HathiTrust Orphan Works Project Grows as University of California, Others Join Up
By David Rapp Sep 1, 2011A number of recent announcements have greatly expanded the research bandwidth available to investigate the pool of orphan works in the HathiTrust collections—a fact that could stir discussion on orphan works access, both in and out of academia.
On August 24, the University of California (UC) Libraries announced that it will join an initiative led by the University of Michigan (UM), which aims to expand access to the in-copyright works in the HathiTrust repository for which no rights holders can be found. The same day, in a separate announcement, Duke University, Raleigh, NC; Cornell University, Ithaca, NY; Emory University, Atlanta; and Johns Hopkins University, Baltimore, also signed on. The institutions join the Universities of Wisconsin and Florida, who joined up since UM first announced the project in May 2011.
Access in October
The UM project currently aims to pin down the status of works in the HathiTrust collection published between 1923 and 1963.
UM has contributed the largest number of digitized books to HathiTrust, more than 4.4 million. UC has contributed nearly three million works, Wisconsin about 488,000, and Cornell about 345,000. Combined, these universities alone have contributed about 87 percent of the 9.5 million works in the HathiTrust repository.
Only 27 percent of HathiTrust’s collections, however, are in the public domain. Up to now, that’s meant that nearly three-quarters of the HathiTrust was inaccessible in digital form—only snippets have been viewable online due to copyright restrictions. But starting in October that will begin to change.
Here’s how the process works: UM researchers look into the rights of a particular title; if they are unable to determine a work’s copyright holder, its bibliographic information is placed on a dedicated page on in the online HathiTrust Digital Library for 90 days. If no rights holder comes forward, the work will be deemed an orphan, and made accessible to the UM community—at least until a rights holder appears and objects. Since the first orphan-works candidates were listed in July, some will become accessible in October.
The current list of 148 potential orphan works on the HathiTrust site has an eclectic mix of mostly out-of-print titles published between 1923 and 1963. (Works copyrighted in the United States before 1923 are automatically in the public domain.) Titles include fiction and nonfiction works, including a few by well-known authors: Pulitzer Prize-winning author James Gould Cozzens’s 1924 debut novel Confusion; Walter Lippmann’s 1959 book The Communist World and Ours; and even a 1933 English translation of French author Jean de Brunhoff’s classic children’s book The Story of Babar.
Newly available UM orphan works will only be accessible to UM students and faculty. Similarly, orphan works from other institutions’ collections would only be viewable by their own communities. This limited usage hinges on an interpretation of the Copyright Act’s “fair use” provision (Section 107), which allows limited copying for scholarly use. Also, if UM finds a work is an orphan, and another university has its own copy of that work in HathiTrust, both institutions could make that work accessible. “Since we often hold the same volumes, then doing the work for one [institution] is doing the work for all,” UM associate university librarian and HathiTrust executive director John Wilkin told LJ back in June.
The plus side
What are the benefits to universities of having full digital access to these orphan works? Anne Kenney, university librarian at Cornell cited full-text searching as an advantage. “We know that such access can breathe new life into older works. It also helps us to care [for and] protect the originals and to move them to the annex for safe storage,” she told LJ.
Laine Farley, executive director of UC’s California Digital Library, expected “to identify rights holders who are UC faculty and who can then make a choice about the terms under which they prefer to make their works available.”
A way for Google?
Though the UM project is relatively limited in scope, it’s a bold step to expand access to digitized orphan works. That issue is at the center of the high-profile Google Books case: a settlement in that case, rejected in March, would have given Google the ability to sell digitized orphan works in its vast Google Books collection.
The Google case has now stalled, but could the UM project provide a model for digital orphan-works access? That’s not so clear.
In general, companies such as Google are the most exposed to potential infringement lawsuits. However, private universities—including Duke, Emory, Cornell and Johns Hopkins—can be at more risk than publicly funded institutions, although officials LJ talked to seemed confident of their solid legal ground.
“I think the core of the fair use case here rests on the educational purpose which we ensure by restricting access only to our campus communities, the fact that digital files will only be available for orphan books we also hold in our physical collections, and the fact that by definition there is no locatable rights holder for these orphan works and thus no market that this limited use will harm,” said Kevin Smith, Duke’s scholarly communications officer. “Michigan has committed to suspend access and talk with any rights holder who may arise at a later date so we think the overall risk of this project is very low.”
Jonathan Band, an intellectual property attorney and a frequent consultant to the library community, pointed out that the academic distinction is a crucial one. “To the extent that the legal theory relied upon is fair use, the privilege is stronger for noncommercial and transformative uses,” he told LJ. “So, a university providing its faculty and students with access to orphan works for research purposes probably would have a stronger fair use argument than a private party that sells access to the orphan works for profit.”
Carrie Russell, a copyright expert at the American Library Association (and a columnist on copyright issues for School Library Journal) said that “the copyright law does favor more uses of works (without authorization) by non-profit, educational institutions. It is arguably fair use because—if the title is a true orphan—it has no commercial viability.”
As the library community awaits imminent copyright decisions—a verdict on the Georgia State University e-reserves case is expected soon, and a hearing on the Google case is slated for September 15—the HathiTrust project, which will be granting access to some orphan works in about a month, will be an interesting one to watch.
