Some Leeway, Some Limits

A federal judge on Friday outlined many ways colleges can continue to cite the doctrine of "fair use" to permit their making electronic copies of books and other materials for use in teaching and scholarship. In a landmark ruling over many issues not previously litigated to this degree in the digital era, the judge rejected many of the claims in a suit by three prominent publishers against Georgia State University. In 94 of the 99 instances cited by the publishers as copyright violations, the judge ruled that Georgia State and its professors were covered by fair use. And the judge also rejected the publishers' ideas about how to regulate e-reserves -- ideas that many academic librarians said would be unworkable.

At the same time, however, the judge imposed a strict limit of 10 percent on the volume of a book that may be covered by fair use (a proportion that would cover much, but by no means all, of what was in e-reserves at Georgia State, and probably at many other colleges). And the judge ruled that publishers may have more claims against college and university e-reserves if the publishers offer convenient, reasonably priced systems for getting permission (at a price) to use book excerpts online. The lack of such systems today favored Georgia State, but librarians who were anxiously going through the decision were speculating that some publishers might be prompted now to create such systems, and to charge as much as the courts would permit.

The 340-page decision by Judge Orinda D. Evans is a pivotal point in years of litigation brought by Cambridge University Press, Oxford University Press and Sage Publications -- with backing from the publishing industry. Many experts expect this case to assume a role that cases against Kinko's (decided in 1991) and Michigan Document Services (decided in 1996) played in defining copyright issues for printed coursepacks. But the Georgia State decision doesn't end the legal hearings (even if there isn't an appeal). Evans ordered the publishers to propose remedies for the violations she found, and new hearings will be held on those proposals.

While some university librarians were so anxious about this case that they stayed up late Friday to tweet their reactions, some of the official reactions aren't coming until later today. The Association of American Publishers would say this weekend only that it was studying the decision. A Georgia State spokeswoman said that its officials were also reviewing the decision and couldn't say much more than "we're reviewing the judge's order but are pleased with our initial assessment." (As the decision notes, the publishers' group recruited the three plaintiffs in the case, and with the Copyright Clearinghouse Center split the legal costs of the three publishers who sued.)

While the legal analysis may take time, both publishers and academic librarians have reacted strongly throughout the case. Publishers argued hat their system of promoting scholarship can't lose copyright benefits. Judge Evans in her decision noted that most book (and permission) sales for student use are by large for-profit companies, not by nonprofit university presses. But the Association of American University Presses has backed the suit by Cambridge and Oxford, saying that university presses "depend upon the income due them to continue to publish the specialized scholarly books required to educate students and to advance university research."

Many librarians, meanwhile, have expressed shock that university presses would sue a university for using their works for teaching purposes. Barbara Fister, a librarian at Gustavus Adolphus College and an Inside Higher Ed blogger, tweeted Friday night: "It still boggles my mind that scholarly presses are suing scholars teaching works that were written to further knowledge."

The reserve readings at the crux of the dispute are chapters, essays or portions of books that are assigned by Georgia State professors to their undergraduate and graduate students. (While the readers are frequently referred to as "supplemental," they are generally required; "supplemental" refers to readings supplementing texts that the professors tell students to buy.) E-reserves are similar to the way an earlier generation of students might have gone to the library for print materials on reserve. The decision in this case notes a number of steps taken by Georgia State (such as password protection) to prevent students from simply distributing the electronic passages to others.

Sorting Out the Law

Judge Evans spends much of the decision focused on whether Georgia State's use of e-reserves was consistent with the principles of fair use. She notes that the fair use exemption in federal law requires consideration of four factors (although the law is vague on exactly how the four factors should be weighed). The four factors are:

1. "The purpose and character of the use," including whether the use is "for nonprofit educational purposes."

2. "The nature of the copyrighted book."

3. "The amount and substantiality of the portion used."

4. The impact of the use on "the market" for sale of the book or other material.

Evans found that the first two factors strongly favored Georgia State. The university is a nonprofit educational institution using the e-reserves for education, she notes. Further, she found for Georgia State on the second factor, noting that the works in question were nonfiction and "informational," categories she said were appropriately covered by fair use.

The analysis of the third and fourth factors was less straightforward to Judge Evans. She starts by rejecting a claim of the publishers that a 1976 agreement between publishers and some education groups should govern fair use for e-reserves. That agreement was "very restrictive," she writes. For example, only work that did not exceed 2,500 words was covered. Still other limits were set on how many times an instructor could invoke fair use in a single course.

While rejecting the 1976 agreement, Judge Evans writes that there are legitimate questions about how much material may be used. In a sign of just how complicated the issues are, she notes that the publishers asked her to base any percentages on only the text portion of a book (excluding introductory pages, footnotes and concluding tables) while Georgia State wanted everything counted. Evans based her percentages on Georgia State's view that the book is the entire book.

Her challenge, she writes, is to determine what size excerpts are "small enough" to justify fair use. Here, after reviewing a range of decisions, Evans settles on 10 percent of a book (or one chapter of a book) as an appropriate measure, allowing professors enough substance to offer students, while not effectively making a large portion of the book available.

On the fourth factor (market impact), Evans writes that there is a clear impact if and only if the publisher has a system for selling access to excerpts that are "reasonably available, at a reasonable price." The reason this prong did not help the publishers more in the case is evidence cited by the judge that much of the material in question was not available through an online licensing program. So Georgia State did not have the "reasonably available option."

At various points in the decision, Evans also weighs the intent of both copyright protection and fair use in the context of this case, generally with an analysis that is sympathetic to Georgia State. "Because the unpaid use of small excerpts will not discourage academic authors from creating new works, will have no appreciable effect on plaintiffs' ability to publish scholarly works, and will promote the spread of knowledge," she writes.

Further, she rejects the idea that Georgia State's actions have had a significant impact on the ability of the publishers to sell books. But she does see an economic cost to students of passing along more of the cost of materials. "Plaintiffs offered no trial testimony or evidence showing that they lost any book sales in or after 2009 on account of any actions by anyone at Georgia State. The court finds that no book sales were lost," Judge Evans writes. The publishers lost only "a small amount" of permissions revenue.

"If students at Georgia State had been required to pay for use of small excerpts of plaintiffs' works in 2009, there would have been some small overall increase in the cost of education, assuming that the charge for excerpts would be included in the tuition and spread across the student body," Judge Evans adds. "If individual students had to pay the cost of excerpts, the total of all permissions payments could be significant for an individual student of modest means.

'Good News for Libraries'

Kevin Smith, scholarly communications officer at Duke University, was among the first to offer a detailed analysis of the ruling. In a blog post receiving praise from many librarians, he wrote that there was "good news for libraries" in the decision, but also some challenges for professors, librarians and publishers. "Most of the extreme positions advocated by the plaintiff publishers were rejected."

Smith noted that this may send a message to publishers. "It suggests that suing libraries is an unprofitable adventure, when 95 percent of the challenged uses were upheld," he wrote. At the same time, however, he added that there are "uncertainties" in the ruling.

The 10 percent rule for share of a book that can be used, he said, "is a less flexible standard than many libraries would like, I think, and it seems too rigid to be a good fit with the overall structure of fair use." (The decision notes that the excerpts examined by the court averaged 10.1 percent of book length, with many well below the 10 percent threshold and some well above it.)

Fister of Gustavus Adolphus said via e-mail that she too was relieved that the judge had rejected many of the publishers' arguments. But she also saw "problematic issues" for librarians and professors.

She noted that the ruling requires colleges to apply analysis of all four factors in determining, case by case, what is permitted. The judge issued her decision nearly a year after the trial in the case started, suggesting that it "requires a lot of time and a lot of legal discovery" to make these determinations. This is particularly difficult, she said, in the requirement of knowing something of publishers' revenue streams to analyze the market impact of making a portion of a given work available for students. Fister said that the finding that 94 of 99 instances were properly covered by fair use might have suggested college libraries be given more leeway.

"The other thing I come away with is that publishers that get their act together and make it easier for institutions to pay licenses will make it much harder for academics to claim fair use for e-reserves," Fister said. "Even large publishers have been cautious about making permissions too easy, as two of the three plaintiffs do not make their content widely available for licensing. Sage has made it slick, and four of the five findings against fair use were for Sage titles. This will favor large publishers and perhaps publishers that require authors to treat their scholarship as work for hire before it's published because it will be easier to license that work and reap new profits from it."