This month, we discuss Authors Guild v. Google,1 in which the U.S. Court of Appeals for the Second Circuit decided whether Google's Library project and Google Books, through which Google creates unauthorized digital copies of books, as well as a search functionality and the display of “snippets” from those books, constitute copyright infringement.

In an opinion by Judge Pierre N. Leval, joined by Judges José A. Cabranes and Barrington D. Parker, the court affirmed the district court's opinion, holding that Google's actions constitute fair use under the Copyright Act and the U.S. Supreme Court's holding in Campbell v. Acuff-Rose Music.2 The Second Circuit explicitly found that Google's purpose in copying is highly transformative, that Google provides a limited public display of text, that its copying does not constitute a significant market substitute for the protected aspects of the original works, and that Google's profit motivations do not justify denial of fair use. The court further held that Google's provision of digitized copies to participating libraries, by way of contracts requiring libraries' compliance with copyright law, also does not constitute infringement.

Background

In 2004, Google initiated its Library Project and the Google Books search engine. Through bilateral agreements, participating research libraries send books to Google. Google then creates a digital scan of each book, extracts a machine-readable text, and prepares an accompanying index. This digital corpus is saved on servers protected by the same security system that safeguards Google's confidential information. Through the search engine, users can search for terms and obtain a list of books from Google's database in which they appear. Its aim is to permit the public to identify which books, among millions, use terms selected by the user, as well as those that do not. Links to purchase books online or names of carrying libraries are sometimes provided, but Google receives no payment in exchange for a user's purchase of the book. The search engine also enables “text mining” and “data mining,” through which Google can identify the frequency of word and phrase usage over time.

The search function displays three snippets of text—each roughly an eighth of a page—from every book containing the search terms. A search for a particular term will always reveal the same three snippets from a book. Google makes unavailable one snippet per page, and one page out of ten. Google also disables snippet view for books such as dictionaries and cookbooks, and allows copyright holders to request exclusion of their works from snippet view.