A History of Fair Use at Universities

The case arose out of a common practice at colleges and universities: the copying or scanning of excerpts from books for use on course websites and in academic library e-reserve services. It is common for faculty members to assign particular readings to their students that may be outside or in lieu of an assigned textbook. In the pre-digital days, this was usually done by the creation of a course pack, a printed and bound set of copied readings that were sold to the students in the class. Justification for this practice arises out of copyright’s Fair Use Doctrine—Title 17, Section 107 of the U.S. Code—which allows for limited uses of copyrighted works for purposes such as teaching, research, or scholarship.

Guidelines developed in the 1970s established certain minimum “safe harbors” for copying educational materials, including the use of a single chapter from a book, articles of 2,500 words or fewer, or book sections made up of no more than 10% of a work. However, those guidelines did not have the force of law and were intended to pertain to in-class handouts, not course packs. Two court decisions from the 1990s tested how fair use and these guidelines applied to course packs. In both cases, the packs were developed by faculty members, but their actual printing, distribution, and sale were done by commercial copy shops. The court in both cases said that the commercial nature of the shops was not supportive of fair use with regard to the amount of the works being used and the availability of permissions or licenses for the works being copied. The cases did not address whether fair use would have applied if the copying had been done by the nonprofit universities.

Following those decisions, it has largely been a standard practice for all course packs—whether printed by commercial copy shops or university copy centers—to be permitted or licensed with royalties included in their cost. Stories were often shared about exorbitant course pack costs in the hundreds or even thousands of dollars, largely due to the royalties.

As electronic distribution of scholarly content became more prevalent, practices changed. Faculty members began distributing readings through course websites, school-sponsored learning tools such as Blackboard or Canvas, and university library e-reserve systems. In some cases, a university might have had a license for an electronic edition of a reading—such as an ejournal article—so no additional permission or royalty was required. However, in other cases, often involving excerpts from books or other materials that were not available electronically, school or library staffers would scan the excerpt and post it to the website, learning tool, or e-reserve system. As this practice became more common, it became less common to pay royalties or to obtain permission for these scanned works.

The acts of scanning and posting the readings both involve copyright restrictions, but given the structure of the copy shop court cases, they did not make the issue of fair use very clear. The purpose of copying was certainly educational, and no commercial company was involved, so the facts that were not supportive of fair use were different. Many institutions also used varying internal policies or guidelines to determine how much copying was “allowed” by fair use.

Fair Use at Georgia State

Georgia State was one such institution, scanning materials from a number of publishers to course web sites and e-reserves under a policy that suggested “up to 20 percent of a work” could be posted under fair use. Concerned about these practices, Cambridge University Press, Oxford University Press, and SAGE sued Georgia State in 2008 for copyright infringement. After a lengthy trial, a court determined in 2012 that most—but not all—of the copying was fair use. It found that the works’ purpose was educational and that they were more informational in nature than creative (both factors favor fair use). The court then set a standard of 10%, or one chapter of a book, as a fair use safe harbor and said that not having a readily available mechanism for digital permissions for an excerpt also favored fair use. The court then applied these four standards to each act of copying, and if at least three of them favored fair use, no infringement was found.

In 2014, an appellate court overturned the trial court’s decision. While it agreed with many of the trial court’s findings, it found that it had applied them too “mechanistically” and that a more “holistic analysis” was required.

This sent the case back to the trial court, which, in spring 2016, found that 44 of 48 specifically identified acts of copying were considered fair use, and only four were considered infringement. The court was more careful in evaluating and weighing the factors, considering not only the educational use, amount used, and availability or non-availability of a licensing system, but also the transformative or nontransformative nature of the use, whether the amount chosen was “narrowly tailored to fit the pedagogical aim of the course,” and the overall impact on the market and value of the work.

For the 44 cases considered fair use, the court generally found that they were educational, informative more than creative, and narrowly tailored to fit the course. It also found that even if copying practices such as Georgia State’s were widespread, there would be little impact on the market for the book or the licensing revenue. For the four items found not to be fair use, the decisions were usually based on the amount being copied or the impact on the market.

The court’s evaluation was extensive—it went through each act of copying, taking up 190 pages of a 212-page opinion—and very thorough. The court invited the publishers to propose terms for a future injunction to ensure that Georgia State complies with the court’s decision. The publishers wanted “rigid and burdensome” requirements that the university maintain extensive records, make modifications to its website, and mandate broad publication of the injunction’s requirements. Instead, the court simply ordered that Georgia State maintain and disseminate a new fair use policy consistent with the court’s rulings.

While it’s possible that the publishers may again appeal the court’s orders, the thoroughness of the evaluation and the diligence in complying with the appeals court’s rules suggest that any appeal would be an uphill battle. Also, the court did state that the availability of a licensing system for excerpts similar to the Copyright Clearance Center’s system for journals would make a fair use finding more difficult. The best solution may be to end the litigation and focus on a system that allows for easy licensing and royalties for all academic scholarship.