GOOGLE HAS DIGITIZED millions of books for its Google Books Library Project, a searchable database that allows a user to search the content of all books that have been scanned into the database. The Authors Guild maintains that the Google Books Library Project constitutes mass copyright infringement, because Google did not obtain licenses from the rights holders for millions of the books. When the Authors Guild sued Google in the Southern District of New York for copyright infringement, Google prevailed, arguing that the Google Books Library Project constitutes fair use, a statutory defense to copyright infringement. The Authors Guild has appealed the ruling to the Second Circuit Court of Appeals where the case is pending and being briefed.
LARB contributor and entertainment attorney Mark Robertson discussed the case, some of its main issues, and its implications with Professor Jay Dougherty of Loyola Law School Los Angeles, where Dougherty is a Professor of Law and the Director of the Entertainment and Media Law Institute and Concentration Program.
MARK ROBERTSON: What is the main issue in this case?
JAY DOUGHERTY: The district court was asked to address whether the Google Books Library Project constitutes fair use. Fair use had been codified in the 1976 Copyright Act, in Section 107, which directs courts to consider a variety of factors, including the four factors that are specifically mentioned in the statute, and actually sort of a fifth one in a way: whether the work is published or not. And so the court considered those factors, and it determined that what Google was doing was, in its view, clearly fair use. That question is what is on appeal right now in the Second Circuit Court of Appeals.
The district court determined that Google Books was highly transformative. Yet Google does not really add new material to the works, which is a characteristic often present in such fair use cases. Is the conclusion that Google Books is transformative well reasoned?
“Transformativeness” is a relatively recent development in fair use analysis, and it derives from a court case involving a rap version of Roy Orbison’s “Oh, Pretty Woman.” In that case, the United States Supreme Court emphasized the importance of considering whether a use is transformative in determining whether or not it is fair. That was in 1994. Since that time, although the Supreme Court instructed judges to carefully balance and weigh all of the factors, courts have tended to emphasize transformativeness. If the use is transformative, they tend to find fair use, and if not, they tend not to find fair use. The language the Court used in the “Pretty Woman” case suggested that they were attending to what used to be called “productive uses,” whether the use “adds something new with a further purpose or different character, altering the [work] with new expression, meaning, or message.” At first blush, you would think that that sounds like they have added some new material or transformed the material that is there in some creative way. But the Court also mentioned that a work might be considered transformative if it is used for a new purpose.
And so, a number of courts have begun to find fair use even when there is not changed expression or new expression added, if a work is being used for a purpose that is different from the purpose of the copyright owner, or the original purpose of the work. The Google Books case is not the first case to do that. For example, there was a recent case involving a number of posters used in a book about the history of the band The Grateful Dead. The posters were shrunken and were being used, arguably, for a very different purpose than the original posters were used for, which was to promote the band and its shows. Also, go back even further to the Sony Betamax case in the ’80s. There the Supreme Court acknowledged that the literal duplication of an entire work could under some circumstances be considered fair use even when nothing was added to it — although that case was about home-taping of over-the-air broadcast TV shows for time-shifting purposes, use that was not commercial in nature.
That is a bit of background on transformativeness, and on the idea that the literal duplication of an entire work or the changing of the size or the resolution of an image can be fair use, especially when the uses are noncommercial. So it is not terribly surprising that the court in the Google Books case considered the use transformative. And concluding that the use is transformative really ends up as conclusive since once the court determines that it is a transformative purpose, a transformative use, all the elements under Section 107 kind of fall in place.
One of the arguments that the Authors Guild is making on appeal is that Google’s use of the books is commercial, that the database that Google has amassed of an estimated 20 million books is a large improvement to its core business and provides Google with a competitive advantage. The district court acknowledged only an indirect commercial benefit to Google. Did the court give adequate consideration to the commercial advantage?
I have to evaluate that the way the court did. Read between the lines, and the court concluded that the benefits to society highly outweigh whatever indirect commercial benefits Google is getting. Of course that leaves the copyright owners out in the cold. But I think that Google has attempted to minimize any direct commercial benefit. I am not sure which argument will prevail, but it certainly can be argued that indirect commercial benefit should be a negative factor in a fair use analysis. At one time, the commerciality of a use was considered very important; in fact, the statute itself mentions it as an important consideration. But since the transformativeness test evolved as an important part of the analysis, courts have downplayed the commerciality aspect if they find a use to be transformative.
In the district court, the fair use analysis was applied to the entire database of books rather than on a work-by-work or an exemplary work basis. Is there precedent for applying the fair use analysis to a large body of works like that? Or is it typically an analysis that is done on a work-by-work basis?
I do not think any precedent that has dealt with this kind of fair use question has involved class actions rather than groups of works. So I do not know that there is precedent for lumping them all together in this way. The irony here is that when it comes to certification of the class, the copyright owners are arguing that the fair use analysis is the same for all of the works, and Google, by contrast, is arguing the opposite, saying no, you cannot certify the class because each work should be considered separately. The judge has no doubt considered that each party is arguing the opposite position in the class certification dispute than they argue in connection with the fair use question.
So, I suppose looking at it a factor at a time, the transformative factor, presumably, is going to be the same for all of the works. Now, of course some of the works are fiction, some are nonfiction; some are published, some are not. And, in that scenario, the court normally would consider each group or would consider an individual work. But here the court dealt with them in generalities and said that they were mostly nonfiction. That tends to favor a little bit broader use analysis sometimes, and the court said that the works have all been published. That favors fair use under the case law. On the third factor — the amount and substantiality of the portion used — presumably varies depending on who is searching what. The way the courts read the transformativeness inquiry into this third factor is to say that you can use the amount that is necessary to make your transformative use. And the court found a transformative purpose in making the entire texts of books available, saying it is “critical to the function of Google Books.” The court also mentioned that the amount of text that is provided in search results is somewhat smaller — “snippets,” the portions are called. So the court ultimately decided this factor (the amount or substantiality of the portion used) weighed slightly against fair use, but ultimately discounted it.
Regarding the effect on potential market value — you think that might be something that would vary from book to book, but the court really did not consider it that way. It treated the books in the aggregate, and I imagine that the Authors Guild may try to challenge that. But again, they argued in the class certification matter that the fair use is essentially the same in all those books.
Regarding the fourth factor (the effect on the marketplace for the copyrighted work), what effect do you believe the existence of Google’s own partner program and the Amazon Look Inside program and other similar programs may have on the Second Circuit’s analysis?
Well there is certainly precedent for the importance of that fourth factor — the effect of the use on potential marketable value of the copyrighted work — but once the courts decide that a use is transformative, they tend to discount its potential to harm a market. It is sort of a catch-22. So, the fourth factor is certainly important, but what I imagine Google will argue is that this cannot really substitute for purchase of the book because you cannot really re-read entire books using Google Books. So, it cannot supersede the object of the original book, it cannot act as a replacement for the books. But of course, couldn’t Google have bought copies of these books to scan? The court said Google needs to digitize the books in order to provide this function, and the court endorsed something that I found a little surprising, which is that it is not unusual at all for parties who want to make use of a copyrighted work and either do not want to or cannot secure permission for it, to assert fair use, using an argument that their use enhances the market. In my experience, it is not unusual in many of these cases for someone to say, “Well, the work I am using is little-known or not widely available. My use will make it more popular, more available. It will actually help sales of the work.” I had thought the courts were generally not terribly receptive to that argument, but the court here definitely emphasized that Google Books could only enhance the sales of books, thereby benefiting the copyright holder. It may be true to some extent. It might create markets, it might increase markets. But in the past, I have not seen that factor weighted like this.
As to market substitutes, maybe Google Books will not serve as a market replacement for the books as a whole, but the other question that is usually raised is whether there is a reasonably likely to-be-developed derivative market or a new market that might be impaired. And so arguably, if this decision is affirmed, presumably there will not be much of a market for searchable digital databases of books because Google will have preempted the field. The authors might argue, “Well look, if we had been left to more typical copyright infringement rules, we would have developed a private market for licensing that type of thing.” I think the court did not seem to deal with whether a private market would have evolved.
A further point is that some scholars, in thinking about what should or should not be fair use, have focused on the idea that there can be market failures in privately negotiated markets, and have argued that fair use sometimes comes into play when there is a market failure. The historical market failure in this area has been parody, where people would refuse or reject a private transaction for noneconomic reasons, basically because no one wants to be parodied. No one wants to be mocked or poked fun at. So there is a market failure and fair use jumps in.
There arguably is a market failure problem in the sheer number of works and authors involved in this case and the difficulty in identifying who the owners of some books are. It is an enormous problem. It is less of a problem with books, frankly, than it is with other works, but the “orphan works” problem is a big problem. And arguably there is a market failure going on. I do not recall the court talking about it in those terms, but I could see that lying in the background of its thinking. And that would support a fair use conclusion, really.
Turning the focus to the libraries involved here, Google has relied heavily on these libraries allowing Google to scan large portions of their collections. Are the libraries operating within the boundaries of copyright law by shipping their collections to Google and allowing Google to scan these books?
There is a whole different decision on that. It is the HathiTrust case, and this is another decision the Southern District of New York issued in October of 2012, which the Second Circuit Court of Appeals affirmed in June of 2014. The HathiTrust case is just focused on the libraries. The district court found that this kind of thing is a poster child for fair use. The judge at the end of that trial said that
I cannot imagine a definition of fair use that would not encompass the transformative uses made by [the defendant], and would require that I terminate this invaluable contribution to the progress of science and cultivation of the arts that at the same time [reflects] ideals espoused by the [Americans with Disabilities Act.]
So, that is a real, clear conclusion or strong holding that what the library is doing is fair use. And then the court concluded that because of that, there is no liability on Google’s part. Now that is an analytical approach that has been criticized, and a number of cases have said that a commercial user cannot benefit from a fair use made by the end user. So, for example, think of professors or students making copies of excerpts from books for uses in classrooms. It is very possible that that kind of reproduction of parts of works would be a fair use. But there have been cases against copy centers that create business around making copies of those works for the professors and the students. And, those cases said, the defendants’ copying is not shielded by their end users’ fair use. In this case the court did not follow that approach.
One of Google’s arguments has been that the Google Books Library Project actually advances the purposes of copyright law and the progress of science and the useful arts. Is this policy argument going to be persuasive enough to overcome the commercial use that Google can also make of the project?
If you step back from the details of the court’s statutory fair use analysis, it seems that what the court was trying to do was evaluate the value to society or the value to the progress of science, the progress of knowledge, of this type of use versus the harm to, or potential harm to, copyright owners. And in a way that is ultimately what fair use is trying to help determine, what uses should be permitted, what exceptions should be allowed in copyright because those exceptions further the purpose of the copyright power more than a finding of copyright infringement would. I think that one problem that is potentially troubling — if it does get affirmed and if the courts continue to push in this direction — is this: What we have here in a way is a new medium, we have a new way that books can be accessed. There are private markets for digital books. There is clearly a market there, and presumably a market could have been developed for searchable databases where they are not. So it is a new medium and the courts are characterizing the use for that purpose in that new medium to be a transformative one. And of course a new media’s uses of works will often be a great public benefit, right? So therefore it creates concern that a court could push this further and say any new media use is a transformative use, and therefore can be done without the copyright holder’s permission, and that does not sound like a very good policy rule. I think that what might distinguish this is that with Google Books, for the most part, only snippets are made available. And so it is not quite the same thing as freely offering up digital copies of entire works without paying the authors. So it is conceivable that my fear is unfounded, that the courts would not go as far as I just said, but it is somewhat troubling that this type of approach might be applied in that way.
What are the broad implications to the publishers specifically and content owners generally if this project is indeed a fair use?
They will probably have a strong incentive to negotiate some kind of publishing arrangement with Google. In other words, if this decision gets affirmed and there are no further proceedings favorable to the Authors Guild, then basically they are not getting paid anything and so it might encourage authors to try to make a deal with Google and try to get paid something for the use of their books.
It seems like this is the kind of thing that a functional democratic government would have tried to provide. In other words, we have the Library of Congress, which is the pre-digital version of a huge comprehensive library of works, and you would have thought that the federal government could have developed a program that would have offered this kind of thing without creating the fear that Google has a potential excessive economic power, monopoly power. So you wonder why that has not happened and I do not know why that has not happened. I am assuming the problem is that it is in a very complex area with very strong conflicting views and that Congress is not at its most functional right now. And so it seems like a massive, powerful corporation has stepped into a project that perhaps should have been a social project. Whether it is better or worse, reasonable minds could differ. Some people might feel that a private organization can provide this service more efficiently than the government could. Certainly they are making those arguments about things like health care, all kinds of other things. Prison, even. On the other hand, some people, I think, fear that it puts too much economic power in an already powerful, giant corporation’s hands.
The Copyright Office tried to get the orphan works problem addressed. So far it has not been successful. That is not the same issue, but, to some extent, overlaps with it, with some similarities in terms of market failure. And our government has not been able to come up with something that could succeed.
This case is now pending before the Second Circuit Court of Appeals. Is there an issue in this case that you think has a possibility of going up to the Supreme Court?
I would not be terribly surprised. The Supreme Court seems to be willing to take on copyright cases. They had three or four this year. And the justices have very divergent views. Justice Breyer has written that copyright should be as little as necessary to achieve its purpose, and others on the Court are much more favorably disposed towards authors and making sure that they have the incentive that copyright is supposed to provide. I think if the district court ruling is affirmed, certainly the plaintiffs will apply for Supreme Court review. It strikes me as a very important copyright topic with important social implications. The Court might want to take an opportunity to clarify what it meant by transformativeness and the place that a transformative purpose should hold in this analysis, rather than a creative transformation. Personally, I would like to see that.
If it is reversed, then presumably the Second Circuit will send it back down to the district court on some basis or other. In that case, Google would have a strong incentive to appeal it back up to the Supreme Court. It would not be the first time that the Supreme Court reversed a circuit decision and went back to the district court decision on fair use. I think in either event it is likely to get appealed to the Supreme Court and I cannot really say whether they would take it or not, but they do seem open to considering important and interesting copyright cases recently.
How do you suspect this case to come out on appeal in the Second Circuit?
It is really difficult to predict any fair use case. I think the Second Circuit does have some recent precedent involving repurposing copies of entire works — the Grateful Dead poster case, for example — that might suggest that they would be open to the district court’s analysis. That they have affirmed the HathiTrust ruling also suggests that they might affirm this decision. Really, it is nothing but pure guesswork when it comes to predicting fair use analysis.
The district court emphasized the advancement of human knowledge: the social benefit of the project versus the appropriation of the full work and any indirect commercialization of it. Is this a convincing balancing of interests?
What that risks is the possibility that if valuable digital uses can be made of works without getting permission, without payment, it might undermine the incentives to create work in the first place. And I guess if I had to venture a guess on that, I do not know if an author will be less likely to write a book because Google is going to digitize it and allow it to be searched, but it is something to keep in mind. And also there is the question of our international relations and international treaties that we belong to. Most other countries do not have a fair use doctrine, and they have more specific lists of types of things that are exempted from copyright infringement, and that is what international law, the TRIPS treaty for example, requires. So arguably this decision is inconsistent with our international treaty obligations. Now I do not know that the Supreme Court really cares about whether our law is inconsistent with international treaties that we are parties to or not in this context anyway. But it is worth mentioning that this kind of a very broad exception might be in conflict with the international law concerning exceptions to copyright.
Mark Robertson’s law practice focuses on copyright, trademark, contract drafting, and other issues common to music business transactions, as well as civil litigation involving contracts and view rights.