DISCLOSURE/DISCLAIMER: I worked at the University of Michigan Libraries for several years during the beginning of the book scanning project with Google that is the subject of this lawsuit. The University of Minnesota Libraries, my current employer, is also affiliated with the Google scanning project and HathiTrust Digital Library. This post represents only my own opinions and thoughts on the case, not that of any current or past employer or co-worker.
This morning, Judge Denny Chin issued an opinion in the massive, many-year lawsuit between the Author’s Guild and Google over the Google book scanning project. Just a couple years ago, many possible outcomes were anticipated – and few of them would have provided much clarity for anyone except Google.
Instead, we got a ruling today which dismissed the case, ruled several different research and public interest activities were fair
use, and in general, was very positive about the public benefits that
this project has produced.
Quick Aside on Court Procedure
Despite the many years of booting around various class action certifications and possible settlements, the end result is that this case was dismissed fairly early in the full court process. As with the HathiTrust ruling from last fall, this opinion dismisses the case on summary judgment – that means the court has determined that there are no substantial factual points in contention, and the legal arguments so clearly favor one party that there is no need for a trial. Appealing a dismissal on summary judgment is generally a weaker place to be than appealing an opinion after a trial. (Though of course, in the HathiTrust case, the Authors Guild has indeed decided to appeal, and it will likely do so here as well.)
Fair Use Analysis
The main issue on summary judgment here was whether Google’s book scanning was fair use. Judge Chin affirms that it is, for several different reasons. He appears to have been influenced both by amicus briefs from a variety of public interest organizations, as well as the HathiTrust ruling from last fall.
Judge Chin’s fair use discussion begins by reminding us that it is “copyright’s very purpose, ‘[t]o promote the Progress of Science and Useful Arts.”” He cites a variety of case law (as well as Leval’s seminal fair use article) affirming that both incentives to authors -and- opportunities for others to use “protected works” are necessary to achieve that purpose (p. 16-17). As an advocate for the public interest in copyright, it is always heartening to see courts acknowledging that progress is the main goal, and that protection for creators is only one part of a considerably larger equation. (A frequent theme of my teaching is that we are all users and creators – creativity is an ecosystem.)
Judge Chin does dig in to the traditional four fair use factors, although on more than one occasion he notes that they are non-exclusive, and that other relevant considerations should also be weighed (p. 18, p. 25).
For the purpose factor, Judge Chin focuses quite a bit on transformative use, and notes approvingly that “Google’s use of the copyrighted works is highly transformative” (p. 19). He approves of “the use of book text to facilitate search through the display of snippets” (p. 19), as well as turning “book text into data for purposes of substantive research, including data mining and text mining” (p. 20) as transformative uses.
Chin’s fair use analysis follows a line of cases from the 9th Circuit that looked at thumbnail uses of images as transformative fair use – and one outcome that the publisher/plaintiffs may have been looking for by bringing this lawsuit in the generally-more-conservative 2nd Circuit was to create a “Circuit split” on this issue (which often opens up avenues for Supreme Court appeal.) This could possibly still happen on appeal, but it seems pretty unlikely at this point. Instead, as commentator/law prof James Grimmelmann summed it up in a tweet: “I feel safe in saying that search indexing and snippet display are now definitely fair uses.”
Another aside: I am very glad to see that these purposes were ruled transformative and in favor of fair use. However, I was also glad to see affirmed that “transformative use is not ‘absolutely necessary’ to a finding of fair use” (p. 19). Multiple avenues for arguing fair use is a good thing.
The court also discusses Google’s commercial purpose, finding that it “does […] benefit commercially” but “does not engage in the direct commercialization of copyrighted works” (p. 22). Even considering Google’s profit motive, the educational and transformative purposes of the scanning project led Judge Chin to conclude “the first factor strongly favors a finding of fair use.” (p. 22)
As to the nature of the copyrighted works, Judge Chin only briefly engages with the factual/creative angle on this question, noting that the “vast majority” of the books are non-fiction. He also notes that all the books have been published, and concludes that this factor favors fair use.
As to the amount used, Judge Chin notes that Google does scan the full text of books, and does reproduce the books verbatim. But, he notes, copying an entire work -can- be fair use sometimes, and “full-work reproduction is critical to the functioning of Google Books” (p. 23). He notes approvingly that only limited amounts of text are ever displayed, and concludes that the amount factor does weigh “slightly against” fair use (p. 24).
Finally, Judge Chin considers the possibility of market harm. He directly rejects the publisher-plaintiffs’ argument that people could use Google Books to piece together a whole book for reading, and states that “a reasonable factfinder could only find that Google Books enhances the sales of books to the benefit of the copyright holders” (p. 25). Notably, he does not engage with the argument discussed in HathiTrust, that scanning without a license harms sales of licenses to scan (although I’m not sure if that was argued in the summary judgment briefs.) He concludes that the fourth factor “weighs strongly in favor” of fair use (p. 25).
After looking to the four factors in detail, Judge Chin enumerates the “significant public benefits” of the project as another relevant consideration. (p. 26) The various benefits he highlights includes that it makes books more findable, that it allows text-mining, that it preserves books (especially out of print ones), that it facilitates access to books for people with print disabilities and for “remote or underserved populations”, and that it generates income for copyright holders. “[A]ll society benefits.” (p. 26)
Because the plaintiffs also argued in their summary judgment motion that Google was committing contributory infringement by providing scanned copies of books to libraries, Judge Chin also (in quite short order) affirms that library uses are lawful uses, transformative uses, and that they advance the arts and sciences, etc. (p. 27). He states that the fair use analysis of the HathiTrust decision “applies here as well” (p. 28), so neither the libraries receiving the scans, nor Google in providing the scans to libraries, are doing anything wrong.
Library and Cultural Institution Implications
The great thing about this case moving away from class action settlement and towards actual court rulings on the substantive legal issues, is that the resulting rulings do not just apply to Google. If it’s fair use for Google to do this kind of stuff, it probably is for others as well.
Like a lot of recent fair use cases, this case affirms the public interest elements of copyright, and how closely fair use is connected to those public interest elements, in kind of screamingly strong language. For institutions that have been reluctant to engage with fair use, this opinion, and the HathiTrust opinion of last year, are extremely strong grounds for contemplating the application of fair use to digitization projects, exhibits, and other such publicly beneficial uses. (The more so because most libraries and cultural institutions are entirely non-profit, unlike Google which has acknowledged commercial purposes.)
This decision, by again affirming the 9th Circuit precedents in a different circuit, also lends weight to many recent best practices or other statements on use of materials in research or museum contexts that have relied on transformative use and related arguments. Examples include the ARL Code of Best Practices in Fair Use for Research Libraries, the AAMD Policy on the use of thumbnail images, the VRA Statement on the Fair Use of Images for Teaching, Research, and Study, and several others.
One fun thing to note is that if you’ve been thinking, “Well, Section 108 gives us broad preservation digitization rights, but then we can’t do anything with the resulting digital files” – hey, making those files searchable, and sharing snippets at least, is looking like a strong fair use! (What’s more, digitizing some things that you don’t think would be covered by section 108 preservation rights might be fine, too! It might have been before these rulings – but if you were feeling timid before, for goodness’ sakes, GET OVER THAT now. If your institution has been reluctant to embrace the risks and uncertainties of fair use, these decisions are reducing the uncertainty, a lot. Or, you know, stick your head in the sand and wait for appeal, or wait for a Supreme Court ruling, or…)
Edited to add: also, negotiating for preservation of statutory use rights in ALL of our licensing agreements is EVEN MORE IMPORTANT than it was before, GLAM folks!!!!
I’m sure there’s more to think about here. I note that several colleagues have posted analyses of the decision that I haven’t yet had time to read. My reading list now:
- Matthew Sag
- Timothy B. Lee
- Ariel Katz
- Eric Goldman
- surely others…
