The line between search and research is disappearing, thanks to Google.
When I first began using computer databases for legal research 30-some years ago, composing the query string took a lot of forethought, and forethought was warranted because every minute on Westlaw or Lexis cost another small fortune.
Now, Google Scholar allows the free search of court cases using any old collection of words. Google Scholar also makes it easy to trawl through the scholarly literature in every academic field. The Google database contains whole books, too.
For example, if you enter the sentence “Brains that are unbought will never serve the public much,” Google will instantly link you to the source for that quote, Anthony Trollope’s autobiography. Clicking on the link will take you into Google Books, where you can finish Trollope’s thought: “Take away from English authors their copyrights, and you would very soon take away from England her authors.”
The very term “copyright” implies a right to control the making of copies of a protected work. So if a corporation – let’s say, Google – were to come along and make a copy of a book under copyright, without asking the author’s permission and without paying a cent in royalties, wouldn’t that be a violation of copyright law? If the same corporation were to copy a million copyrighted books, wouldn’t that be a million violations?
Well, no, according to a decision by the federal 2nd Circuit Court of Appeals that the Supreme Court recently refused to review. (Many Americans don’t realize that the Supreme Court’s primary activity is deciding which cases not to hear.) Because the 2nd Circuit covers New York, it has a special authority over America’s book publishing industry. A group of authors and publishers, along with the Authors Guild, sued Google for copying their books without permission and making excerpts freely available on the Internet.
Google couldn’t very well deny that it copied the copyrighted works. Nor did it deny that it used them for its own profit in multiple ways. The augmented database gives its search engine unrivaled breadth and sophistication. Its artificial intelligence programs analyze the scanned books to understand language usage, just as its Go-playing program analyzed historical Go games. And every page of Google Books contains a plug for Google Play, which the company bills as “the world’s largest eBookstore.”
But while making unauthorized copies for commercial purposes might seem an obvious copyright infringement, Google claimed the “fair use” defense. The copyright statute specifically allows fair use, but does so without defining the term. Many forms of fair use are familiar and uncontroversial, such as quotations in a book review or a teacher copying passages into course materials. As explained by the 2nd Circuit, all forms of fair use have two things in common: They create something new, rather than merely replicating what already exists, and they don’t function in the marketplace as a substitute for the original.
The court concluded that Google Books satisfied both criteria. It characterized the search engine’s results as information about the copyrighted books, which sounds plausible right up until the moment you realize the “information” about the text is the text itself. Indeed, the search engine would be useless if it failed to reproduce the original exactly.
The court was on firmer ground when it held that Google Books didn’t provide a substitute for the copyrighted original because it provides such a truncated view. As Google Books users know, older books appear in their entirety, while books under copyright appear only in excerpts.
The 2nd Circuit explained that Google divides each page of a standard-sized book into eight parts, which it calls snippets. The court asserted as a “fact that no more than three snippets are shown – and no more than one per page – for each term searched.” In addition, according to the court, one snippet on every page and one complete page out of ten are never displayed, a process Google calls “blacklisting.” According to the court, nobody could ever read more than 16 percent of a book in snippet view. It held that, under those stringent conditions, Google Books don’t diminish the commercial value of the copied books.
That conclusion seems sensible enough on the facts as given. But real-world conditions have changed considerably since the lawsuit was filed in 2005. In 2012, Google reached an agreement with American publishers, who generally don’t own the copyright to the books they put out. The agreement “allows users to browse up to 20% of books,” according to Google’s own press release. So much for the 2nd Circuit’s 16 percent. And when I looked up a recent book I admire, Jessica Stern’s “Denial: A Memoir,” and searched for a word I know appears in it frequently, Google Books presented me with 17 snippets, not the three the 2nd Circuit assured me would be the maximum.
In theory, the authors could file another lawsuit based on the changed conditions today, pitting their resources against those of the largest corporation in the world (by the fickle measure of market capitalization), but conditions would just change again before that case was ever decided.
More broadly, if copyright holders have to sue in order to influence the distribution of their own work, it can only mean some degree of control over their work has already passed from them. They didn’t agree to give it up, they received no compensation for it, but it’s Google’s now.
In law school, students are taught that property rights are a bundle of sticks, each of which can be sold or given away separately. Google has successfully acquired a stick or two from every author in the country, and it did so by the simple device of taking them.
Joel Jacobsen is an author and has recently retired from a 29-year legal career. If there are topics you would like to see covered in future columns, please write him at email@example.com.