Plaintiffs filed a motion last night asking Judge Baer to rule on the fair use defense in the HathiTrust lawsuit. At issue is a mass book digitization program through which Google converted millions of copyright-protected library books into machine-readable digital files that were duplicated and distributed to university libraries and HathiTrust, an online digital repository.
This is the first motion that squarely places before a court the question of whether the unauthorized mass digitization of library books is a fair use under U.S. law.
Background on the HathiTrust Lawsuit
Authors groups from Australia, Canada, Norway, Sweden, Quebec, the U.K., and the U.S., along with twelve individual authors, brought suit against the University of Michigan, the University of California, the University of Wisconsin, Indiana University, and Cornell University and HathiTrust last fall. The lawsuit seeks impoundment of the unauthorized scans, pending appropriate Congressional action.
Last June, the University of Michigan, which oversees HathiTrust, announced plans to permit unlimited downloads by its students and faculty members of “orphaned” books (some consider works whose rights-owners cannot be found after a diligent search to be “orphans”). Michigan devised a set of procedures — including a protocol for searching for an author and posting the names of “orphan work candidates” at the HathiTrust website for 90 days – to determine whether it would deem a work an “orphan.” Several other schools joined the project in August.
The first set of so-called orphans, 27 works by French, Russian, and American authors, were scheduled to be released to an estimated 250,000 students and faculty members on October 13th. An additional 140 books, including works in Spanish, Yiddish, French, and Russian, were to be released starting in November.
Within days of the suit’s filing on September 12th, the Authors Guild, its members, and others commenting on its blog had developed strong leads to dozens of authors and estates holding rights to the first 167 works listed as “orphan candidates” at HathiTrust’s website. Four living authors were on HathiTrust’s list. So were significant literary estates, such as those of Pulitzer Prize winners James Gould Cozzens and Walter Lippmann and the philosopher Sidney Hook. Foreign authors were also on the list, including André Missenard, who died in Paris in August. At least three of the works on the list were found to be still in print.
Michigan announced on September 16th that it was suspending, but not ending, its “orphan works” program. Its online servers continue to host an estimated 7 million digitized, copyright-protected books. Millions of those books are believed to be in print, with e-book versions available for many of them.
Fair Use Motion
Yesterday, plaintiffs filed a motion for “partial judgment on the pleadings” that the mass book digitization program isn’t protected as a fair use. The Preliminary Statement from the brief spells out the argument:
Plaintiffs submit this memorandum in support of their motion pursuant to Federal Rule of Civil Procedure 12(c) for partial judgment on the pleadings and an order that Defendants’ admitted systematic reproduction, distribution, and use of millions of copyright-protected books are not shielded by the First Amendment, the fair use defense, or any other provision of the Copyright Act. Plaintiffs rarely move for judgment on the pleadings, but this is a highly unusual case in which Defendants, a group of university libraries and their joint online venture, not only concede that they have engaged in the mass book digitization project at issue, but admit to acts that Congress has expressly prohibited under Section 108 of the Copyright Act.
Section 108 provides the ground rules under which libraries and archives may reproduce and sometimes distribute copyright-protected works. Those ground rules are the product of years of discussions and debate among representative academics, archivists, authors, educators, librarians, and publishers; of hundreds of hours of fact-gathering sessions organized by the Copyright Office and the Patent and Trademark Office; and of testimony before House and Senate committees in the 1960s, 1970s, and 1990s. The story of Section 108 as a legislative proposal in the 1960s, its enactment in 1976, and its amendment in 1992, 1998 (twice), and 2005
is the story of stakeholders and legislators grappling with the promise to researchers and peril to literary markets posed by a set of emerging technologies that for the first time made it convenient to create inexpensive, high quality reproductions of literary works.
Broadly, Section 108 permits libraries and archives to reproduce copyright-protected literary works for two purposes: to maintain a library’s existing collection (for example, to permit a library to replace damaged, lost, or stolen books that are no longer commercially available) and to fulfill the requests of readers and researchers using a library or, through interlibrary loan arrangements, to fulfill the requests of researchers using other libraries. At every turn, Section 108 carefully limits the scope of permissible copying — for example, prohibiting the reproduction of an entire book that is commercially available, never permitting a library to retain a reproduction of an entire book for its collection, except as a replacement, and never authorizing the distribution of a digital replacement of a published work outside of the physical premises of the library that houses the original. These limitations are carefully designed to protect the vital cultural markets that copyright is intended to foster.
Defendants admit to a mass book digitization program and “orphan works” project that violate nearly every restriction imposed by Section 108. Where Section 108 permits a library to create at most, and under limited circumstances, three duplicates of a work, Defendants admit to routinely making five (actually ten, as shown below). Where Section 108 allows a library to copy entire books for its own collection solely for the purpose of replacing badly damaged or lost copies, Defendants authorized the copying of entire libraries of books, without regard to their condition, and then retained those copies for their collections. Where Section 108 permits a library to create such a replacement copy only if the book is not commercially available, Defendants admit to not bothering to check: they authorized the copying of everything. Where Section 108 permits only the reproduction (not the distribution) of published books except to fulfill the requests of users, and expressly requires that any digital replacement copy be used strictly on the library’s premises, Defendants admit to loading digital copies of published books by the million onto servers in Michigan and Indiana and concede that those servers are connected to the Internet. Where Section 108 explicitly forbids systematic copying, Defendants embraced what is surely the most systematic, comprehensive scheme of reproducing copyright-protected books ever undertaken, creating unauthorized, machine-readable versions of the literary work of millions of authors from all over the world. And where Section 108 authorizes a library to make uses of “orphan works” during the last twenty years of their copyright term, Defendants threaten to begin exploiting purported orphan works decades before they are permitted to do so.
Given Defendants’ admitted conduct, there is no reason for the parties to engage in
extensive discovery regarding the defenses they assert. Plaintiffs respectfully urge the Court to hold that Defendants’ mass book digitization and orphan works projects are not protected by any defense recognized by copyright law.
Plaintiffs’ entire brief is here.