Michael J. Boni (pro hac vice)
Joshua D. Snyder
John E. Sindoni
BONI & ZACK LLC
15 St. Asaphs Road
Bala Cynwyd, PA 19004
(610) 822-0200 (phone)
(610) 822-0206 (fax)
MBoni@bonizack.com
UNITED STATES DISTRICT COURT
SOUTHERN DISTRICT OF NEW YORK
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - x
Case No. 05 CV 8136-DC
FILED UNDER SEAL
PUBLIC VERSION
The Authors Guild, Inc., Associational Plaintiff,
Betty Miles, Joseph Goulden, and Jim Bouton,
individually and on behalf of all others similarly
situated,
Plaintiffs,
v.
Google Inc.,
Defendant.
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - x
PLAINTIFFS’ MEMORANDUM OF LAW IN OPPOSITION
TO GOOGLE’S MOTION FOR SUMMARY JUDGMENT
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 1 of 57
i
TABLE OF CONTENTS
Page
TABLE OF AUTHORITIES ..................................................................................................... iii
I. INTRODUCTION ................................................................................................................. 1
II. RESPONSE TO GOOGLE’S STATEMENT OF FACTS...................................................... 6
A. Google’s Massive Reproduction and Distribution to the Libraries .................................... 6
B. Google Displays Significant Portions of Books ................................................................ 9
C. Security Risks ................................................................................................................ 11
D. Google’s Pursuit of a Competitive Advantage and Increased Advertising Revenues ...... 12
III. ARGUMENT ...................................................................................................................... 13
A. Separate Uses Must Be Analyzed Separately ................................................................. 13
B. Google Has Infringed Plaintiffs’ Reproduction, Distribution and Display Rights ........... 14
C. Google’s Infringing Conduct Cannot Be Justified As a “Fair Use” of Plaintiffs’ Books . 18
1. Copyright Law Reflects a Careful Balance Between the Rights of Authors and
Users ....................................................................................................................... 19
2. The Four-Factor Test Is of Little Help in a Mass Digitization Use of this
Magnitude ............................................................................................................... 20
3. The Four-Factor Test ............................................................................................... 21
a. Google Cannot Satisfy Factor One ..................................................................... 21
i. Google’s Uses Are Plainly Commercial........................................................ 21
ii. The Challenged Uses Are Not Transformative .............................................. 23
(a) Courts Have Rejected Fair Use Defenses Involving Less Verbatim
Copying, Display and Distribution .......................................................... 23
(b) Google’s Attempted Justification of Its Index Does Not Render Its
Other Uses Transformative ..................................................................... 27
iii. Google’s “Customary Practice” Argument Lacks Merit ................................ 33
b. Factor Two Does Not Favor Google ................................................................... 36
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 2 of 57
ii
c. Google Cannot Satisfy Factor Three ................................................................... 36
d. Google Cannot Satisfy Factor Four .................................................................... 42
4. Aggregate Assessment ............................................................................................. 46
D. Google’s Library Distribution Is Not Fair Use ............................................................... 46
IV. CONCLUSION ................................................................................................................... 49
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 3 of 57
iii
TABLE OF AUTHORITIES
Page(s)
FEDERAL CASES
A.V. ex rel. Vanderhye v. iParadigms, LLC,
562 F.3d 630 (4th Cir. 2009) ............................................................................................... 32
American Geophysical Union v. Texaco, Inc.,
60 F.3d 913 (2d Cir. 1995) ....................................18, 20-22, 24, 28, 30, 33, 34, 39, 41, 43-46
American Geophysical Union v. Texaco Inc.,
802 F. Supp. 1 (S.D.N.Y. 1992) ...........................................................................................20
Anderson v. Liberty Lobby, Inc.,
477 U.S. 242 (1986) ............................................................................................................ 19
Authors Guild v. Google Inc.,
770 F. Supp. 2d 666 (S.D.N.Y. 2011) ..................................................................................20
Authors Guild, Inc. v. HathiTrust, Inc.,
902 F. Supp. 2d 445 (S.D.N.Y. 2012) ..................................................................................25
Basic Books, Inc. v. Kinko’s Graphics Corp.,
758 F. Supp. 1522 (S.D.N.Y. 1991) ................................................................................24-25
Bill Graham Archives v. Dorling Kindersley Ltd.,
448 F.3d 605 (2d Cir. 2006) ............................................................................ 3, 30-31, 38-39
Blackwell Publishing, Inc. v. Excel Research Group, LLC,
661 F. Supp. 2d 786 (E.D. Mich. 2009) ...............................................................................16
Blanch v. Koons,
467 F.3d 244 (2d Cir. 2006) ..................................................................................................4
Bouchat v. Baltimore Ravens Ltd. Partnership,
619 F.3d 301 (4th Cir. 2010) ............................................................................................... 14
Campbell v. Acuff-Rose Music, Inc.,
510 U.S. 569 (1994) ........................................................................... 4, 21, 23, 27, 32, 36, 37
Capitol Records, LLC v. ReDigi, Inc.,
No. 12 Civ. 95, 2013 WL 1286134 (S.D.N.Y. Mar. 30, 2013) ...................................... 18, 20
Cariou v. Prince,
714 F. 3d 694 (2d Cir. 2013) ................................................................................ 4, 26-27, 32
Cartoon Network LP, LLLP v. CSC Holdings, Inc.,
536 F.3d 121 (2d Cir. 2008) ...........................................................................................15-17
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 4 of 57
iv
Castle Rock Entertainment, Inc. v. Carol Publishing Group, Inc.,
150 F.3d 132 (2d Cir. 1998) .....................................................................................24-26, 42
Daubert v. Merrell Dow Pharmaceuticals, Inc.,
509 U.S. 579 (1993) ............................................................................................................ 35
Ehrens v. Lutheran Church,
385 F.3d 232 (2d Cir. 2004) ................................................................................................ 35
Ford Motor Co. v. Summit Motor Products, Inc.,
930 F.2d 277 (3d Cir. 1991) ............................................................................................... 15
Harper & Row Publishers, Inc. v. Nation Enterprises,
471 U.S. 539 (1985) ............................................................................................ 5, 24, 33, 38
Hotaling v. Church of Jesus Christ of the Latter-Day Saints,
118 F.3d 199 (4th Cir. 1997) ............................................................................................... 15
Infinity Broadcast Corp. v. Kirkwood,
150 F.3d 104 (2d Cir. 1998) ....................................................................... 4-5, 22, 24, 37, 48
Kelly v. Arriba Soft Corp.,
336 F.3d 811 (9th Cir. 2003) ................................................................................ 5, 31-32, 37
Kirtsaeng v. John Wiley & Sons, Inc.,
133 S. Ct. 1351 (2013) ........................................................................................................ 19
Metro-Goldwyn-Meyer Studios Inc v. Grokster, Ltd.,
545 U.S. 913 (2005) ............................................................................................................ 18
National Football League v. Prime Time, 24 Joint Venture,
211 F.3d 10 (2d Cir. 2000) .................................................................................................. 17
Nihon Keizai Shimbun, Inc. v. Comline Bus. Data, Inc.,
166 F.3d 65 (2d Cir. 1999) .................................................................................................. 24
On Davis v. The Gap, Inc.,
246 F.3d 152 (2d Cir. 2001) .....................................................................................24, 29-30
Paramount Pictures Corp. v. Labus,
No. 89-C-797-C, 1990 WL 120642 (W.D. Wis. Mar. 23, 1990) ...........................................15
Perfect 10, Inc. v. Amazon.com, Inc.,
508 F.3d 1146 (9th Cir. 2007) .............................................................................. 5, 14, 31-32
Radji v. Khahbaz,
607 F. Supp. 1296 (D.D.C. 1985) ........................................................................................25
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 5 of 57
v
Ringgold v. Black Entertainment Television, Inc.,
126 F.3d 70 (2d Cir. 1997) ............................................................................................ 24, 30
Salinger v. Colting,
607 F.3d 68 (2d Cir. 2010) .................................................................................................. 24
Sega Enterprises Ltd. v. Accolade, Inc.,
977 F.2d 1510 (9th Cir. 1992) ........................................................................................40-41
Sony Corporation of America v. Universal City Studios, Inc.,
464 U.S. 417 (1984) ................................................................................................ 16, 20, 39
Twin Peaks Productions v. Publications International Ltd.,
996 F.2d 1366 (2d Cir. 1993)............................................................................. 14, 24, 26, 29
UMG Recordings, Inc. v. MP3.com, Inc.,
92 F. Supp. 2d 349 (S.D.N.Y. 2000) .............................................................................. 24, 29
United Public Workers of America (C.I.O.) v. Mitchell,
330 U.S. 75 (1947) .............................................................................................................. 13
Universal City Studios, Inc. v. Nintendo Co.,
746 F.2d 112, 118 (2d Cir. 1984) ......................................................................................... 35
Video Pipeline, Inc. v. Buena Vista Home Entertainment, Inc.,
342 F.3d 191, 199 (3d Cir. 2003) ......................................................................................... 28
WNET, Thirteen v. Aereo, Inc.,
712 F.3d 676 (2d Cir. 2013) ...........................................................................................19-20
FEDERAL STATUTES
17 U.S.C. § 106(1) .................................................................................................. 14, 15, 16, 18
17 U.S.C. § 106(3) ........................................................................................................ 14, 15, 16
17 U.S.C. § 106(4) .................................................................................................................... 15
17 U.S.C. § 106(5) .................................................................................................................... 15
17 U.S.C. § 107 ............................................................................................................. 13, 20, 22
17 U.S.C. § 107(1) .................................................................................................................... 21
17 U.S.C. § 107(2) .............................................................................................................. 36, 39
17 U.S.C. § 107(3) .............................................................................................................. 36, 39
17 U.S.C. § 107(4) .................................................................................................................... 42
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 6 of 57
vi
17 U.S.C. § 108 ......................................................................................................................... 19
RULES
Fed. R. Civ. P. 56(c)(4) ............................................................................................................. 35
OTHER AUTHORITIES
Campuses Face Rising Threat from Hackers,
N.Y. TIMES, July 17, 2013, at A1 ..................................................................................... 3, 12
Shari Seidman Diamond, Reference Guide on Survey Research, in REFERENCE MANUAL
ON SCIENTIFIC EVIDENCE (SECOND) (2000) .......................................................................... 35
Pierre N. Leval, Toward a Fair Use Standard,
103 HARV. L. REV. 1105 (1990).............................................................. 23, 24, 27, 29, 32, 42
MANUAL FOR COMPLEX LITIGATION (FOURTH) § 11.493 (2004) .................................................35
2 MELVILLE B. NIMMER & DAVID NIMMER, NIMMER ON COPYRIGHT § 8.11 ............................. 15
4 MELVILLE B. NIMMER & DAVID NIMMER, NIMMER ON COPYRIGHT § 13.05 ............... 20, 23, 37
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 7 of 57
1
I. INTRODUCTION
Google argues that its mass digitization of millions of copyright-protected books, its
display of much of those books to users of its search engine, and its distribution of digital books
to its library partners are transformative and harmless to the potential market or value of the
copyright-protected works. This is false, as we will demonstrate below. A single example here,
however, belies Google’s claim.
Google digitized plaintiff Jim Bouton’s baseball memoir Ball Four, as it did millions of
other in-print books, without the permission of the author or his publisher. Google claims that
displaying “snippets” of Ball Four to users of its search engine is transformative and cannot
possibly harm the market for Mr. Bouton’s literary property. But before Google came along to
copy and display his book, Mr. Bouton, through his publisher, licensed Amazon to display full
pages of Ball Four to those searching the retailer’s online bookstore. Whatever Google believes
“transformative” means, it cannot logically be defined to include a less useful display of text than
the one licensed by the rightsholder to a major online bookseller. In other words, Google’s
digitization and display of snippets cannot be viewed as transforming Ball Four by using the
work for a new purpose or adding something new, when Mr. Bouton and his publisher had
already licensed the book for limited online display.
What really happened here is that Google decided to use portions of Mr. Bouton’s book
to enhance its search engine, drive advertising revenue and gain an advantage over its search
engine competitors. Thus the only thing “transformative” about Google’s display of snippets of
in-print books is that it transforms online browsers of book retailers to online users of Google’s
search engine. Google “transforms” Amazon customers into Google ad-clickers. (See
Declaration of Paul Aiken (“Aiken Decl.”) ¶¶ 4-7, 33.)
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 8 of 57
2
Google also falsely asserts that its copying and display cannot harm the market for
copyrighted books. After all, it argues, Mr. Bouton and countless other authors and publishers
that have permitted other online retailers to display portions of their books necessarily must be
fine with what Google has done. What harm could there be in Google displaying snippets of
those same books? Plenty. Readers browsing books at online retailers are always at a virtual
checkout counter, just one click away from purchasing the author’s book. Through its display of
snippets, Google lures online browsers of Ball Four away from Amazon, which is optimized for
selling books, and draws them to Google’s webpages, which, no doubt, are optimized for
encouraging users to stay within Google’s ecosystem and click on the ads that Google sells. (Id.
¶¶ 4-5.)
Then there is the matter of Google’s decision to distribute its unauthorized digital copy of
Ball Four, and millions of other books, to university libraries. These distributions are gratuitous,
completely unrelated to Google’s supposed “transformative” use of the books it spent hundreds
of millions of dollars to digitize.1 By choosing to distribute the digitized books to university
libraries without retaining any effective means of controlling the digital book database, including
such rudimentary matters as assuring that the works are protected by commercial-grade security
and not stored online, Google has put much of the world’s most valuable literary property at risk.
One need only pick up a daily newspaper to appreciate how serious this security risk is. Stories
of break-ins, hacking and theft appear almost daily. Even in the days immediately following the
return of this case to the Court for a determination of fair use, front page headlines described
1 These distributions are, of course, contractually mandated—Google agreed to give the digitized
books to the libraries as payment for being allowed to scan the stacks of the libraries—but
Google could have negotiated a fee to compensate the universities. Google’s pockets are deep,
and every university’s need for funding is nearly endless.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 9 of 57
3
break-ins into major university websites across the country. See, e.g., Campuses Face Rising
Threat from Hackers, N.Y. Times, July 17, 2013, at A1.
Fair use cannot possibly condone such actions, and Google cannot possibly meet its
burden of showing that its distribution of digital books does not harm the potential market or
value of the copyrighted works. Google’s Library Project bestows on university libraries digital
copies that they otherwise would have to pay for.
What about out-of-print books? Out-of-print books have commercial value, and that
value is increasingly being realized, as print-on-demand and e-book technologies lower barriers
to entry and thousands upon thousands of out-of-print books become commercially available
again. Google’s actions recklessly put the value of those millions of books, representing billions
of hours of effort by their authors, at risk. (Aiken Decl. ¶ 34.)
Until the Google Library Project came along, authors and other copyright owners
primarily controlled the way that others could use their works. Rightsholders decided whether
and when their works were duplicated, put into new formats or used to create derivative works.
Rightsholders had the right to decide when to grant rights in their works, and when to withhold
those rights for possible future exploitation. They also had the critical right to decide how to
balance their interest in promoting sales against the security risks posed by digital availability.
Rightsholders know, of course, that there are limits to their rights. Educators and
researchers may use portions of copyrighted works for scholarly or educational purposes; artists
may use others’ works as building blocks for new creations; parodists can make fun of older
books, songs and art. One may use copies of a few images in a book such as the biography of
the Grateful Dead at issue in Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605,
614 (2d Cir. 2006), where seven small images of concert posters were included in a 480-page,
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 10 of 57
4
2000 illustration history of the band. An artist may alter pre-existing works with entirely new
expression, such as the photographs that were the subject of the recent Second Circuit cases
Blanch v. Koons, 467 F.3d 244 (2d Cir. 2006), and Cariou v. Prince, 714 F.3d 694, 706 (2d Cir.
2013) (while a work need not “comment on the original or its author to be considered
transformative,” it still must “alter the original with ‘new expression, meaning or message’”)
(quoting Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579 (1994)).
Nothing in these or other cases, however, comes close to justifying what has happened
here. Google has turned copyright on its head by subverting the copyright interests of hundreds
of thousands of authors of millions of books, and grabbing those rights for itself. Whatever
mantra Google may chant about the public value of making all of the world’s works available to
everyone, Google is heart and soul an enormous commercial enterprise, and its various uses of
the copyrighted books in its Library Project are designed to gain a competitive advantage over
other search engines and to generate even greater advertising revenues.
Google chose not to seek the authors’ permission to include the books in its digital
database. In spite of its vast resources, spending hundreds of millions of dollars on digitization
alone, Google decided to leave authors entirely out of the process. Instead, it chose to purchase
the ability to make unauthorized copies from university libraries by reproducing and distributing
a digital copy of the authors’ works to the libraries, bypassing the authors’ rights. In doing so,
Google used millions of copyrighted works as currency to buy the “right” to digitize the
libraries’ hard copies, in order to further Google’s commercial interests.
Google cannot hide its commercial purpose by pointing to possible benefits to
hypothetical end users. Almost any act of copyright infringement likely benefits someone, but it
is Google’s use—not the use of hypothetical end users—that must be analyzed. Infinity Broad.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 11 of 57
5
Corp. v. Kirkwood, 150 F.3d 104, 108 (2d Cir. 1998) (focusing on “Kirkwood’s own
retransmission of the broadcasts, not the acts of his end-users”). Google indisputably used the
books to enhance the value of its search engine, gain an advantage over its competitors and
generate even greater advertising revenues.
In Google’s brief in support of its motion for summary judgment (“Google Br.”), Google
not only denigrates the rights of authors, but it also completely disregards the huge security risks
caused by the online maintenance of digital files. Google relies on two Ninth Circuit decisions
in very different situations. Kelly v. Arriba Soft Corp., 336 F.3d 811 (9th Cir. 2003), and Perfect
10, Inc. v. Amazon.com, Inc., 508 F.3d 1146 (9th Cir. 2007), involve search engine access to
works whose rightsholders made available on the Internet. Those cases are of little relevance
here, where it was Google, not the individual rightsholders, that: (a) made the decision to
digitize print books, exposing them to the risks and perils of Internet security compromises, and
(b) provided digital copies to the libraries with virtually no control or assurance of security.
In short, enabled by its enormous wealth and emboldened by its position as the world’s
leading search engine, Google has taken it upon itself to upset the balance between the rights of
copyright creators and copyright users that Congress so carefully established in copyright law.
Just because Google is the first business entity with the capacity and temerity to digitize, copy,
distribute and display millions of copyrighted works without authority, does not mean that it
should be allowed to do so. See Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539,
559 (1985) (“[T]o propose that fair use be imposed whenever the social value [of dissemination]
. . . outweighs any detriment to the artist, would be to propose depriving copyright owners of
their right in the property precisely when they encounter those users who could afford to pay for
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 12 of 57
6
it.”) (alteration in original). Through its actions, Google completely disregards the risks it is
taking on behalf of authors, and the role of authors in the copyright balance.
Google has a heavy burden to prove that its mass digitization Library Project is a fair use.
Plaintiffs demonstrate herein that it has not met its burden, and that summary judgment should be
denied to Google and granted to plaintiffs.
II. RESPONSE TO GOOGLE’S STATEMENT OF FACTS
In its statement of facts, Google concentrates on two discrete uses of Google Books.
(Google Br. at 4-18.) First, Google places heavy reliance on the card-catalogue, or “indexing”
feature of Google Books. Second, Google extols the virtues of a computer application called the
“n-gram viewer.” These are beside the point. What really is at issue is Google’s massive
digitization and reproduction of millions of copyrighted books for its own competitive gain.
Google minimizes (if not misrepresents) the degree to which it displays verbatim content from
the millions of books it digitized and placed online, and the deals it made with its library partners
providing them with millions of digital reproductions of copyright-protected books. It also
ignores that it is putting millions of valuable copyrighted works at risk by exposing them to the
all too real risks of hacking, theft and widespread distribution.
A. Google’s Massive Reproduction and Distribution to the Libraries
To date, in its Library Project, Google has digitally copied over four million in-copyright
English language books, and displayed verbatim expression as snippets from millions of in-
copyright books over the Internet in response to search requests from its users. (Plaintiffs’
Statement of Undisputed Facts in Support of Their Motion for Summary Judgment, ECF No.
1054 (“Pls.’ 56.1 Stmt.”) ¶ 53.) Google had distributed approximately 2.7 million scanned books
to the partnering libraries, as of March 26, 2012. (Pls.’ 56.1 Stmt. ¶ 5.) There is no dispute that
Google has made multiple complete copies of plaintiffs’ works (“Reproduction”), and has
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 13 of 57
7
provided them to libraries as payment for the libraries permitting Google to scan the libraries’
hard copies (“Library Distribution”). Pursuant to their agreements with Google, attached as
Exhibit 23 to the Declaration of Joanne Zack in Support of Plaintiffs’ Motion for Partial
Summary Judgment (ECF No. 1053) (“Zack Decl.”), each library provides books to Google,
Google scans and digitizes the books, retains a digital file of the books for Google’s use, and
distributes digital copies of scanned books to the providing library. (Pls.’ 56.1 Stmt.¶ 57.)
That digital distribution—Google’s provision of millions of unauthorized digital copies to
cooperating libraries—is a critical component of the Library Project. (Pls.’ 56.1 Stmt. ¶¶ 66-67.)
As Google explains, in exchange for access to a library’s print books, “[e]ach library will receive
a digital copy of every book we scan … from their respective collections.” (Pls.’ 56.1 Stmt. ¶
30.) Google cares little what the libraries do with the digitized, copyright-protected books. Its
agreement with Stanford University, for example, provides
(Pls.’ 56.1 Stmt. ¶ 58; emphasis supplied.) Agreements with other
libraries
(Pls.’ 56.1 Stmt. ¶ 64.) Google’s library agreements also provide that Google
transfers the ownership of the digital copies it has created to the libraries. (Pls.’ 56.1 Stmt. ¶¶ 68,
69, 70.)
Google’s library agreements presume the libraries’ further use of the digitized works.
(Pls.’ 56.1 Stmt. ¶¶ 71-72.)
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 14 of 57
8
When Google states that “[n]o library may obtain a digital copy created from another
library’s book—even if both libraries own identical copies of that book” (Google Br. at 15),
Google seeks to confuse the specific technology by which it transmits digital copies, on the one
hand, with the libraries’ contractual rights to share with third parties and access each other’s
works, on the other. Paragraph eight of Google’s Jaskiewicz declaration explains how Google
uses its GRIN system to transmit an encrypted digital copy only to the library that submitted the
book.2 The contracts between Google and the libraries, however,
2 GRIN stands for “Google Return Interface.” The libraries can download digital copies of every
book they submit for scanning. To download the copy, the library sends a request through the
GRIN system. The request triggers the creation of an encrypted copy of the book from the
(footnote continued)
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 15 of 57
9
As further examples,
B. Google Displays Significant Portions of Books
Since 2005, pursuant to uniform rules of its own devising, Google has displayed verbatim
expression from these books on the Internet in response to search requests by users of its search
engine (hereinafter defined as “Verbatim Display”). (Pls.’ 56.1 Stmt. ¶ 43.) Google generally
divides each page into eighths, each of which Google calls a snippet. (Pls.’ 56.1 Stmt. ¶ 44.) By
performing multiple searches using different search terms (including multiple search terms
suggested by Google), a single user can view far more than three snippets from a Library Project
book. (Pls.’ 56.1 Stmt. ¶ 45.) For example, once a searcher retrieves a book (such as searching
for “Steve Hovley” and finding as a search result Ball Four, Google Br. at 13), Google Books
does not stop there. Rather, the Internet user can then rummage through Ball Four for any other
search terms he or she chooses (“fastball,” “curveball,” “slider,” “spitball,” “walk,” “Mickey
Mantle”) and Mr. Bouton’s verbatim words on these subjects are displayed in three “snippets”
for each search. No commentary, criticism, or new content is added by Google at all—rather, the
sentences are displayed exactly as Mr. Bouton wrote them. Google itself lists dozens of
suggested terms for the user to search. See Boni Dec. Ex 1; Zack Decl. Exs. 4, 5, 6, 7 & 8.
central copy that Google retains. The library then receives a unique encryption key that is used to
download the encrypted digital copy. (Jaskiewicz Decl. ¶ 6-8, ECF No. 1041.)
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 16 of 57
10
Google downplays its Verbatim Display. It claims: “In some cases, users will also see a
small amount of text from the book.” (Google Br. at 9, citing Clancy Decl. ¶ 10.) The use of “in
some cases” and “small amount” grossly understates the number of books that are displayed
online, and the amount of content that is actually displayed. The reality is that Google makes
available approximately 78% of the verbatim text of each book.3 Moreover, by excluding what it
characterizes as reference works, such as dictionaries and cookbooks, from snippet view (Google
Br. at 10-11, citing Clancy Decl. ¶ 11), Google implicitly acknowledges that the Verbatim
Display of even a small portion of a work could be detrimental to the sales of that work. But
Google offers no explanation as to why the line should be drawn where Google has arbitrarily
chosen to draw it. Take for example, Jim Bouton’s Ball Four. There is no rational distinction
between a user who can learn what Mr. Bouton had to say about Steve Hovley, or Mickey
Mantle, or another player, by reading the relevant snippet and a user who could obtain the entire
relevant portion of a “reference” work by reading the relevant snippet. In other words, a reader
might want to know what Mr. Bouton had to say about Mickey Mantle, and Google’s display of
what it calls snippets would allow such reader to find out everything Mr. Bouton had to say
about his former teammate without buying the book.
Examples of snippet displays for one book each of the three named plaintiffs are attached
in the accompanying declaration. It bears note that these are representative, and are far less than
the total of what Google admits it makes available for public display.
3 Google concedes that it makes available for display 7 out of 8 snippets per page, and 9 out of
10 pages of each book. (Google Br. at 10.) That means that 7/8 of the snippets are available for
display on each of the 90% of the pages that are not blacklisted, or approximately 78% of each
book is available for display.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 17 of 57
11
Google displays from Jim Bouton’s Ball Four, including Google’s suggested
terms to search (Boni. Decl. Ex. 1). This exhibit contains 222 unique snippets and
about 11,611 words of verbatim text;
Google displays from Joseph Goulden’s history Super-Lawyers, including
Google’s suggested terms to search (Boni Decl. Ex. 2). This exhibit contains 220
unique snippets and about 9,676 words of verbatim text;
Google displays from Betty Miles’ novel The Trouble with Thirteen, including
Google’s suggested terms to search (Boni Decl. Ex. 3). This exhibit contains 61
unique snippets and about 1,760 words of verbatim text.
Google’s Verbatim Display is not only far greater than Google would have us believe, it
is not logically (or otherwise) connected to the indexing or the n-gram functions upon which
Google seeks to justify the Library Project. Google admits as much. (See Boni Decl. Ex. 4
(
).)
C. Security Risks
Through its Verbatim Display, Reproduction, and Library Distribution, Google has
subjected millions of copyrighted works to unacceptable security risk. Google acknowledges
there is always a risk that its “security measures may be breached due to the actions of outside
parties, employee error, malfeasance, or otherwise, and, as a result, an unauthorized party may
obtain access to our data or our users’ or customers’ data.” (Pls.’ 56.1 Stmt. ¶ 107, quoting
Google 10-K.) Additionally, Google is not contractually obligated to, and does not in practice,
monitor or control the security of the digital copies of books it provided to its library partners,
and the security measures of libraries who receive digital copies of books from Google are
subject to similar breaches. (Pls.’ 56.1 Stmt. ¶ 108.) Subsequent copying by the libraries of the
digital files provided by Google creates an even greater risk of a potential security breach. See,
e.g., Campuses Face Rising Threat from Hackers, N.Y. Times, July 17, 2013, at A1. Google’s
primary response is that—in the first several years of the Library Project—a security breach has
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 18 of 57
12
not happened yet. However, it is undeniable that just one security breach could result in
devastating losses to the rightholders of the books Google has subjected to the risk of such a
breach by digitizing them and placing them on the Internet. (Pls.’ 56.1 Stmt. ¶¶ 109-118, citing
Edelman Report, attached as Exhibit 39 to the Zack Decl; see also Pls.’ Opening Br. at 44-46.)
D. Google’s Pursuit of a Competitive Advantage and Increased Advertising
Revenues
The motivation behind Google’s massive digitization of millions of copyrighted works,
for which it has invested $180 million in scanning alone (Pls.’ 56.1 Stmt. ¶ 81), is its own
competitive advantage. Google operates the largest Internet search engine in the world, which it
monetizes by running advertisements in response to search queries. (Pls.’ 56.1 Stmt. ¶ 79.)
Google reported over $36.5 billion in “advertising revenues” for 2011 and $29 billion in revenue
generated “primarily by delivering relevant, cost-effective online advertising” the previous year.
(Pls.’ 56.1 Stmt. ¶¶ 11-12.) Google undertook the Library Project for commercial reasons in
order to gain a competitive advantage, drive traffic to its already lucrative search engine, and
gain even greater advertising revenues. (Pls.’ 56.1 Stmt. ¶ 75-77.) As one Google witness
testified, “to the extent other search engines were not similarly investing in such a project, then
the belief [within Google] would be that this would enhance our experience for users and,
therefore, improve the product we might offer, which would tend to give us an advantage.” (Pls.’
56.1 Stmt.¶ 78, quoting Clancy Dep. at 141.) As one Google internal presentation states, Google
“want[s] web searchers interested in book content to come to Google not Amazon,” and
“[e]verything else is secondary . . . but make money.” (Pls.’ 56.1 Stmt. ¶ 76.)
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 19 of 57
13
III. ARGUMENT
Google’s motion focuses upon fair use, effectively conceding literal infringement.
Plaintiffs addressed the fair use argument in their opening brief (ECF No. 1050), which is
incorporated by reference, but make these additional points to respond to the arguments raised in
Google’s brief.
A. Separate Uses Must Be Analyzed Separately
Google’s brief proceeds on the implicit assumption that if any of Google’s uses constitute
“fair use,” then all of them must. This is as fallacious as arguing that if one use of a gun is legal,
every use of that gun must also be legal. Translated to this case, Google focuses its argument
almost exclusively upon indexing and its n-gram program. But these are irrelevant if the Court
determines that Reproduction, Verbatim Display and Library Distribution do not constitute fair
use. Those uses must be analyzed separately from indexing and the n-gram program.4 In a
copyright infringement case, a Court does not weigh or balance the n-gram program versus
Verbatim Display to make a determination of whether “on balance” all of Google’s actions are
fair use. Rather, each separate use must be examined to determine whether it is justified as a fair
use or not. The Copyright Act is written in the singular, and requires ascertaining “whether the
use made of a work in any particular case is a fair use.” 17 U.S.C. § 107 (emphasis supplied). It
4 In this regard, the amicus brief of The Digital Humanities and Law Scholars (ECF No. 1055) is
irrelevant to any issue in this lawsuit. They specifically disclaim any interest in the “expressive
uses” of any work and state they are concerned only with “metadata” and focus on Google’s n-
gram viewer. (Id. at 1, 3, 7-8.) Such uses are not at issue in this lawsuit. Yet the amicus Digital
Scholars appear to be seeking an advisory opinion from this Court that the metadata research
they do does not infringe copyrights. However, it is well established that federal courts do not
render advisory opinions. United Pub. Workers of Am. (C.I.O.) v. Mitchell, 330 U.S. 75, 89
(1947). For this reason, the arguments in the amicus brief of the American Library Association
et al. (ECF No. 1048), focused on indexing, pp. 5-6, and the n-gram viewer, pp. 7-8, are similarly
irrelevant.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 20 of 57
14
does not ask whether “all possible uses, taken in the aggregate, are fair.” See, e.g., Twin Peaks
Prods. v. Publ’ns Int’l Ltd., 996 F.2d 1366, 1375 (2d Cir. 1993) (finding just two chapters of a
book about a TV series to be non-transformative and therefore not fair use, and not weighing or
balancing the transformative and non-transformative chapters against each other); Bouchat v.
Baltimore Ravens Ltd. P’ship, 619 F.3d 301, 306, 308-17 (4th Cir. 2010) (where artist sued
football team using his copyrighted logo in three ways—in season highlight films, in film clips
played at football games, and in photographs and memorabilia at team’s headquarters—the
Fourth Circuit considered each use distinctly and concluded that only the use of the logo at the
team’s headquarters was fair use).
The principal that each alleged use must be analyzed separately is well illustrated by a
case upon which Google relies heavily, Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146 (9th
Cir. 2007). While the Ninth Circuit upheld as fair use one of Google’s uses—thumbnail displays
of images—the court determined that Google could be liable for another use—the in-line linking
to full-sized infringing images. Id. at 1169-73. In sum, Google’s two separate uses of these same
Internet images were analyzed separately.
Similarly, whether or not Google’s indexing and n-gram viewing uses constitute fair use
do not bear upon whether Google’s Verbatim Display, Reproduction, and Library Distribution
constitute fair use. They must be analyzed separately.
B. Google Has Infringed Plaintiffs’ Reproduction, Distribution and Display
Rights
As plaintiffs explained in their opening brief in support of their motion for summary
judgment, Google has infringed plaintiffs’ copyrights by the unauthorized reproduction of in-
copyright books under 17 U.S.C. § 106(1) (“Reproduction”), see Pls.’ Opening Br. at 13-14;
unauthorized distribution of in-copyright books under 17 U.S.C. § 106(3) (“Library
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 21 of 57
15
Distribution”), see Pls.’ Opening Br. at 15-16; and unauthorized display of in-copyright books
under 17 U.S.C. § 106(5) (“Verbatim Display”), see Pls.’ Opening Br. at 16-17.
In its motion for summary judgment, Google does not dispute that it has scanned,
digitized and copied millions of copyrighted books in violation of plaintiffs’ reproduction rights
under Section 106(1) or that it has displayed portions of those works in violation of Section
106(5), focusing solely on whether such reproduction or display constitutes fair use.
Google argues at length that its provision of books to various libraries does not infringe
plaintiffs’ distribution right under Section 106(3) of the Copyright Act. (See Google Br. at 32-
39.) Google argues, citing the Second Circuit’s Cablevision decision, Cartoon Network LP,
LLLP v. CSC Holdings, Inc., 536 F.3d 121 (2d Cir. 2008), that Google’s provision of digital
copies of books for the libraries through the GRIN system, see p. 8, note 2, above, does not
implicate plaintiffs’ distribution right because the volitional act to generate a copy is undertaken
by the library and not by Google. It also argues that providing copies of the books to the
libraries is not a distribution to the “public” and therefore cannot violate Section 106(3).5
5 At the threshold, the “to the public” requirement of Section 106(3) is not steep. The Third
Circuit held it meant distribution to a single person. Ford Motor Co. v. Summit Motor Prods.,
Inc., 930 F.2d 277, 299-300 (3d Cir. 1991). Although the Second Circuit indicated in
Cablevision, 536 F.3d at 139, that this was likely too expansive a reading, without stating how
many more persons were necessary, that statement was dicta because Cablevision did not involve
distribution under § 106(3), but rather “public performance” under § 106(4). Nimmer on
Copyright suggests an interpretation in which Ford Motor may be correct. 2 MELVILLE B.
NIMMER & DAVID NIMMER, NIMMER ON COPYRIGHT § 8.11[C][1][b] at p. 8-154.16 n.139.
Besides Ford Motor, there have not been many decisions construing what “to the public” means
in § 106(3). See, e.g., Paramount Pictures Corp. v. Labus, 1990 WL 120642, at *4-5 (W.D.
Wis. Mar. 23, 1990) (distribution to the public found when owner of a small motel rented a small
number of videotapes to a few guests); see also Hotaling v. Church of Jesus Christ of the Latter-
Day Saints, 118 F.3d 199, 203 (4th Cir. 1997) (“When a public library adds a work to its
collection, lists the work in its index or catalog system, and makes the work available to the
borrowing or browsing public, it has completed all the steps necessary for distribution to the
public.”).
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 22 of 57
16
Google’s argument presents the reddest of red herrings. Not only is Google wrong about
its violation of plaintiffs’ distribution right under Section 106(3), but it does not matter. Even if,
arguendo, Google were to prevail on its distribution argument, the undisputed facts show that
Google engaged in repeated violations of plaintiffs’ Reproduction right under Section 106(1) by
itself reproducing, or by authorizing and enabling the libraries, to reproduce millions of books to
be stored on the libraries’ own servers for the libraries’ own uses.
First, Google’s use of its GRIN program plainly implicates plaintiffs’ distribution right.
See Blackwell Publ’g, Inc. v. Excel Research Group, LLC, 661 F. Supp. 2d 786, 791-92 (E.D.
Mich. 2009) (rejecting argument by defendant Excel, provider of photocopied coursepacks
containing copyrighted materials, that because the students actually did the physical copying, it
could not be held responsible as an infringer). There the court noted that “[w]hile ‘[t]he
Copyright Act does not expressly render anyone liable for infringement committed by another,’
Excel’s argument ignore[d] the circumstances of the copying.” Id. at 791 (quoting Sony Corp. of
Am. v. Universal City Studios, Inc., 464 U.S. 417, 434 (1984)). The court stated:
[A]lthough students press the start button and make a copy of the coursepack,
Excel is the source of the reproduction. Excel controls the entire copying process.
It retains the “master,” maintains its quality, gives it to a student to copy, and
accepts payment. Excel also owns and supplies all of the elements of
reproduction-the venue, the copy machines, the paper, and the utilities. Excel staff
members are available to assist students in the copying process and also provide
binding services if requested.
Id. The court held that Excel violated the plaintiff’s distribution right under Section 106(3) when
it gave the student the coursebook to copy and when the student paid for the coursebook. Id.
Google’s reliance on Cablevision is misplaced. There, an individual customer of
Cablevision’s services could choose to make a copy of a television program that he or she
already was entitled to receive, by “pushing a button” that would trigger the making of a copy of
that program on a portion of a hard drive dedicated to that particular customer. Until the
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 23 of 57
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 24 of 57
18
of providing the digital copies to its library partners, Google violates the rightsholders’
reproduction right under Section 106(1).
First, Google itself makes at least one digital copy of the entirety of each book as part of
its placement of the book on the GRIN system, whether or not that book has been requested by
its library partner. It then authorizes the libraries to access the GRIN system to “take” their own
digital copies. In so doing, Google causes the creation of additional digital copies of the books.
See Capitol Records, LLC v. ReDigi, Inc., No. 12 Civ. 95, 2013 WL 1286134, at *4-6 (S.D.N.Y.
Mar. 30, 2013) (reproduction right implicated by creation of digital copies; by definition, bits
cannot be “moved” from one server to another). And this is not some theoretical process. The
undisputed evidence establishes that Google has provided—and the libraries have accessed and
downloaded—millions of copies of books.
Accordingly, in addition to violating plaintiff’s distribution rights, Google’s provision of
digital copies to the libraries is also a blatant infringement of plaintiffs’ Reproduction right in
their copyrighted works. Whether such infringement is found under an “inducement” theory,
Metro-Goldwyn-Meyer Studios Inc v. Grokster, Ltd., 545 U.S. 913 (2005), or under a theory of
secondary liability does not matter. Google is not permitted to make copies of copyrighted
works or to authorize the libraries to make additional copies. See 17 U.S.C. § 106 (copyright
holder has exclusive right “to do and to authorize” the reproduction of their copyright protected
works) (emphasis added).
C. Google’s Infringing Conduct Cannot Be Justified As a “Fair Use” of
Plaintiffs’ Books
Google bears the burden of proof on its fair use defense. Am. Geophysical Union v.
Texaco, Inc., 60 F.3d 913, 918 (2d Cir. 1995) (“[T]he party claiming that its secondary use of the
original copyrighted work constitutes a fair use typically carries the burden of proof as to all
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 25 of 57
19
issues in the dispute.”). Thus, for Google to prevail, the record evidence must be construed in
the light most favorable to plaintiffs, the burden of proof shifted to Google, and Google must
then show that no reasonable jury could reject the fair use defense. See Anderson v. Liberty
Lobby, Inc., 477 U.S. 242, 248 (1986). Google does not even begin to meet this burden.
1. Copyright Law Reflects a Careful Balance Between the Rights of
Authors and Users
Throughout the history of American copyright law, Congress has sought to balance the
interests of copyright owners against those of users. In the 1976 Copyright Act, as well in
subsequent amendments such as the DMCA legislation, Congress carefully considered the
impact of new technologies on this balance, including that developing and anticipated
technologies would permit reproduction on a previously unprecedented scale and put
copyrighted works at risk of unauthorized reproduction and dissemination. Section 108 of the
Copyright Act, 17 U.S.C. § 108, is a good example of this balance. That provision gave libraries
a circumscribed right to make digital copies of a specific subset of books in their collections
under certain carefully delineated conditions.
Google’s Library Project upsets this fundamental balance. Just because Google has the
financial wherewithal and economic incentive to digitize, copy and disseminate much of the
world’s copyrighted works does not mean that it should be permitted to do so. Until Congress
addresses these critically important issues, courts must defer to the choices already made by
Congress. See Kirtsaeng v. John Wiley & Sons, Inc., 133 S. Ct. 1351, 1371 (2013) (“Whether
copyright owners should, or should not, have more than ordinary commercial power to divide
international markets is for Congress to decide. We do no more here than try to determine what
decision Congress has taken.”); WNET, Thirteen v. Aereo, Inc., 712 F.3d 676, 695 (2d Cir. 2013)
(noting that “unanticipated technological developments” may require legislative action); ReDigi,
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 26 of 57
20
Inc., 2013 WL 1286134, at *11 (“It is left to Congress, not this Court” to determine whether first
sale principles should apply to digital copies); Authors Guild v. Google Inc., 770 F. Supp. 2d
666, 677 (S.D.N.Y. 2011) (“[T]the establishment of a mechanism for exploiting unclaimed
books is a matter better suited for Congress than the Court.”).
2. The Four-Factor Test Is of Little Help in a Mass Digitization Use of
this Magnitude
A case such as this, which involves the digitization and use of millions of works, is vastly
different than the typical fair use case, which involves the use of one or a small handful of pre-
existing copyrighted works in the creation of a new work. As such, this case requires a different
type of analysis than a methodical application of the four factors set forth in Section 107. In
light of the Supreme Court’s ruling in Sony Corp of America v. Universal City Studios, Inc., 464
U.S. 417 (1984), then District Judge Pierre Leval observed that the first factor had developed
into a “two-track” pattern by which secondary users prevail by reason of either “(1)
transformative (or productive) nonsuperseding use of the original, or (2) noncommercial use,
generally for a socially beneficial or widely accepted purpose.” Am. Geophysical Union v.
Texaco Inc., 802 F. Supp. 1, 12 (S.D.N.Y. 1992), aff’d, 60 F.3d 913, 923 (2d Cir. 1995). Those
in the first category are the typical fair use cases triggering typical four-factor analysis. 4
MELVILLE B. NIMMER & DAVID NIMMER, NIMMER ON COPYRIGHT § 13.05[A][1][b] at 13-165
n.75 (“Virtually all fair use cases reflect the ‘productive use’ principle in their facts if not their
stated rationale.”). In the second type of case, the Court must determine whether a use that is
purely “noncommercial” and is “generally for a socially beneficial or widely accepted purpose.”
Texaco, 802 F. Supp. at 12. Google loses under either test. As explained in the following
sections, if the traditional four-factor analysis is conducted, Google cannot prevail because the
lion’s share of its uses of copyrighted works is not productive or transformative and because it
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 27 of 57
21
cannot prevail on any of the other traditional fair use factors. If, as plaintiffs submit, this case
really belongs in the second category where non-transformative uses are analyzed, then Google
loses because its activities are highly commercial and because Google’s use runs contrary to
Congress’s careful balance of the rights of the various stakeholders, including authors,
publishers, users, and libraries. A determination that this balance is outdated or needs
adjustment should not be made by the Courts, but rather must be made by Congress.
3. The Four-Factor Test
a. Google Cannot Satisfy Factor One
The first fair use factor is “the purpose and character of the use, including whether such
use is of a commercial nature or is for nonprofit educational purposes.” 17 U.S.C. § 107(1).
Google cannot satisfy this factor because its mass digitization of millions of copyrighted works
was indisputably for the purpose of gaining a competitive advantage for its own search engine in
order to reap advertising revenues and nothing about its Reproduction, Verbatim Display and
Library Distribution is remotely transformative.
i. Google’s Uses Are Plainly Commercial
A finding of commercial use, while not dispositive, “tends to weigh against a finding of
fair use.” Texaco, 60 F.3d at 921 (quoting Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569,
585 (1994)). Plaintiffs’ opening brief at 12 showed, by reference to Google’s documents, that
Google’s purpose was plainly commercial. Google’s brief does not address those documents or
rebut their contents. Rather, Google argues its purposes are not commercial because “a student
or professor (or indeed anyone who finds a Library Project book on Google Books) is engaging
in precisely the sort of use historically favored as noncommercial under the first fair use factor.”
(Google Br. at 24.)
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 28 of 57
22
Google’s argument was flatly rejected by the Second Circuit in Texaco, which held that
the “research” referred to in 17 U.S.C. § 107 does not refer to the activities of end users of the
secondary materials:
Moreover, the concept of a “transformative” use would be extended beyond
recognition if it was applied to Chickering’s [the Texaco scientist’s] copying
simply because he acted in the course of doing research. The purposes illustrated
by the categories listed in section 107 refer primarily to the work of authorship
alleged to be a fair use, not to the activity in which the alleged infringer is
engaged.
60 F.3d at 924. This means it is the work of the infringer, e.g., Google and Texaco, not end
users, e.g., the scientists in Texaco or Internet searchers here, that is the subject of the fair use
analysis.7 It is undisputed that Google, a public company with a fiduciary duty to maximize
profits, has undertaken the Library Project for a commercial purpose.
Texaco further held that “courts will not sustain a claimed defense of fair use when the
secondary use can fairly be characterized as a form of ‘commercial exploitation,’ i.e., when the
copier directly and exclusively acquires conspicuous financial rewards from its use of the
copyrighted material.” 60 F.3d at 922. Here, Google’s purpose was clearly commercial
exploitation, to increase the traffic to the Google website, and thereby increase its advertising
revenues. Google’s internal documents reveal this purpose: “We want web searchers interested
in book content to come to Google not Amazon;” “Everything else is secondary . . . but make
money.” (Pls.’ 56.1 Stmt. ¶ 76; see also id. ¶¶ 77-84; see generally Aiken Decl.)
7 Google argues that its distribution to the libraries constitutes fair use because it facilitates
transformative or noninfringing uses by the libraries. (Google Br. at 33-34.) But Google cannot
justify its conduct based upon the conduct of the libraries. The Second Circuit could not have
been clearer on this point when it rejected the argument made by a defendant broadcaster that its
users transformed the plaintiff’s broadcasts by using them for their own purposes, concluding
that it was the defendant’s “own retransmission of the broadcasts, not the acts of his end-users,
that [was] at issue.” Infinity Broad. Corp. v. Kirkwood, 150 F.3d 104, 108 (2d Cir 1998).
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 29 of 57
23
ii. The Challenged Uses Are Not Transformative
Under factor one and fair use generally, the question whether the challenged uses are
“transformative” is often the most important factor. (See Pls.’ Opening Br. at 21-34.) A
“transformative” use means that it “adds something new, with a further purpose or different
character, altering the first with new expression, meaning, or message.” Campbell, 510 U.S. at
579 (citing Pierre N. Leval, Toward a Fair Use Standard, 103 HARV. L. REV. 1105, 1111
(1990)).
Merely articulating a new “purpose” for a copyrighted work, without changing or adding
anything new, is not enough to render the work transformative. See Campbell, 510 U.S. at 579
(a transformative use “adds something new, with a further purpose or different character, altering
the first with new expression, meaning or message”). The phrase “with a different purpose or
different character” does not stand on its own, but rather modifies the ultimate requirement that
the secondary use “adds something new.” If a “different purpose” were enough, a secondary
user could always claim transformation by articulating some different purpose for making a
copy. In this regard, “for the ‘transformative’ moniker to guide, rather than follow, the fair use
analysis, it must amount to more than a conclusory label.” 4 MELVILLE B. NIMMER & DAVID
NIMMER, NIMMER ON COPYRIGHT § 13.05[A][1][b].
(a) Courts Have Rejected Fair Use Defenses Involving Less
Verbatim Copying, Display and Distribution
There is nothing “transformative” about Reproduction, Verbatim Display and Library
Distribution. All use the expressive content of the authors’ words exactly as the authors wrote
them, to convey exactly the same meaning and content that the authors intended. Google’s
mechanical conversion of printed books into digital form is itself not transformative because it
does not add any “new information, new aesthetics, [or] new insights and understandings,” to the
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 30 of 57
24
books. Leval, supra, at 1111. Rather, “photocopying merely transforms the material object
embodying the intangible article that is the copyrighted original work.” Texaco, 60 F.3d at 924.
Google’s computers add no criticism, commentary, or any other “transformative” element.
Many cases have rejected a fair use defense where the secondary use involved far less
verbatim copying, display and distribution than do the uses here. See, e.g., Harper & Row
Publishers, Inc. v. Nation Enters., 471 U.S. 539 (1985) (verbatim quote of 300-400 words from
book length manuscript not fair use); Salinger v. Colting, 607 F.3d 68, 83 (2d Cir. 2010) (novel
based on characters in Catcher in the Rye likely not fair use); On Davis v. The Gap, Inc., 246
F.3d 152, 174-76 (2d Cir. 2001) (display of copyrighted jewelry in Gap advertisement not fair
use); Nihon Keizai Shimbun, Inc. v. Comline Bus. Data, Inc., 166 F.3d 65, 72 (2d Cir. 1999)
(abridged English translations of copyrighted Japanese news stories not fair use); Infinity Broad.
Corp. v. Kirkwood, 150 F.3d 104, 108-09 (2d Cir. 1998) (retransmission by telephone of local
radio broadcasts not fair use where “retransmissions leave the character of the original
broadcasts unchanged” and “[t]here is neither new expression, new meaning nor new message”);
Castle Rock Entm’t, Inc. v. Carol Publ’g Group, Inc., 150 F.3d 132, 141-43 (2d Cir. 1998) (trivia
questions and small selection of verbatim quotes derived from Seinfeld television show not fair
use); Ringgold v. Black Entm’t Television, Inc., 126 F.3d 70, 78-79 (2d Cir. 1997) (short displays
in television program of copyrighted poster of quilt not fair use); Am. Geophysical Union v.
Texaco Inc., 60 F.3d 913, 923 (2d Cir. 1994) (copying and distribution of short scientific articles
from larger collective publication not fair use); Twin Peaks Prods. v. Publ’ns Int’l Ltd., 996 F.2d
1366, 1374-77 (2d Cir. 1993) (plot summaries and short verbatim quotes not fair use); UMG
Recordings, Inc. v. MP3.com, Inc., 92 F. Supp. 2d 349, 350-51 (S.D.N.Y. 2000) (website that
permitted downloads of CD’s not fair use); Basic Books, Inc. v. Kinko’s Graphics Corp., 758 F.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 31 of 57
25
Supp. 1522, 1530-31 (S.D.N.Y. 1991) (sale of course packets to college students with excerpts
from copyright works not fair use); Radji v. Khahbaz, 607 F. Supp. 1296, 1300-01 (D.D.C. 1985)
(publishing a Farsi translation of English book excerpts that constituted only 10% of the book
not fair use).8
Significantly, Google’s discussion of transformative use (Google Br. at 20-23) does not
cite or mention, let alone discuss or analyze, any of the above decisions, including the eight
controlling decisions by the Second Circuit cited above. It is instructive to compare Google’s
practices with Castle Rock, supra, because what Google does is far less transformative than what
occurred in that case. The infringing book, The Seinfeld Aptitude Test, consisted of five
chapters, a sample of which is reprinted at Boni Decl. Ex. 5, that posed questions about the
Seinfeld television show. Comparing the Seinfeld book with Google’s available snippets of Ball
Four shows the following. First, the Seinfeld book consists of 643 questions in total, which used
41 actual verbatim quotations—all of them a sentence or less—drawn from 84 separate
television episodes of Seinfeld. 150 F.3d at 135-36. In contrast, with Ball Four, Google
publishes hundreds of actual verbatim quotations—and all of them are multiple sentences, not
just a phrase—drawn from only one book. (See Boni Decl. Ex. 1; Zack Decl. Ex. 4.) This
means that Google has made far more concentrated use of its source than the Seinfeld book.
Second, the author of the Seinfeld book has used her imagination to select what information she
deemed relevant and to invent wrong answers to multiple choice questions. 150 F.3d at 136,
8 But see Authors Guild, Inc. v. HathiTrust, Inc., 902 F. Supp. 2d 445 (S.D.N.Y. 2012) (Baer, J.)
(in a case currently on appeal to the Second Circuit and fully briefed, court found that Libraries’
use of digitized works for text-searching, preservation and access to visually disabled persons
was non-commercial, transformative and not likely to cause harm to authors. The case did not
involve the display of “snippets” or other visual displays and the Libraries were not seeking to
obtain a competitive advantage or to obtain advertising revenue).
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 32 of 57
26
143. Google in contrast simply displays verbatim the words written by Mr. Bouton. The
Seinfeld book was held not to be “transformative.” Id. at 142-43. Indeed, the Second Circuit
affirmed summary judgment against the defendant’s assertion of fair use, holding it was not fair
use as a matter of law. Id. at 146.
A similar comparison can be made with Twin Peaks Productions. The Second Circuit
held that two portions of the book infringed the original television series: chapter three, which
summarized plots, and chapter seven, which consisted of 15 questions and answers and one page
of verbatim quotations. 996 F.2d at 1372-77. To put this in context, three pages from each
chapter in the book, and the entirety of chapter seven’s three pages are submitted as Boni Decl.
Ex. 6. There was only an identity of 89 lines of dialogue between the eight television episodes
and the entire book. Id. at 1372. Most of those lines occur on the final page of the book, titled
“The Wit and Wisdom of Agent Cooper.” (Boni Decl. Ex. 6, p. 128.) That page contains six
quotes taken from four different episodes. In other words, the page offers the same amount of
verbatim text that appears in only two snippet searches on Google Books. The Second Circuit
affirmed summary judgment for plaintiff, rejecting the fair use defense as a matter of law. 996
F.2d at 1377-78.
In its recent decision in Cariou v. Prince, 714 F.3d 694 (2d Cir. 2013), the Second Circuit
reaffirmed that in order to be transformative, the later use must add some new expression,
meaning or message. There the plaintiff photographer, Mr. Cariou, lived among and took
photographs of Rastafarians in Jamaica. He published a book of these photographs. The
“appropriation artist” Mr. Prince tore pages out of the book and used the photographs as a basis
for paintings and a series of collages. Id. at 695-703. The Court re-affirmed the meaning of
“transformative”: “[f]or a use to be fair, it ‘must be productive and must employ the quoted
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 33 of 57
27
matter in a different manner or for a different purpose from the original.’” Id. at 706 (emphasis
supplied) (quoting Leval, supra, at 1111). The Second Circuit held that many of Prince’s
paintings were “transformative” under this standard because:
These twenty-five of Prince’s artworks manifest an entirely different aesthetic
from Cariou’s photographs. Where Cariou’s serene and deliberately composed
portraits and landscape photographs depict the natural beauty of Rastafarians and
their surrounding environs, Prince’s crude and jarring works, on the other hand,
are hectic and provocative. Cariou’s black-and-white photographs were printed in
a 9½” by 12” book. Prince has created collages on canvas that incorporate color,
feature distorted human and other forms and settings, and measure between ten
and nearly a hundred times the size of the photographs. Prince’s composition,
presentation, scale, color palette, and media are fundamentally different and new
compared to the photographs, as is the expressive nature of Prince’s work.
Id. at 706 (emphasis supplied). However, five (5) of Prince’s paintings or collages could not
qualify as fair use as a matter of law because it was not clear they “present a new expression,
meaning or message.” Id. at 711 (quoting Campbell, 510 U.S. at 579). All of the works at issue
are displayed at http://www.ca2.uscourts.gov/11-1197apx.htm.
(b) Google’s Attempted Justification of Its Index Does Not
Render Its Other Uses Transformative
Google claims its uses are “transformative,” but its arguments do not withstand scrutiny.
(Google Br. at 20-23.) It seeks to justify its use as an “index.” (Id. at 21.) That does not speak to
whether its Verbatim Display of substantial portions of the authors’ words, its Reproduction as
currency to pay the libraries, and its distribution to libraries of complete copies are fair uses.
Google has virtually nothing to say about these uses. Google devotes only one paragraph
to explaining why Verbatim Display of snippets is supposedly transformative. (Google Br. at 23,
first full paragraph.) Google avoids mentioning that it allows the display of approximately 78%
of each author’s verbatim words. Google claims that it helps readers “decide whether a
particular book is of interest to them” and to “decide whether to pursue more information about it
(for example by clicking a nearby link to buy the book).” (Id.) Google claims this is a “different
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 34 of 57
28
purpose for the [book content],” and that Google is involved in “an entirely new use of the
content.” (Id.)
Google’s argument fails for many reasons. First, using a book’s contents to “decide
whether a particular book is of interest” to the reader is not a “different purpose for the book’s
content” nor an “entirely new use of the content.” Rather, it is the very same “purpose” and
“use” for which the book’s content was created in the first place. See Texaco, 60 F.3d at 923
(“[A]n untransformed copy is likely to be used simply for the same intrinsic purpose as the
original . . . .”).
Second, Google’s argument is that it is promoting sales of the same original book by
pointing people to an Amazon.com link where the book can be purchased.9 But promoting sales
of the original copyright book is, as a matter of law, not “transformative.” Indeed, it has nothing
to do with transformative use. Rather, it is the same precise purpose for which the author wrote
the words in the first place. Promoting sales of the original work is one of the incidents of
copyright ownership. Typically the author and publisher jointly determine the best manner to
promote the copyrighted work. It is not the province of a third party interloper to decide to
“promote” someone else’s book without the author’s or publisher’s authorization.10 In fact the
9 A similar argument was rejected by the Third Circuit in Video Pipeline, Inc. v. Buena Vista
Home Entm’t, Inc., 342 F.3d 191, 199 (3d Cir. 2003), which is discussed further in Pls.’ Opening
Br. at 32.
10 Google’s bogus “promotion” claim fails for another critical reason. Many of the books in issue
are out of print. (In print books are mainly in the Partner Program with the authorization of the
publishers.) Thus, even if the Library Project could be seen as a device for promoting books, the
authors would receive nothing as a result of any sales. Because the books are out of print, the
only sales taking place are in the secondary market, under the first sale doctrine, and the author
gets no payment for such sales. There is a potential market for out-of-print books (see Aiken
Decl. ¶ 29), but not one that is advanced by Google’s use of Verbatim Display.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 35 of 57
29
evidence suggests that Google’s Verbatim Display of snippets could have a deleterious effect on
book sales. (See Aiken Decl. ¶¶ 4-5, 32-34, and generally.) In any case, it is undisputed that
Google is unaware of snippet display resulting in the sale of a single book. (Zack Decl. Ex. 16,
Clancy Dep. at 163:10-18; Zack Decl. Ex. 18, Turvey Dep. at 96:11-20.)
Third, case law establishes that helping to publicize or promote sales of an original
copyrighted item does not constitute fair use. See Twin Peaks, 996 F.2d at 1377 (“Though
appellants may be correct in arguing that works like theirs provide helpful publicity and thereby
tend to confer economic benefit on the copyright holder, we nevertheless conclude that the Book
competes in markets in which [plaintiff] has a legitimate interest . . . .”); UMG Recordings, Inc.
v. MP3.com, Inc., 92 F. Supp. 2d 349, 352 (S.D.N.Y. 2000) (“Any allegedly positive impact of
defendant’s activities on plaintiffs’ prior market in no way frees defendant to usurp a further
market that directly derives from reproduction of the plaintiffs’ copyrighted works.”); see also
Leval, supra, 103 HARV. L. REV. at 1124 n.84 (noting that “[a]n unjustified taking that enhances
the market for the copyrighted work is easy to imagine,” but “if the taking is unjustified under
the first factor, it should be considered an infringement, regardless of the absence of market
impairment”).
In each of the cases cited on pp. 24-25 above, the Supreme Court or the Second Circuit
rejected the fair use defense under circumstances in which the secondary uses could quite
plausibly be viewed as “promoting sales” of the original. The Nation magazine could claim that
by publishing excerpts, it was promoting sales of the full manuscript of President Ford’s
memoirs, and helping readers determine “whether that book was of interest” to them. The
Seinfeld question and answer book could plausibly be viewed as promoting interest in the TV
series and therefore enhancing the market for re-runs. In On Davis v. Gap, where The Gap used
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 36 of 57
30
the plaintiff’s copyrighted jewelry in its publicity shoots for clothing, The Gap could argue it was
giving publicity to and promoting sales of the On Davis jewelry. In Ringgold, where the
defendant television program used the copyrighted poster as background in its television show,
the defendant could logically claim it was giving the poster (which was sold by a very small
museum in Atlanta) far more publicity than it ever would have otherwise received. It is for the
copyright owner, however, to make decisions about promotion, not the copyright infringer.
Google argues that its snippets are useful formats for extracting information. (Google Br.
at 23.) But its argument is contradicted by its own internal communications,
(Boni Decl. Ex. 4.) Moreover, the
Second Circuit rejected that argument in Texaco, which, as here, involved verbatim copying.
The defendant claimed that copying small articles from a much larger and bulkier volume
allowed its research scientists ready access to the particular articles they needed to do research.
The Second Circuit disagreed:
Texaco suggests that its conversion of the individual Catalysis articles through
photocopying into a form more easily used in a laboratory might constitute
transformative use. However, Texaco’s photocopying merely transforms the
material object embodying the intangible article that is the copyrighted original
work.
60 F.3d at 923. Similarly, Google’s display of snippets is a mere conversion of the material
object in which the verbatim words are conveyed. The substance is the same.
Google relies upon Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605, 608-
11 (2d Cir. 2006). (Google Br. at 21.) This involved a 480-page book about the Grateful Dead.
Id. at 607. It displayed 2,000 images in chronological order, combining those images with text
and graphic art to convey the history of the Grateful Dead. Id. At issue were seven (7) of those
images. Id. In those images, defendant had photographed concert posters and greatly reduced
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 37 of 57
31
them in size for the book format. Id. These images were interspersed with narrative text and
graphic art about those concerts. Id. The Court held this use was transformative because its
purpose—to illuminate history—was different from the originals. Id. at 608-11. The images
were used for “enhancing the biographical information in Illustrated Trip, a purpose separate and
distinct from the original artistic and promotional purpose for which the images were created.”
Id. at 610. Here, Google has copied, displayed and distributed language verbatim, without
creating any new work. Google did not, for example, publish an original biography about Jim
Bouton’s major league baseball career, and use as part of that biography the cover page and brief
excerpts from Ball Four to illustrate that career, which is the analogue to Bill Graham Archives.
Google also relies upon two decisions from the Ninth Circuit, Kelly v. Arriba Soft Corp.,
336 F.3d 811 (9th Cir. 2003), and Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146 (9th Cir.
2007). (Google Br. at 22-23.) Both cases involved defendant’s indexing of visual materials
already placed on the Internet by their copyright holders through use of “thumbnails.” In neither
case did the defendant digitally reproduce, distribute and display millions of offline printed
works and place them on the Internet without permission of copyright owners, as Google has
done here. Of course, the fatal distinction between these Ninth Circuit cases and the present case
is that the books Google digitized were not previously available “on the Internet.” As part of its
mass digitization program, Google took it upon itself to make this significant decision on behalf
of thousands of rightsholders, many of whom might not ever have wanted their books placed on
the Internet and subjected to potential piracy, hacking and other security breaches.
Further, the holdings in the Ninth Circuit cases do not follow Second Circuit law. Kelly
involved an Internet search engine that copied photographs posted on other websites and
converted them to small, low-resolution “thumbnail” images as part of its search results. 336
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 38 of 57
32
F.3d at 815-16. In that case, the Ninth Circuit concluded that such use was a fair use, finding
that the use of the thumbnails for indexing on the Internet was for a different purpose than the
plaintiff’s purpose of artistic expression. Id. at 818-20. Perfect 10 similarly involved the use of
thumbnail images as part of Internet search results, in particular images of nude models sold by
the plaintiff on the Internet. 508 F.3d at 1155-57. In Perfect 10, the Ninth Circuit found the use
of thumbnail images to be “substantially the same use” as in Kelly, and followed its conclusion
that it was a fair use. Id. at 1163-68. In both cases, the secondary user may have used the work
for a new purpose, but in neither case did the copier add “new expression, meaning or message.”
While that may rise to the level of transformative use in the Ninth Circuit, it does not in this
Circuit. See Cariou v. Prince, 714 F. 3d 694, 706 (2d Cir. 2013) (while a work need not
“comment on the original or its author to be considered transformative,” it still must “alter the
original with ‘new expression, meaning or message’”) (citing Campbell, 510 U.S. at 579).11
Additionally, none of those cases presented any risk of potential market harm.
Google’s attempts to mischaracterize these cases as justifying a very different project
involving very different facts must fail. Indeed, in a different portion of its decision, the Perfect
10 Court determined that Google was liable for in-line linking to full-sized infringing images,
508 F.3d 1170-73. This was using the same images on the Internet for a different purpose than
thumbnails. Clearly Perfect 10 is limited to thumbnails, and nothing broader.
11 A.V. ex rel. Vanderhye v. iParadigms, LLC, 562 F.3d 630 (4th Cir. 2009), which involved
copying and archiving of student term papers (which the students had submitted on the Internet),
and which Google relies for the proposition that “[t]he use of a copyrighted work need not alter
or augment the work to be transformative in nature” (Google Br. at 21 n.3), is easily
distinguishable and similarly inconsistent with Second Circuit law. As Cariou re-emphasized,
“[f]or a use to be fair, it must be productive and must employ the quoted matter in a different
manner or for a different purpose from the original.” 714 F.3d at 707 (emphasis added) (quoting
Leval, supra, 103 HARV. L. REV. at 1111).
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 39 of 57
33
For all the above reasons, Google’s practices at issue are not transformative, the most
important of the four fair-use factors.
iii. Google’s “Customary Practice” Argument Lacks Merit
A theme throughout Google’s brief is that its uses are blessed as supposed customary
trade usages, or are assented to by the victims. (See, e.g., Google Br. at 11-12, 17-18, 29.) In
analyzing the first factor in Texaco, the Second Circuit rejected a similar argument concerning
supposed “Reasonable and Customary Practice.” 60 F.3d at 924. Google argues that certain
books are in the Partner Program, where the rightsholder can agree with Google as to the amount
of display. (Google Br. at 11-12, 17.) Google stresses that “[o]ver 45,000 publishers have
included works within the Partner Program,” including prominent publishers such as
HarperCollins, Penguin, Simon & Schuster and Macmillan. (Id. at 12.) Google also advises that
Amazon has a “Search Inside the Book” display in which “[r]ightsholders give permission for
online browsing.” (Id. at 17.) Google then states that a “leading representative of authors,” the
William Morris Endeavor agency, concurs that Google Books is a “fair use.” (Id. at 18.) Google
argues that the Authors Guild has implicitly endorsed Google’s practices, because it endorses the
Partner Program and Amazon’s display. (Id. at 17, 29.) Finally, Google argues that a survey it
commissioned demonstrates that authors themselves support Google’s practices. (Id. at 18.)
First, the fact that certain rightsholders have programs in which Google or Amazon has
solicited and obtained rightsholders’ consent to display portions of a work has no bearing on the
present action, where Google did not solicit or obtain any such consent in connection with the
Library Project. Indeed, as explained above, the fact that rightsholders have agreed to such
displays under negotiated and circumscribed conditions cuts against any argument that Google
has the unilateral right to do this.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 40 of 57
34
Second, Google’s implication that publishers have consented en masse to Google’s
actions is false. In fact, a number of prominent publishers filed suit against Google for copyright
infringement, in a parallel action that challenged the same practices at issue in this lawsuit.
McGraw-Hill et al. v. Google, Inc., No. 05-CV-8881 (S.D.N.Y. Oct. 19, 2005). Several
prominent publishers (McGraw-Hill, Penguin, Simon & Schuster, John Wiley) who filed suit are
also in the Partner Program. If anything, this proves the opposite of Google’s point: these
prominent rightsholders consented with Google to agreed-upon displays in the Partner Program
(in which rightsholders give permission for display and are compensated through a share of
revenue from ads, see Pls.’ 56.1 Stmt. ¶¶ 17-21; Clancy Decl. ¶ 14), but challenged Google in a
lawsuit for the nonconsensual uses at issue here. This hardly proves any “trade practice”
favoring Google and disfavoring present plaintiffs. (As this Court is aware, the publishers’
lawsuit was settled in 2012 on undisclosed terms.)
Third, the Second Circuit in Texaco rejected a similar argument. Texaco argued that its
copying of verbatim articles from the research journal Catalysis was justified as a “reasonable
and customary practice.” 60 F.3d at 924. The Court held that this was legally irrelevant,
because licensing arrangements existed whereby Texaco could have paid for licenses rather than
copying without permission. See id. (“[W]hatever validity this argument might have had before
the advent of the photocopying licensing arrangements discussed below in our consideration of
the fourth use factor, the argument today is insubstantial.”). Here, too, Google’s argument is
unavailing, for reasons further explained in discussion of the fourth factor, discussed below.
To the extent Google claims the Authors’ Guild implicitly endorses Google’s actions, the
deposition testimony Google quotes pertains to promotional programs where rightsholders give
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 41 of 57
35
permission, not those done without permission. Google misleadingly attempts to use it to
endorse the latter (without permission). (See Aiken Decl. ¶ 3 and generally.)
As to the William Morris Agency, in the testimony Google cites (Gratz Decl. Ex. 6, p.
18) the witness is speaking indiscriminately about all of “Google Books,” i.e., both the Partner
Program (permission display) and the Library Project (non-permission display).12 The witness’
views therefore have little if any relevance to this case. Google also cites a memorandum by the
agency claiming Google’s practices are “fair use,” (Gratz Decl. Ex. 7), but this memo concerns
the proposed class action settlement (it is titled “Regarding the Google Book Settlement”).
Finally, Google’s reliance upon its “survey” of authors (Google Br. at 18) is irrelevant to
this motion, fatally flawed, and entitled to no weight for the reasons set forth in Plaintiffs’ Reply
Brief in Support of Class Certification, pp. 9-12 (ECF No. 1008).13
12 The amicus brief of the American Library Association et al. (ECF No. 1048), pp. 3-4, also
focuses upon a generalized discussion of something it terms “Google Book Search,” which
includes all books that Google displays on the Internet, including those displayed with
rightsholders’ permission in “full view” or “preview” under the Partner Program. This
muddying of the waters invalidates any useful contribution the ALA might be able to make on
the topic at hand.
13 The “informal survey” of certain largely anonymous librarians relied upon in the amicus brief
of the American Library Association et al. (ECF No. 1048), pp. 5-9 & Appendix, is similarly
unreliable. There is no effort to explain the methodology used in this “informal” survey, in
contravention of Shari Seidman Diamond, “Reference Guide on Survey Research,” Reference
Manual on Scientific Evidence (Second) 229 (2000). Surveys used in litigation must meet
rigorous standards for methodological reliability under Daubert v. Merrell Dow
Pharmaceuticals, Inc., 509 U.S. 579 (1993), and are frequently held inadmissible when they do
not. See, e.g., Universal City Studios, Inc. v. Nintendo Co., 746 F.2d 112, 118 (2d Cir. 1984)
(survey inadmissible that was not “fairly prepared”); accord Manual for Complex Litigation
(Fourth) § 11.493 (2004). This “informal survey” could not pass this test, because it describes
no methodology. Further, the cherry-picked verbatim responses statements from certain
librarians relied upon in the brief are not made under oath and are hearsay. Thus, they cannot be
considered in this summary judgment adjudication. See Fed. R. Civ. P. 56(c)(4); Ehrens v.
Lutheran Church, 385 F.3d 232, 235 (2d Cir. 2004) (Court may “rely only on admissible
evidence” in adjudicating summary judgment).
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 42 of 57
36
In sum, Google cannot satisfy the first factor of the fair use defense. Its use is plainly for
the commercial purpose of gaining a competitive advantage for its search engine in order to reap
substantial advertising revenues. It has engaged in the Reproduction, Verbatim Display, and
Library Distribution of plaintiffs’ works without transforming those works, and to the extent
there is anything transformative about its index or n-gram viewer, such uses do not justify its
massive Reproduction, Verbatim Display, and Library Distribution.
b. Factor Two Does Not Favor Google
The second statutory fair use factor is “the nature of the copyrighted work.” 17 U.S.C.
§ 107(2). This factor “calls for recognition that some works are closer to the core of intended
copyright protection than others, with the consequence that fair use is more difficult to establish
when the former works are copied.” Campbell, 510 U.S. at 586.
Google’s brief correctly states that because Google scans and displays non-fiction and
fiction identically, this factor has no bearing on the fair use analysis. Google agrees that the
works “are of every conceivable nature.” (Google Br. at 25.) In Google’s words:
Google Books encompasses almost every imaginable type of book . . . . It follows
that no conclusions may be drawn as to whether all the books within Google
Books are closer to the core or the periphery of the copyright law . . . .
(Google Br. at 25.) Plaintiffs agree with Google that the second factor is a non-factor. Under the
unique circumstances of this case, where Google indiscriminately scanned millions of books in
assembly-line fashion, Google cannot meet its burden of showing that the nature of the
copyrighted work favors a finding of fair use. If anything, this factor favors plaintiffs’ position
as shown in plaintiffs’ opening brief at pp. 36-39.
c. Google Cannot Satisfy Factor Three
The third statutory fair use factor is “the amount and substantiality of the portion used in
relation to the copyrighted work as a whole.” 17 U.S.C. § 107(3). This factor “recognizes that
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 43 of 57
37
the more of a copyrighted work that is taken, the less likely the use is to be fair, and that even a
less substantial taking may be unfair if it captures the essence of the copyrighted work.” Infinity
Broad., supra, 150 F.3d at 109. “Though not an absolute rule, ‘generally, it may not constitute a
fair use if the entire work is reproduced.’” Id. (quoting 4 MELVILLE B. NIMMER & DAVID
NIMMER, NIMMER ON COPYRIGHT, § 13.05[A][3] at 13-178 (1997)). There is no dispute about
this item and it is uniform: Google: (1) made numerous Reproductions of 100% of each work;
(2) makes available for Verbatim Display in snippet view approximately 78% of each work; and
(3) provides 100% of each work for Library Distribution.
Google claims this is “no more of a work than is necessary.” (Google Br. at 27.) This
begs the question: necessary to what? Google’s argument is limited to indexing. (“Google
scans the full text of works because full-text scanning is necessary for full-text search.” Id.)
Google does not explain why it is “necessary” to retain complete image and text files comprising
every page of every book, display substantial portions of each work in snippet view, or to
reproduce and distribute complete copies of all works to its library partners. It plainly is not.
See, e.g., Campbell, 510 U.S. at 589 (while defendants’ use of lyrical excerpts from Roy
Orbison’s song “Oh Pretty Woman” was justified to achieve a parodic purpose, case was
remanded to determine whether the repetition of the original’s iconic bass riff was excessive);
see also Kelly, 336 F.3d at 815 (once the search engine created a thumbnail version of the
indexed image, the full-sized originals were deleted).
Google’s two sentences pertaining to Verbatim Display state: “Google likewise displays
no more of a work than is necessary to allow users to assess whether a book is responsive to their
query. The amount of text actually displayed to users is only a very small portion of the
copyrighted work.” (Google Br. at 27-28.) The first sentence is circular (whatever a reader
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 44 of 57
38
wants to read from the book is necessary to display to that reader) and proves too much. Under
that logic Google could display 100% of the book’s verbatim contents without permission,
because that is the amount “necessary to allow users to assess whether a book is responsive to
their query.” The second sentence misleadingly focuses upon a single search, not the totality of
searches Google permits each user to make for each book in snippet display. (Google Br. at 28
n.5.)14 However, as the Supreme Court has made clear, even “[i]solated instances of minor
infringements, when multiplied many times, become in the aggregate a major inroad on
copyright that must be prevented.” Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S.
539, 569 (1985). One has to only examine Boni Decl. Exs. 1, 2, and 3 to see the force of
Harper’s statement.
Google relies upon Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605 (2d
Cir. 2006), for the proposition that “copying the entirety of a work may be fair use where that
copying is necessary to the purpose of the fair use.” (Google Br. at 27.) This seriously misstates
that case. As discussed on pp. 30-31, above, in that case a 480 page book did not infringe
copyrights by including, among 2,000 images, small photographs of seven (7) concert posters,
used to chronicle the history of the Grateful Dead. Id. at 608-11. The decision approved of
“copying the entirety of” that miniscule number of concert posters, because they were used in the
book for a very different purpose than the originals—not to promote a concert, but rather to help
depict the history of the band.
14 Even as to that, Google’s math is wrong. It claims that one search in a 100 page book displays
only “0.0075%” of the book’s text. (Google Br. at 28 n.5.) That is seven and one-half
thousandths of one percent. Rather, each 100 page book consists of 800 snippets (a snippet is
one-eighth of a page). Each search yields three snippets. Therefore each search yields 3/800 of
the book, or .375%.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 45 of 57
39
Google also relies upon Sony Corporation of America v. Universal City Studios, Inc., 464
U.S. 417, 449-50 (1984), for the same proposition. (Google Br. at 27-28). That case considered
whether Sony could sell video tape recorders that were used to record public-air broadcasts for
private, non-commercial viewing in consumers’ homes, for the purpose of time shifting. The
Court held:
Moreover, when one considers the nature of a televised copyrighted audiovisual
work, see 17 U.S.C. § 107(2), and that timeshifting merely enables a viewer to see
such a work which he had been invited to witness in its entirety free of charge, the
fact that the entire work is reproduced, see id. at § 107(3), does not have its
ordinary effect of militating against a finding of fair use.
Id. at 448 (emphasis supplied). In a footnote, the Court added that “the timeshifter no more
steals a program by watching it once than does the live viewer.” Id. at 448 n.33. In sum, that
case stands for the opposite proposition than Google advances: ordinarily large-scale copying
and verbatim display from a copyrighted work militates against fair use, but did not in the very
special circumstances of that case, where it enabled the consumer to watch the same program
once, either in the afternoon or the evening, at his or her choice.
Texaco rebuts Google’s reliance upon Sony. There, the defendant copied the entirety of
selected journal articles, which, however, were a small fraction of the total articles in each
volume of the journal Catalysis itself (from which the articles were taken). 60 F.3d at 925.
There, as here, the defendant cited Sony “for the proposition that the copying of entire
copyrighted works can still constitute fair use.” Id. The Second Circuit disagreed with this
reading of Sony, finding the case stood for the opposite proposition:
Despite Texaco’s claims that we consider its amount of copying “miniscule” in
relation to the entirety of Catalysis, we conclude, as did the District Court, that
Texaco has copied entire works. Though this conclusion does not preclude a
finding of fair use, it militates against such a finding, see Sony, 464 U.S. at 449-
50, and weighs the third factor in favor of the publishers.
Id. at 926.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 46 of 57
40
In arguing that it is “necessary” to copy the entirety of each book, Google in effect
invokes the doctrine of “intermediate use.” See Texaco, 60 F.3d at 921 (citing Sega Enters. Ltd.
v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992)). That does not assist Google. That doctrine
holds that “intermediate” copying may be justified only where it is the only means possible to
achieve a lawful end product. Sega Enters., 977 F.2d at 1518-28. In Sega, the Ninth Circuit
began by affirming its long-standing rule that copying at any stage (whether intermediate or
final) constitutes infringement. Id. at 1518 (“We have previously held that the Copyright Act
does not distinguish between unauthorized copies of a copyrighted work on the basis of what
stage of the alleged infringer’s work the unauthorized copies represent.”). The Ninth Circuit,
however, made a limited fair use exception in the unique circumstances of that case. Id. at 1520-
28. Plaintiff Sega manufactured video game consoles. Id. at 1514. It also manufactured video
games that ran on these consoles. The computer code for the video games was protected by
copyright. Id. There was no copyright on the console itself, meaning that competing video game
makers were free to make and sell their own video games to run on the Sega console. Defendant
Accolade was one such company. Its engineers, at great time and expense, reverse engineered
Sega’s computer code so that they could make their own independent code that would be
compatible with the Sega console. Id. at 1514-16. They “decompiled” Sega’s machine-readable
object code back to human-readable source code, taking care to make only “functional”
descriptions, and not to copy the actual code. This disassembly was done for the sole purpose of
independently designing Accolade’s own code. Id.
In those limited circumstances, the Ninth Circuit held that this “intermediate”
disassembly of Sega’s code was protected by fair use, because there was no other possible means
of learning how to make a non-infringing video game:
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 47 of 57
41
Accolade copied Sega’s software solely in order to discover the functional
requirements for compatibility with the Genesis console—aspects of Sega’s
programs that are not protected by copyright . . . . On these facts, we conclude
that Accolade copied Sega’s code for a legitimate, essentially non-exploitative
purpose, and that the commercial aspect of its use can best be described as of
minimal significance.
Id. at 1522-23.
This doctrine does not assist Google. First, Texaco rejected that an “intermediate use”
meant a “fair use.” It described Texaco’s scientist’s conduct as “more appropriately labeled an
‘intermediate use,’” 60 F.3d at 921, but rejected the fair use defense for this conduct.
Second, “intermediate use” is not relevant to the practices challenged in this lawsuit. An
example of true “intermediate use” might be Google’s digitizing books to use in the n-gram
viewer—n-grams will not function without the underlying source material of digitized books that
are fully searchable by computers.
However, “intermediate use” cannot justify Verbatim Display. Nor can it justify Library
Distribution. Rather, if those two uses are illegal (as they are), the fact that Google has scanned
(i.e., Reproduced) full works in order to display them to the public and distribute them does not
absolve Google. Quite the contrary, an “intermediate” use (scanning/Reproducing full texts) that
enhances illegal ends (Verbatim Display and Library Distribution) is just a further act of
illegality, as Sega held, 977 F.2d at 1518.
Most importantly, making Reproductions of the entirety of millions of books as currency
to pay the libraries for permission to scan their print books—all for Google’s commercial ends—
cannot possibly be considered an intermediate use. It is certainly not the only means possible, as
Google could have paid the libraries money to digitize the books.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 48 of 57
42
d. Google Cannot Satisfy Factor Four
The fourth statutory factor is the “effect of the use upon the potential market for or value
of the copyrighted work.” 17 U.S.C. § 107(4). Google claims that its uses have no adverse
effects. (“Plaintiffs have adduced no evidence that Google books has displaced the sale of even a
single book,” Google Br. at 18.) Google further implies that its uses have a positive effect on
plaintiffs’ book sales. (“Browsing Books Promotes Sales,” Google Br. at 17.)
Google knows that there is no evidentiary support for the latter proposition.
(See Boni Decl. Ex. 4.) In
addition, Google cannot identify the sale of a single book as a result of snippet display. (Zack
Decl. Ex. 16, Clancy Dep. at 163:10-18; Zack Decl. Ex. 18, Turvey Dep. at 96:11-20; see
generally Aiken Decl. (showing how snippets can harm book sales).)
In any event, as Judge Leval wrote in his seminal law review article: “The fact that the
secondary use does not harm the market for the original gives no assurance that the secondary
use is justified.” Leval, supra, 103 HARV. L. REV. at 1124. Indeed, as he points out, an
unjustified taking can enhance the original and still be an infringement:
An unjustified taking that enhances the market for the copyrighted work is easy to
imagine. If, for example, a film director takes an unknown copyrighted tune for
the score of a movie that becomes a hit, the composer may realize a windfall from
the aftermarket for his composition. Nonetheless, if the taking is unjustified
under the first factor [transformative use], it should be considered an
infringement, regardless of the absence of market impairment.
Id. at 1124 n.84; see also Castle Rock, 150 F.3d at 145 (rejecting defendants’ claim that the
fourth factor favors defendant because there is no proof of actual market harm to the original
work).
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 49 of 57
43
Google’s argument about the market addresses the incorrect legal standard. As Texaco
makes clear (a case Google cites on p. 29 of its opening brief but then ignores), the correct issue
is whether “unrestricted and widespread conduct of the sort engaged in by the defendant . . .
would result in a substantially adverse impact on the potential market for or value of the
plaintiff’s present work.” 60 F.3d at 927 n.12 (alternation and first emphasis by Second Circuit,
second emphasis supplied) (citations omitted). In other words, if everyone were free to engage
in Google’s conduct, would this have an adverse effect upon a potential market for plaintiffs’
works?
Plaintiffs have addressed this by submitting two expert reports, one by Dr. Daniel
Gervais (Zack Decl. Ex. 37) and one by Dr. Ben Edelman (Zack Decl. Ex. 39). In Dr. Gervais’
opinion:
I believe that, if Google’s uses are determined not to be fair uses, the market
would intervene and one or more CMO’s (with proper authorization from rights
holders) would license Google (and potentially others) to scan, distribute and
display copyrighted works. In fact, as discussed further below, the type of
copyrighted content that Copyright Clearance Center, Inc. presently licenses is
essentially printed content, much of the same nature as the material scanned by
Google. The rights involved are also essentially the same. In other words, this
type of licensing is already a reality.
(Zack Decl. Ex. 37, p. 3 ¶ 11 (emphasis supplied).) Dr. Gervais’ extensive opinion, and the
reasoning behind it, are described more fully in plaintiffs’ opening brief at pp. 42-43,
incorporated by reference.
Similarly, Dr. Edelman, coming at the issue from a different perspective, reaches a
similar conclusion:
If Google’s practices of digitally copying, distributing and displaying books
without rights-holder permission are found to be fair uses and become
widespread, the market for books will be adversely impacted by the potential for
security breaches. Conversely, requiring Google and others to obtain the
permission of rights-holders before engaging in such practices could prompt
negotiations between rights-holders and those who seek to digitally use their
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 50 of 57
44
works, thereby fostering standards for the allocation of the costs and risks of any
harm flowing from such security breaches.
(Zack Decl., Ex. 39, p. 8, ¶ 39.)
These two opinions plainly demonstrate that this factor favors plaintiffs. At the very
least, the expert opinions create a genuine issue of disputed fact, precluding Google from
obtaining summary judgment on the fourth factor. While Google’s brief vaguely argues that
“harms that are ‘theoretical and speculative’ are not considered in the fourth factor analysis”
(Google Br. at 30), Dr. Gervais and Dr. Edelman’s opinions are neither theoretical nor
speculative. To the contrary, they are extensively buttressed with concrete, real-world evidence
to support their conclusions.
Dr. Gervais’ opinion is squarely in line with evidence that Texaco held tilted the fourth
factor analysis away from fair use. In Texaco, at issue were discrete articles from the journal
Catalysis that the scientist copied and kept in his office to enhance and expedite his laboratory
research. Texaco argued there was no “market” for these individual articles, because they were
only copyrighted and sold as part of a larger volume of Catalysis that contained many additional
articles. See 60 F.3d at 928 (“Significantly, publishers have traditionally produced and marketed
individual articles only in a journal format, i.e., in periodical compilations of numerous
articles.”) Moreover, the Court noted that these scholarly articles were not written for monetary
compensation in the first place:
Quite significantly, though, in the unique world of academic and scientific
articles, the effect on the marketability of the composite work in which individual
articles appear is not obviously related to the effect on the market for or value of
the individual articles. Since (1) articles are submitted unsolicited to journals; (2)
publishers do not make any payment to authors for the right to publish their
articles or to acquire their copyrights; and (3) there is no evidence in the record
suggesting that publishers seek to reprint particular articles in new composite
work . . . .
60 F.3d at 928.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 51 of 57
45
Nonetheless, the Second Circuit endorsed and adopted the District Judge’s (Judge
Leval’s) conclusion that if Texaco’s practices were not deemed fair use, licensing for the
individual articles would develop through the Copyright Clearance Center, Inc.:
Specifically, the [District] Court pointed out that, if Texaco’s unauthorized
photocopying was not permitted as fair use, the publishers’ revenues would
increase significantly since Texaco would (1) obtain articles from document
delivery services (which pay royalties to publishers for the right to photocopy
articles), (2) negotiate photocopying licenses directly with individual publishers,
and/or (3) acquire some form of photocopying licenses from the Copyright
Clearance Center (“CCC”).
Id. at 929. The Court then elaborated:
Though the publishers still have not established a conventional market for the
direct sales and distribution of individual articles, they have created, primarily
through the CCC, a workable market for institutional users to obtain licenses for
the right to produce their own copies of individual articles via photocopying.
Id. at 930. The Court held this “likely to be developed” market to conclude that the fourth factor
weighed in favor of the publishers, and against Texaco’s fair use defense. Id. at 931.
Here, Dr. Gervais explains that licensing mechanisms could and would develop through
the same CCC cited in Texaco, and through similar collective rights organizations, absent a
finding of fair use. (Zack Decl. Ex. 37 at ¶¶ 10-42.)
But this Court need not rely solely upon the Gervais and Edelman opinions in deciding
that Google is not entitled to summary judgment on the “market” factor. Rather, perhaps the
most probative evidence defeating Google’s argument is Google’s own Partner Program.
Google repeatedly touts its Partner Program, in which more than 45,000 rightsholders have given
Google permission to display partial book content on the Internet in exchange for compensation
from ad revenues. (Google Br. at 11-12.) As Google explains:
Rightsholders may also request that Google display more text through the Partner
Program. The rightsholder can choose what percentage of the text of the book to
display—usually at least 20%. For example, Jim Bouton’s book Foul Ball is in
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 52 of 57
46
the Partner Program, and his publisher has chosen to allow some pages to be
viewed.
(Google Br. at 11 (citations omitted).) Google compensates the rightsholders for this limited
display through advertising revenue that Google receives. “Since 2004, the Partner Program has
allowed publishers and other rightsholders to permit Google to display their works in exchange
for a split of ad revenue.” (Pls.’ 56.1 Stmt. ¶ 17.)
Hence, there is in existence an actual and viable market for paying rightsholders for
partial electronic display on the Internet of their books—Google’s own Partner Program. This
is directly analogous to the CCC market described in Texaco, and eviscerates Google’s
argument.
4. Aggregate Assessment
Google’s Reproduction, Verbatim Display, and Library Distribution are not
transformative as a matter of law (factor one). Factor two is neutral or favors plaintiffs because
Google admits it makes no distinction between fiction and non-fiction and copies, displays and
distributes uniform amounts of all types of works. Factor three strongly favors plaintiffs,
because Google cannot justify Verbatim Display of approximately 78% of each work and
Library Distribution and Reproduction of 100%. Factor four strongly favors plaintiffs because of
the Gervais and Edelman opinions and Google’s own Partner Program. In short, Google cannot
prevail on summary judgment where not a single one of the factors favors fair use.
D. Google’s Library Distribution Is Not Fair Use
There is no colorable argument that providing to libraries complete verbatim copies of
millions of books is transformative. It is analytically no different than copying full texts of
books at a Xerox machine and distributing those copies to libraries.
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 53 of 57
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 54 of 57
48
(Zack Dec. Ex. 23.)
(Id.)
As to factor two, Google concedes the same analysis is applicable as for the rest of
Google Books. (Google Br. at 34.) That analysis leads to the conclusion that because Google
digitizes and distributes fiction and non-fiction in the same manner, it does not affect the fair use
analysis.
As to factor three, Google acknowledges that “library copies include the entirety of the
books,” but tries to justify distributing 100% of each work with the misleading argument that
“the use of the whole is necessary to the transformative purpose of making an index.” (Google
Br. at 34.) First, as stated, although Google tries to rely upon the libraries’ use of the copies to
make an index, it is Google’s secondary use that is considered in determining transformative use,
not the libraries or other downstream users. Infinity Broad. Corp. v. Kirkwood, 150 F.3d 104,
108 (2d Cir. 1998). As we demonstrate above (see pp. 6-9), distributing 100% of the books to
the libraries was done for the purpose of paying the libraries for the right to scan their print
books. Second, even as to the libraries own uses, it was clear they are not limited to indexing; in
fact, the contracts do not even mention indexing. The contracts do, however, provide for the
display of the full digitized works. (See pp. 7-9, above.)
Finally, as to the market factor, Google makes sweeping statements that “the undisputed
evidence shows that the availability of digital copies has not reduced libraries’ purchase of books
or their spending on licenses—indeed, that spending has increased. The libraries have not used
the scans in any way that substitutes for copies they could have purchased or licenses they could
have paid for.” (Google Br. at 35, emphasis in original.) But no record evidence is cited in
support of these statements. In contrast, in the factual portion of its brief, Google makes much
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 55 of 57
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 56 of 57
50
Sanford P. Dumain
MILBERG LLP
One Pennsylvania Plaza
New York, NY 10119
Tel: (212) 594-5300
Fax: (212) 868-1229
sdumain@milberg.com
Counsel for Plaintiffs
Case 1:05-cv-08136-DC Document 1070 Filed 08/26/13 Page 57 of 57