On October 16, 2015, the Second Circuit issued an opinion holding that the actions of the Google Library Project, which creates and retains digital copies of books and allows users to search and see snippets of text, constituted a “fair use” under 17 U.S.C. § 107.


Google’s Library Project is a service that creates and retains digital copies of books sent to it through agreements with major libraries. Google scans each book sent to it, creates a machine readable file of the book, and indexes the text of the book. These digital copies are then indexed on the Google Books website, where users can search the library for specific words or terms.  The site returns information including a list of books containing the terms and term frequency within each book. Google Books also allows users to see “snippets” of the text surrounding the search terms. Google’s use of millions of books for this project is done without Google obtaining authorization from the copyright holders.

In this long-running dispute, Plaintiffs, authors of various work scanned by Google, brought suit alleging copyright infringement. The district court granted Google’s motion for summary judgment that its use of the works constituted protected fair use.

Second Circuit Opinion

On appeal, the Second Circuit affirmed. Among other arguments that were addressed, the opinion (written by Judge Leval, with Judges Cabranes and Parker on the panel)  closely analyzed the four established fair use factors, as applied to both the “search” and “snippet” functions in Google Books.

As to fair use, the Court heavily relied on the Supreme Court’s analysis of the factors inCampbell v. Acuff-Rose Music, 510 U.S. 569 (1994), and found that all four statutory fair use factors -- the purpose and character of the use, the nature of the copyrighted work, the amount and substantiality of the portion used, and the effect of the use on the potential market -- supported fair use.

As to the first factor, the Court focused on whether Google’s use of the works was transformative.  It stated that “[w]e have no difficulty concluding that Google’s making of a digital copy of Plaintiffs’ books for the purpose of enabling a search for identification of books containing a term of interest . . . involves a highly transformative purpose”, in that the purpose is “to make available significant information about those books”. The Court also found that the snippet function  “adds important value to the basic transformative search function” by being  “designed to show the searcher just enough context… to help her evaluate whether the book falls within the scope of her interest (without revealing so much as to threaten the author’s copyright interests).” The Court rejected the argument that Google’s overall profit motivation in enhancing its market position undercut its “highly convincing transformative purpose”.

As to the second factor, the nature of the copyrighted works, the Court stated that this factor had no influence “in isolation”, but it “necessarily combines” with the purpose and character of the secondary work to help determine whether there is a transformative use.  From that standpoint, the Court found the second factor favored fair use because Google’s use “transformatively provides valuable information about the original” rather than replicating protected expression to provide a substitute for the original.

With respect to the third factor, the amount and substantiality of the portion used, the Court found that Google’s copying of the totality of the works was both “reasonably appropriate to Google’s transformative purpose” and “literally necessary to achieve that purpose”.  As to the snippet view function, the Court noted that the relevant inquiry is not the portion of a work used in making a copy, but instead the amount “made accessible to the public.” In finding that this factor favored fair use, the Court relied on Google’s use of measures that “substantially protects against its serving as an effectively competing substitute for Plaintiffs’ books.”  Although 78% of a book is theoretically accessible to a searcher on Google Books, in practice a searcher could access only in aggregate about 16% of the text of a book in “fragmentary and scattered” snippets that communicated little sense of the original work.  The Court noted that if snippet view could be used to show “a coherent block amounting to 16% of a book” that would raise a very different question.

In addressing the fourth factor, effect upon the potential market, while the Court recognized that the snippet function could cause some loss of sales, it could not provide a searcher with an “effectively competing substitute” for the original book that would have a “meaningful or significant effect” on its sales. In addition, the Court observed that the type of loss of sales envisioned would generally involve interests not protected by copyright, such as researchers searching for unprotected facts gleaned from a snippet, not the author’s copyright expression.