Blog

October 26, 2015 -

Google Books Case Transforms the Fair Use Standard

In Authors Guild v. Google, Inc., No. 13-4829-cv (2nd Cir.  Oct. 16, 2015), a groundbreaking case involving the Google Books project, the U.S. Court of Appeals for the Second Circuit has “transformed” and extended the doctrine of copyright fair use for digital media.

For its Google Books project, Google has made digital copies of tens of millions of books submitted to it by participating libraries, scanned the digital copies and created a public search function allowing Internet users to search for specified words or terms.  Google also enables users to see brief snippets of text from the books which provide contextual information on how the books used the searched-for terms.  Google allows the participating libraries to download and retain digital copies of the books they submit, subject to contracts requiring the libraries not to use their digital copies in violation of the copyright laws.  The Authors Guild had originally brought suit against Google in 2005 as a class action on behalf of rights-owning authors.  In 2011 the federal district court for the Southern District of New York rejected a proposed settlement as unfair to authors who were members of the class.  In 2013, however, the district court granted Google’s motion for summary judgment dismissing the suit and held that the uses made by Google of the copyrighted books were non-infringing transformative fair uses.

On October 16, the Second Circuit upheld the lower court’s ruling, and in so doing, essentially revamped the legal standard for transformative fair use so that the driver is now whether or not the use of copyrighted material being complained of is a meaningful substitute for the original.  The court also clarified the difference between a (non-infringing) transformative fair use and an (infringing) derivative work, holding that a derivative work (which the copyright holder retains the exclusive right to create under the Copyright Act) represents a transformation only in form (for example, the adaptation of a book into a screenplay or a movie, or the translation of a book from one language to another), whereas a use is genuinely “transformative” in the fair use sense only where it so fundamentally changes the nature and/or purpose of the original material that there is unlikely to be substantial substitution of the two (or for derivative works of the original) by consumers.  The court held that an author’s rights over derivative works do not include an exclusive right to provide information about a copyrighted work of the type provided by Google’s search function.  Finally, the court raised the possibility that data security risks could negate a finding of fair use if hacking results in widespread access to digital copies of copyrighted works created without the author’s permission (although the court found for Google on this particular issue because, not surprisingly, Google’s data security measures were robust).

Google’s snippet function significantly limits a user’s ability to read text from the books being searched.  For example, each snippet is roughly one-eighth of a page, no more than three snippets are shown in response to a search (and no more than one per page), one snippet per page is blocked from snippet view, and approximately 22% of each book’s text is permanently blocked.  Through aggressive and cumbersome efforts, the plaintiffs’ searchers were only able to reveal 16% of the text of their books through snippet view, and the displayed elements were scattered and could not be assembled into anything resembling a coherent, readable whole (i.e., anything that could substitute for the original book or a licensed derivative).  In addition to the snippet function, Google’s search function also provides research tools such as the “ngrams” tool that furnish statistical information about word and phrase usage and how this changes over time and across linguistic regions.

Applying the Fair Use Standard

17 U.S.C §107 sets out four factors for courts to evaluate and balance against one another in determining whether a particular unauthorized use of copyrighted material is a fair use:  (1) “the purpose and character of the use,” including whether such use is commercial or non-profit, (2) the nature of the copyrighted work, (3) the “amount and substantiality of the portion used” in relation to the copyrighted work, and (4) the effect of the use upon the potential market for or value of the copyrighted work.  The fourth factor, which the appeals court in Authors Guild stressed as the most important one, also requires a consideration of the impact of the unauthorized use on the market for licensing derivative works.  Overall, the purpose of the fair use standard is to strike a balance between the author’s and the public’s interests (in the words of the Second Circuit, “to expand public learning while protecting the incentives of authors to create for the public good”).   The court noted that transformative uses tend to favor a fair use finding “because a transformative use is one that communicates something new and different from the original or expands its utility, thus serving copyright’s overall objective of contributing to public knowledge.”

In analyzing the first fair use factor (purpose and character of the use), the court drew an important distinction between transformative uses and derivative works, upping the ante in a controversy bubbling within the federal court system over transformative fair use.  The U.S. Court of Appeals for the Seventh Circuit (covering Indiana, Illinois and Wisconsin) had previously criticized the Second Circuit’s emphasis on transformativeness (particularly in the search context) as dominating the whole fair use analysis, pointing to the fact that the statutory definition of a derivative work (which is included in the exclusive rights of copyright) in 17 U.S.C §101 includes not only a dramatization, translation, motion picture version, sound recording, abridgement or condensation, but also “any other form in which a work may be recast, transformed or adapted.”  The Second Circuit responded that derivative works generally involve transformations in the nature of changes of form (such as reproducing a painting in the form of a postcard or a cartoon character in the form of a three-dimensional plush toy), whereas “copying from an original for the purpose of criticism or commentary on the original or provision of information about it” shows a transformative purpose.  Based on this standard, and extending previous caselaw finding that the digitization of copyrighted material for purposes of creating a searchable repository was highly transformative, the court had no problem concluding that the first fair use factor weighed strongly for Google in terms of its search and snippet view functionality and associated analytics.  Finally, the court rejected the notion that Google’s commercial purpose should defeat a finding of fair use; the more transformative the use, the less the existence of a profit move matters.

Heavily influencing the court’s analysis of all four fair use factors (and not just the first and fourth, as might be expected) was its sense that none of Google’s activities competed with the plaintiffs’ commercialization of their copyrighted books or offered any kind of meaningful substitute.  Time after time the court returned to the “absence of significant substitutive competition” in determining that a particular factor weighed in favor of Google.  If a secondary use “transformatively provides valuable information about the original, rather than replicating protected expression in a manner that provides a meaningful substitute for the original [or a licensed derivative work],” this drives the analysis toward a finding of fair use.   According to the court, neither Google’s search function nor the snippet view reveals enough text (or text in a sufficiently ordered manner) to present readers with a coherent, workable substitute for the creative and expressive (i.e., protected) content of a book.  This line of reasoning represents both a clarification and an expansion of the transformative fair use doctrine.

In applying the fourth (market impact) factor of the fair use analysis to Google’s search and snippet view functions, the court acknowledged that the snippet view could result in some loss of book sales.  For example, if all a searcher was looking for was an historical fact such as a date, the snippet view might satisfy this need, thereby eliminating any need to buy the book or acquire it from a library.  However, the court found that this did not prevent the fourth fair use factor from weighing in favor of Google, since any market harm was speculative and non-substantial, and (as in the above example) might occur in relation to interests that are not even protected by copyright.  (Facts like historical dates are not copyrightable, and thus there is no infringement even if they are taken from a copyrighted book.)   The court also rejected the plaintiffs’ contention that copyright holders have exclusive derivative rights in the application of search and snippet view functions to their works, including through queries of digitized copies.  Owning the copyright in a work does not equate to owning the right to supply search information and analytics about it in the way that Google does.

Data Security/Copyright Convergence

The plaintiffs also raised a novel argument:  Google’s storage of digitized copies of the plaintiffs’ books exposed them to the risk that hackers might gain access and make the books widely available, thus destroying the value of the plaintiffs’ copyrights.  Interestingly, the court agreed that this claim had a “reasonable theoretical basis,” and posited that if a secondary user failed to use reasonable efforts to secure digital material that had been copied without permission, “this might well furnish a substantial rebuttal to the secondary user’s claim of fair use.”  In Google’s case, however, any risk was speculative, since Google showed that the digital scans of the plaintiffs’ books were protected by the same “impressive” security measures as it uses to guard its own confidential information.  Furthermore, Google’s contracts with the participating libraries under which it provided the digital copies to them committed them to use the copies only in a manner consistent with copyright law and to take precautions to prevent public dissemination of the copies, which satisfied the court.

Nevertheless, the court’s opinion indicates that the adequacy of a secondary user’s data security measures is a legitimate consideration in the fair use analysis, since the widespread release of unauthorized digital copies employed in what might otherwise be a transformative fair use could cause market harm to the copyright owner and thereby tip the fourth fair use factor in its favor.  Examination of the sufficiency of a defendant’s data security measures has the potential to open a new dimension in future digital fair use cases.  Not only is its intentional dissemination of copyrighted material now relevant, but so are the protections it uses to prevent unintended uses by unrelated third parties.

A Transformed Standard?

On its face the Second Circuit’s opinion does not change the fair use standard or upend any existing caselaw (other than criticizing the Seventh Circuit’s recent line of opinions skeptical of an expansive transformative fair use doctrine).   However, one does get the sense that the universe of considerations has been distilled and, at least in the Second Circuit, transformative fair use now seems to boil down largely to whether a transformation beyond a mere change in form has occurred and whether this is meaningful enough to render the secondary use non-competing with the original material and its potential derivatives.  It will be particularly interesting to see whether the influential Ninth Circuit (which includes the tech-heavy states of California and Washington) follows the Second Circuit’s reasoning in the Authors Guild case.

Please email andrew@baercrossey.com for more information.