The Authors Guild v. Google, Inc.

Filing 73

AMICUS BRIEF, on behalf of Amicus Curiae Malcolm Gladwell, J. M. Coetzee, Michael Pollan, Margaret Atwood, Peter Carey, Karen Russell, Ursula Le Guin, Ron Larson, Thomas Keneally, Taylor Branch, Tracy Chevalier, Lawrence Hill, Michael Frayn, Diane McWhorter, Robert Christopherson, Tracy Kidder and Yann Martel, FILED. Service date 04/14/2014 by CM/ECF. [1201502] [13-4829]--[Edited 04/15/2014 by KG]

13-4829-cv

d
IN THE

United States Court of Appeals
FOR THE SECOND CIRCUIT

THE AUTHORS GUILD, BETTY MILES, JIM BOUTON, JOSEPH GOULDEN,
individually and on behalf of all others similarly situated,
Plaintiffs-Appellants,
HERBERT MITGANG, DANIEL HOFFMAN, individually and on behalf of all others
similarly situated, PAUL DICKSON , THE MCGRAW-HILL COMPANIES , INC.,
PEARSON EDUCATION, INC., SIMON & SCHUSTER, INC., ASSOCIATION OF
AMERICAN PUBLISHERS, INC., CANADIAN STANDARD ASSOCIATION, JOHN
WILEY & SONS, INC., individually and on behalf of all others similarly situated,
Plaintiffs,

—against—
GOOGLE, INC.,

Defendant-Appellee.
ON APPEAL FROM THE UNITED STATES DISTRICT COURT
FOR THE SOUTHERN DISTRICT OF NEW YORK

BRIEF FOR AUTHORS MALCOLM GLADWELL, J.M. COETZEE,
MICHAEL POLLAN, MARGARET ATWOOD, PETER CAREY,
KAREN RUSSELL, URSULA LE GUIN, RON LARSON,
THOMAS KENEALLY, TAYLOR BRANCH, TRACY CHEVALIER,
LAWRENCE HILL, MICHAEL FRAYN, DIANE MCWHORTER,
ROBERT CHRISTOPHERSON, TRACY KIDDER AND YANN MARTEL
AS AMICI CURIAE IN SUPPORT OF PLAINTIFFS -APPELLANTS
GLORIA C. PHARES
CHRISTOPHER M. STRONG
HOFFMANN MARSHALL STRONG LLP
116 West 23rd Street, Suite 500
New York, New York 10011
( 212) 851-8403
Attorneys for Amici Curiae Authors

CORPORATE DISCLOSURE STATEMENT

Each of the amici curiae is a natural person and is not a
nongovernmental corporation.

TABLE OF CONTENTS
Page
TABLE OF AUTHORITIES .......................................................................................................... ii
IDENTITY AND INTEREST OF THE AMICI CURIAE ........................................................ 1

STATEMENT OF RELEVANT FACTS ..................................................................................... 7
.
ARGUMENT ................................................................................................................................... 11

GOOLGE’S COPYING OF BOOKS TO CREATE A VAST
DATABASE OF BOOKS AND ITS DISTRIBUTION OF
COPIES OF THOSE BOOKS ARE NOT A FAIR USE ............................................. 11

A.
Google’s Vast Unauthorized Copying and

Reproduction of Authors’ Copyrighted Works

Do Not Justify the Harm to Authors………………………………………...14

1.

Google’s Library Project, Founded on
Contracts to Make Millions of Unauthorized
Copies, is Not Justified. …………………………………………………14

2.

Google’s Distribution to the Libraries of Digital
Copies of All the Books it Copies Is Not a Fair Use………….17

3.

The Utility of a Full‐Text Search Function
Does Not Justify Copying Over 20 Million Books……………20

4.

Google’s “Use is of a Commercial Nature.”……………………..22

B.

The District Court Gave Short Shrift to
the Nature of the Copyrighted Works……………………………………..24

C.

Google Copies and Maintains Books in Its Book
Database That It Does Not Use in the Library Project.………………25

D.

Google’s Conduct Harms Authors. …………………………………………. 27

i

E.

Google Cannot Justify its Use of
Copyrighted Works in the Library Project. ………………………………30

CONCLUSION ................................................................................................................................ 32

TABLE OF AUTHORITIES
Page
Cases
American Geophysical Union v. Texaco, Inc.,
37 F.3d 881 (2d Cir. 1994) ...................................................................................................... 12
Authors Guild, Inc. v. Google, Inc.,
954 F. Supp. 2d, 282 (S.D.N.Y. 2013) ............................................................ 10, 23‐24, 28
Bill Graham Archives v. Dorling Kindersley Ltd.,
448 F.3d 605 (2d Cir. 2006) ........................................................................................... 13, 23

Blanch v. Koons, 467 F.3d 248 (2d Cir. 2006) ................................................................. 23
Castle Rock Entm’t, Inc. v. Carol Publ’g Grp., Inc.,
150 F.3d 132 (2d Cir. 1998) ................................................................................................... 23
Diversey v. Schmidly, 738 F.3d 1196 (10th Cir. 2013) ................................................. 18
eBay, Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006) .............................................. 13
Hotaling v. Church of Jesus Christ of Latter‐Day Saints,
118 F.3d 199 (4th Cir. 1997) ................................................................................................. 18
Kelly v. Arriba Soft Corp., 336 F.3d 811 (9th Cir. 2003) .............................................. 16
Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146
(9th Cir. 2007) .............................................................................................................................. 16
Salinger v. Colting, 607 F.3d 88 (2010) ............................................................................. 13
.
Sony Corporation of America v. Universal City Studios, Inc.,
464 U.S. 417 (1984) ................................................................................................................... 12
ii

Statutes
17 U.S.C. §106 ................................................................................................................................. 7
17 U.S.C. §106(3) ......................................................................................................................... 18
17 U.S.C. §107 ....................................................................................................................... 11, 16

Other Authorities
Edwin G. Burrows and Mike Wallace, Gotham,
A History of New York City to 1898 (Oxford U. Press 1999) ...................................... 21
Richard Dannay, Factorless Fair Use? Was Melville Nimmer Right?,
60 J. Copyright Soc. of the U.S.A., 127 (2003).................................................................. 13
Richard Dannay, Copyright Injunctions and Fair Use:
Enter eBay–Four‐Factor Fatigue or Four‐Factor Freedom?,
55 J. Copyright Soc. of the U.S.A 449 (2008) ................................................................... 13
.
Gallica Experiment: Offering Copyrighted Digital Documents,
(available at http://www.bnf.fr/en/collections_and_
services/digital_libraries_gallica/a.gallica_experimentation_
digital_offer.html) ....................................................................................................................... 28

Georgetown Law Library’s Guide to Legal History Databases
(available at https://www.law.georgetown.edu/library/
research/guides/legalHistory.cfm) .................................................................................... 20

Google, Inc., Form 10‐K , submitted to the Securities and Exchange
Commission for the Fiscal Year Ending Dec. 31, 2013,
(available at https://investor.google.com/pdf/
20131231_google_10K.pdf) ........................................................................... 7, 8, 14, 22‐23
.
Google, Inc. Information regarding the Library Project
available at http://books.google.com/intl/en‐US/
googlebooks/publisher_library.html#options4 ............................................................ 20
Google Inc. Opposition to Author Guild’s Motion for Summary
Judgment [Doc. 1072] ............................................................................................................... 17
iii

Google, Inc. Procedure for Excluding Books from Scanning,
(available at https://support.google.com/books/partner/
answer/3365282?hl‐en&ref_topic_3396243) ............................................................... 19
Google, Inc. Procedure for Not Displaying Scanned Books,
available at https://support.google.com/books/partner/
answer/2520009 ........................................................................................................................ 19
Peter S. Menell, In Search of Copyright’s Lost Ark:
Interpreting the Right to Distribute in the Internet Age,
59 J. Copyright Soc. of the U.S.A. 1 (2011) ........................................................................ 18
Melville B. Nimmer & David Nimmer, Nimmer on Copyright ............................. 12, 18
Readers Guide to Periodical Literature, (H.W. Wilson (available
at http://www.hwwilsoninprint.com/periodicals.php) ........................................... 20
.
University of Connecticut Law Library’s List of Legal Periodical
Indexes (U.S. and foreign) (available at http://library.law.
uconn.edu/research‐resources/legal‐periodical‐indexes) ...................................... 21

iv

IDENTITY AND INTEREST OF THE AMICI CURIAE
The amici are American and foreign authors and copyright
owners of bestselling works ranging from short stories to historical nonfiction
to calculus textbooks to fantasy and science fiction novels. 1 Works authored
by the amici have won nearly every major literary prize, including the Nobel
Prize in Literature, multiple Pulitzer Prizes and Booker Prizes.
Under the balance set by Congress under the Copyright Act, and
with limited exceptions, no one can exercise any of the Section 106 rights
without prior permission from the author. Technological advances are
challenging this balance. The development of the ability to create digital
copies and the proliferation of search engines and file sharing on the internet
have introduced new concerns for authors: the problem of controlling
unauthorized digital copies of works and the associated problem of
unauthorized distribution and display of those digital copies. In addition to

1

This brief was authored entirely by counsel for amici curiae. No party or
party’s counsel contributed money that was intended to fund the preparation
or submission of this brief. The Textbook and Academic Authors Association,
an organization to which some amici curiae belong, contributed money to
partially fund the preparation of this brief.

Counsel for both parties have consented in writing to the filing of this
brief.

widespread copyright infringement, the unchecked mass digitization of
written works interferes with an author’s legitimate exploitation of works in
new electronic formats and jeopardizes an author’s reputation by making
widely available earlier editions of works, such as textbooks, that have been
updated and superseded. These concerns—and whether the balance set by
Congress in the Copyright Act is being irreparably disrupted—are paramount
to any consideration of Google’s Library Project. All the amici curiae share an
interest that authors’ copyrights be protected from a for‐profit company’s
unjustified commercial uses.
Malcolm Gladwell, a staff writer at the New Yorker, is the author
of five bestselling books, including the nonfiction works The Tipping Point,
Blink, and Outliers. Mr. Gladwell has been named one of the 100 most
influential people by TIME magazine.
J.M. Coetzee is the recipient of the 2003 Nobel Prize in Literature.
A native of South Africa and now an Australian national, Mr. Coetzee is a
playwright and novelist, including Life & Times of Michael K and Disgrace, both
of which won the Booker Prize. Mr. Coetzee has held a number of positions at
universities in the United States, South Africa, and Australia.
Michael Pollan is the author of five New York Times bestsellers,
including The Omnivore’s Dilemma, which was named one of the ten best
2

books of 2006 by both the New York Times and the Washington Post, won the
California Book Award, the Northern California Book Award, and the James
Beard Award. Mr. Pollan is a journalism professor at the University of
California.
Margaret Atwood, a founder of the Writers’ Union of Canada, is
the author of essays, poetry, short‐stories and novels. Her novel The Blind
Assassin won the Booker Prize in 2000, and Ms. Atwood has received
countless other awards and honors. Her 2008 Massey Lecture series,
Payback: Debt and the Shadow Side of Wealth was adapted into the
documentary film Payback.
Peter Carey, an Australian, is one of only three writers to have
received the Booker Prize twice—for Oscar and Lucinda and The True History
of the Kelly Gang. Mr. Carey has won Australia’s Miles Franklin Award three
times, the Commonwealth Writers’ Prize twice, and the Prix du Meilleur Livre
Étranger. He has taught creative writing at several American universities, and
currently directs the MFA program in creative writing at Hunter College.
Karen Russell was named a Pulitzer Finalist for her first novel,
Swamplandia!, which was included in the New York Times’s 10 Best Books of
2011. Ms. Russell was chosen as one of Granta’s Best Young American
Novelists in 2007, and was named one of the best 20 under 40 writers by The
3

New Yorker. In 2013, she received a MacArthur Fellowship. Her short fiction
has appeared in The New Yorker, Granta, Oxford American, Zoetrope and
Conjunctions.
Ursula Le Guin, fiction writer and poet, is one of the world’s most
respected authors of imaginative writing. She has received many honors and
awards, including twenty‐one Locus Awards, four Nebula Awards, two Hugo
Awards, and the PEN/Malamud Award for short fiction. In April 2000, the U.S.
Library of Congress recognized Ms. Le Guin as a “Living Legend” for her
significant contributions to America's cultural heritage.
Millions of students from middle school through graduate school
rely on Ron Larson’s algebra, trigonometry, and calculus textbooks. Dr.
Larson has published more than 400 books, including the first completely
interactive online calculus textbook. Mr. Larson is a professor at Pennsylvania
State Erie, and his textbooks have won countless awards for pedagogy,
innovation, and design.
Thomas Keneally is an Australian writer of novels and non‐fiction
with an international publishing record and reputation. He is best known for
writing Schindler's Ark, the Booker Prize‐winning novel of 1982, later adapted
to become the Oscar‐winning movie, Schindler's List.

4

Taylor Branch is a historian, whose works include a three‐volume
narrative history of the civil rights era, America in the King Years. The trilogy’s
first book, Parting the Waters: America in the King Years, 1954‐63, won the
Pulitzer Prize. Mr. Branch has received lifetime achievement awards from the
Dayton Literary Peace Prize and the Anisfield‐Wolf Book Awards.
Tracy Chevalier is an author of seven historical novels, including
Girl with a Pearl Earring, is a Fellow of the Royal Society of Literature, and the
recipient of honorary doctorates from Oberlin College and the University of
East Anglia.
Lawrence Hill, a Canadian novelist and memoirist, is the author of
nine books, including The Book of Negroes (published in the United States as
Someone Knows My Name), which won the Rogers Writers’ Trust Fiction Prize
and the Commonwealth Writers Prize for Best Book. Mr. Hill’s 2013 Massey
Lectures were drawn from his non‐fiction book Blood: the Stuff of Life.
Michael Frayn is an English novelist, translator, memoirist,
nonfiction author and Tony Award winning playwright (for Copenhagen, in
2000). His novels have won countless honors, including the Somerset
Maugham Award, the Whitbread Novel Award, and the Commonwealth
Writers Prize. Mr. Frayn’s plays also include Noises Off and Democracy.

5

Diane McWhorter, an author and journalist, won the Pulitzer Prize
for General Nonfiction, the J. Anthony Lukas Book Prize, and many other
awards for her book Carry Me Home: Birmingham, Alabama, the Climactic
Battle of the Civil Rights Revolution. She is a regular contributor to The New
York Times and USA Today.
Robert Christopherson is the author of the leading physical‐
geography textbooks in the US and Canada, which since their 1992 publication
have been used by millions of geography students in 39 countries in English
and in translation. Mr. Christopherson taught geography for thirty years at
American River College.
Tracy Kidder won the Pulitzer Prize for General Nonfiction and
the National Book Award for his book The Soul of a New Machine, which
describes the race to design the next‐generation of computers. Mr. Kidder
also has written short fiction, essays and articles for publications including
The Atlantic, The New Yorker, The New York Times, and Granta.
Yann Martel is a Canadian author of novels and short stories. His
novel Life of Pi won the 2002 Booker Prize, was an international bestseller,
and was adapted into a blockbuster movie. In 2005 Martel was a visiting
scholar at the University of Saskatchewan.

6

STATEMENT OF RELEVANT FACTS
The amici curiae, U.S. and foreign, are the authors (“Authors”) of
works that are in copyright and were first published in the United States or in
Berne Convention countries. Some of the Authors’ works are not currently in
print, which does not mean they are not in copyright or that the Authors’
rights in those works are diminished in any way. The Authors are entitled to
all of the rights granted to an author under 17 U.S.C. §106 (“Section 106”).
Google is a for‐profit company that is currently valued at over
$350,000,000,000. Google does not generate its own books of history, fiction,
poetry, or other expressive works of authorship. (Google 10‐K at 32) Google
has assembled its massive wealth “primarily by delivering relevant, cost‐
effective online advertising.” (Id.) Its “business is primarily focused around
the following key areas: search and display advertising, the Android
operating system platform, consumer content through Google Play, enterprise,
commerce and hardware products.” (Id.) Google acknowledges that its search
function “continues to evolve and improve as more information comes online,

2

”Google 10‐K” refers to Google’s Form 10‐K submitted to the Securities
and Exchange Commission for the period ending December 2013, available at
https://investor.google. com/pdf/131231_google_10K.pdf.

7

and as people increasingly look to their mobile devices for answers
throughout the day.” (Id. at 4.)
One way that Google has “evolve[d]” is by increasing the volume
of information it makes available online—and thereby improving its search
function—by copying over 20 million books that it obtained from several
research libraries, and by displaying digital copies of these books through
Google Books.
There are two parts of the Google Books Program: (1) the
Partners Program, where an author’s Publisher has licensed Google to include
the full text of designated portions of specified works in the search results,
while also maintaining a full copy in the Google book database; use of copies
of works provided to Google under the Partners Program is governed by
contractual restrictions between Google and rightsholders;3 and (2) the
Library Project, where Google has not received permission from any
rightsholder of a work, but still copies the work and returns searches on the
internet in a form that Google calls a “snippet,” which it defines as
approximately 1/8 of a page. (A1616‐1617 at ¶¶ 43‐45.) Google has nowhere
stated that it will not ever revise its current definition of a “snippet” to include

3

Some of the Authors’ works are in the Partners Program.
8

a larger portion of a page’s text. Only the Library Project is at issue in this
case.
Google’s desire to build a national database of books was matched
by the desire of many university libraries to digitize the volumes in their
collections, a time‐consuming and very costly process. To achieve the goals of
both Google and the libraries, they entered into agreements in which each
library agreed to make the books in its collections available to Google for
copying. Some libraries allowed Google to copy only public domain works.
Other libraries allowed Google to scan books in copyright. (See, e.g., A593‐
604.) As part of the consideration for that access (Google also offered
indemnities), Google agreed to provide to each library digital copies of all the
books that that library had provided to Google—i.e., unauthorized digital
copies of millions of books in copyright. (A1618‐19 at ¶¶ 53‐54.)
Libraries have no right to authorize the copying of the physical
copies of works in their collections that are still in copyright, unless they also
own the copyrights. Google, similarly, has no right to reproduce and
distribute to the libraries digital copies of works that it was not authorized to
copy in the first place.
After Google scans a book, it applies optical character recognition
technology (OCR) to produce a machine‐readable text. (A1622 at ¶ 62.) The
9

result is not the kind of well‐organized book index included in copyright‐
protected non‐fiction works identifying names, places, concepts, subjects, etc.
and directing the reader to a location in the book. Google’s full‐text search
identifies words or phrases, but there is no selectivity or ability to search by
topics or concepts.
Google says that it does not display any text of a work that,
because of its brevity, might appear completely in the snippet display‐format.
This excludes works such as encyclopedias, almanacs, dictionaries, thesauri,
trivia books, books of quotations, bibliographies, poetry books, sheet music,
pricing guides, travel guides, joke books, recipe books, catalogs, and indexes.
Authors Guild, Inc. v. Google, Inc., 954 F. Supp. 2d 282, 285 (S.D.N.Y. 2013). But
having decided not to include these works in the search results, there is no
evidence that Google deletes the original scans from its own database, and no
explanation for why it is necessary for Google to retain unauthorized copies of
those books if they are not used for readers and researchers.
Google “generated 91% of its revenues from [its] advertisers in
2013[,]” amounting to $46,025,980,000 in advertising revenues (Google 10K
at 9, 27, 28.) Google does not currently display advertising on the specific
“About a Book” pages of the Library Project (but it does display advertising on
search results that link to these pages), but Google does not represent that it
10

never will display advertising on those pages. Google also does not provide
any insight about how the Authors’ works and the countless other copyrighted
materials in the Library Project will be used in the future.
ARGUMENT
GOOGLE’S COPYING OF BOOKS TO CREATE A VAST DATABASE OF BOOKS
AND ITS DISTRIBUTION OF COPIES OF THOSE BOOKS
ARE NOT A FAIR USE
When a for‐profit organization worth hundreds of billions of
dollars, which depends on evolving forms of advertising to return a profit for
its investors, sets out to copy every book both in and out of copyright, invests
the kinds of money necessary to do so, and then rolls out its project without
making representations concerning how it will use these copies in the future,
a more searching examination is required than a rote application of the four
fair‐use factors set out in 17 U.S.C. § 107. A fair use by definition does not
require a copyright owner’s prior approval. But when the creation of a
massive digital library is required for the fair use, it is incumbent on this Court
to consider whether any interpretation of fair use has ever approved
unauthorized copying (and subsequent distribution) on such a massive scale
and whether approving such a scheme destroys the appropriate balance
between fair use and the rights of a copyright owner.

11

Twenty years ago, Judge Newman noted his concern that
photocopying “creates a pressing need for the law ‘to strike an appropriate
balance between the authors’ interest in preserving the integrity of copyright,
and the public’s right to enjoy the benefits that photocopying technology
offers.’” American Geophysical Union v. Texaco Inc., 37 F.3d 881, 885 (2d Cir.
1994) (quoting 3 Melville B. Nimmer & David Nimmer, Nimmer on Copyright
§ 13.05 [E][1] at 13‐226 [“Nimmer on Copyright”] and citing Justice
Blackmun’s dissent in Sony Corporation of America v. Universal City Studios,
Inc., 464 U.S. 417, 467‐68 n.16 (1984)). Even then Judge Newman was
skeptical about applying the mechanical photocopying process to the
traditional fair use analysis developed to “adjust the competing interest of
authors—the author of the original copyrighted work and the author of the
secondary work that ‘copies’ a portion of the original work in the course of
producing what is claimed to be a new work.” Photocopying, he observed, “is
not concerned with creative authorship.” 37 F.3d at 886.
Nor is the process of scanning a book and converting the scan to
machine readable text concerned with creative authorship. But if Sony
requires this Court to apply the common law (now statutory) doctrine of fair
use to the facts of this case, a more apt formulation of the inquiry might be
whether “[i]n light of the purposes of copyright law and the public interest, is
12

there sufficient justification for the use to outweigh the copyright owner’s
interests in prohibiting the use or at least in being compensated for that use, if
an injunction is not warranted.”4 This does not mean the Court ignores the
four factors, the preamble purposes, transformativeness, and other important
considerations, but it focuses the inquiry on whether the ultimate goal
justifies the inroads on the copyright owner’s interests.5
Applied here, the question is whether Google, a for‐profit
company that profits commercially from the Library Project, has advanced
sufficient justification for digitizing and making searchable over 20 million
books and providing to the contributing libraries digital copies of those books
that outweighs the copyright owner’s right to prohibit those uses or, where an
injunction would be judged inappropriate, being compensated for the use.

4

Richard Dannay, Factorless Fair Use? Was Melville Nimmer Right?, 60 J.
Copyright Soc. of the U.S.A. 127, 144, 148 (2013) (“Factorless Fair Use?”). This
formulation also takes account of the holdings of eBay, Inc. v. MercExchange,
L.L.C., 547 U.S. 388 (2006), and Salinger v. Colting, 607 F.3d 88 (2010), and,
when applied in the fair‐use context, could result in damages, not an
injunction. Richard Dannay, Copyright Injunctions and Fair Use: Enter eBay—
Four‐Factor Fatigue or Four‐Factor Freedom?, 55 J. Copyright Soc. of the U.S.A.
449, 456‐60 (2008).

5
See Factorless Fair Use?, supra, n.4 at 144, 148.
13

A.

Google’s Vast Unauthorized Copying and
Reproduction of Authors’ Copyrighted
Works Do Not Justify the Harm to Authors.
1.

Google’s Library Project, Founded on
Contracts to Make Millions of
Unauthorized Copies, is Not Justified.

The first fair‐use factor asks the Court to examine the purpose and
character of the use. The district court was persuaded by Google’s message
that it is providing an apparently philanthropic service that permits
researchers to use the Google Library Project to find books that justifies its
copying of over 20 million books, many of them in copyright. But helping
people find books and data mining are not Google’s reason for digitizing more
than 20 million books: that may be how those researchers use the Library
Project, but as Google candidly admits, that is not its principal purpose.
Google’s purpose is to sustain its economic growth, which
depends upon the continued acquisition of more “information,” because it
attracts more people to its site increasing Google’s value and making it more
successful than its competitors (Google 10‐K at 3, 4). A project aimed at
copying all the books in as many libraries as possible and making those copies
available to the public for search—whether the authors of these books like it
or not—is an ideal way of achieving that goal.

14

While Google does not currently offer advertising on the specific
“About a Books” page, it does include advertising on search results that link to
this page, and nothing guarantees that Google will not change its advertising
model in the future or that Google will share advertising income fairly with
the copyright owners whose works have driven that additional income to
Google’s bottom line. Authors should not be forced to be vigilant about
Google’s “next move” and to bring another case if those events occur; that
would be a gross distortion of the statutory assumption that a user of a
copyrighted work must seek permission in advance.
What undermines Google’s fair use claim is that the entire project
is founded on agreements that create unauthorized copies of copyrighted
works. In exchange for access to books for copying—which the universities
were not authorized to give—Google distributed to the universities digital
copies of all the books copied from their collections—which Google was not
authorized to do. The result of this transaction was the copying of over 20
million books without the permission of any copyright owner and the
distribution to the libraries of digital copies that were not critical to Google’s
Library Project. Those digital copies were Google’s payment for the extensive
access it was given to copy the libraries’ books, a kind of rental fee. That
rental fee for the books was not a transformative use as that term has been
15

interpreted under Section 107. In addition, that reproduction and distribution
deprived authors of potential royalties for the copying of their books.6
It is not accurate for Google to say that the Library Project permits
readers and researchers to “find” books to purchase, because many of the
works in the Library project are very old and have not been available for sale
for centuries. Many other books are not for sale because they are out‐of‐print.
Old and out‐of‐print books can be “identified,” and some of them may be
“found” on a used‐book site, but as a practical matter, the only place they are
“found” is in a library. Nascent licensing of out‐of‐print books dried up when
Google‘s preemptive, extensive, and royalty‐free copying drove it away.

6

The facts of Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146, 1168
(9th Cir. 2007), Kelly v. Arriba Soft Corp., 336 F.3d 811 (9th Cir. 2003), and Bill
Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605, 609‐11 (2d Cir.
2006), on which the district court relied, are not comparable to the facts of
this case. Instead of textual works, all three cases involved visual images,
which cannot be identified unless displayed in full. To diminish the harm to
the copyright owners of the images, all required the display of images in
reduced size—“thumbnail images” online or reduced sizes on paper—
sufficient to identify the work but without retaining full‐size images once the
thumbnails were created. In addition, Amazon.com and Arriba Soft did not
retain copies in their databases. Kelly, 336 F.3d at 815; Perfect 10, 508 F.3d at
1156‐57.
16

2.

Google’s Distribution to the Libraries of Digital
Copies of All the Books it Copied Is Not a Fair Use.

Even if Google were able to advance a justification for the massive
unauthorized copying it claims is required for searching, that exception
cannot apply to Google’s creation of unauthorized digital copies, which it then
distributed to libraries as in‐kind payment for access to the books. Google
acknowledged the force of this argument when it argued before the District
Court that distribution to a library of digital copies of the books from that
library was not a distribution under Section 106. (Google Opp. to AG’s MSJ
[Doc. 1072] at 13‐14.)
Google created the Google Return Interface (GRIN) for the
purpose of distributing digital copies to libraries. Through GRIN, Google
makes available for downloading the digital copies made from the books that
it rented from a library, and the library can then download the digital version.
Because the library downloads the copy, Google claims that it has not
reproduced or distributed the copy. However after the download, the library
owns both the physical paper book and a digital copy of the book. It owns one
more copy of a “book” than it owned before its agreement with Google.

17

Google’s theory is not supported by the case law or the legislative history.7
The Court should reject this elevation of form over reality.
Other evidence supports the conclusion that Google was more
concerned with acquiring more “information” to sustain its dominant position
in the search business and avoiding an infringement claim than with creating
a philanthropic service for readers and scholars. Google apparently believed
that displaying works in copyright in full text as part of its search function
would be infringing, because it took steps to display those works only in
“snippets,” to return only three snippets per search from each work, and to
black out certain snippets and pages of a book.

7

Google’s argument is no different from those made and rejected in two
cases involving libraries: Diversey v. Schmidly, 738 F.3d 1196, 1202 (10th Cir.
2013) (placing unpublished dissertation on library shelves available to the
public was an unauthorized distribution to the public); Hotaling v. Church of
Jesus Christ of Latter‐Day Saints, 118 F.3d 199, 203 (4th Cir. 1997) (holding
that a library distributes a published work when it places an unauthorized
copy of the work in its collection, includes the work’s title in its catalog or
index system, and makes the copy available to the public); 2 Nimmer on
Copyright § 8.11[B][4][d] at 8‐154.10 (2013) (“No consummated act of actual
distribution need be demonstrated . . . to implicate the copyright owner's
distribution right."); see Peter S. Menell, In Search of Copyright’s Lost Ark:
Interpreting the Right to Distribute in the Internet Age, 59 J. Copyright Soc. of
the U.S.A. 1, 56‐58 (2011) (analyzing the legislative history of the distribution
right, §106(3), and concluding that it was intended to include “offers to
distribute” previously included in the 1909 Copyright Act’s rights to publish
and vend (emphasis in original)).
18

Google also excluded whole categories of works, such as poetry,
travel guides, encyclopedias, trivia books, indexes, etc. Yet, despite excluding
them from all displays to users, Google does not delete copies of these works
from its database. It provides no explanation why it maintains unauthorized
copies of copyrighted works whose display in the Library Project would be
infringing, and it makes no promises about what it may do with these copies
in the future. The irony is that Google’s decision results in the exclusion of
“indexes,” even though most professionally prepared indexes would often
provide more utility to readers and researchers than the word searches
Google offers.
Google offers copyright owners (although not prominently or very
clearly) procedures for excluding their works from scanning as part of the
Library program8 or for not displaying at all books that have been scanned9 by
submitting a form that includes a list of all works that should not be scanned.
This turns on its head the usual procedure by which a party seeking to
exercise Section 106 rights seeks the consent of the copyright owner in
advance. Instead, the burden is placed upon the copyright owner, who may

8

https://support.google.com/books/partner/answer/
3365282?hl‐en&ref_topic_3396243.
9
https://support.google.com/books/partner/answer/
2520009.
19

not even be aware of the conduct until long after it occurs. Even if a copyright
owner follows the procedure for excluding a scanned book, Google does not
agree to remove it from Google’s database. An earlier version of Google’s
support pages states that “We're happy to remove your book from our search
results at any time,”10 leaving the clear implication that the work remains in
Google’s database.
3.

The Utility of a Full‐Text Search Function
Does Not Justify Copying Over 20 Million Books.

Google campaigns for the utility of its full‐text word searches,
which return thousands of word or phrases, by comparing them to the library
catalogue MARC cards prepared by librarians, which provide bibliographic
and limited indexing information. But this comparison overlooks the many
professional and more sophisticated indices and guides to books, newspapers,
literature, and journals—none that involves widespread copyright
infringement—on which readers and researchers have relied for decades
before the arrival of Google’s Library Project (and still do).11

10

http://books.google.com/intl/en‐US/googlebooks/publisher_library.
html#options4 (emphasis added). 
11
See, e.g., Readers Guide to Periodical Literature, (H.W. Wilson (available
at http://www.hwwilsoninprint.com/periodicals.php)); Georgetown Law
Library’s Guide to Legal History Databases (available at https://www.law.
georgetown.edu/library/research/).
20

The examples Google offered the district court to show the alleged
superiority of its full‐text search mechanism are contrived. It observes that
the Library of Congress card catalogue has no entry for the search term, “500
Pearl Street” and using the word‐search in the Library Project leads to
information about a particular Chinese factory owner on the site in the late
19th century. But a search in that same catalogue for “New York City/history”
would turn up a reference to Gotham, A History of New York City to 1898 by
Edwin G. Burrows and Mike Wallace, which won the 1999 Pulitzer Prize for
History. The book has an extensive index, including a long entry for “Pearl
Street,” and references to the expansion of the Chinese community into the
Pearl Street area.
Much of the claimed utility of the Library Project concerns the
ability to search and discover ancient or otherwise long‐forgotten texts. If
making such books available to the public was Google’s true goal, it could have
worked with libraries to scan only books in the public domain, which includes
anything published before 1923. This approach—which Google rejected—

guides/legalHistory.cfm), including indexes to several foreign and domestic
newspapers; and the University of Connecticut Law Library’s List of Legal
Periodical Indexes (U.S. and foreign), including the Index to Legal Periodicals
1908‐1981 (available at http://library.law.uconn.edu/research‐resources/
legal‐periodical‐indexes).  
21

would have permitted Google to create a vast database of works while
respecting the rights of copyright owners. Instead, Google copied all books as
a way to maximize its profitability, and chose to display works in copyright in
“snippets,” thereby sacrificing utility in an ill‐fated attempt to avoid
infringement claims.
4. Google’s “Use is of a Commercial Nature.”
Addressing the question of whether Google’s use is of a
commercial nature, the district court acknowledged that “Google does, of
course, benefit commercially in the sense that users are drawn to the Google
websites by the ability to search [the more than 20 million] Google Books.”
954 F. Supp. 2d at 292‐93. This is exactly what Google describes as the
principal source of its income. Despite that admission, the district court
adopted verbatim Google’s argument that it “does not engage in the direct
commercialization of copyrighted works,” apparently because Google does not
include advertising on the specific “About a Book” pages of its website. In
other words, because Google has found a way to market this particular search
function as a philanthropic benefit for the public, it claims that it is not
commercializing copyrighted works. This Court should not be distracted by
Google’s marketing message.

22

Of course Google benefits commercially from the addition of
copies of over 20 million books to its search database. As it explained to the
SEC (Google 10‐K at 3‐4), the addition of additional “information” to its search
database is what attracts more and more people to Google—and its
advertising. And even if Google does not currently include advertising on its
“About a Book” pages, the search results that link to these results are full of
paid advertising,
Further supporting its conclusion about the commercial nature of
Google’s use, the district court relied on this Court’s observation that “fair use
has been found even where a defendant benefitted commercially from the
unlicensed use of copyrighted works,” (954 F. Supp. 2d at 291), relying on
Blanch v. Koons, 467 F.3d 244, 248, 258 (2d Cir. 2006) (defendant used only
the image of a woman’s leg from plaintiff’s photograph, which she had never
licensed, in a new work of art); Bill Graham Archives, 448 F.3d at 607, 609 (in a
480‐page pictorial biography of the Grateful Dead band, publisher included
for historical context reduced images of seven posters advertising the band’s
concerts); and Castle Rock Entm’t, Inc. v. Carol Publ’g Grp., Inc., 150 F.3d 132,
142, 145 (2d Cir. 1998) (finding defendant’s use of fictional facts from the
Seinfeld TV show in a trivia‐book repacked plaintiff’s work for the same
audience and preempted plaintiff’s creation of a similar trivia book). The
23

commercial character of the defendants’ uses in none of these cases compares
with Google’s use of multiple copies of over 20 million books, without
authorization or payment, to provide additional “information” for its multi‐
billion dollar search engine business. The district court’s suggestion that
Google’s commercial interest is not “direct” is insupportable; but even if
“indirect,” the commercial benefit to Google is so overwhelming that it should
result in a finding of commercial use.
Even if the Court concludes that Google’s Library Project offers a
useful online search mechanism, Google’s enormous income from the addition
of tens of millions of books to its search database and its unauthorized
distribution to the contributing libraries of digital copies of the books as
payment for the books’ rental do not justify the uncompensated use of so
many copyrighted books. The first fair‐use factor does not favor Google.
B.

The District Court Gave Short Shrift to
the Nature of the Copyrighted Works.
The district court acknowledged that Google has copied all kinds

of books—fiction, non‐fiction, children’s books, cookbooks, etc. Then, because
the court concluded that the “vast majority of the books in Google Books [not
the Library Project] are non‐fiction” (954 F. Supp. 2d at 24), it concluded that

24

the second factor favored fair use. This analysis of the value of non‐fiction
works is disturbing.
That non‐fiction books include facts and ideas does not mean that
a non‐fiction work does not also include extensive creative, moving,
persuasive, and enlightening expression that is every bit as copyrightable as is
a fictional work. The district court’s dismissal of fact works would have been
legitimate had an almanac been at issue, but it is not justified when the works
Google copied included a wide variety of expressive non‐fiction works.
The district court also took no account at all of the presence of
fictional works in Google’s Library Project. Even accepting the district court’s
finding about the high percentage (93%) of non‐fiction works in the Library
Project, this still means that the other 7% of works Google copied amount to
nearly 1.5 million fictional works, yet the district court made no mention of
them in its analysis of factor two. Before this factor can be weighed with the
others, it requires a more searching examination.
C.

Google Copies and Maintains Books in its Book
Database That It Does Not Use in the Library Project.
The district court concluded that the third fair‐use factor—the

amount and substantiality of the portion used in relation to the copyrighted
work as a whole—weighed slightly in favor of fair use based on the fact that

25

full‐text searching depends on copying the full text and that Google limits the
amount of text it displays in response to a search. This presumably referred
to Google’s display of snippets for works that are in copyright.
The district court did not explain why the display of snippets of
text justified copying an entire work and maintaining it in Google’s database.
The district court also completely ignored examples of Google’s policy of
copying and maintaining in its database copies of works that it does not use at
all—the certain categories of works Google excludes from display because
they are short and could be revealed completely in a snippet. Even though
these works are not revealed to any researcher or reader, Google still copies
and maintains copies of them in its database. Similarly, if an author
successfully manages to complete the papers necessary to exclude a scanned
work from the Library Project, Google does not exclude the work from its
database; it simply excludes it from the search results. (See links at nn. 8‐10,
supra.) Nor did Google exclude any works from the copies it made and
distributed to the libraries as payment for the original loan of the books.
The third fair‐use factor is supposed to examine whether a second
user has used more of the first work than was necessary to achieve its alleged
fair use. Google does not explain, and the district court ignored, why Google’s

26

copying far exceeds its needs for its alleged fair use, as well as Google’s
unnecessary distribution of millions of books to its library partners.
D.

Google’s Conduct Harms Authors.
The district court’s analysis of the fourth factor focused on the

possibility that the Library Project will help readers locate a book, which
could then lead to a purchase of the book and an economic benefit to the
author. But this assumption ignores the many ways that books can be located
without resorting to widespread copyright infringement. (See pp. 20‐21 &
n.11, supra.) This assumption also ignores the fact that most of the copyright‐
protected books in the Library Project are out of print and not available for
sale. By including, without compensation, millions of books in its database
that are not readily available for purchase, Google increased the performance
of its search results, but it did not benefit authors. It also interfered with the
developing licensing market for out‐of‐print books because Google forced
those licensees out of the market when it began copying without paying
licensing fees. (A1299‐1301.)
Due to the dominant position it has achieved in the marketplace,
Google has left little room for authors and publishers to develop partnerships
and licensing opportunities of their own. The clearest example of Google’s
actions leading to an economic loss for authors is Google’s decision to provide
27

libraries with complete digital copies of the books Google scans. The district
court wrongly implies that libraries are entitled to make copies of books they
own (954 F. Supp. 2d at 293), and thus ignored these lost sales. Because
Google provides these copies to libraries (as payment from Google for the
right to copy the books in the first place), libraries have no incentive to
purchase or license digital copies of the books from the copyright owners,
such as through print‐on‐demand programs or other programs that allow
authors to sell or license out‐of‐print books.
In contrast to the model of allowing a for‐profit company to
digitize millions of books for its own gain, many libraries and non‐profits
outside the United States have worked with authors and publishers to make
digital copies of books available online. These efforts—such as the Gallica
project which has made over 1,600,000 documents and 320,000 books in
French available online through the National Library of France—have been
accomplished while respecting national copyright laws. See www.bnf.fr/en/
collections_and_services/digital_libraries_gallica/a.gallica_experimentation_di
gital_offer.html.
The district court also ignored the threat to authors posed by the
potential loss of the digital copies of books Google distributed to libraries.
954 F. Supp. 2d at 287. While Google may be confident of its own ability to
28

secure copies of books on its servers, no evidence supports that conclusion for
Google’s library partners. A security breach of Google or its library partners
could result in the distribution of electronic copies of millions of copyrighted
books, thereby destroying the market for electronic copies of these works.
Authors should have at least the right to participate in the oversight of
procedures to provide security and prevent theft and piracy of their works.
The district court did not consider other adverse consequences of
Google’s Library Project. When Google copies and displays numerous prior
editions of books that are frequently updated, such as textbooks, Google can
mislead its users—and harm the reputation of authors—by presenting
information that is stale or no longer accurate. In addition, using so many
authors’ works as the subject of free, online searching may have the effect of
fostering the belief (if it has not done so already) that these works are or
should be available for free. While this may be an unintended consequence of
the general availability of material on the internet, Google’s very prominent
Books Program, which it promotes as a public good (despite Google’s financial
rewards) gives rise to an assumption of an entitlement to free access to
copyrighted works.
The district court considered none of these issues, undermining
its conclusion that the fourth fair‐use factor favors Google.
29

E.

Google Cannot Justify its Use of Works in the Library Project.
To decide this case, the Court must consider all the many factors

that weigh on either side of the equation to determine whether Google has
borne its burden of showing that its use of copyrighted works as part of the
Library Project is justified.12 Google (and the district court) often confuse the
works that are at stake by referring to “Google Books,” which includes works
in the Partners Program.
On one side of that balance is Google’s claim that its Library
Project helps users identify and locate books. This Court must consider
whether the public benefit such a project may have is justified by the harm to
copyright owners if Google’s Library Project is sanctioned as a fair use.
On the other side of the equation is the following: (1) Google, one
of the largest technology companies in the world, has made unauthorized
copies of millions of books for its own commercial benefit. (2) Without
authority to do so, Google has distributed unauthorized digital copies to each

This balancing does not involve any of the works in the Partners
Program, which are governed by contract. At issue here are copyrighted
works that Google displays in snippets, works Google has decided to exclude
because it deems the works too short to be displayed even as snippets, and
works excluded by authors who have directed Google to remove their works
from the Library Project. But these millions of “excluded” works remain in
Google’s book database.
12  

30

library of all the books it received from those libraries, disavowing by contract
any responsibility for the libraries’ uses of those unauthorized copies (A601 at
¶10.1). (3) The mere presence of Google’s royalty‐free program has
destroyed potential licensing opportunities for authors. (4) The creation of
millions of digitized copies for itself and the libraries has created security
risks concerning the control of those digital copies. (5) Google’s display of text
disembodied from its original context devalues the original works and fuels
the notion that creative works are commodities available for other uses. And
(6), despite being a company that has created enormous wealth by developing
new ways to deliver advertising to users, Google has made no representations
about how it will seek to profit further from the vast database of copyrighted
works it has created.
Given this unequal balance of competing factors, Google cannot
justify its mammoth Library Project.
This Court should not fall victim to Google’s attempt to avoid the
limits of the law by presenting the broader “Books Program” as a fait accompli
that is too big to fail. No example of fair use allows the degree of copying
undertaken by Google. The Library Project is not the type of creative
authorship the fair use doctrine was designed to protect, and the district

31

court’s effort to stretch the fair use doctrine beyond its limits should not be
sustained.
CONCLUSION
For the foregoing reasons, the district court’s order granting
summary judgment in favor of Google should be reversed.
New York, New York
April 14, 2014
Respectfully submitted,

______s/ Gloria C. Phares____________

Gloria C. Phares

Christopher M. Strong

HOFFMANN MARSHALL STRONG LLP
116 West 23rd Street, Suite 500
New York, NY 10011
(646) 741‐4503

Attorneys for Amici Curiae Authors

32

CERTIFICATE OF COMPLIANCE
This brief complies with the type‐volume limitation of Fed. R. App.
P. 32(a)(7)(B) because, excluding the parts of the brief exempted by Fed. R.
App. P. 32(a)(7)(B)(iii), it contains 6959 words as measured by the word‐
processing system used to prepare the brief (Microsoft Word); and
This brief complies with the typeface requirements of Fed. R. App.
P. 32(a)(5) and the type style requirements of Fed. R. App. P. 32(a)(6) because
it has been prepared in a proportionally spaced typeface,14‐point Cambria.

_______/s/ Gloria C. Phares___________
Gloria C. Phares

Attorneys for Amici Curiae Authors

Dated: April 14, 2014

33