that matters for determining whether a use is transformative is whether the use is for a different purpose than that for which the copyrighted work was created. It can help if the defendant modifies or adds new expressive form or content as well, but different expressive purpose, not new expressive content, is almost always the key. Using Samuelson’s proffered nomenclature and policy clusters, courts as transformative primarily those uses that are “productive” or “orthogonal.” What Samuelson defines as “transformative” uses — uses that modify a preexisting work in creating a new one — are deemed “transformative” by the courts only if the defendant creates the new work for a different expressive purpose than the preexisting work.
Netanel lists some of these different expressive purposes that courts have deemed transformative:
[R]eplication of literary or graphic works to serve as an information tool; replication of artistic works to illustrate a biography; reproducing a fashion photograph originally made for a lifestyle magazine in a painting to make a comment about the mass media; copying and displaying a photographic portrait originally made as a gift item for the subject’s family and friends for purposes of entertainment and information; a football team’s display of artwork that the team previously used as its logo without the artist’s permission in a “museum-like setting” in the lobby of the team’s corporate headquarters; copying a work to criticize its author; and, of course, copying a work to parody or criticize the work.
The example of the fashion photograph is especially relevant since that is the same case as the one cited by Baio, the “2006 verdict” involving artist Jeff Koons. But his likening of his cover art to the Koons case misses the mark. When you actually read the case, you can see the court explain how Koons’ use of the fashion photograph was fair because it had a transformative purpose, and you can also see how Baio’s use lacked such a purpose. (I’ve left in some of the citations that shed further light on transformative purposes.)
Koons does not argue that his use was transformative solely because Blanch’s work is a photograph and his a painting, or because Blanch’s photograph is in a fashion magazine and his painting is displayed in museums. He would have been ill advised to do otherwise. We have declined to find a transformative use when the defendant has done no more than find a new way to exploit the creative virtues of the original work. See Davis, 246 F.3d at 174 (use of plaintiff’s eyewear in a clothing advertisement not transformative because it was “worn as eye jewelry in the manner it was made to be worn”); Castle Rock Entm’t, 150 F.3d at 142-43 (quiz book called the “Seinfeld Aptitude Test” not transformative when its purpose was “to repackage [the television show] Seinfeld to entertain Seinfeld viewers”); Ringgold v. Black Entm’t Television, Inc. 126 F.3d 70, 79 (2d Cir.1997) (copy of plaintiff’s painting used as decoration for a television program’s set not transformative because it was used for “the same decorative purpose” as the original).
But Koons asserts — and Blanch does not deny — that his purposes in using Blanch’s image are sharply different from Blanch’s goals in creating it. The sharply different objectives that Koons had in using, and Blanch had in creating, “Silk Sandals” confirms the transformative nature of the use. See Bill Graham Archives, 448 F.3d at 609 (finding transformative use when defendant’s purpose in using copyrighted concert poster was “plainly different from the original purpose for which they were created”).
Koons is, by his own undisputed description, using Blanch’s image as fodder for his commentary on the social and aesthetic consequences of mass media. His stated objective is thus not to repackage Blanch’s “Silk Sandals,” but to employ it “‘in the creation of new information, new aesthetics, new insights and understandings.'” When, as here, the copyrighted work is used as “raw material,” in the furtherance of distinct creative or communicative objectives, the use is transformative.
The test for whether “Niagara’s” use of “Silk Sandal” is “transformative,” then, is whether it “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.” The test almost perfectly describes Koons’s adaptation of “Silk Sandals”: the use of a fashion photograph created for publication in a glossy American “lifestyles” magazine — with changes of its colors, the background against which it is portrayed, the medium, the size of the objects pictured, the objects details and, crucially, their entirely different purpose and meaning — as part of a massive painting commissioned for exhibition in a German art-gallery space. We therefore conclude that the use in question was transformative. (Some citations removed.) 6Blanch v. Koons, 467 F.3d 244, 252-53 (2nd Cir. 2006).
Baio used Maisel’s photo from the cover of a Miles Davis album for the same purpose as the original — copying cover art to use as cover art. Aesthetics has no bearing on this lack of a transformative purpose; one could easily say the new cover is “art”, and one could certainly say the new cover has new expression or meaning. But since it serves the same purpose, it is a derivative work, and only Maisel can authorize its use.
Jeff Koons himself knows this, as he had lost in another case over a decade before the one above. Koons claimed fair use for a sculptural copy of a photo from professional photographer Art Rogers without Rogers’ permission, asserting that his sculpture conveyed an altogether different meaning than the photo, but the court didn’t agree:
If an infringement of copyrightable expression could be justified as fair use solely on the basis of the infringer’s claim to a higher or different artistic use — without insuring public awareness of the original work — there would be no practicable boundary to the fair use defense. Koons’ claim that his infringement of Rogers’ work is fair use solely because he is acting within an artistic tradition of commenting upon the commonplace thus cannot be accepted. … Of course, while our view of this matter does not necessarily prevent Koons’ expression, although it may, it does recognize that any such exploitation must at least entail “paying the customary price.” 7Rogers v. Koons, 960 F.2d 301, 310 (2nd Cir. 1992).
In the same way, Baio’s artistic motivations for his cover do not justify his copying of Maisel’s photo as fair use. It is instead clearly a derivative work, a “translation” of the original into 8-bit language, if you will — the equivalent of the 8-bit cover versions of Miles Davis’s music that accompany the work.
Protectable Elements of a Photograph
I also wanted to comment on Baio’s discussion of the third fair use factor: “the amount and substantiality of the portion used in relation to the copyrighted work as a whole.”
Baio first states, “With regard to the third factor, although the illustration does represent the cover of Kind of Blue, it does so at a dramatically reduced resolution that incorporates few of the photograph’s protectable elements.” One has to wonder based on this, what Baio thinks would constitute protectable elements of a photograph. (And, if he’s correct, he wouldn’t need a fair use defense at all — a work that isn’t substantially similar to another work isn’t infringing in the first place.)
The protectable elements of a photograph are those elements derived from the photographer’s creativity and expression, as opposed to factual or functional elements. Admittedly, no bright line exists between protectable and non-protectable elements, but courts have listed protectable elements as including things like subject matter and posing of subjects; layout, positioning, and arrangement; selection of angles, lighting, shading, and coloring. 8See e.g., Burrow-Giles Lithographic Co. v. Sarony, 111 US 53, 60 (1884); Ets-Hokin v. Skyy Spirits, 225 F.3d 1068, 1077 (9th Cir. 2000); Rogers v. Koons at 308; Eastern America Trio Products v. Tang Electronic, 97 F.Supp.2d 395, 417-18 (SDNY 2000).
Comparing the two images with these protectable elements in mind, it’s difficult to point to any protectable elements of the original photograph that weren’t copied. There’s certainly little argument that a substantial portion of them were copied — this is the key factor in infringement and fair use questions; it’s not a defense to show how much someone didn’t copy from the original work. 9Harper & Row v. Nation Enterprises, 471 US 539, 565 (1985), quoting Sheldon v. Metro-Goldwyn Pictures, 81 F.2d 49, 56 (2nd Cir. 1936).
One also has to wonder if Baio really believes that the illustration incorporates few of the photo’s protectable elements since he follows this statement immediately with his remark that “Courts routinely find fair use even where the entirety of an image is used.”
This remark is easily proven wrong. From Beebe’s study of fair uses:
Of the 99 opinions that addressed facts in which the defendant took the entirety of the plaintiff’s work, 27.3% found fair use (albeit with 9 of these 27 opinions finding a transformative use, and 4 finding a nontransformative use). The story is more extreme in situations where the court finds that the defendant did or did not take the “essence” or the “heart” of the plaintiff’s work. Courts explicitly found that the defendant took the heart of the plaintiff’s work in 37 opinions, and found no fair use in 35 of these. Courts made the opposite finding — that the defendant did not take the heart of the plaintiff’s work — in 25 opinions, and found fair use in 23 of these.
It is hardly “routine” when courts reject a finding of fair use when the entire work was used three out of four times.
Kind of Unfortunate
The mobbing of a celebrated photographer is inexcusable, but unfortunately it is becoming all too common for artists and creators to find themselves receiving abuse whenever they speak up for their rights or seek to protect their work online. It is much worse when, like here, the story used to encourage such behavior was so far from the truth.
In a broader sense, there is no broader sense to this event. The right exercised by Maisel to authorize works derived from his own work is much the same right a creator in 1909 had. Fair use is not disappearing — Netanel graphs defendant win rates in fair use cases in his paper; the graph bounces up and down at times but displays a remarkably constant trend since 1978 (when the doctrine was codified into the US code). Baio chose not to defend his use as fair in court because of the cost of litigating, but lawsuits are expensive for both parties.
|↑1||Ty, Inc. v. Publications Intern., 292 F.3d 512, 517 (7th Cir. 2002).|
|↑2||Castle Rock Entertainment v. Carol Publishing Group, 150 F. 3d 132, 143 (1998).|
|↑3||156 University of Pennsylvania Law Review 549 (2008).|
|↑4||77 Fordham Law Review 2537 (2009).|
|↑5||Unpublished, July 6, 2011.|
|↑6||Blanch v. Koons, 467 F.3d 244, 252-53 (2nd Cir. 2006).|
|↑7||Rogers v. Koons, 960 F.2d 301, 310 (2nd Cir. 1992).|
|↑8||See e.g., Burrow-Giles Lithographic Co. v. Sarony, 111 US 53, 60 (1884); Ets-Hokin v. Skyy Spirits, 225 F.3d 1068, 1077 (9th Cir. 2000); Rogers v. Koons at 308; Eastern America Trio Products v. Tang Electronic, 97 F.Supp.2d 395, 417-18 (SDNY 2000).|
|↑9||Harper & Row v. Nation Enterprises, 471 US 539, 565 (1985), quoting Sheldon v. Metro-Goldwyn Pictures, 81 F.2d 49, 56 (2nd Cir. 1936).|