Georgia State and the Boundaries of Academic Fair Use: From Bright Line Test to…Maybe

Stephen Carlisle October 24, 2014

Almost a year in the making and weighing in at a massive 129 pages, the 11^th Circuit Court of Appeals finally released its opinion in the closely watched case of Cambridge University Press v. Patton on October 17^th. ¹ This case, popularly referred to as the “Georgia State” case, was a copyright infringement test case brought by three academic textbook publishers against Georgia State University for their practice of digitizing excerpts from textbooks and placing them on university servers where they could be viewed by students.

I say “test case” for two reasons. First, Georgia State, being an arm of the State Government of Georgia, has sovereign immunity from damages being assessed against it under the U.S. Constitution and 11^thCircuit precedent, ² and second, the behind-the-scenes Plaintiff, who was paying 50% of the attorneys fees, was the Copyright Clearance Center, a licensing agency for publishers. ³

The copying by Georgia State was rather small, on average around 9.6% of the total book pages and the majority were a single chapter from a book that averaged 18 chapters. ⁴ Yet the publishers sued for 126 counts of copyright infringement.

In deciding the case, the Trial Court followed a consistent line of reasoning. Factor one of Section 107, the “purpose and character of the use,” always favored Georgia State as it was engaging in not for profit educational uses. Factor two, the “nature of the work used,” also favored Georgia State as the excerpted works were primarily works of facts, science and research, and therefore were not entitled to as strong protection as works that were strictly commercial. With two out of the four factors already favoring Georgie State, you got the feeling that the publishers were playing catch up. For factor three, ‘the amount and substantiality of the taking” test, the Trial Court announced a bright line rule in an area of Copyright Law that is not supposed to rely on bright line rules. If the book had more than 10 chapters, then up to one chapter was permissible. If the book had less than ten chapters, then up to 10% of the total pagination was permissible. The majority of the counts brought against Georgia State failed under this test as well, making it three to zero against the publishers, though in one case the Judge did side with the publishers because the excerpt took the “heart of the book.” ⁵

As to the fourth factor, the “effect on the market,” the Trial Court’s analysis began with the question of, did the excerpts harm the market for the books. In almost all of the cases, the answer was no because the excerpts were so small that no professor would require the purchase of the entire book only to have the students read a small portion. ⁶ The trickier question was whether the excerpts harmed the market for paid digital excerpts from the Copyright Clearance Center. There, the Trial Court split the difference. If an excerpt was “available from CCC or the publisher…at a reasonable price and in a convenient format…and permissions are not paid, factor four weighs heavily in Plaintiffs’ favor.” ⁷ But if a digital excerpt was not available, factor four tilted towards Georgia State. ⁸

The Trial Court examined each count individually, using the analysis summarized above. This took the Trial Court 177 pages to fully explain her reasoning on each count, so certainly one cannot fault her for lack of thorough consideration. Out of the 126 counts initially brought, only 74 went to trial. ⁹ Of those 74, the Court ruled that 26 counts failed to prove a prima facie case of infringement, 43 were fair use and the remaining 5 were infringing. ¹⁰ Adding salt into the wounds of the publishers, the Judge declared Georgia State the prevailing party and assessed $2.5 million in attorneys fees and costs against the publishers. ¹¹

Needless to say, the publishers appealed. The parties presented oral arguments to the 11^th Circuit Court of Appeals on November 19, 2013. ¹² The fact that it took 11 months to generate an opinion indicates to me that there was a lot of disagreement amongst the three judge panel, and perhaps there was more than a little “horse trading” on certain key points. Add to the fact that one Judge wrote a concurring opinion that read more like a stinging dissent from the majority position makes the decision even more curious.

The Court of Appeals partially reversed the Trial Courts’ decision, and sent the case back down to the Trial Judge to reconsider the case according to their instructions. What the Court of Appeals found the most fault with was the mechanical approach of the Trial Court. So, immediately, the “10% or one chapter” test goes out the window as does the “checkbox” approach that treats all four fairs use factors equally. ¹³ This is not surprising given that there is Supreme Court of the United States authority for this line of reasoning. ¹⁴

So, here are the major bullet points of the 11^th Circuit opinion:

• Fair use requires a case by case analysis. ¹⁵
• Rigid, arithmetic approaches are improper and not a valid method of analysis. ¹⁶
• Individual Fair Use factors may increase or decrease in importance depending upon the case. ¹⁷
• Fair Use factors vary in importance and cannot be given equal weight. ¹⁸
• The fact that you are a not-for-profit educational institution is favorable for factor one, but does not automatically mean a “win” on that factor. ¹⁹
• Digitizing analog works and placing them online was not a “transformative use” that would work in favor of Georgia State in factor one. ²⁰
• Despite the use not being “transformative,” the Court ruled that “use for teaching purposes by a non-profit education institution such as [Georgia State] favors a finding of fair use under the first factor…” ²¹
• Factor two is of relatively little importance in analyzing Fair Use in this situation. ²²
• Factual works, such as textbooks, can have expressive qualities, so the Judge was wrong to rule that because the works contained mainly facts that factor two always fell in favor of Georgia State. ²³
• The 10% or one chapter rule, even if it represents a general industry “best practice” for electronic course reserves, is not a law and is “not relevant to an individualized fair use analysis.” ²⁴ The 1976 Classroom Copying Guidelines pushed by the publishers, which would have limited copying to the lesser of one chapter or 1,000 words, fails as well. ²⁵
• The amount of allowable copying varies with the purpose and character of the use. ²⁶
• The use for a pedagogical purpose may expand the limits of permissible copying, but an entire work may not be copied unless the use is “highly transformative.” ²⁷
• Factor four is the most important and “market substitution” is the primary concern for the Court. ²⁸
• Georgia State’s excerpting did not act as a substitute for the full books. ²⁹
• The availability of electronic excerpts weighs against Fair Use, but does not preclude Fair Use. ³⁰
• The non-availability of electronic excerpts weighs in favor of Fair Use. ³¹
• The fourth factor should be analyzed in the context of the effect it has on the publishers’ incentive to publish, not on an academic authors’ incentive to write. ³²
• The fourth factor should carry the most weight in this case. ³³

I cannot disagree with the emphasis being placed on factor four, the “market effect.” If one reads between the lines of the three Supreme Court cases that have dealt with fair use, the Justices sided with the copyright holder where there was solid evidence on negative market effect and with the alleged infringer where there was no evidence of negative market effect. What I do find fault with is the instruction that the negative market effect be viewed through the eyes of the publisher, not the author.

The Copyright enabling clause talks of granting exclusive rights to the “authors,” not publishers. This is in order to promote the “progress of …the useful arts.” In the academic world, authors may write for a number of reasons, some for profit, but others write to enhance their academic reputation and that of their institution of higher learning. They may not only not object to the small amount of copying being done here, but may be in favor of it.

Next, I think the Court of Appeals’ dismissal of the importance of the second factor “the nature of the work used” to be a serious error. The Copyright Act is very specific that “[i]n no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.” ³⁴ So, a work consisting of facts or theories, whether scientific or historical, enjoys significantly less protection against copying than does a work that is purely commercial in nature. While the Court of Appeals gives lip service to “the proposition that applying fair use in a manner which promotes the dissemination of knowledge is consistent with the goals of copyright,” ³⁵ it should not have demoted the factor in its’ level of importance.

It could have been worse, as I stated above, the “concurring” opinion of Judge Vinson reads more like a dissent, than a concurrence. He would have ruled against Georgia State on a blanket basis saying, “Checking the four statutory factors to ensure that they have been considered merely affirms the conclusion that what GSU is doing is not fair use.” ³⁶

While I understand why the Court of Appeals nixed the “10% or one chapter” rule, apparently lost on the Court was the utility it afforded an institution of higher learning. It was an easy, bright line rule that can be communicated simply and effectively to faculty members that bear the brunt of the fair use analysis. Now, they are being asked to inquire as to the “negative market effect” to a business they have limited understanding of and not a whole lot of resources with which to make a determination. There was a chance here to make fair use a whole lot easier to understand. There was a chance to set the acceptable parameters within which to make an intelligent decision, on an action that can have serious legal consequences, but have chosen not to do so. If Georgia State did not have sovereign immunity, and the Court found for the publishers on all 126 counts initially brought, at the maximum level of regular statutory damages, the amount awarded would have been $3.78 million dollars. Add to that the threat of paying prevailing party attorneys fees (Georgia State spent $2.5 million) and you have a very sobering decision to make. Namely, whether to license everything, even when the use might be fair use, or take the risk of a copyright infringement lawsuit of this type.

Even the “license everything” does not solve the problem. Many faculty members have used the same materials for years, and may not remember from where or from whom it was obtained. What if the work is an “orphan work” for which the current copyright owner cannot be identified or located? The possibility also exists that the faculty member may decide that the analysis required by this Court’s opinion is too burdensome and disregards it.

So, in essence, the Court of Appeals’ opinion does not change a whole lot, it just rewinds us back to before the publishers brought this suit in 2008. Unfortunately, the era to which it rewinds means that the answer to the question “is this fair use?” is always “maybe.”

I don’t often agree with Larry Lessig, but I do when he states that “fair use in America simply means the right to hire a lawyer… ³⁷

In sum, if your university has not previously created a position equivalent to mine, and your location is in Florida, Georgia or Alabama, you’d better do so now, especially if you are a private not-for-profit and absolutely if you are a for profit university.

Because what we have here is the fair use equivalent to Justice Potter Stewarts’s famous pronouncement about pornography… “I shall not today attempt further to define the kinds of material I understand to be embraced within that shorthand description; and perhaps I could never succeed in intelligibly doing so. But I know it when I see it.” ³⁸