holding that universities' systematic digitization of copyrighted books was transformative because the digital copies were used to create a searchable database which supplied users with lists of page numbers and not with copies of the original works, and so the copies served a different purpose than the original worksSummary of this case from Cambridge Univ. Press v. Patton
Edward H. Rosenthal (Jeremy S. Goldman, Anna Kadyshevich, on the brief), Frankfurt Kurnit Klein & Selz, P.C., New York, NY, for Plaintiffs–Appellants. Joseph Petersen (Robert Potter, Joseph Beck, Andrew Pequignot, Allison Scott Roach, on the brief), Kilpatrick Townsend & Stockton LLP, New York, NY, for Defendants–Appellees.
Edward H. Rosenthal (Jeremy S. Goldman, Anna Kadyshevich, on the brief), Frankfurt Kurnit Klein & Selz, P.C., New York, NY, for Plaintiffs–Appellants. Joseph Petersen (Robert Potter, Joseph Beck, Andrew Pequignot, Allison Scott Roach, on the brief), Kilpatrick Townsend & Stockton LLP, New York, NY, for Defendants–Appellees.
Daniel F. Goldstein (Jessica P. Weber, on the brief), Brown Goldstein & Levy, LLP, Baltimore, MD; Robert J. Bernstein, New York, NY, on the brief; Peter Jaszi, Chevy Chase, MD, on the brief, for Intervenor Defendants–Appellees.
Jennifer M. Urban, Pamela Samuelson, David Hansen, Samuelson Law, Technology & Public Policy Clinic, University of California, Berkeley, School of Law, Berkeley, CA, for Amici Curiae 133 Academic Authors.
Blake E. Reid, Brian Wolfman, Institute for Public Representation, Georgetown University Law Center, Washington, DC, for Amicus Curiae American Association of People with Disabilities.
Jonathan Band, Jonathan Band PLLC, Washington, DC, for Amicus Curiae American Library Association.
David Leichtman, Hillel I. Parness, Shane D. St. Hill, Robins, Kaplan, Miller & Ciresi L.L.P., New York, NY, for Amicus Curiae American Society of Journalists and Authors, Inc.
Brian G. Joseph, Karyn K. Ablin, Wiley Rein LLP; Ada Meloy, General Counsel, American Council on Education, Washington, DC, for Amici Curiae American Council on Education, Association of American Universities, et al.
Elizabeth A. McNamara, Alison B. Schary, Colin J. Peng–Sue, Davis Wright Tremaine LLP, New York, NY, for Amicus Curiae the Associated Press.
Mary E. Rasenberger, Nancy E. Wolff, Eleanor M. Lackman, Nicholas J. Tardif, Cowan DeBaets, Abrahams & Sheppard LLP, New York, NY, for Amicus Curiae Association of American Publishers.
Jo Anne Simon, Mary J. Goodwin, Amy F. Robertson, Jo Anne Simon, P.C., Brooklyn, NY, for Amici Curiae Association on Higher Education and Disability, Marilyn J. Bartlett, et al.
Brandon Butler, Washington, DC, for Amici Curiae Beneficent Technology, Inc., and Learning Ally, Inc.
Susan M. Kornfield, Bodman PLC, Ann Arbor, MI, for Amici Curiae Board of Trustees of the University of Illinois, Board of Trustees of Michigan State University, et al.
Jason Schultz, Berkeley, CA; Matthew Sag, Chicago, IL, for Amici Curiae Digital Humanities and Law Scholars.
Michael Waterstone, Los Angeles, CA; Robert Dinerstein, Washington, DC; Christopher H. Knauf, Knauf Associates, Santa Monica, CA; Michael Stein, Cambridge, MA, for Amici Curiae Disability Law Professors.
Roderick M. Thompson, Stephanie P. Skaff, Deepak Gupta, Rochelle L. Woods, Farella Braun & Martel LLP, San Francisco, CA; Corynne McSherry, Daniel Nazer, Electronic Frontier Foundation, San Francisco, CA; John Bergmayer, Public Knowledge, Washington, DC; David Sohn, Center for Democracy & Technology, Washington, DC, for Amicus Curiae Electronic Frontier Foundation.
Stephen M. Schaetzel, Meunier Carlin & Curfman, LLC, Atlanta, GA, for Amicus Curiae Emory Vaccine Center.
Frederick A. Brodie, Pillsbury Winthrop Shaw Pittman LLP, New York, NY, for Amicus Curiae the Leland Stanford Junior University.
Eric J. Grannis, The Law Offices of Eric J. Grannis, New York, NY, for Amici Curiae Medical Historians.
Steven B. Fabrizio, Kenneth L. Doroshow, Steven R. Englund, Jenner & Block LLP, Washington, DC, for Amicus Curiae Motion Picture Association of America, Inc.
Before: WALKER, CABRANES, and PARKER, Circuit Judges.
BARRINGTON D. PARKER, Circuit Judge:
Beginning in 2004, several research universities including the University of Michigan, the University of California at Berkeley, Cornell University, and the University of Indiana agreed to allow Google to electronically scan the books in their collections. In October 2008, thirteen universities announced plans to create a repository for the digital copies and founded an organization called HathiTrust to set up and operate the HathiTrust Digital Library (or “HDL”). Colleges, universities, and other nonprofit institutions became members of HathiTrust and made the books in their collections available for inclusion in the HDL. HathiTrust currently has 80 member institutions and the HDL contains digital copies of more than ten million works, published over many centuries, written in a multitude of languages, covering almost every subject imaginable. This appeal requires us to decide whether the HDL's use of copyrighted material is protected against a claim of copyright infringement under the doctrine of fair use. See17 U.S.C. § 107.
A. The HathiTrust Digital Library
HathiTrust permits three uses of the copyrighted works in the HDL repository. First, HathiTrust allows the general public to search for particular terms across all digital copies in the repository. Unless the copyright holder authorizes broader use, the search results show only the page numbers on which the search term is found within the work and the number of times the term appears on each page. The HDL does not display to the user any text from the underlying copyrighted work (either in “snippet” form or otherwise). Consequently, the user is not able to view either the page on which the term appears or any other portion of the book.
Below is an example of the results a user might see after running an HDL full-text search:
J.A. 681 ¶ 80 (Wilkin Decl.).
Second, the HDL allows member libraries to provide patrons with certified print disabilities access to the full text of copyrighted works. A “print disability” is any disability that prevents a person from effectively reading printed material. Blindness is one example, but print disabilities also include those that prevent a person from physically holding a book or turning pages. To use this service, a patron must obtain certification of his disability from a qualified expert. Through the HDL, a print-disabled user can obtain access to the contents of works in the digital library using adaptive technologies such as software that converts the text into spoken words, or that magnifies the text. Currently, the University of Michigan's library is the only HDL member that permits such access, although other member libraries intend to provide it in the future.
Third, by preserving the copyrighted books in digital form, the HDL permits members to create a replacement copy of the work, if the member already owned an original copy, the member's original copy is lost, destroyed, or stolen, and a replacement copy is unobtainable at a “fair” price elsewhere.
The HDL stores digital copies of the works in four different locations. One copy is stored on its primary server in Michigan, one on its secondary server in Indiana, and two on separate backup tapes at the University of Michigan. Each copy contains the full text of the work, in a machine readable format, as well as the images of each page in the work as they appear in the print version.
Separate from the HDL, one copy is also kept by Google. Google's use of its copy is the subject of a separate lawsuit currently pending in this Court. See Authors Guild, Inc. v. Google, Inc., 721 F.3d 132 (2d Cir.2013), on remand, 954 F.Supp.2d 282 (S.D.N.Y.2013), appeal docketed, No. 13–4829 (2d Cir. Dec. 23, 2013).
B. The Orphan Works Project
Separate and apart from the HDL, in May 2011, the University of Michigan developed a project known as the Orphan Works Project (or “OWP”). An “orphan work” is an out-of-print work that is still in copyright, but whose copyright holder cannot be readily identified or located. See U.S. Copyright Office, Notice of Inquiry, Orphan Works and Mass Digitization, 77 Fed.Reg. 64555 (Oct. 22, 2012).
The University of Michigan conceived of the OWP in two stages: First, the project would attempt to identify out-of-print works, try to find their copyright holders, and, if no copyright holder could be found, publish a list of orphan works candidates to enable the copyright holders to come forward or be otherwise located. If no copyright holder came forward, the work was to be designated as an orphan work. Second, those works identified as orphan works would be made accessible in digital format to the OWP's library patrons (with simultaneous viewers limited to the number of hard copies owned by the library).
The University evidently became concerned that its screening process was not adequately distinguishing between orphan works (which were to be included in the OWP) and in-print works (which were not). As a result, before the OWP was brought online, but after the complaint was filed in this case, the University indefinitely suspended the project. No copyrighted work has been distributed or displayed through the project and it remains suspended as of this writing.
C. Proceedings in the District Court
This case began when twenty authors and authors' associations (collectively, the “Authors”) sued HathiTrust, one of its member universities, and the presidents of four other member universities (collectively, the “Libraries”) for copyright infringement seeking declaratory and injunctive relief. The National Federation of the Blind and three print-disabled students (the “Intervenors”) were permitted to intervene to defend their ability to continue using the HDL.
The Libraries initially moved for partial judgment on the pleadings on the ground that the authors' associations lacked standing to assert claims on behalf of their members and that the claims related to the OWP were not ripe. SeeFed.R.Civ.P. 12(c). The Libraries then moved for summary judgment on the remaining claims on the ground that their uses of copyrighted material were protected by the doctrine of fair use, see17 U.S.C. § 107, and also by the Chafee Amendment, see id. § 121. The Intervenors moved for summary judgment on substantially the same grounds as the Libraries and, finally, the Authors cross-moved for summary judgment.
D. The District Court's Opinion
The district court granted the Libraries' and Intervenors' motions for summary judgment on the infringement claims on the basis that the three uses permitted by the HDL were fair uses. In this assessment, the district court gave considerable weight to what it found to be the “transformative” nature of the three uses and to what it described as the HDL's “invaluable” contribution to the advancement of knowledge, Authors Guild, Inc. v. HathiTrust, 902 F.Supp.2d 445, 460–64 (S.D.N.Y.2012). The district court explained:
Although I recognize that the facts here may on some levels be without precedent, I am convinced that they fall safely within the protection of fair use such that there is no genuine issue of material fact. I cannot imagine a definition of fair use that would not encompass the transformative uses made by [the HDL] and would require that I terminate this invaluable contribution to the progress of science and cultivation of the arts that at the same time effectuates the ideals espoused by the [Americans With Disabilities Act of 1990, Pub.L. No. 101–336, 104 Stat. 327 (codified as amended at 42 U.S.C. §§ 12101, et seq.)].
Id. at 464.
Next, the district court addressed the Libraries' Chafee Amendment defense. Under the Amendment, “authorized entities” are permitted to reproduce or distribute copies of a previously published, nondramatic literary work in specialized formats exclusively for use by the blind or other persons with disabilities. See17 U.S.C. § 121; HathiTrust, 902 F.Supp.2d at 465. Under § 121, an “ ‘authorized entity’ means a nonprofit organization or a governmental agency that has a primary mission to provide specialized services relating to training, education, or adaptive reading or information access needs of blind or other persons with disabilities.” 17 U.S.C. § 121(d)(1).
The district court stated that the ADA requires that libraries of educational institutions, such as the Libraries in this case, “have a primary mission to reproduce and distribute their collections to print-disabled individuals,” which, according to Judge Baer, made “each library a potential ‘authorized entity’ under the Chafee Amendment.” HathiTrust, 902 F.Supp.2d at 465. As a result, the district court concluded that “[t]he provision of access to previously published non-dramatic literary works within the HDL fits squarely within the Chafee Amendment, although Defendants may certainly rely on fair use ... to justify copies made outside of these categories or in the event that they are not authorized entities.” Id.
The district court held that certain associational plaintiffs lacked standing under the Copyright Act and dismissed them from the suit. Id. at 450–55. The district court also held that the OWP claims were unripe for judicial review in the absence of crucial information about what the program would look like and whom it would affect should it be implemented, and because the Authors would suffer no hardship by deferring litigation until such time as the Libraries released the details of a new OWP and a revised list of orphan work candidates. Id. at 455–56. The court entered judgment against the Authors, and this appeal followed.
We review de novo under well-established standards the district court's decisionsgranting summary judgment and judgment on the pleadings. See Maraschiello v. City of Buffalo Police Dep't, 709 F.3d 87, 92 (2d Cir.2013) (summary judgment); LaFaro v. N.Y. Cardiothoracic Grp., PLLC, 570 F.3d 471, 475 (2d Cir.2009) (judgment on the pleadings).
As a threshold matter, we consider whether the authors' associations have standing to assert infringement claims on behalf of their members.
Three of these authors' associations—Authors Guild, Inc., Australian Society of Authors Limited, and Writers' Union of Canada—claim to have standing, solely as a matter of U.S. law, to seek an injunction for copyright infringement on their members' behalf. But, as we have previously explained, § 501 of “the Copyright Act does not permit copyright holders to choose third parties to bring suits on their behalf.” ABKCO Music, Inc. v. Harrisongs Music, Ltd., 944 F.2d 971, 980 (2d Cir.1991); see also Itar–Tass Russian News Agency v. Russian Kurier, Inc., 153 F.3d 82, 92 (2d Cir.1998) ( “United States law permits suit only by owners of ‘an exclusive right under a copyright’....” (quoting 17 U.S.C. § 501(b))). Accordingly, we agree with the district court that these associations lack standing to bring suit on behalf of their members, and they were properly dismissed from the suit.
The remaining four authors' associations—Union des Écrivaines et des Écrivains Québécois, Authors' Licensing and Collecting Society, Sveriges Författarförbund, and Norsk faglitteraer forfattero og oversetterforening—assert that foreign law confers upon them certain exclusive rights to enforce the copyrights of their foreign members (an assertion that the Libraries do not contest on this appeal). These four associations do have standing to bring suit on behalf of their members. See Itar–Tass, 153 F.3d at 93–94 (recognizing that an association authorized by foreign law to administer its foreign members' copyrights has standing to seek injunctive relief on behalf of those members in U.S. court).
Plaintiffs argue that the fair use defense is inapplicable to the activities at issue here, because the Copyright Act includes another section, 108, which governs “Reproduction [of copyrighted works] by Libraries ...” 17 U.S.C. § 108. However, section 108 also includes a “savings clause,” which states, “Nothing in this section in any way affects the right of fair use as provided by section 107....” § 108(f)(4). Thus, we do not construe § 108 as foreclosing our analysis of the Libraries' activities under fair use, and we proceed with that analysis.
As the Supreme Court has explained, the overriding purpose of copyright is “ ‘[t]o promote the Progress of Science and useful Arts....' ” Campbell v. Acuff–Rose Music, Inc., 510 U.S. 569, 574, 114 S.Ct. 1164, 127 L.Ed.2d 500 (1994) (quoting U.S. Const. art. I, § 8, cl. 8); see also Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 156, 95 S.Ct. 2040, 45 L.Ed.2d 84 (1975). This goal has animated copyright law in Anglo–American history, beginning with the first copyright statute, the Statute of Anne of 1709, which declared itself to be “[a]n Act for the Encouragement of Learning, by Vesting the Copies of Printed Books in the Authors ... during the Times therein mentioned.” Act for the Encouragement of Learning, 8 Anne, ch. 19. In short, our law recognizes that copyright is “not an inevitable, divine, or natural right that confers on authors the absolute ownership of their creations. It is designed rather to stimulate activity and progress in the arts for the intellectual enrichment of the public.” Pierre N. Leval, Toward a Fair Use Standard, 103 Harv. L.Rev.. 1105,1107 (1990).
The Copyright Act furthers this core purpose by granting authors a limited monopoly over (and thus the opportunity to profit from) the dissemination of their original works of authorship. See17 U.S.C. §§ 102, 106, 302–305. The Copyright Act confers upon authors certain enumerated exclusive rights over their works during the term of the copyright, including the rights to reproduce the copyrighted work and to distribute those copies to the public. Id. § 106(1), (3). The Act also gives authors the exclusive right to prepare certain new works—called “derivative works”—that are based upon the copyrighted work. Id. § 106(2). Paradigmatic examples of derivative works include the translation of a novel into another language, the adaptation of a novel into a movie or a play, or the recasting of a novel as an e-book or an audiobook. See id. § 101. As a general rule, for works created after January 1, 1978, copyright protection lasts for the life of the author plus an additional 70 years. Id. § 302.
At the same time, there are important limits to an author's rights to control original and derivative works. One such limit is the doctrine of “fair use,” which allows the public to draw upon copyrighted materials without the permission of the copyright holder in certain circumstances. See id. § 107 (“[T]he fair use of a copyrighted work ... is not an infringement of copyright.”). “From the infancy of copyright protection, some opportunity for fair use of copyrighted materials has been thought necessary to fulfill copyright's very purpose, ‘[t]o promote the Progress of Science and useful Arts....' ” Campbell, 510 U.S. at 574, 114 S.Ct. 1164.
Under the fair-use doctrine, a book reviewer may, for example, quote from an original work in order to illustrate a point and substantiate criticisms, see Folsom v. Marsh, 9 F. Cas. 342, 344 (C.C.D.Mass.1841) (No. 4901), and a biographer may quote from unpublished journals and letters for similar purposes, see Wright v. Warner Books, Inc., 953 F.2d 731 (2d Cir.1991). An artist may employ copyrighted photographs in a new work that uses a fundamentally different artistic approach, aesthetic, and character from the original. See Cariou v. Prince, 714 F.3d 694, 706 (2d Cir.2013). An internet search engine can display low-resolution versions of copyrighted images in order to direct the user to the website where the original could be found. See Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146, 1165 (9th Cir.2007); Kelly v. Arriba Soft Corp., 336 F.3d 811, 818–22 (9th Cir.2003). A newspaper can publish a copyrighted photograph 18 (taken for a modeling portfolio) in order to inform and entertain the newspaper's readership about a news story. See Nunez v. Caribbean Int'l News Corp., 235 F.3d 18, 25 (1st Cir.2000). A viewer can create a recording of a broadcast television show in order to view it at a later time. See Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 447–450, 104 S.Ct. 774, 78 L.Ed.2d 574 (1984). And a competitor may create copies of copyrighted software for the purpose of analyzing that software and discovering how it functions (a process called “reverse engineering”). See Sony Comp. Entertainment, Inc. v. Connectix Corp., 203 F.3d 596, 599–601 (9th Cir.2000).
The doctrine is generally subject to an important proviso: A fair use must not excessively damage the market for the original by providing the public with a substitute for that original work. Thus, a book review may fairly quote a copyrighted book “for the purposes of fair and reasonable criticism,” Folsom, 9 F. Cas. at 344, but the review may not quote extensively from the “heart” of a forthcoming memoir in a manner that usurps the right of first publication and serves as a substitute for purchasing the memoir, Harper & Row, Publishers, Inc. v. Nation Enters., 471 U.S. 539, 105 S.Ct. 2218, 85 L.Ed.2d 588 (1985).
In 1976, as part of a wholesale revision of the Copyright Act, Congress codified the judicially created fair-use doctrine at 17 U.S.C. § 107. See Copyright Act of 1976, Pub.L. No. 94–553, § 107, 90 Stat. 2541, 2546 (1976) (codified as amended at 17 U.S.C. § 107). Section 107 requires a court to consider four nonexclusive factors which are to be weighed together to assess whether a particular use is fair:
(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.
An important focus of the first factor is whether the use is “transformative.” A use is transformative if it does something more than repackage or republish the original copyrighted work. The inquiry is whether the work “adds something new, with a further purpose or different character, altering the first with new expression, meaning or message....” Campbell, 510 U.S. at 579, 114 S.Ct. 1164 (citing Leval, 103 Harv. L. RevV. at 1111). “[T]he more transformative the new work, the less will be the significance of other factors ... that may weigh against a finding of fair use.” Id. Contrary to what the district court implied, a use does not become transformative by making an “invaluable contribution to the progress of science and cultivation of the arts.” HathiTrust, 902 F.Supp.2d at 464. Added value or utility is not the test: a transformative work is one that serves a new and different function from the original work and is not a substitute for it.
The second factor considers whether the copyrighted work is “of the creative or instructive type that the copyright laws value and seek to foster.” Leval, 103 Harv. L.Rev.. at 1117; see also Folsom, 9 F. Cas. at 348 (“[W]e must often ... look to the nature and objects of the selections made....”). For example, the law of fair use “recognizes a greater need to disseminate factual works than works of fiction or fantasy.” Harper & Row, 471 U.S. at 563, 105 S.Ct. 2218.
The third factor asks whether the secondary use employs more of the copyrighted work than is necessary, and whether the copying was excessive in relation to any valid purposes asserted under the first factor. Campbell, 510 U.S. at 586–87, 114 S.Ct. 1164. In weighing this factor, we assess the quantity and value of the materials used and whether the amount copied is reasonable in relation to the purported justifications for the use under the first factor. Leval, 103 Harv. L.Rev.. at 1123.
Finally, the fourth factor requires us to assess the impact of the use on the traditional market for the copyrighted work. This is the “single most important element of fair use.” Harper & Row, 471 U.S. at 566, 105 S.Ct. 2218. To defeat a claim of fair use, the copyright holder must point to market harm that results because the secondary use serves as a substitute for the original work. See Campbell, 510 U.S. at 591, 114 S.Ct. 1164 (“cognizable market harm” is limited to “market substitution”); see also NXIVM Corp. v. Ross Inst., 364 F.3d 471, 481–82 (2d Cir.2004).
As discussed above, the Libraries permit three uses of the digital copies deposited in the HDL. We now consider whether these uses are “fair” within the meaning of our copyright law.
1. Full–Text Search
It is not disputed that, in order to perform a full-text search of books, the Libraries must first create digital copies of the entire books. Importantly, as we have seen, the HDL does not allow users to view any portion of the books they are searching. Consequently, in providing this service, the HDL does not add into circulation any new, human-readable copies of any books. Instead, the HDL simply permits users to “word search”—that is, to locate where specific words or phrases appear in the digitized books. Applying the relevant factors, we conclude that this use is a fair use.
Turning to the first factor, we conclude that the creation of a full-text searchable database is a quintessentially transformative use. As the example on page 7, supra, demonstrates, the result of a word search is different in purpose, character, expression, meaning, and message from the page (and the book) from which it is drawn. Indeed, we can discern little or no resemblance between the original text and the results of the HDL full-text search.
There is no evidence that the Authors write with the purpose of enabling text searches of their books. Consequently, the full-text search function does not “supersede[ ] the objects [or purposes] of the original creation,” Campbell, 510 U.S. at 579, 114 S.Ct. 1164 (internal quotation marks omitted). The HDL does not “merely repackage[ ] or republish[ ] the original[s],” Leval, 103 Harv. L.Rev.. at 1111, or merely recast “an original work into a new mode of presentation,” Castle Rock Entm't, Inc. v. Carol Publ'g Grp., Inc., 150 F.3d 132, 143 (2d Cir.1998). Instead, by enabling full-text search, the HDL adds to the original something new with a different purpose and a different character.
Full-text search adds a great deal more to the copyrighted works at issue than did the transformative uses we approved in several other cases. For example, in Cariou v. Prince, we found that certain photograph collages were transformative, even though the collages were cast in the same medium as the copyrighted photographs. 714 F.3d at 706. Similarly, in Bill Graham Archives v. Dorling Kindersley Ltd., we held that it was a transformative use to include in a biography copyrighted concert photos, even though the photos were unaltered (except for being reduced in size). 448 F.3d 605, 609–11 (2d Cir.2006); see also Blanch v. Koons, 467 F.3d 244, 252–53 (2d Cir.2006) (transformative use of copyrighted photographs in collage painting); Leibovitz v. Paramount Pictures Corp., 137 F.3d 109, 114 (2d Cir.1998) (transformative use of copyrighted photograph in advertisement).
Cases from other Circuits reinforce this conclusion. In Perfect 10, Inc., the Ninth Circuit held that the use of copyrighted thumbnail images in internet search results was transformative because the thumbnail copies served a different function from the original copyrighted images. 508 F.3d at 1165; accord Arriba Soft Corp., 336 F.3d at 819. And in A.V. ex rel. Vanderhye v. iParadigms, LLC, a company created electronic copies of unaltered student papers for use in connection with a computer program that detects plagiarism. Even though the electronic copies made no “substantive alteration to” the copyrighted student essays, the Fourth Circuit held that plagiarism detection constituted a transformative use of the copyrighted works. 562 F.3d 630, 639–40.
The second fair-use factor—the nature of the copyrighted work—is not dispositive. The HDL permits the full-text search of every type of work imaginable. Consequently, there is no dispute that the works at issue are of the type that the copyright laws value and seek to protect. However, “this factor ‘may be of limited usefulness where,’ as here, ‘the creative work ... is being used for a transformative purpose.’ ” Cariou, 714 F.3d at 710 (quoting Bill Graham Archives, 448 F.3d at 612). Accordingly, our fair-use analysis hinges on the other three factors.
The third factor asks whether the copying used more of the copyrighted work than necessary and whether the copying was excessive. As we have noted, “[t]here are no absolute rules as to how much of a copyrighted work may be copied and still be considered a fair use.” Maxtone–Graham v. Burtchaell, 803 F.2d 1253, 1263 (2d Cir.1986). “[T]he extent of permissible copying varies with the purpose and character of the use.” Campbell, 510 U.S. at 586–87, 114 S.Ct. 1164. The crux of the inquiry is whether “no more was taken than necessary.” Id. at 589, 114 S.Ct. 1164. For some purposes, it may be necessary to copy the entire copyrighted work, in which case Factor Three does not weigh against a finding of fair use. See Bill Graham Archives, 448 F.3d at 613 (entire image copied); Arriba Soft, 336 F.3d at 821 (“If Arriba only copied part of the image, it would be more difficult to identify it, thereby reducing the usefulness of the visual search engine.”).
In order to enable the full-text search function, the Libraries, as we have seen, created digital copies of all the books in their collections. Because it was reasonably necessary for the HDL to make use of the entirety of the works in order to enable the full-text search function, we do not believe the copying was excessive.
The HDL also creates digital copies of the images of each page of the books. As the Libraries acknowledge, the HDL does not need to retain these copies to enable the full-text search use. We discuss the fair-use justification for these copies in the context of the disability-access use, see infra pp. 102–03.
The Authors also contend that the copying is excessive because the HDL creates and maintains copies of the works at four different locations Appellants' Br. 27–28. But the record demonstrates that these copies are also reasonably necessary in order to facilitate the HDL's legitimate uses. In particular, the HDL's services are offered to patrons through two servers, one at the University of Michigan (the primary server) and an identical one at the University of Indiana (the “mirror” server). Both servers contain copies of the digital works at issue. According to the HDL executive director, the “existence of a[n] [identical] mirror site allows for balancing the load of user web traffic to avoid overburdening a single site, and each site acts as a back-up of the HDL collection in the event that one site were to cease operation (for example, due to failure caused by a disaster, or even as a result of routine maintenance).” J.A. 682–83 ¶ 88–89 (Wilkin Decl.). To further guard against the risk of data loss, the HDL stores copies of the works on two encrypted backup tapes, which are disconnected from the internet and are placed in separate secure locations on the University of Michigan campus. Id. at 683 ¶ 90. The HDL creates these backup tapes so that the data could be restored in “the event of a disaster causing large-scale data loss” to the primary and mirror servers. Id.
We have no reason to think that these copies are excessive or unreasonable in relation to the purposes identified by the Libraries and permitted by the law of copyright. In sum, even viewing the evidence in the light most favorable to the Authors, the record demonstrates that these copies are reasonably necessary to facilitate the services HDL provides to the public and to mitigate the risk of disaster or data loss. Accordingly, we conclude that this factor favors the Libraries.
The fourth factor requires us to consider “the effect of the use upon the potential market for or value of the copyrighted work,” 17 U.S.C. § 107(4), and, in particular, whether the secondary use “usurps the market of the original work,” NXIVM Corp., 364 F.3d at 482.
The Libraries contend that the full-text-search use poses no harm to any existing or potential traditional market and point to the fact that, in discovery, the Authors admitted that they were unable to identify “any specific, quantifiable past harm, or any documents relating to any such past harm,” resulting from any of the Libraries' uses of their works (including full-text search). Defs.-Appellees' Br. 38 (citing Pls.' Resps. to Interrogs.). The district court agreed with this contention, as do we.
At the outset, it is important to recall that the Factor Four analysis is concerned with only one type of economic injury to a copyright holder: the harm that results because the secondary use serves as a substitute for the original work. See Campbell, 510 U.S. at 591, 114 S.Ct. 1164 (“cognizable market harm” is limited to “market substitution”). In other words, under Factor Four, any economic “harm” caused by transformative uses does not count because such uses, by definition, do not serve as substitutes for the original work. See Bill Graham Archives, 448 F.3d at 614.
To illustrate why this is so, consider how copyright law treats book reviews. Book reviews often contain quotations of copyrighted material to illustrate the reviewer's points and substantiate his criticisms; this is a paradigmatic fair use. And a negative book review can cause a degree of economic injury to the author by dissuading readers from purchasing copies of her book, even when the review does not serve as a substitute for the original. But, obviously, in that case, the author has no cause for complaint under Factor Four: The only market harms that count are the ones that are caused because the secondary use serves as a substitute for the original, not when the secondary use is transformative (as in quotations in a book review). See Campbell, 510 U.S. at 591–92, 114 S.Ct. 1164 (“[W]hen a lethal parody, like a scathing theater review, kills demand for the original, it does not produce a harm cognizable under the Copyright Act.”).
The Authors assert two reasons why the full-text-search function harms their traditional markets. The first is a “lost sale” theory which posits that a market for licensing books for digital search could possibly develop in the future, and the HDL impairs the emergence of such a market because it allows patrons to search books without any need for a license. Thus, according to the Authors, every copy employed by the HDL in generating full-text searches represents a lost opportunity to license the book for search. Appellants' Br. 43.
This theory of market harm does not work under Factor Four, because the full-text search function does not serve as a substitute for the books that are being searched. See Campbell, 510 U.S. at 591–92, 114 S.Ct. 1164; Bill Graham Archives, 448 F.3d at 614. Thus, it is irrelevant that the Libraries might be willing to purchase licenses in order to engage in this transformative use (if the use were deemed unfair). Lost licensing revenue counts under Factor Four only when the use serves as a substitute for the original and the full-text-search use does not.
Next, the Authors assert that the HDL creates the risk of a security breach which might impose irreparable damage on the Authors and their works. In particular, the Authors speculate that, if hackers were able to obtain unauthorized access to the books stored at the HDL, the full text of these tens of millions of books might be distributed worldwide without restriction, “decimat[ing]” the traditional market for those works. Appellants' Br. 40.
The record before us documents the extensive security measures the Libraries have undertaken to safeguard against the risk of a data breach. Some of those measures were described by the HDL executive director as follows:
First, [HDL] maintains ... rigorous physical security controls. HDL servers, storage, and networking equipment at Michigan and Indiana University are mounted in locked racks, and only six individuals at Michigan and three at Indiana University have keys. The data centers housing HDL servers, storage, and networking equipment at each site location are monitored by video surveillance, and entry requires use of both a keycard and a biometric sensor.
Second, network access to the HDL corpus is highly restricted, even for the staff of the data centers housing HDL equipment at Michigan and Indiana University. For example, two levels of network firewalls are in place at each site, and Indiana University data center staff do not have network access to the HDL corpus, only access to the physical equipment. For the backup tapes, network access is limited to the administrators of the backup system, and these individuals are not provided the encryption key that would be required to access the encrypted files on the backup tapes.
Web access to the HDL corpus is also highly restricted. Access by users of the HDL service is governed by primarily by [ sic] the HDL rights database, which classifies each work by presumed copyright status, and also by a user's authentication to the system (e.g., as an individual certified to have a print disability by Michigan's Office of Services for Students with Disabilities).
Even where we do permit a work to be read online, such as a work in the public domain, we make efforts to ensure that inappropriate levels of access do not take place. For example, a mass download prevention system called “choke” is used to measure the rate of activity (such as the rate a user is reading pages) by each individual user. If a user's rate of activity exceeds certain thresholds, the system assumes that the user is mechanized (e.g., a web robot) and blocks that user's access for a set period of time.
J.A. 683–85 ¶¶ 94–96, 98 (Wilkins Decl.).
This showing of the security measures taken by the Libraries is essentially unrebutted. Consequently, we see no basis in the record on which to conclude that a security breach is likely to occur, much less one that would result in the public release of the specific copyrighted works belonging to any of the plaintiffs in this case. Cf. Clapper v. Amnesty Int'l USA, –––U.S. ––––, ––––, 133 S.Ct. 1138, 1143, 1149, 185 L.Ed.2d 264 (2013) (risk of future harm must be “certainly impending,” rather than merely “conjectural” or “hypothetical,” to constitute a cognizable injury-in-fact); Sony Corp., 464 U.S. at 453–54, 104 S.Ct. 774 (concluding that time-shifting using a Betamax is fair use because the copyright owners' “prediction that live television or movie audiences will decrease” was merely “speculative”). Factor Four thus favors a finding of fair use.
Without foreclosing a future claim based on circumstances not now predictable, and based on a different record, we hold that the balance of relevant factors in this case favors the Libraries. In sum, we conclude that the doctrine of fair use allows the Libraries to digitize copyrighted works for the purpose of permitting full-text searches.
2. Access to the Print–Disabled
The HDL also provides print-disabled patrons with versions of all of the works contained in its digital archive in formats accessible to them. In order to obtain access to the works, a patron must submit documentation from a qualified expert verifying that the disability prevents him or her from reading printed materials, and the patron must be affiliated with an HDL member that has opted-into the program. Currently, the University of Michigan is the only HDL member institution that has opted-in. We conclude that this use is also protected by the doctrine of fair use.
In applying the Factor One analysis, the district court concluded that “[t]he use of digital copies to facilitate access for print-disabled persons is [a] transformative” use. HathiTrust, 902 F.Supp.2d at 461. This is a misapprehension; providing expanded access to the print disabled is not “transformative.”
As discussed above, a transformative use adds something new to the copyrighted work and does not merely supersede the purposes of the original creation. See Campbell, 510 U.S. at 579, 114 S.Ct. 1164. The Authors state that they “write books to be read (or listened to).” Appellants' Br. 34–35. By making copyrighted works available in formats accessible to the disabled, the HDL enables a larger audience to read those works, but the underlying purpose of the HDL's use is the same as the author's original purpose.
Indeed, when the HDL recasts copyrighted works into new formats to be read by the disabled, it appears, at first glance, to be creating derivative works over which the author ordinarily maintains control. See17 U.S.C. § 106(2). As previously noted, paradigmatic examples of derivative works include translations of the original into a different language, or adaptations of the original into different forms or media. See id. § 101 (defining “derivative work”). The Authors contend that by converting their works into a different, accessible format, the HDL is simply creating a derivative work.
It is true that, oftentimes, the print-disabled audience has no means of obtaining access to the copyrighted works included in the HDL. But, similarly, the non-English-speaking audience cannot gain access to untranslated books written in English and an unauthorized translation is not transformative simply because it enables a new audience to read a work.
This observation does not end the analysis. “While a transformative use generally is more likely to qualify as fair use, ‘transformative use is not absolutely necessary for a finding of fair use.’ ” Swatch Grp. Mgmt. Servs. Ltd. v. Bloomberg L.P., 756 F.3d 73, 84, 2014 WL 2219162, at *7 (2d Cir.2014) (quoting Campbell, 510 U.S. at 579, 114 S.Ct. 1164). We conclude that providing access to the print-disabled is still a valid purpose under Factor One even though it is not transformative. We reach that conclusion for several reasons.
First, the Supreme Court has already said so. As Justice Stevens wrote for the Court: “Making a copy of a copyrighted work for the convenience of a blind person is expressly identified by the House Committee Report as an example of fair use, with no suggestion that anything more than a purpose to entertain or to inform need motivate the copying.” Sony Corp. of Am., 464 U.S. at 455 n. 40, 104 S.Ct. 774.
Our conclusion is reinforced by the legislative history on which he relied. The House Committee Report that accompanied codification of the fair use doctrine in the Copyright Act of 1976 expressly stated that making copies accessible “for the use of blind persons” posed a “special instance illustrating the application of the fair use doctrine....” H.R. Rep. No. 94–1476, at 73 (1976), reprinted in 1976 U.S.C.C.A.N. 5659, 5686. The Committee noted that “special [blind-accessible formats] ... are not usually made by the publishers for commercial distribution.” Id. In light of its understanding of the market (or lack thereof) for books accessible to the blind, the Committee explained that “the making of a single copy or phonorecord by an individual as a free service for a blind persons [ sic ] would properly be considered a fair use under section 107.” Id. We believe this guidance supports a finding of fair use in the unique circumstances presented by print-disabled readers.
Since the passage of the 1976 Copyright Act, Congress has reaffirmed its commitment to ameliorating the hardships faced by the blind and the print disabled. In the Americans with Disabilities Act, Congress declared that our “Nation's proper goals regarding individuals with disabilities are to assure equality of opportunity, full participation, independent living, and economic self-sufficiency for such individuals.” 42 U.S.C. § 12101(7). Similarly, the Chafee Amendment illustrates Congress's intent that copyright law make appropriate accommodations for the blind and print disabled. See17 U.S.C. § 121.
Through the HDL, the disabled can obtain access to copyrighted works of all kinds, and there is no dispute that those works are of the sort that merit protection under the Copyright Act. As a result, Factor Two weighs against fair use. This does not preclude a finding of fair use, however, given our analysis of the other factors. Cf. Davis v. Gap, Inc., 246 F.3d 152, 175 (2d Cir.2001) (“The second statutory factor, the nature of the copyrighted work ..., is rarely found to be determinative.”).
Regarding Factor Three, as previously noted, the HDL retains copies as digital image files and as text-only files, which are then stored in four separate locations. The Authors contend that this amount of copying is excessive because the Libraries have not demonstrated their need to retain the digital image files in addition to the text files.
We are unconvinced. The text files are required for text searching and to create text-to-speech capabilities for the blind and disabled. But the image files will provide an additional and often more usefulmethod by which many disabled patrons, especially students and scholars, can obtain access to these works. These image files contain information, such as pictures, charts, diagrams, and the layout of the text on the printed page that cannot be converted to text or speech. None of this is captured by the HDL's text-only copies. Many legally blind patrons are capable of viewing these images if they are sufficiently magnified or if the color contrasts are increased. And other disabled patrons, whose physical impairments prevent them from turning pages or from holding books, may also be able to use assistive devices to view all of the content contained in the image files for a book. For those individuals, gaining access to the HDL's image files—in addition to the text-only files—is necessary to perceive the books fully. Consequently, it is reasonable for the Libraries to retain both the text and image copies.
The Authors also complain that the HDL creates and maintains four separate copies of the copyrighted works at issue. Appellants' Br. 27–28. For reasons discussed in the full-text search section, this does not preclude a finding of fair use. See supra pp. 98–99.
The fourth factor also weighs in favor of a finding of fair use. It is undisputed that the present-day market for books accessible to the handicapped is so insignificant that “it is common practice in the publishing industry for authors to forgo royalties that are generated through the sale of books manufactured in specialized formats for the blind....” Appellants' Br. 34. “[T]he number of accessible books currently available to the blind for borrowing is a mere few hundred thousand titles, a minute percentage of the world's books. In contrast, the HDL contains more than ten million accessible volumes.” J.A. 173 ¶ 10 (Maurer Decl.). When considering the 1976 Act, Congress was well aware of this problem. The House Committee Report observed that publishers did not “usually ma[ke]” their books available in specialized formats for the blind. H.R. Rep. No. 94–1476, at 73, 1976 U.S.C.C.A.N. at 5686. That observation remains true today.
In light of our holding, we need not consider whether the disability-access use is protected under the Chafee Amendment, 17 U.S.C. § 121.
By storing digital copies of the books, the HDL preserves them for generations to come, and ensures that they will still exist when their copyright terms lapse. Under certain circumstances, the HDL also proposes to make one additional use of the digitized works while they remain under copyright: The HDL will permit member libraries to create a replacement copy of a book, to be read and consumed by patrons, if (1) the member already owned an original copy, (2) the member's original copy is lost, destroyed, or stolen, and (3) a replacement copy is unobtainable at a fair price. The Authors claim that this use infringes their copyrights.
Even though the parties assume that this issue is appropriate for our determination, we are not convinced that this is so. The record before the district court does not reflect whether the plaintiffs own copyrights in any works that would be effectively irreplaceable at a fair price by the Libraries and, thus, would be potentially subject to being copied by the Librariesin case of the loss or destruction of an original. The Authors are not entitled to make this argument on behalf of others, because § 501 of “the Copyright Act does not permit copyright holders to choose third parties to bring suits on their behalf.” ABKCO Music, 944 F.2d at 980; see also our discussion of standing, supra pp. 93–94.
Because the record before us does not reflect the existence of a non-speculative risk that the HDL might create replacement copies of the plaintiffs' copyrighted work, we do not believe plaintiffs have standing to bring this claim, and this concern does not present a live controversy for adjudication. See Clapper, ––– U.S. at ––––, 133 S.Ct. at 1147; Jennifer Matthew Nursing & Rehab. Ctr. v. U.S. Dep't of Health & Human Servs., 607 F.3d 951, 955 (2d Cir.2010) (noting that we have an “independent obligation” to evaluate subject matter jurisdiction, including whether there is “a live controversy”). Accordingly, we vacate the district court's judgment insofar as it adjudicated this issue without first considering whether plaintiffs have standing to challenge the preservation use of the HDL, and we remand for the district court to so determine.
II. Ripeness of Claims Relating to the Orphan Works Project
The district court also held that the infringement claims asserted in connection with the OWP were not ripe for adjudication because the project has been abandoned and the record contained no information about whether the program will be revived and, if so, what it would look like or whom it would affect. HathiTrust, 902 F.Supp.2d at 455–56. We agree.
In considering whether a claim is ripe, we consider (1) “the fitness of the issues for judicial decision” and (2) “the hardship to the parties of withholding court consideration.” Murphy v. New Milford Zoning Comm'n, 402 F.3d 342, 347 (2d Cir.2005) (quoting Abbott Labs. v. Gardner, 387 U.S. 136, 149, 87 S.Ct. 1507, 18 L.Ed.2d 681 (1967)).
The fitness analysis is concerned with whether the issues sought to be adjudicated are contingent on unknowable future events. N.Y. Civil Liberties Union v. Grandeau, 528 F.3d 122, 132 (2d Cir.2008). The Authors assert that their OWP claim is fit for judicial decision because it “will not change based upon the particular procedures that [the University of Michigan] ultimately employs to identify orphan works.” Appellants' Br. 13. According to the Authors, the legality of the OWP does not depend upon the specific means the Libraries ultimately employ to identify orphan candidates or the time the Libraries wait before making works available. Rather, the Authors believe that any iteration of the OWP that results in the publication of complete copyrighted works is an infringement of copyright.
We are not persuaded that these concerns create a ripe dispute. Even assuming, arguendo, that “[a]ny iteration of the OWP under which copyrighted works are made available for public view and download” would infringe someone's copyright, id., it does not follow that the OWP will inevitably infringe the copyrights held by the remaining plaintiffs in this case. It is conceivable that, should the University of Michigan ever revive the OWP, the procedures it ultimately implements to identify orphan works would successfully identify and exclude works to which a plaintiff in this suit holds a copyright. Consequently, we cannot say that any of the plaintiffs face a “certainly impending” harm under our ripeness analysis, Clapper, ––– U.S. at ––––, 133 S.Ct. at 1147; see also Grandeau, 528 F.3d at 130 n. 8.
We note that, in addition to our conclusion about ripeness, the same reasoning leads us to conclude that the remaining plaintiffs lack standing to bring this claim, see our discussion of standing, supra pp. 93–94.
Nor do we perceive any hardship if decision is withheld. See Grandeau, 528 F.3d at 134. The Authors argue that they would suffer hardship because “there is nothing to stop the Libraries from reinstituting the OWP and then, if owners of the listed works come forward, suspending it again.” Appellants' Br. 16.
We disagree. As indicated above, it is far from clear that the University of Michigan or HathiTrust will reinstitute the OWP in a manner that would infringe the copyrights of any proper plaintiffs. If that occurs, the Authors may always return to court. Suffice it to say that “[t]he mere possibility of future injury, unless it is the cause of some present detriment, does not constitute hardship.” Grandeau, 528 F.3d at 134 (internal quotation marks omitted). For these reasons, we conclude that the OWP claims are not ripe for adjudication.
The judgment of the district court is AFFIRMED, in part, insofar as the district court concluded that certain plaintiffs-appellants lack associational standing; that the doctrine of “fair use” allows defendants-appellees to create a full-text searchable database of copyrighted works and to provide those works in formats accessible to those with disabilities; and that claims predicated upon the Orphan Works Project are not ripe for adjudication. We VACATE the judgment, in part, insofar as it rests on the district court's holding related to the claim of infringement predicated upon defendants-appellees' preservation of copyrighted works, and we REMAND for further proceedings consistent with this opinion.