Last week, Judge Denny Chin handed down the latest opinion in the now-eight year battle between Google and the Author’s Guild (among others) over Google’s massive book scanning project. If the Author’s Guild fails to overturn the Judge’s decision on appeal, it will mark an enormous watershed in the ability of Web site owners to display copyrighted works without the prior permission of the owners of those works.
At issue was the appropriate application of the “fair use” doctrine under U.S. law to the Google project, a rationale that allows certain types of copying to be permissible that would otherwise be actionable. As applied by Judge Chin, the scope of that doctrine has seemingly been expanded by orders of magnitude. Indeed, in the case at hand, the judge has broadened its scope so dramatically that it’s difficult not to conclude that he was struggling to find sufficient legal precedents to justify a favorable outcome for Google. Many will contend that he fell short in that effort, and that his intent was instead to rebalance, if not rewrite, the doctrine itself in order to bring it into the Internet age.
The need for authoritatively clarifying the bounds of the fair use doctrine has been more than apparent for some time, given the vagueness of the applicable legal test, the conservative historical limits of the fair use defense, and the radically transformed opportunities for copying, repackaging, and re-publishing the Internet provides. Prior to the digital era, fair uses of text were limited in a commercial context to just a few words, or (in the case of a book review) a paragraph or so. In the case of music, more than a quick “quote” by a jazz musician could land the instrumentalist in court. And the means to exercise fair use were similarly limited: printed text media, recorded music, art reproductions and the like. Finally, where the borrowing was significant, value was often clearly going to the seller doing the borrowing rather than the author of the original.
Have YOU Discovered the Alexandria Project?
Unfortunately, under the strict letter of copyright law, copying any amount of text is an open and shut violation of the author’s rights. If that author (in the U.S.) has filed a copyright registration, such minimal copying can make the infringer liable to the content owner for up to $150,000 in statutory damages without the owner being required to show that she has suffered any actual economic harm at all. That is, unless the borrower of the text can prove that it should be excused under the fair use doctrine – an accretive creature of judicial opinions, eventually codified by Congress in 17 USC § 107. As stated in § 107:
[T]the fair use of a copyrighted work, including such use by reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright.
The statute goes on to direct that in determining whether a given use is a “fair” use, the factors to be taken into account are to include:
(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
(2) the nature of the copyrighted work; [e.g., fiction, non-fiction, image, etc.]
(3) the amount and substantiality [i.e., percentage] of the portion used in relation to the copyrighted work as a whole; and
(4) the effect of the use upon the potential market for or value of the copyrighted work.
With this by way of prelude, let's take a look at what Google has actually done. If a book is out of copyright, the entire book is viewable on line, and can even be downloaded for free. But if it is still protected by copyright, unless the content owner has given permission to display more, it can be viewed only in "snippet view" - meaning that only a few instances of the text containing searched words are displayed, as well as 50 or so surrounding words. Google likens the project to the creation of a card catalog of books (although the creation of a card catalog entry hardly requires the copying of an entire book - only brief access to it).
As you'll see, snippets aren’t particularly useful (scroll down the page), and certainly not in comparison to buying the entire book. That said, if the owner of an in-copyright work requests, the book will be removed entirely. Conversely, an owner can give permission to display more text (several pages, or even the entire book), and Google will also provide a link to a page where the book can be purchased. Here's an example.
One might wonder why the Author's Guild is so upset, if only low-value snippets can be displayed without the content owner's permission, and if Google is helping authors of out of print books be able to gain some income from their back list? One reason is that although Google is currently only displaying snippets, there's nothing to prevent it from deciding to begin displaying complete pages tomorrow. Or for some future owner (or hacker) of the database to begin doing so. And importantly, from a legal perspective, Google had to scan - and therefore copy - each book before it could index it. This is a clear violation of copyright law, giving the Author's Guild and other content owners a legal argument to wield. Given the fact that under copyright law the owner of a work has the exclusive right to grant copying privileges, it's admittedly a bit difficult to square copying 20 million books with fair use, no matter how little is displayed without owner permission. But Judge Chin rose to the challenge, accepting Google's assertion that the mass copying was a necessary first step to the actual fair use of displaying the content.
He was clearly swayed by the first test, and the undeniable public benefit that must follow from making virtually all of recorded knowledge available to the peoples of the world at the click of a mouse. He also noted that Google neither sells nor runs ads against accessed text. At best, however, this part of the test is useful only as evidence of whether the republisher’s motive was truly educational or otherwise public spirited, since the economic impact on the seller is the same either way. After this first conclusion, Judge Chin’s efforts become more labored.
For example, with respect to the second test, Judge Chin noted that the copying of non-fiction has generally been accorded greater respect under the fair use doctrine than fiction, and that “most” of the books copied comprised non-fictional work. But citing that factor as meaningful is at best disingenuous. Just because more of the works copied are non-fiction doesn’t justify copying even one of the fictional works, which could easily have been excluded. And clearly making only the 13 million or so works of non-fiction accessible would still be enormously beneficial even if readers still needed to visit Amazon or a bookstore to scan a section of novel.
Judge Chin takes a similarly glib approach to the third test, asserting that the Google book viewer is not suitable for reading a book, because only "snippets" of text are displayed. That’s true so long as Google displays only snippets, but if Google wanted to change its idea of what fair use should allow, it could begin displaying more tomorrow. And while the text might not be quite as pleasantly displayed as on a Kobe or a Nook, it is certainly readable.
Instead, Judge Chin focuses on the “highly transformative” nature of the Google service, which allows searching for specific topics or text and, if the book is out of copyright or the owner has been given permission, reading more than just a snippet. But if the books are in copyright and that permission has not been given, it's questionable whether the very low value of a few random snippets justifies fair use at all, much less the previous copying of the entire book.
In another creative leap, Judge Chin cites an earlier decision permitting fair use of an image as precedent for copying snippets of text, analogizing them not to repurposed sentences, but to an image of those sentences. But a sentence is not converted from text to an image simply because it is viewed on a screen rather than on the page of illegally copied book. To your eye, it is a sentence and an image in both cases. He also cites a 1984 Supreme Court decision that held that it was a fair use to copy a TV show for later viewing. But allowing a single viewer to make a single copy for their own use is a far cry from exposing text to the billions of people with Web access. Instead, it is closer to the type of sampling that has in the past been found to be copyright infringement, generating myriad take down notices honored by Google at the Youtube section of its site.
On the last test, Judge Chin returns to somewhat firmer ground. He is not so naïve as to presume that Google does not expect to reap indirect economic benefits from the scanning project. But he also fairly observes that millions of the books copied by Google have been forgotten and are no longer available for sale. Absent a service such as Google’s, most would likely never be widely accessed again, whereas at least some percentage of the researchers that use the Google service will go on to purchase a copy of the occasional book that is still available. On this point, Judge Chin concluded that “a reasonable fact finder could only find that Google Books enhances the sales of books to the benefit of copyright holders.”
All in all, Judge Chin’s opinion is breathtaking in its willingness to break new ground. Whether you regard his holding as the act of a visionary judge willing to take a necessary step to update the application of the fair use doctrine or a case of judicially assisted piracy is a matter upon which people will certainly differ. Certainly an argument could be made that such a significant extension of the four factors of the fair use doctrine might be a more appropriate task for Congress than a single judge (that of course assumes the existence of a working Congress). But to the extent that Congress has revealed any indication of whether it is inclined to favor the rights of content owners or content consumers, it did so by substantially extending the period of copyright protection. But for that action, millions of the books subject to the Author’s Guild suit would already be in the public domain. That’s hardly an invitation to the courts to widen the scope of fair use.
Be that as it may, if other federal judges follow Judge Chin’s lead, the rules regarding the publishing of content on the Internet have been radically expanded. Indeed, where there is no obvious commercial benefit to be derived by a site owner from the placement of borrowed content, it’s difficult to see why that site owner should be any more at risk to a copyright claim than Google. After all, a snippet of text is a snippet of text no matter where it is hosted, so long as it’s equally accessible to a browser. Either way, the economic impact, if any, on the content owner are the same. And why stop at a snippet? If there's a magic line to be drawn, surely it must be on the other side of larger blocks of text than before.
It will be interesting to see what follows from this decision. Certainly the level of dismay of content owners will be matched by the exuberance of the delight of advocates of free content. Given that the former are better funded than the latter, though, it’s difficult to believe that Judge Chin’s conclusions will be allowed to propagate into other Federal circuits without a fight.