.
The Google Library Project: Is Digitization for
Purposes of Online Indexing Fair Use Under
Copyright Law?
Kate M. Manuel
Legislative Attorney
September 25, 2009
Congressional Research Service
7-5700
www.crs.gov
R40194
CRS Report for Congress
P
repared for Members and Committees of Congress
c11173008
.
The Google Library Project
Summary
The Google Book Search Library Project, announced in December 2004, raised important
questions about infringing reproduction and fair use under copyright law. Google planned to
digitize, index, and display “snippets” of print books in the collections of five major libraries
without the permission of the books’ copyright holders, if any. Authors and publishers owning
copyrights to these books sued Google in September and October 2005, seeking to enjoin and
recover damages for Google’s alleged infringement of their exclusive rights to reproduce and
publicly display their works. Google and proponents of its Library Project disputed these
allegations. They essentially contended that Google’s proposed uses were not infringing because
Google allowed rights holders to “opt out” of having their books digitized or indexed. They also
argued that, even if Google’s proposed uses were infringing, they constituted fair uses under
copyright law.
The arguments of the parties and their supporters highlighted several questions of first
impression. First, does an entity conducting an unauthorized digitization and indexing project
avoid committing copyright infringement by offering rights holders the opportunity to “opt out,”
or request removal or exclusion of their content? Is requiring rights holders to take steps to stop
allegedly infringing digitization and indexing like requiring rights holders to use meta-tags to
keep search engines from indexing online content? Or do rights holders employ sufficient
measures to keep their books from being digitized and indexed online by publishing in print?
Second, can unauthorized digitization, indexing, and display of “snippets” of print works
constitute a fair use? Assuming unauthorized indexing and display of “snippets” are fair uses, can
digitization claim to be a fair use on the grounds that apparently prima facie infringing activities
that facilitate legitimate uses are fair uses?
On October 28, 2008, Google, authors, and publishers announced a proposed settlement, which, if
approved by the court, could leave these and related questions unanswered. However, although a
court granted preliminary approval to the settlement on November 17, 2008, final approval is still
pending. Class members had until September 4, 2009, to file objections with the court. The court
will consider these objections, as well as conduct an independent review of the proposed
agreement, in determining whether to grant final approval to the settlement. The final hearing on
the settlement is presently scheduled for October 7, 2009.
Congressional Research Service
.
The Google Library Project
Contents
Introduction ................................................................................................................................ 1
The Google Library Project ......................................................................................................... 1
The Litigation and the Parties’ Positions...................................................................................... 2
Legal Issues Raised by the Litigation .......................................................................................... 3
“Opt Out” Programs and Liability for Infringement ............................................................... 3
Digitization, Indexing, and Display as Fair Uses.................................................................... 4
The Purpose and Character of the Use ............................................................................. 5
The Nature of the Copyrighted Work............................................................................... 7
The Amount and Substantiality of the Portion Used......................................................... 8
The Effect of the Use Upon the Potential Market or Value of the Work ............................ 8
The Proposed Settlement Agreement ........................................................................................... 9
Opposition to the Proposed Settlement ................................................................................ 10
Support for the Proposed Settlement.................................................................................... 13
Potential Effects of Rejection or Approval of the Settlement................................................ 14
Contacts
Author Contact Information ...................................................................................................... 14
Acknowledgments .................................................................................................................... 14
Congressional Research Service
.
The Google Library Project
Introduction
Authors and publishers sued Google Inc. in 2005, shortly after Google announced plans to
digitize books in the collections of several major libraries, index them in its search engine
(http://www.google.com), and allow searchers to view “snippets” of the digitized books. Google’s
proposed reproduction and display of copyrighted books was not authorized by the rights holders,
who alleged that the Google Library Project infringed their copyrights. Google’s
counterarguments—that allowing rights holders to “opt out” of having their books digitized or
indexed kept its proposed uses from being infringing, or that, if found to be infringing, its
proposed uses were fair—raised important questions about reproduction and fair use under
copyright law. Namely, does an entity engaged in unauthorized digitization and indexing avoid
committing copyright infringement by offering rights holders the opportunity to request removal
or exclusion of their content? And, assuming unauthorized indexing and display of “snippets” are
fair uses, can digitization claim to be a fair use on the grounds that apparently prima facie
infringing activities that facilitate legitimate uses are fair uses? The proposed settlement
agreement between Google and rights holders could mean that litigation over the Library Project
does not help to answer these questions. However, final court approval of the settlement is still
pending, and future digitization and indexing projects may raise similar questions.
This report provides background on the Library Project, legal issues raised by digitization and
indexing projects, and the proposed settlement. It will be updated as developments warrant. It
supersedes CRS Report RS22356, The Google Book Search Project: Is Online Indexing a Fair
Use Under Copyright Law? by Robin Jeweler.
The Google Library Project
In December 2004, Google initiated its Library Project by announcing partnerships with five
libraries.1 Under the partnership agreements, the libraries would allow Google to digitize the print
books in their collections, and Google would (1) index the contents of the books; (2) display at
least “snippets” of the books among its search results; and (3) provide partner libraries with
digital copies of the print books in their collections.2 Google and its partners never planned to
make the full text of any digitized and indexed books that are still within their terms of copyright
protection available to searchers.3 Rather, by digitizing and indexing books, Google and its
partners sought to make the contents of print books more accessible to searchers, who could
potentially buy or borrow books after seeing “snippets” of them among the results of Google
searches.4 Google also intended to sell advertising “keyed” to results lists incorporating the
digitized books.5
1 Google Checks Out Library Books, Dec. 14, 2004, available at http://www.google.com/press/pressrel/
print_library.html. Participating libraries included those at the University of Michigan, Harvard University, Stanford
University, and Oxford University, as well as the New York Public Library.
2 Id.
3 Id. Copyright protection for books generally lasts “for a term consisting of the life of the author and 70 years after the
author’s death.” 17 U.S.C. § 302(a).
4 Google Checks Out Library Books, supra note 1.
5 Id.
Congressional Research Service
1
.
The Google Library Project
Google’s Library Project was itself part of a larger initiative initially known as Google Print and
later renamed Google Book Search.6 The Google Partner Program was also part of this initiative.7
The Partner Program allowed authors and publishers to submit copies of their books for indexing
in Google’s search engine.8 However, because rights holders affirmatively chose to have their
books digitized or indexed through the Partner Program, the Program was not subject to
allegations of copyright infringement like those made against the Library Project.
The Litigation and the Parties’ Positions
Authors and publishers objected to the Google Library Project from its inception on the grounds
that it infringed their copyrights.9 Generally, copyrights in books initially vest in the books’
authors.10 Many authors later transfer their copyrights to publishers under contract in exchange
for payment and the publisher’s manufacturing and selling copies of the book.11 Regardless of
whether they are the books’ authors or publishers, however, copyright holders have exclusive
rights “to reproduce the copyrighted work in copies,” or, in the case of literary works such as
books, “to display the copyrighted work publicly.”12 The authors and publishers who objected to
the Library Project claimed that Google infringed these exclusive rights by making digital copies
of print books and presenting snippets from the digitized books without rights holders’
permission.13 Google initially responded to these concerns by allowing rights holders who did not
want their books included in Google Book Search to “opt out.”14 If rights holders notified Google,
Google would ensure that digitized versions of their books were not included in its database.
The ability to “opt out” of the Library Project did not satisfy authors and publishers, however.
They sued to enjoin Google’s digitization and indexing and to recover monetary damages for
Google’s alleged copyright infringement. In September 2005, the Authors Guild filed a class
action suit in U.S. District Court for the Southern District of New York on behalf of “all persons
or entities that hold the copyright to a literary work that is contained in the library of the
University of Michigan.”15 Shortly thereafter, five publishing companies also sued in the
6 Association of American University Presses, Google Book Search, Neé Google Print, available at
http://www.aaupnet.org/aboutup/issues/gprint.html.
7 Google Books Partner Program: Promote Your Books on Google—For Free, 2009, available at
http://books.google.com/googlebooks/book_search_tour.
8 Id.
9 See, e.g., Anandashankar Mazumdar, University Press Group Expresses Concern Over Google Print’s Digitization of
Works, 70 Pat., Trademark & Copyright J. 109 (June 3, 2005).
10 17 U.S.C. § 201(a). There are exceptions to this general rule, such as when a book is “made for hire” or is a “work of
the United States Government.” See 17 U.S.C. § 105 & 201(b).
11 See, e.g., Example Author Contract, available at http://www.writecontent.com/Publishing_Tools/Author_Contract_/
author_contract_.html (“The Author hereby grants to the Publisher exclusive rights to reproduce and/or publish or
adapt and sell, and/or license third parties to publish or adapt and sell said Work.”).
12 17 U.S.C. § 106(1) & (5).
13 See, e.g., Mazumdar, supra note 9.
14 See, e.g., Christine Mumford, Google Library Project Temporarily Halted to Allow Copyright Owner Response, 70
Pat., Trademark & Copyright J. 461 (Aug. 19, 2005).
15 Authors Guild v. Google Inc., Class Action Complaint, No. 05 CV 8136 (S.D.N.Y. Sept. 20, 2005) at ¶ 20. The
University of Michigan’s library was the focus because Google began digitizing its books first. Id. at ¶ 31. Under
copyright law, “literary works” are any “works, other than audiovisual works, expressed in words.” 17 U.S.C. § 101.
Congressional Research Service
2
.
The Google Library Project
Southern District of New York.16 The suits were consolidated, and additional plaintiffs, including
the Association of American Publishers, joined the suit. Because the consolidated case was a class
action, the court must approve any settlement of it.17
In responding to the suit, Google essentially contended that its conduct was not infringing
because it gave rights holders the opportunity to “opt out” of having their books digitized and
indexed.18 Google also claimed that, even if a court found its conduct to be infringing, this
conduct represented a fair use of the rights holders’ works.19 Google and supporters of its Library
Project specifically cited the decision by the U.S. Court of Appeals for the Ninth Circuit in Kelly
v. Arriba Soft Corporation as support for the proposition that the indexing activities of Internet
search engines constitute fair uses.20
Legal Issues Raised by the Litigation
The litigation over the Google Library Project raised important questions about infringing
reproduction and fair use under copyright law. Namely, can an entity engaging in unauthorized
digitization and indexing avoid liability for copyright infringement by offering rights holders the
opportunity to request removal or exclusion of their content from its database? And, assuming
unauthorized indexing and display of “snippets” of digitized works are fair uses, can digitization
itself claim to be a fair use on the grounds that apparently prima facie infringing activities that
facilitate legitimate uses are fair uses? These questions will arguably persist, and their answers
remain important, even if the parties ultimately settle the litigation over the Library Project.
“Opt Out” Programs and Liability for Infringement
Google’s first line of defense against the authors and publishers was essentially that it was not
liable for copyright infringement because it gave rights holders the opportunity to “opt out” of
having their works digitized and indexed. In making this argument, Google relied on the related
claim that no one would conduct multi-library digitization and indexing projects like the Library
Project if they had to clear the copyrights for every book with the rights holders.21 Identifying and
locating the rights holder(s) for one book can be difficult enough, supporters of the Google
Library Project noted, without repeating this process millions of times, as would be necessary
with a major library collection.22 The publishers, in contrast, noted that Google’s offer to let rights
holders “opt out” of having their books digitized and indexed “stands copyright law on its
head.”23 They argued that one cannot generally announce one’s intention to infringe multiple
16 McGraw Hill Cos. v. Google Inc., Complaint, No. 05 CV 8881 (S.D.N.Y. Oct. 19, 2005). These companies were
McGraw-Hill Companies; Pearson Education; Penguin Group; Simon & Schuster; and John Wiley and Sons.
17 Fed. R. Civ. P. 23(e).
18 See, e.g., Susan Wojcicki, Google Print and the Authors Guild, Sept. 20, 2005, available at
http://googleblog.blogspot.com/2005/09/google-print-and-authors-guild.html.
19 Id.
20 See 336 F.3d 811 (9th Cir. 2003). For more background on Kelly, see CRS Report RL33810, Internet Search
Engines: Copyright’s “Fair Use” in Reproduction and Public Display Rights, by Robin Jeweler and Brian T. Yeh.
21See, e.g., Wojcicki, supra note 18.
22 The Harvard University Libraries (HUL), for example, contain over 15 million books. HUL, About the HOLLIS
Catalog, June 25, 2007, available at http://lib.harvard.edu/catalogs/hollis.html.
23 Anandashankar Mazumdar, Publishers: Value of Book Search Project Shows That Scanning Is Not Fair Use, 71 Pat.,
(continued...)
Congressional Research Service
3
.
The Google Library Project
copyrighted works and collectively offer rights holders the opportunity not to have their works
infringed.24
It is impossible to predict what a court would find based on such arguments, and this report does
not attempt to do so. This report does, however, highlight some of the considerations that could
factor in the court’s consideration of the issue. On the one hand, the requirement that a copyright
owner act affirmatively to stop non-willful infringement is not without precedent. The “notice and
takedown” procedures of the Digital Millennium Copyright Act (DMCA),25 for example, require
content owners to notify Internet Service Providers (ISPs) of the existence of infringing content
and can immunize ISPs from liability for infringement when they serve as “passive conduits” for
infringing content transmitted by third parties.26 Similarly, at least one court has found that
content owners are responsible for taking affirmative measures, such as using meta-tags within
the computer code of a Web page, to prevent Internet search engines from automatically indexing
and displaying their content.27 On the other hand, plaintiffs could argue that comprehensive
digitization projects, like that proposed by Google, willfully infringe copyright28 and differ from
the “passive conduits” protected by the DMCA. Likewise, rights holders in print books could
argue that their situations differ from that of Web page authors because Google had to digitize
their books before indexing them. They could claim that they took sufficient affirmative measures
to protect their works by not making them available for free on the Web.29
Digitization, Indexing, and Display as Fair Uses
Google also attempted to defend against the rights holders’ allegations of copyright infringement
by claiming that the Library Project, if found to be infringing, constituted a fair use.30 The “fair
use” exemption within copyright law limits rights holders’ exclusive rights by providing that uses
for “certain purposes”—including, but not limited to, criticism, comment, news reporting,
teaching, scholarship, and research—do not infringe copyright even if they are made without the
rights holders’ consent.31 In determining whether challenged conduct constitutes a fair use, a court
considers the following factors, which were developed under the common law and later codified
in the Copyright Act of 1976:
(1) the purpose and character of the use, including whether such use is of a commercial
nature or is for nonprofit educational purposes;
(...continued)
Trademark & Copyright J. 94 (Nov. 25. 2005).
24 Id.
25 P.L. 105-304. For more information on the DMCA generally, see CRS Report 98-943, Digital Millennium Copyright
Act, P.L. 105-304: Summary and Analysis, by Dorothy M. Schrader.
26 11 U.S.C. § 512(b)-(c).
27 See Field v. Google Inc., 412 F. Supp. 2d 1106 (D. Nev. 2006).
28 See, e.g., Class Action Complaint, supra note 15, at ¶ 23.d and ¶ 41 (alleging Google’s infringement was willful);
Complaint, supra note 16, at ¶ 2 (same).
29 See, e.g., Complaint, supra note 16, at ¶ 29 (arguing that Web pages differ from print books because rights holders in
Web pages can rely on technological measures to prevent indexing, while authors of print books can take no such
measures to prevent digitization).
30 See, e.g., Wojcicki, supra note 18.
31 17 U.S.C. § 107.
Congressional Research Service
4
.
The Google Library Project
(2) the nature of the copyrighted work;
(3) the amount and substantiality of the portion used in relation to the copyrighted work as a
whole; and
(4) the effect of the use upon the potential market for or value of the copyrighted work.32
These four factors must not be “treated in isolation, one from another.”33 Rather, “[a]ll are to be
explored, and the results weighed together, in light of the purposes of copyright,”34 which is to
“Promote the Progress of Science and useful Arts” and serve the public welfare.35 Also, because
fair use is an “equitable rule of reason” to be applied in light of copyright law’s overall purposes,
other relevant factors may be considered.36 The court hearing the case makes findings of fact and
assigns relative value and weight to each of the fair use factors. The court can also look to prior
cases for guidance even though determining whether a challenged activity constitutes a fair use
“calls for a case-by-case analysis.”37
Although it is impossible to predict what a court would find when confronted with an actual case,
and this report will not attempt to do so, it does highlight some of the many questions that the
Google Library Project raised regarding each of the four statutory “fair use” factors. The report
does so in order to illustrate the potential importance of the Library Project—or similar
digitization and indexing projects—in establishing the scope of infringing reproduction and fair
use under copyright law.
The Purpose and Character of the Use
First, as regards the purpose and character of the use, copyright law generally presumes that
commercial uses are not fair,38 and that transporting a work to a new medium is not a fair use.39
These presumptions would seem to work against digitization and indexing projects like the
Library Project. The Project was implemented by a for-profit corporation that proposed, among
other things, to sell ads “keyed” to the digitized content. The Project was also intended to migrate
content from print to digital format. These presumptions can, however, be overridden when the
use is sufficiently transformative.40 A copy’s use of the original is transformative when the copy
does not “merely supersede[]” the original but rather “adds something new, with a further purpose
or a different character” to the original.41
32 Id.
33 Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 578 (1994).
34 Id.
35 Perfect 10, Inc. v. Amazon.com, Inc., 487 F.3d 701, 720 (9th Cir. 2007) (quoting the U.S. Constitution, art. I, § 8, cl.
8, as well as Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 429 n.10 (1984)).
36 Stewart v. Abend, 495 U.S. 207, 237 (1990).
37 Campbell, 510 U.S. at 577-78.
38 See, e.g., Sony, 464 U.S. at 451 (“Every commercial use of copyrighted materials is presumptively an unfair
exploitation of the monopoly privilege that belongs to the owner of the copyright.”).
39 See, e.g., Infinity Broad. Corp. v. Kirkwood, 150 F.3d 104, 108 (2d Cir. 1998) (retransmission of radio broadcast
over telephone lines not a fair use); UMG Recordings, Inc. v. MP3.com, Inc., 92 F. Supp. 2d 349, 351 (S.D.N.Y. 2000)
(reproducing analog audio CDs as MP3s not a fair use).
40 Campbell, 510 U.S. at 578-79.
41 Id. at 579.
Congressional Research Service
5
.
The Google Library Project
The transformative nature of the Library Project would arguably be more easily established if it
merely indexed books and displayed “snippets” of them. Were Google’s uses so limited, it could
probably rely on the precedent of two cases from the U.S. Court of Appeals for the Ninth Circuit
which found that indexing and abridged displays of copyrighted content were fair uses. In the first
case, Kelly v. Arriba Soft Corporation, the court held that a company operating a search engine,
which had indexed a rights holder’s online photographs and displayed “thumbnail” versions of
them, was not liable for copyright infringement because its uses were fair.42 Key to this holding
was the court’s finding that indexing represented a transformative use of the original photographs.
While the original photographs were intended “to inform and to engage the viewer in an aesthetic
experience,” Arriba used its copies of them for a different function: “improving access to
information on the internet.”43 The court also emphasized that Arriba indexed and displayed
“thumbnail” versions of the photographs.44 The thumbnails had much lower resolution than the
originals and thus could not substitute for them because “enlarging them sacrifices their clarity.”45
The Ninth Circuit reached a similar conclusion in Perfect 10, Inc. v. Amazon.com, Inc.46 There,
the court also considered a use’s benefit to society in finding the use to be transformative. The
court noted that “a search engine provides social benefit by incorporating an original work into a
new work, namely, an electronic reference tool.”47
The digitization involved in the Library Project complicates the analysis, however. Admittedly,
the prior cases that found indexing and abridged displays of copyrighted content to be fair uses
also involved copying of originals.48 However, in these cases, the copying was of originals posted
on the Internet and resulted in copies that were “inferior” to the originals for all purposes except
their use in indexing. The first difference is potentially significant because courts have held that
rights holders confer limited licenses to copy their content for purposes of indexing and abridged
display by posting it on the Internet without taking affirmative measures to prevent copying.49
The second difference could also be significant because digitized books are arguably superior to
print ones when it comes to locating specific information within them.50
Because digitization was so central to the Library Project, and arguably could not be directly
paralleled to the copying in cases involving indexing and display of Internet materials, Google
might have had to rely on the proposition that apparently prima facie infringing activities (such as
digitization) that facilitate legitimate uses (such as indexing and limited displays) are fair uses.
The Supreme Court’s decision in Sony Corporation of America v. Universal City Studios could
arguably provide broad support for this principle.51 In Sony, the Court held that the sale of the
42 336 F.3d 811 (9th Cir. 2003).
43 Id. at 818-19.
44 Id. at 818.
45 Id. at 819.
46 487 F.3d 701, 721 (9th Cir. 2007), rev’g Perfect 10, Inc. v. Google Inc., 416 F. Supp. 2d 828 (C.D. Cal. 2006)
(holding that Google’s use of thumbnail versions of Perfect 10’s copyrighted photographs was not fair, in part, because
Google’s thumbnails could potentially substitute for the reduced-size versions of these photographs that Perfect 10 had
licensed another company to reproduce and distribute for display on cell phones).
47 Perfect 10, Inc., 487 F.3d at 721.
48 See, e.g., Kelly, 336 F.3d at 816.
49 Field, 412 F. Supp. 2d at 1115-16.
50 A digital version of a print book would display poorer resolution than the original. However, it would enable
researchers to locate specific content more easily by using the “search” or “find” functions of their Web browsers.
51 464 U.S. 417 (1984).
Congressional Research Service
6
.
The Google Library Project
video recording machine, which was used to “time shift” broadcast television for personal home
viewing, was not contributory copyright infringement.52 Although the factual underpinnings and
legal precedent of Sony are not particularly relevant to or controlling in a case like Google’s, the
Sony decision itself stands as a landmark in copyright law demonstrating the willingness of the
Court to balance new technological capabilities against traditional principles of copyright law and
to recognize new categories of fair use. Many copyright experts saw analogies to the
technological considerations inherent in Sony in Google’s case.53 Such experts noted that
Google’s allegedly infringing activity in digitizing print books was incidental to the valid and
socially useful function of indexing.
The analogy to Sony might not be enough to persuade a court that digitizing for purposes of non-
infringing indexing constitutes a fair use, however. Digitizing and indexing print books are
arguably far removed from making and selling devices that consumers use to record broadcast
television programming and replay it later. Additionally, courts have shown little inclination to
recognize categories of judicially created fair uses other than time shifting. In UMG Recordings v.
MP3.com, Inc., for example, a U.S. district court rejected out-of-hand the defendant’s proffered
fair use defense as a justification for unauthorized copying of plaintiffs’ audio CDs.54 The
defendant had claimed that its unauthorized copying enabled CD owners to “space shift” because
they could access the music on their CDs from any location through MP3.com’s subscription
service.55
The Nature of the Copyrighted Work
Comprehensive digitization and indexing projects, such as the Google Library Project, raise
similar questions when the second fair use factor is considered. Projects that digitize library
collections potentially encompass diverse types of materials. Some of these materials may be
works of fiction, which are among the creative works accorded the highest level of copyright
protection.56 Other materials may be reference books or compendiums of facts, which are
afforded the “thinnest” copyright protection.57 Yet other materials may be nonfiction and mix
unprotected ideas with protected expressions of these ideas.58 This diversity of materials makes
possible the arguments of both proponents and opponents of the view that projects like Google
Book Search constitute fair uses. The nature of the work can, however, be less important than the
purpose and character of the use, at least in situations where the use can be clearly recognized as
transformative.59
52 Id. at 442.
53 See, e.g., Jonathan Band, The Google Print Library Project: Fair or Foul?, 9 J. of Internet L. 1, 4 (Oct. 2005);
Christopher Heun, Courts Unlikely to Stop Google Book Copying, InternetWeek (Sept. 2, 2005), available at
http://internetweek.cmp.com/showArticle.jhtml?articleID=170700329.
54 92 F. Supp. 2d 349, 352 (S.D.N.Y. 2000) (“[D]efendant’s ‘fair use’ defense is indefensible and must be denied as a
matter of law.”).
55 Id.
56 Kelly, 336 F.3d at 820.
57 See, e.g., Feist Publ’ns, Inc. v. Rural Tel. Serv., 499 U.S. 340 (1991) (requiring originality in the selection or
arrangement of facts for copyrightability).
58 See, e.g., Baker v. Selden, 101 U.S. 99 (1879) (distinguishing non-protectable ideas from their protectable
expressions).
59 Campbell, 510 U.S. at 577-78.
Congressional Research Service
7
.
The Google Library Project
The Amount and Substantiality of the Portion Used
The amount and substantiality of the portion used in relation to the copyrighted work as a whole
is another factor that could potentially cut either way in cases involving digitization and indexing
projects. As a general rule, “[w]hile wholesale copying does not preclude fair use per se, copying
an entire work militates against a finding of fair use.”60 Copying entire works can, however, be
found to constitute a fair use when doing so is reasonable given the purpose and character of the
use.61 Digitization projects, such as the Google Library Project, would clearly be engaged in
wholesale copying, including copying any segments comprising the “heart” of the copied work.62
The question would thus become whether such wholesale copying was reasonable for an indexing
project. Proponents of the project could argue that courts have found copying entire works in
order to digitize them reasonable,63 and that searchers would see only “snippets” of the work in
any case. Opponents, in contrast, could argue that, in all cases where courts protected wholesale
copying for purposes of indexing, the authors had placed their works online, thereby creating
implied licenses for others to copy and index them.64 Moreover, in at least some of these cases,
the copies were deleted after the indexing was completed.65 In no case did the copier propose to
give copies to third parties, as Google did when contracting to provide digital copies of the books
in their collections to libraries.
The Effect of the Use Upon the Potential Market or Value of the Work
Finally, digitization and indexing projects could be seen as either promoting or inhibiting the
potential markets or values of the copyrighted works. Proponents of digitization could argue that
indexing and display of “snippets” of print books increases the markets for the originals by
alerting researchers to books on their topics. If researchers purchase books of which they would
otherwise have been unaware, the markets for these books could potentially be improved by the
unauthorized digitization. Opponents, in contrast, could argue that unauthorized digitization and
indexing usurps markets that the rights holders are developing;66 that viewing “snippets” of print
books sometimes can substitute for purchases of them; and that rights holders should be free to
determine whether, when, and how their print works are digitized.67 The outcome of any findings
by the court on this factor may hinge upon the degree of harm to their markets that plaintiffs must
show. Some courts have required plaintiffs to show only that the markets in which they alleged
harm are “likely to be developed,”68 while others have required proof of actual losses in
60 Worldwide Church of God v. Philadelphia Church of God, Inc., 227 F.3d 1110, 1118 (9th Cir. 2000) (internal
quotations omitted).
61 See, e.g., Kelly, 336 F.3d at 821; Sega Enters., Ltd. v. Accolade, Inc., 977 F.2d 1510, 1523 (9th Cir. 1992).
62 See, e.g., Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 564 (1985).
63 See, e.g., Kelly, 336 F.3d 811; Perfect 10, Inc., 487 F.3d 701.
64 See Field, 412 F. Supp. 2d at 1115-16.
65 See Kelly, 336 F.3d at 816.
66 See Complaint, supra note 16, at ¶ 5 (noting that publishers were already making their print books available online in
various ways, including a partnership with the search engine Yahoo!).
67 Cf. BMG Music v. Gonzalez, 430 F.3d 888, 891 (7th Cir. 2005) (“Copyright law lets authors make their own
decisions about how best to promote their works.”).
68 See, e.g., Bill Graham Archives, LLC v. Dorling Kindersley, Ltd., 386 F. Supp. 2d 324, 332 (S.D.N.Y. 2005).
Congressional Research Service
8
.
The Google Library Project
established markets.69 The fact that a use is transformative can, however, outweigh even
inhibition of or harm to plaintiffs’ markets.70
The Proposed Settlement Agreement
On October 28, 2008, Google and the rights holders announced a proposed settlement
agreement.71 Under this agreement, Google would compensate rights holders for prior and future
uses of their work.72 Google would also fund the establishment and initial operations of a not-for-
profit entity, called the Registry, which would represent rights holders in negotiating future uses
of their content with Google.73 Google, in turn, would receive a non-exclusive license74 to (1)
“Digitize all Books and Inserts” published before January 5, 2009, and (2) make certain uses of
the digitized materials, including displaying “snippets” of them among its search results, subject
to the terms of the agreement.75 By allowing Google to digitize and display books, the agreement
would pave the way for Google to expand Google Book Search, selling subscriptions to
institutions and electronic versions of books to individuals.76 The agreement would also create
certain rights and responsibilities for libraries that allow Google to digitize their books,77 as well
as make certain provisions for institutional subscribers to, or individual users of, commercialized
versions of Google’s Book Search database.78
The agreement will not take effect until certain conditions are met, one of which requires final
court approval of the settlement agreement.79 The court granted preliminary approval of the
agreement on November 17, 2008.80 However, final approval is still pending. Class members had
69 Perfect 10, Inc., 487 F.3d at 725.
70 See, e.g., Campbell, 510 U.S. at 591.
71 Authors Guild, Inc. v. Google Inc., Settlement Agreement, Case No. 05 CV 8136-JES (S.D.N.Y. Oct. 28, 2008).
72 Id. at ¶ 2.1(a) (providing that Google would pay 70% of the net revenue earned from uses of Google Book Search in
the United States to rights holders); ¶ 2.1(b) (providing that Google would pay at least $45 million into a “Settlement
Fund,” whose proceeds would pay rights holders whose books or “inserts” were digitized prior to January 5, 2009).
73 Id. at ¶ 2.1(c). Among other functions, the Registry could negotiate the terms of “New Revenue Models” (e.g., print-
on-demand) with Google and negotiate pricing categories and percentages for sale of digitized materials to users.
74 Because this license is non-exclusive, the Registry could license other entities to digitize, index, or display the works
of rights holders. However, if the Registry were to enter into a similar agreement within 10 years of the settlement’s
effective date, it must extend comparable economic and other terms to Google. Id. at ¶ 3.8(a).
75 Id. at ¶ 3.1. Google’s rights to use books within their terms of copyright protection would hinge upon whether they
were “commercially available,” or available “for sale new through one or more then-customary channels of trade in the
United States.” See id. at ¶ 1.28. If a book is commercially available, Google could not make “display uses” without the
copyright holders’ consent. Id. at ¶¶ 3.3-3.5. Conversely, if a book is not commercially available, Google could make
“display uses” unless the rights holder objects. Id. This distinction between commercially available and non-
commercially available books would significantly vary the legal protections of copyright law, which protects all works
equally, regardless of their commercial availability, during their terms of copyright protection. See 17 U.S.C. § 106 and
§ 302.
76 See, e.g., Settlement Agreement, supra note 71, at ¶ 3.7.
77 Id. at ¶ 7.2(f)(i)-(ii) and Article X.
78 See, e.g., James Grimmelmann, Principles and Recommendations for the Google Book Search Settlement, Nov. 8,
2008, available at http://www.laboratorium.net/archive/2008/11/08/
principles_and_recommendations_for_the_google_book (noting the existence, as well as the potential inadequacy, of
provisions regarding users).
79 Settlement Agreement, supra note 71, at ¶1.49.
80 Authors Guild, Inc. v. Google Inc., Order Granting Preliminary Settlement Approval, Case No. 05 CV 8136-JES
(continued...)
Congressional Research Service
9
.
The Google Library Project
until September 4, 2009, to file objections with the court.81 The court will consider these
objections, as well as conduct an independent review of the proposed agreement, in determining
whether to grant final approval to the settlement.82 The final hearing on the settlement is presently
scheduled for October 7, 2009.83
Opposition to the Proposed Settlement
Commentators have voiced numerous concerns about the proposed settlement, some of which
could recur in class members’ objections or the court’s review of the agreement. One major
concern is that the settlement would effectively grant Google an exclusive license to digitize the
books covered by the agreement and display them to individual users and libraries.84 The
settlement states that Google’s digitization and display rights under the agreement are non-
exclusive,85 and nothing, under the settlement or otherwise, would preclude another entity from
undertaking a digitization and indexing project like Google Book Search.86 However, some
commentators worry that these factors might not suffice to prevent Google’s effectively
monopolizing the book-search field given that the settlement agreement (1) includes a clause
granting Google the right to any more favorable terms that the Registry negotiates with third-
parties over the next ten years87 and (2) protects Google from the litigation risks likely to confront
those initiating new book-digitization projects. Commentators fear that these factors could
significantly diminish the willingness of potential competitors to enter the field and result in a
monopolized market where individual consumers and libraries, in particular, are vulnerable to
(...continued)
(S.D.N.Y. Nov. 17, 2008). The final hearing is presently scheduled to be held on October 4, 2009. It was originally set
for June 11, 2009, but the court changed the date in an order issued on April 28, 2009. See Court Delays Google Book
Search Settlement Hearing, Extends Opt-Out Period, 78 Pat., Trademark & Copyright J. 9 (May 1, 2009).
81 Court Delays Google Book Search Settlement Hearing, supra note 80. The deadline for filing objections was
originally May 5, 2009. See Order Granting Preliminary Settlement Approval, supra note 80.
82 Under the Federal Rules of Civil Procedure, judges must ensure that settlements in class action lawsuits are “fair,
reasonable, and adequate.” Fed. R. Civ. P. 23(e). In conducting this review, the court could reject a proposed settlement
because of concerns not raised by class members or other parties to the agreement. See, e.g., Muchnick v. Thompson
Corp., 509 F.3d 116 (2d Cir. 2007) (quashing the proposed settlement agreement resolving the litigation in Tasini v.
New York Times because some members of the proposed settlement class had not registered their works with the U.S.
Copyright Office and so lacked standing to bring suit in federal court). The district court had previously approved the
settlement, and no class member or party had raised this objection.
83 Nathan Pollard, Public Interest Organization Offers Privacy Recommendations to Google in Book Search, BNA.
84 See, e.g., James Gibson, Google’s New Monopoly? How the Company Could Gain by Paying Millions in Copyright
Fees, Wash. Post, Nov. 3, 2008, at A21 (“[S]ettling probably puts Google in a better position than it would have been
in if it had won its case in court. ... Google’s concession has made it more difficult for anyone to invoke fair use for
book searches [and] [b]y settling the case, Google has made it much more difficult for others to compete with its Book
Search service.”).
85 See, e.g., Settlement Agreement, supra note 71, at ¶ 3.1(a).
86 See, e.g., Miguel Helft, Some Raise Alarms as Google Resurrects Out-of-Print Books, New York Times, Apr. 4, 2009,
at A1 (“Nothing prevent[s] a potential rival from following in [Google’s] footsteps—namely, by scanning books
without explicit permission, waiting to be sued and working to secure a similar settlement.”). Microsoft, for example,
has explored offering book-search capabilities in the past. See Andrea L. Foster, Microsoft’s Book-Search Project Has
a Surprise Ending, Chron. of Higher Educ., May 29, 2008, available at http://chronicle.com/free/2008/05/
3022n.htm.
87 See, e.g., Settlement Agreement, supra note 71, at ¶ 3.8(a) (“The Registry ... will extend economic and other terms to
Google that, when taken as a whole, do not disfavor or disadvantage Google as compared to any other substantially
similar authorizations granted to third parties by the Registry.”).
Congressional Research Service
10
.
The Google Library Project
price increases.88 On July 2, 2009, the U.S. Department of Justice confirmed that it is conducting
an antitrust investigation into the Google settlement because of such concerns.89 The Justice
Department reportedly has written a letter to the federal judge reviewing the proposed settlement
indicating that the settlement’s terms could violate the Sherman Act of 1890, which prohibits,
among other things, any “contract, combination ... or conspiracy” that constitutes a “restraint of
trade or commerce.”90 A coalition of potential Google competitors, including Microsoft, Amazon,
and Yahoo, has also formed to lobby against and otherwise oppose the settlement, in part because
a “single organization should not control public access to the books.”91
A related concern focuses upon the settlement’s potential effects on “orphan works,” or
copyrighted works whose owners are difficult or impossible to identify or locate.92 Some
commentators worry that the settlement would grant Google a “unique lock”93 on orphan works
by making the Registry, which is to be created under the agreement, the effective trustee of rights
in “orphaned” books.94 The Registry has already opted to deal with Google regarding these
books, and some commentators are concerned that the absence of known or identifiable rights
holders effectively means that these books could not be further used without the agreement of the
Registry and/or Google.95 Commentators concerned about orphan works also suggest that the
settlement agreement’s treatment of them constitutes “a kind of legislation, stepping on
congressional prerogatives.”96 The 110th Congress considered, but did not enact, orphan works
legislation that would have authorized use of orphan works under certain conditions and would
have made the U.S. Copyright Office responsible for receiving and recording notices tracking
users and uses of orphan works.97 Similar legislation may be introduced in the 111th Congress.
Some commentators feel that any “changes” involving orphan works of the magnitude allegedly
made in the settlement agreement ought to be made by Congress, not the federal courts.98
Yet other concerns center upon the settlement agreement’s effects on public welfare, especially its
provisions regarding public access and user privacy. Several library groups have stated that, while
88 See, e.g., Laura G. Mirviss, Harvard-Google Online Book Deal at Risk, The Harvard Crimson, Oct. 30, 2008,
available at http://www.thecrimson.com/article.aspx?ref=524989 (“The settlement provides no assurance that the
prices charged for access will be reasonable.”).
89 See, e.g., Letter to the Honorable Denny Chin, Re. The Authors Guild Inc., et al. v. Google, Inc., Civil No. 1:05-CV-
8136, available at http://pub.bna.com/ptcj/DOJLetterEtc.pdf; Miguel Helft, U.S. Inquiry Is Confirmed into Google
Books Deal, N.Y. Times, July 2, 2009, available at http://www.nytimes.com/2009/07/03/technology/companies/
03google.html?_r=1&ref=technology.
90 15 U.S.C. § 1.
91 Nathan Pollard, Opposition Group to Google Book Settlement Includes Big Names, Authors and Libraries, 78
Patent, Trademark & Copyright Journal 539 (August 28, 2009).
92 See, e.g., Anandashankar Mazumdar, Internet Archive Warns of Online Book Market Consolidation from Google
Settlement, Pat., Trademark & Copyright L. Daily, Apr. 23, 2009. For more on orphan works generally, see CRS
Report RL33392, “Orphan Works” in Copyright Law, by Brian T. Yeh.
93 Grimmelmann, supra note 78.
94 See, e.g., Mazumdar, supra note 92.
95 Id.
96 See, e.g., Anandashankar Mazumdar, Google Books Settlement Prompts Questions about Effect on Readers,
Libraries, Others, Pat., Trademark & Copyright L. Daily, Mar. 18, 2009 (quoting Marybeth Peters, Register of
Copyrights).
97 Orphan Works Act of 2008, H.R. 5889, 110th Cong. There was a similar bill in the Senate, but it did not include the
notice requirements of the House bill. See Shawn Bentley Orphan Works Act of 2008, S. 2913, 110th Cong.
98 See, e.g., Helft, supra note 86 (“They are doing an end run around the legislative process.”) (quoting Brewster Kahle,
founder of the Internet Archive and the Open Content Alliance).
Congressional Research Service
11
.
The Google Library Project
they do not oppose the settlement, they are concerned that some of its provisions could
“compromise fundamental library values such as equity of access to information, patron privacy,
and intellectual freedom.”99 These groups would like the court to oversee the pricing of individual
subscriptions and the selection of board members for the Registry to ensure that pricing is
equitable and the Registry acts in the public interest. Other library groups have expressed similar
concerns regarding the restrictions on printing and networking imposed on the free subscriptions
to Book Search that would be provided to public libraries under the settlement agreement.100
Relatedly, privacy groups have noted that the proposed agreement would allow or require Google
to collect extensive information about users, but makes few provisions for the protection of this
information. As one group stated,
... [T]he settlement gives Google the potential, and in some instances the need, to collect
substantial quantities of sensitive reader information. Google will also need to share some
usage data with the Registry. Specifically, Google will share sales and subscription usage
data for calculating and distributing payment to rights holders, market research data
concerning various Preview options, and data pertaining to audits and security breaches.
Some collection and sharing is of course necessary to effectuate the settlement, but the
settlement does not—and should not—require sharing anything other than aggregate data.
However, the settlement does not contain a broad restriction on the sharing of user data. The
agreement does state that Google cannot be forced to disclose “confidential or personally
identifiable information except as compelled by law or valid legal process” in the case of a
security breach, but it does not address voluntary disclosure by Google. More generally, it
also does not address Google’s collection, use, retention, and sharing of user data outside the
specific context of a security breach. More formal privacy safeguards would ensure that
readers maintain the privacy they have traditionally enjoyed, preserving the right to read
anonymously and allowing readers to feel free to access and read books of any sensitive
sort.101
A challenge to the settlement class itself has also reportedly been filed by an attorney and author
who alleges that the settlement represents “an abuse of the class-action process” because the
settlement class is so broad and the named plaintiffs do not adequately represent all class
members.102 This challenge also alleges that the proposed settlement does not adequately
compensate authors,103 a concern which has also been raised by some writers groups.104 One such
group has noted that “of the $125 million offered by Google, only $45 million is for writers,” an
amount allegedly “way short of the amount needed to compensate authors of millions of
99 American Library Association, Association of Research Libraries, and Association of College and Research
Libraries, Google Library Project Settlement, July 29, 2009, available at http://librarian.lishost.org/?p=2731.
100 Nathan Pollard, Urban Libraries Council Suggests Revisions to Google Book Search Settlement, 78 Patent,
Trademark & Copyright Journal 540 (August 28, 2008).
101 Center for Democracy and Technology, Privacy Recommendations for the Google Book Search Settlement, July 27,
2009, available at http://cdt.org/copyright/20090727_GoogleRecs.pdf. CDT filed an amicus brief with the court on
September 4, 2009, expressing similar concerns. See Authors Guild v. Google, Inc.: Brief Amicus Curiae of the Center
for Democracy and Technology in Support of the Approval of the Settlement and Protection of Reader Privacy,
available at http://www.cdt.org/copyright/CDT-GoogleAmicusFinal.pdf
102 Miguel Helft & Motoko Rich, Lawyer and Author Adds His Objections to Settling the Google Book Lawsuit, New
York Times, Aug. 18, 2009, available at http://www.nytimes.com/2009/08/19/technology/internet/19google.html. Some
commentators have described this challenge as the “most fundamental challenge to the settlement yet.” See id. (quoting
James Grimmelmann, associate professor at the Institute for Information Law and Policy at New York Law School).
103 Id.
104 See, e.g., National Writers Union Opposes Settlement of Google Books Dispute, 78 Patent, Trademark & Copyright
Journal 515 (Aug. 21, 2009).
Congressional Research Service
12
.
The Google Library Project
books.”105 This group also alleges that the settlement is “grossly unfair” to authors by requiring
them to “opt out” of Google’s proposed uses of their works and interferes in the relationship
between authors and publishers.106
Support for the Proposed Settlement
Supporters of the proposed settlement agreement dispute many of the previous claims, as well as
note the potential benefits of the project. Some supporters have claimed that, far from being
anticompetitive, the settlement is pro-competitive because it “would make it easier for
competitors to enter the book search market by resolving uncertainty over intellectual property
rights, allowing competitors to follow in Google’s footsteps.”107 Such claims might get some
support from a recently announced partnership between the University of Michigan and
Amazon’s subsidiary BookSurge, under which 400,000 titles owned by the university will be
made available through print on demand.108 Although the university created the digital versions of
some of these titles itself, others were produced by Google as part of its Library Project and will
now be made available through an alleged Google competitor. Other supporters have similarly
noted that many alleged “orphan works” are not, in fact, orphans and that the settlement will
prompt their owners to identify themselves, or reveal that the works are in the public domain.109
Relatedly, they noted that the agreement would not create an “orphan works monopoly,” as some
critics fear, because the “scanned books naturally remain in libraries throughout the country, and
competitors can use them according to copyright law.”110
Other supporters note that the agreement may help “solve disparities in resources available to
minorities, the disabled, and the impoverished” by “remov[ing] some of the physical obstacles we
have had to acquiring knowledge.”111Advocates for the persons with visual disabilities assert that
the agreement would “allow blind people to have access to millions of books for the first time in
history,”112 while various civil rights leaders claim it could help resolve the “digital divide,” or
differences in access to information between minority and other communities; “democratize
knowledge”; and transform education.113 The European Union’s Telecoms and Information
Society Commissioner, Viviane Reding, has also reportedly voiced support for the agreement.114
105 Id.
106 Id.
107 See, e.g., Nathan Pollard & Christine You, Benefits of Google Book Deal Praised, But Orphan Works Impact Still in
Doubt, 14 Electronic Commerce & Law Report 1184 (Aug. 19, 2009).
108 See, e.g., Josh Hadro, Michigan Deal a New Twist on Access to Scanned Book Content: The Effort Marks Another
Repurposing of Digital Editions Scanned In-house and by Google, Library Journal, July 23, 2009, available at
http://www.libraryjournal.com/article/CA6672693.html.
109 See, e.g., Nathan Pollard, Social Justice IP Panel Says Google Book Settlement Will “Level the Playing Field,” 78
Patent, Trademark & Copyright Journal 408 (July 31, 2009).
110 See, e.g., Benefits of Google Book Deal Praised, supra note 107 (quoting Ed Black, President of the Computer &
Communications Industry Association).
111 See, e.g., Justice IP Panel Says Google Book Settlement Will “Level the Playing Field,” supra note 109.
112 Id. (quoting Charles S. Brown, advisor to the president, National Association of the Blind).
113 Id. (quoting Lateef Mtima, Professor of Law, Howard University; Wade Henderson, President, Leadership
Conference on Civil Rights; and Brent A. Wilkes, National Executive Director, League of United Latin American
Citizens).
114 See, e.g., Joe Kirwin, EU Commissioner Endorses Google Digital Book Settlement with U.S. Publishers, Patent,
Trademark & Copyright Law Daily, Aug. 31, 2009.
Congressional Research Service
13
.
The Google Library Project
In calling on EU member states to “reconsider their positions and … back similar arrangements in
the European Union,” Reding reportedly called for an overhaul of EU copyright laws to help
expand digitization of orphan works and stated that “[i]t is good to see that new business models
are evolving that could allow bringing more content to an increasing number of consumers.”115
Potential Effects of Rejection or Approval of the Settlement
Rejection of the proposed settlement agreement could place the parties’ claims and defenses back
before the court. Moreover, even if eventually approved by the courts, the settlement agreement
only governs claims against Google over its Library Project within the United States. Litigation in
other jurisdictions remains possible.116
Author Contact Information
Kate M. Manuel
Legislative Attorney
kmanuel@crs.loc.gov, 7-4477
Acknowledgments
Robin Jeweler, Legislative Attorney, original author of RS22356 The Google Book Search Project: Is
Online Indexing a Fair Use Under Copyright Law?
115 Id.
116 See, e.g., Editions du Seuil v. Google Inc., Tribunal de Grand Instance de Paris (alleging that Google Book Search
infringes copyright under French law); Joe Kirwin, EU States Urge Investigation into Google’s Digital Book Plan, Pat.,
Trademark & Copyright L. Daily, May 29, 2009 (noting that the Federation of European Publishers has also raised
concerns about the potential anticompetitive effects of the Google Library Project with the European Commission).
Congressional Research Service
14