By Jason Huang. Jason is a 1L at the Sandra Day O’Connor College of Law and is a McCarthy Fellow. He was born in Texas but grew up in Shanghai, China, where he lived until he graduated high school. He spent a gap year hiking across the United States, then enrolled at Thomas Aquinas College to study theology and graduated in 2020. After a brief stint in political campaigning, he enlisted in the United States Army National Guard as a combat engineer. During that time, he worked first as a construction worker and then as an emergency medical technician. Afterward, he worked in graphic design until deciding to enter law school.
I. Introduction
The Copyright Act, 17 U.S.C. § 106, protects copyright owners by granting them the exclusive right to reproduce, distribute, and display copies of their work.[1] However, Congress has created a fair use exemption to protect creativity.[2] The factors are:
“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.[3]“
Supposedly determined through a four-factor balancing test, fair use analysis has increasingly been dominated by just one subfactor: transformative use. Jiarui Liu’s 2019 paper, An Empirical Study of Transformative Use in Copyright Law,[4]chronicles this change and provides a framework for understanding the rise of transformative use. This essay uses Liu’s framework to predict the outcome of Hachette Book Group, Inc. v. Internet Archive based on the strength of the transformative use subfactor.[5]
II. Hachette Book Group, Inc. v. Internet Archive
Internet Archive is a non-profit digital library that scans print books (both copyrighted and non-copyrighted) and lends out digital copies. Before 2020, the non-profit maintained a one-to-one owned-to-lent books ratio. If all copies of an electronic book were lent out, prospective borrowers had to wait for the original borrowers to return them before checking those books out. When the 2020 pandemic occurred, however, Internet Archive removed this restriction and allowed up to 10,000 borrowers to check out the same book. In response, several publishers sued the organization in the lawsuit Hachette Book Group, Inc. v. Internet Archive.[6] Internet Archive relied on the Fair Use Clause of the Copyright Act as an affirmative defense for their actions. In 2024, however, the U.S. Court of Appeals for the Second Circuit ruled against Internet Archive. It held that scanning copyrighted books and distributing them to the public without permission from the copyright holders was not fair use, even if Internet Archive maintained a one-to-one owned-to-loan ratio between physical and digital books.[7]
While Liu’s paper was published five years before Hachette, it remains an accurate guide to predict and explain the case’s outcome. Though courts purportedly base their arguments on legal precedent, in reality, policy factors strongly determine decision outcomes.[8] Based on empirical data, Liu’s framework presents a useful guide for attorneys involved in fair use litigation to predict court decisions accurately.
III. Liu’s Analysis of Transformative Use
Nowhere is transformative use explicitly mentioned in fair use’s four-factor test. It was not elucidated until 1990, when Judge Leval coined it in his Harvard Law Review commentary, Toward a Fair Use Standard.[9] Even after that, transformative use was not widely used until 1994, when Justice Souter utilized it as a factor in deciding Campbell.[10] After Campbell, the proportion of transformative use analysis in fair use cases skyrocketed—in just one year, it jumped from 8% to 41% and has continued increasing ever since, hovering around 90% in recent years.[11]
The share of transformative use analysis has increased, and transformative use has become increasingly determinative. Among the 238 dispositive decisions in the sample Liu researched, 121 found transformative use—of these decisions, 94% found fair use.[12] Even though the U.S. Supreme Court asserted that the fourth factor, market effect, is “undoubtedly the single most important element of fair use,”[13] Liu found that a finding of transformative use correlated more to a finding of fair use.[14] This discrepancy has arisen because competing Supreme Court rulings find that 1) market harms did not necessarily weigh against the fourth factor, 2) derivative markets should be considered in addition to primary markets, and 3) transformative use counts against inferring an adverse market effect.[15] As a result, lower courts tend to find fair use where there was a finding for transformative use. This is unsurprising as the Court has held that “transformative works…lie at the heart of the fair use doctrine.”[16]
IV. Applying Lui to Hachette
From Justice Souter’s definition in Campbell, which says transformative use adds “something new, with a further purpose or different character,”[17] Liu teases out two general elements of transformation—purposive (“further purpose”) and physical (“different character”).[18] Physical transformation occurs when actors use copyrighted material in their works, such as including a copyrighted album cover in a mural.[19] In contrast, purposive transformation occurs when actors use copyrighted material differently than the copyright holders (such as Google scanning books to improve their search engine).[20] By examining these two elements, transformative use can weed out “garden-variety infringements that serve as a substitute in the primary market [because] an allegedly infringing use that makes neither physical nor purposive transformation is likely to compete directly with the original work.”[21] However, even if an infringing work is not a substitute in the primary market, “a court needs to determine whether it falls within a derivative market that the copyright owner is entitled to license.”[22]
The Internet Archive does not meet either the purposive or physical transformation elements in Liu’s framework. In Hachette, Internet Archive argued that scanning and lending books was transformative because it makes “1) lending more efficient” and enables “2) other uses not possible with print books and physical borrowing.”[23] Scanning and lending books do not meet the physical transformation element because it does not add to their character or creatively incorporate them into a larger, differing work. The purposive element is more difficult to discern due to its subjective nature. Liu suggests, however, that courts have implicitly used a two-part test to determine purposive transformation by examining 1) whether a copied work served the same purpose as the original and, if not, 2) whether a new audience was reached.[24] If either part is satisfied, a finding of purposive transformation usually follows.
In examining Hachette, both parts of Liu’s suggested test for purposive transformation are discernible—however, neither part is fulfilled for the Internet Archive. The organization copied and displayed the works of book publishers to entertain and inform the public. This fails the first part of Liu’s two-part test because Internet Archive’s offerings do not fulfill a different function than the publishers they copied.
If the Internet Archive could prove that their services reached a different audience, Liu’s research suggests that the court might have still found fair use. However, the Internet Archive could not prove this in Hachette. The Circuit Court found that the organization’s market audience was substantially the same as the one served by the publishers.[25] This fails the second part of Liu’s two-part test. Unsurprisingly, the Second Circuit Court of Appeals ruled against Internet Archive.
V. Conclusion
In the years since it was coined, transformative use has grown increasingly crucial in determining fair use cases. Though published in 2019, Liu’s An Empirical Study of Transformative Use in Copyright Law still provides a valuable framework for analyzing transformative use, and it proves remarkably prescient in explaining the Second Circuit Court’s ruling in Hachette. Internet Archive’s scanning and lending program could not qualify as transformative use because it failed to meet either the physical or purposive transformation elements that Liu identified. The service neither altered the character of the works nor served a fundamentally different purpose than the publishers’ service. Given the strong correlation Liu found between transformative use findings and overall fair use determinations, the Second Circuit Court’s ruling against the Internet Archive was consistent with the empirical patterns he identified. Perhaps the Internet Archive should have read Liu’s paper before they created the National Emergency Library.
[1] Mattel, Inc. v. Walking Mountain Prods., 353 F.3d 792 (9th Cir. 2003).
[2] Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569 (1994).
[3] 17 U.S.C. § 107 (2012).
[4] Jiarui Liu, An Empirical Study of Transformative Use in Copyright Law, 22 Stan. Tech. L. Rev. 163 (2019).
[5] Hachette Book Group, Inc. v. Internet Archive, 115 F.4th 163 (2d Cir. 2024).
[6] Aaron Schwabach, THE INTERNET ARCHIVE’S NATIONAL EMERGENCY LIBRARY: IS THERE AN EMERGENCY FAIR USE SUPERPOWER?, 18 Nw. J. Tech. & Intell. Prop. 187 (2021).
[7] Hachette, supra note 4, at 196.
[8] Liu, supra note 1, at 172.
[9] Pierre N. Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. 1105, 1111 (1990).
[10] Campbell, supra note2, at 579.
[11] Liu, supra note 1, at 174.
[12] Id. at 180; (Liu searched Westlaw for all reported transformative use decisions up until January 1st, 2017; he describes his search methodology in a footnote. See Liu, supra note 1, fn 47 at 173.)
[13] Fox News Network, LLC v. Tveyes, Inc., 883 F.3d 169 (2d Cir. 2018).
[14] Liu, supra note 1, at 198 (factor two, the nature of the copyrighted work, played almost no role in determining fair use).
[15] Id. at 199-200.
[16] Harper & Row Publishers, Inc. v Nation Enterprises, 471 U.S. 539 (1985).
[17] Campbell, supra note 2, at 579.
[18] Liu, supra note 1, at 204.
[19] Id. at 206.
[20] Id. at 207.
[21] Id. at 171.
[22] Id.
[23] Hachette, supra note 4, at 182.
[24] Liu, supra note 1, at 209.
[25] Id. at 189.