The Authors Guild, Inc. et al v. Hathitrust et al

Filing 112

MEMORANDUM OF LAW in Support re: 100 MOTION for Summary Judgment.. Document filed by Hathitrust. (Petersen, Joseph)

KILPATRICK TOWNSEND & STOCKTON LLP
Joseph Petersen (JP 9071)
Robert Potter (RP 5757)
1114 Avenue of the Americas
New York, New York 10036
Telephone: [phone]
Facsimile: [phone]
Email: [e-mail]
Joseph M. Beck (admitted pro hac vice)
W. Andrew Pequignot (admitted pro hac vice)
Allison Scott Roach (admitted pro hac vice)
1100 Peachtree Street, Suite 2800
Atlanta, Georgia 30309-4530
Telephone: [phone]
Facsimile: [phone]
Email: [e-mail]
Attorneys for Defendants
UNITED STATES DISTRICT COURT
SOUTHERN DISTRICT OF NEW YORK
THE AUTHORS GUILD, INC., ET AL.,
Plaintiffs,

Case No. 11 Civ. 6351 (HB)

v.
HATHITRUST, ET AL.,
Defendants.

MEMORANDUM IN SUPPORT OF
THE LIBRARIES’ MOTION FOR SUMMARY JUDGMENT
ON FAIR USE AND LACK OF INFRINGEMENT
UNDER SECTION 106 OF THE COPYRIGHT ACT

TABLE OF CONTENTS
A.

Digitization for Purposes of Preservation. ...............................................................3

B.

Google Vastly Expedited Digitization. ....................................................................4

C.

The Formation of HathiTrust for Purposes of Improving Scholarship. ...................4

D.

The Limited Uses of the Works within the HDL. ....................................................5

E.

The Benefits of the HDL’s Full-Text Search Functionality. ...................................6

F.

The Groundbreaking Benefits of the HDL for Individuals with Print
Disabilities. ..............................................................................................................8

I.

Legal Standard. ....................................................................................................................9

II.

The Libraries’ Uses of Plaintiffs’ Works Are Protected Fair Uses. ..................................10
A.

Fair Use and the Purpose of Copyright Law..........................................................10

B.

Section 107 and the Fair Use Factors. ...................................................................10
1.

The Purpose and Character of the Use Factor Favors Fair Use. ................11
a.
b.

Transformative Uses Are Particularly Favored. ............................11

c.

The HDL’s Search Functionality is Unquestionably
Transformative. ..............................................................................12

d.
2.

The Libraries’ Uses Are Favored Preamble Uses and
Presumptively Fair. ........................................................................11

Access to the Disabled and Preservation Also Are
Transformative Uses and Are Otherwise Fair Uses. ......................14

The Nature of the Copyrighted Works Factor Favors the Libraries. .........17
a.
b.

The Majority of Works in the HDL Are Factual Works. ...............19

c.
3.

The Majority of Works in the HDL Are Published and Out
of Print. ..........................................................................................17

In-Print and Fictional Works Have Limited Significance
Given the Transformative Nature of the Use. ................................20

The Libraries Copy No More of the Books Than Necessary.....................20

i

4.
C.
III.

The Libraries’ Uses Do Not Affect the Market For or Value of
Plaintiffs’ Books. .......................................................................................22

Balancing All the Factors.......................................................................................26

The Orphan Works Project. ...............................................................................................27

ii

TABLE OF AUTHORITIES
Cases
A.V. v. iParadigms, LLC,
562 F.3d 630 (4th Cir. 2009) .............................................................................................. 12, 14
Anderson v. Liberty Lobby, Inc.,
477 U.S. 242 (1986) .................................................................................................................... 9
Bill Graham Archives v. Dorling Kindersley Ltd.,
448 F.3d 605 (2d Cir. 2006) ........................................................... 12, 13, 15, 20, 22, 23, 24, 26
Blanch v. Koons,
467 F.3d 244 (2d Cir. 2006) ....................................................................... 10, 12, 17, 19, 21, 24
Cambridge Univ. Press v. Becker,
No. 1:08-CV-1425-ODE,
2012 WL 1835696 (N.D. Ga. May 11, 2012) ..................................................................... 16, 21
Campbell v. Acuff-Rose Music, Inc.,
510 U.S. 569 (1994) .......................................................................................... 10, 11, 14, 19, 20
Duffy v. Penguin Books USA Inc.,
4 F. Supp. 2d 268 (S.D.N.Y. 1998) .......................................................................................... 25
Feist Publ’ns, Inc. v. Rural Tel. Serv. Co.,
499 U.S. 340 (1991) .................................................................................................................. 10
Field v. Google, Inc.,
412 F. Supp. 2d 1106 (D. Nev. 2006) ...................................................................................... 15
Folsom v. Marsh,
9 F. Cas. 342 (C.C.D. Mass. 1841) (No. 4901) ........................................................................ 14
Hofheinz v. A & E Television Networks,
146 F. Supp. 2d 442 (S.D.N.Y. 2001) ...................................................................................... 23
Hofheinz v. Discovery Commc’ns, Inc.,
No. 00 Civ. 3802, 2001 WL 1111970 (S.D.N.Y. Sept. 20, 2001) ................................ 11, 23, 24
Kelly v. Arriba Soft Corp.,
336 F.3d 811 (9th Cir. 2003) ........................................................................................ 13, 20, 21
Lennon v. Premise Media Corp.,
556 F. Supp. 2d 310 (S.D.N.Y. 2008) ...................................................................................... 12
Maxtone-Graham v. Burtchaell,
803 F.2d 1253 (2d Cir. 1986) ................................................................................... 9, 17, 19, 20

iii

NXIVM Corp. v. Ross Inst.,
364 F.3d 471 (2d Cir. 2004) ............................................................................................... 11, 22
Perfect 10 v. Amazon.com, Inc.,
508 F.3d 1146 (9th Cir. 2007) ...................................................................................... 13, 21, 24
Rosemont Enters., Inc. v. Random House, Inc.,
366 F.2d 303 (2d Cir. 1966) ..................................................................................................... 19
Sega Enters. Ltd. v. Accolade, Inc.,
977 F.2d 1510 (9th Cir. 1993) .................................................................................................. 20
Sony Computer Entm’t, Inc. v. Connectix Corp.,
203 F.3d 596 (9th Cir. 2000) .................................................................................................... 20
Sony Corp. of America v. Universal City Studios, Inc.,
464 U.S. 417 (1984) .................................................................................... 15, 16, 20, 22, 24, 26
Sundeman v. Seajay Soc’y, Inc.,
142 F.3d 194 (4th Cir. 1998) .................................................................................................... 20
Swatch Grp. Mgmt. Servs. Ltd. v. Bloomberg LP,
No. 11 Civ. 1006, 2012 WL 1759944 (S.D.N.Y. May 17, 2012) ............................................. 15
The Authors Guild, Inc. v. Google Inc.,
No. 05-cv-8136 (S.D.N.Y. Oct. 28, 2008) ................................................................................ 17
Twentieth Century Music Corp. v. Aiken,
422 U.S. 151 (1975) .................................................................................................................. 10
Statutes
17 U.S.C. § 107(1) .................................................................................................................. 11, 26
17 U.S.C. § 108(e) .................................................................................................................. 27, 28
Rules
Fed. R. Civ. P. 56(c) ....................................................................................................................... 9
Other Authorities
4 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.05[B][5] (2011) .............. 10
H.R. Rep. No. 94-1476 (1976) ...................................................................................................... 16
Matthew Sag, Copyright and Copy-Reliant Technology,
103 Nw. U. L. Rev. 1607 (2009) .............................................................................................. 21
Pierre N. Leval, Toward a Fair Use Standard,
103 Harv. L. Rev. 1105, 1110 (1990) ....................................................................................... 10

iv

Rebecca J. Rosen, The Missing 20th Century:
How Copyright Protection Makes Books Vanish, The Atlantic, Mar. 30, 2012 ....................... 18
Staff of S. Comm. on the Judiciary (Barbara Ringer),
86th Cong., Renewal of Copyright 31 (Comm. Print 1960) ....................................................... 5
U.S. Const. Art. I § 8, cl. 8.............................................................................................................. 2
Wendy J. Gordon, Fair Use As Market Failure, A Structural and Economic Analysis of the
Betamax Case and Its Predecessors,
82 Colum. L. Rev. 1600 (1982) ................................................................................................ 25

v

INTRODUCTION
For over three centuries, academic libraries have purchased and preserved books—they
purchase millions of books each year—and have recorded and displayed basic information about
their holdings to help scholars identify relevant works.
About 20 years ago, in support of their efforts to preserve and catalog their books,
academic libraries—including Defendants in this action (the “Libraries”)—began making digital
copies of portions of their collections, including those in copyright. These digital copies serve
profoundly important functions for libraries and, more importantly, their patrons:
•

Digitization preserves books. All printed books are imperiled; they literally
disintegrate from use and exposure to oxygen, a process that is greatly hastened
for books printed on paper with high acid content (which constitutes the majority
of the holdings of academic libraries). Books also are at risk of loss from theft,
fire, flood, and other causes.

•

Digitization enables search—without providing direct access to in-copyright
material. Billions of pages of text can be searched in milliseconds, allowing
librarians, researchers, scholars, and students to identify relevant books amidst
vast collections. Previously, search was limited to a work’s bibliographic
information—for example, a patron could find the title, author, and general
subject matter but little or nothing about the actual contents of the work.

•

Digitization enables students and scholars with print disabilities to read or
listen to books. Through digitization, an authorized patron with a print disability
can have immediate access to a work in a format that can be made accessible
through a variety of technologies, including software that translates the text into
spoken words.

The Libraries’ early digitization efforts were inadequate. Because of the size of their
collections, the Libraries could not digitize their collections fast enough to preserve them.
Eight years ago, that began to change. First, with the assistance of Google, digitization
that would have taken the Libraries literally centuries to complete could be accomplished in less
than a decade. Second, with the establishment of the HathiTrust service, more than sixty
university and research libraries could store, secure, and search their digital collections in a

1

shared space that ensured the highest degree of security while reducing cost and increasing
scholarship. This collaboration permits library patrons to find works that they would otherwise
not find and preserves a significant portion of the world’s cultural record. It also opens, as never
before, the doors of scholarship to individuals with print disabilities.
Not one line of actual text treated as in copyright is ever displayed to patrons
searching through the HathiTrust service (except those who have certified print disabilities).
For all works treated as in copyright, scholars must check the books out of the library (as library
patrons have done for centuries) or purchase a new or used copy if they want to read a book
identified through the HathiTrust.
The Libraries’ uses for preservation, search, and access to the blind, which are no
different in kind than the functions long performed by libraries, serve the Constitutional
requirement that copyright law “promote the Progress of Science.” U.S. Const. Art. I § 8, cl. 8.
These uses do not harm authors; they increase discovery of their works and stimulate new
purchases. They are fair uses and lawful reproductions for people with disabilities. These uses
are rightfully made without the authorization of the Plaintiffs because they have the authorization
of Congress.
The Plaintiffs’ belated demand for an extraordinary court order—nearly seven years
after the collaboration with Google began and decades after the Libraries began digitizing
works—impounding and freezing the HathiTrust corpus presents the Court with a stark choice.
The Libraries respectfully submit that the Court should hold that the limited uses at issue in this
proceeding fall squarely within the Libraries’ right of fair use and right to reproduce works for
people with disabilities.

2

FACTUAL BACKGROUND
A.

Digitization for Purposes of Preservation.

Until recently, the Libraries and other large academic and research institutions were
annually losing hundreds of thousands of “out-of-print” books, especially those that existed on
paper with high acid content.1 (Wilkin Decl. ¶¶ 25-29.) As of 2004, Michigan estimated that
approximately 3.5 million of its books were at the most immediate risk of deterioration and,
ultimately, loss.2 (Id. ¶ 25.) Nor was disintegration the only threat to books in academic libraries.
(Id. ¶¶ 30-32.) Millions of volumes in libraries were destroyed during the World Wars, and the
collection of the National Library in Sarajevo lost over one million volumes due to shelling in
the 1990s.3 (Id. ¶ 32.) Hurricane Katrina devastated Tulane University’s Howard-Tilton
Memorial Library, with flooding destroying 90% of the 500,000 volumes in one of the library’s
collections. (Wilkin Decl. ¶ 31.) Earlier this month, the University of Wisconsin also suffered
significant losses to its collection due to severe flooding. (Hensrud Decl. ¶¶ 6-20.) Images of the
destruction appear below. Many of these works may be lost forever. (Id. ¶ 19.)

1

As John Wilkin, Associate University Librarian at the University of Michigan and Executive Director of
HathiTrust, explains, paper with high acid content is particularly prone to disintegration and loss as a result of a
process called “acid hydrolysis reaction.” (Id. ¶ 23.) Through this process, paper fibers are repeatedly broken down
to the point where the book ultimately disintegrates. (Id.)
2

The University of Michigan’s experience is typical of the other Libraries and is set out in detail in the
accompanying declaration of John Wilkin.
3

Perhaps the most famous example of a loss of a library is the destruction of the Library of Alexandria.

3

B.

Google Vastly Expedited Digitization.

Prior to Google’s involvement, it would have taken Michigan over 1,000 years to digitize
its collection. (Wilkin Decl. ¶ 44.) Google digitized most of the collection in less than a decade,
(id. ¶ 45), providing both timely preservation and access to books for the blind and print
disabled. (Id. ¶ 47.) “[W]e bargained for this right because it was important to us that we had the
right to control our own uses and satisfy one of our primary missions of providing specialized
services to the blind or other persons with disabilities.” (Id.)
C.

The Formation of HathiTrust for Purposes of Improving Scholarship.

In 2008, Michigan formed HathiTrust, named for the Hindi word for elephant, “hathi,”
evoking the qualities of memory, wisdom, and strength symbolized by elephants. (Id. ¶ 53.)
Participating members of the HathiTrust project pooled their digitized collections (whether
initially digitized by the respective libraries themselves or Google) into a single collection that
came to be known as the HathiTrust Digital Library (“HDL”). (Id. ¶ 54-55.) There are currently
more than sixty institutions participating in HathiTrust, including Michigan, and membership is
open to institutions worldwide. (Id. ¶ 56.)
The concept underlying the formation of HathiTrust is simple: it did not make sense for
each research library to maintain its own digitized collection of works. (Id. ¶ 54-55.) By pooling
resources, HathiTrust members could better serve their common goals of providing secure, longterm storage for the works, comprehensive research and discovery tools, and access to works in
the public domain and to students and faculty with print disabilities. (Id.)
The HDL now totals more than ten million works published over many centuries, in a
multitude of languages, covering almost every subject imaginable. (Id. ¶¶ 57-61.)

4

D.

The Limited Uses of the Works within the HDL.

The Libraries only make the following uses of works presumed to be in copyright:4
•

Full-Text Search. The Libraries’ patrons may search for one or more terms or
phrases across all works within the HDL. For those works that are not in the
public domain or for which the copyright holder has not expressly authorized use,
the search results indicate only the page numbers on which a term is found within
a particular book and the number of times it appears on each page. Search results
do not show sentences, “snippets,” or other selections of text, and patrons do
not have electronic access to any copyrighted content within such works
(unless they are users with certified print disabilities). In other words, there is no
copyrighted text displayed on the computer screen or available for print.

•

Preservation. As noted, the HDL is a safeguard against the ongoing loss of print
books and enables the Libraries to make copies for, among other Section 108
purposes, replacing a work that is damaged, deteriorated, lost, or stolen, and a
replacement copy cannot be obtained at a fair price.

•

Access for persons with print disabilities. As authorized entities under Section
121 and consistent with Section 107, the Libraries have, for many years,
converted works into alternative formats for the blind and other persons with
disabilities that prevent them from accessing printed materials.5

(Id. ¶ 68.)
These uses do not harm markets—whether actual or potential—for in-copyright material.
“[I]f the HDL has any impact whatsoever on the University Library’s acquisition of in-copyright
material, it has a positive effect on our purchasing.” (Id. ¶ 70.) This is because “scholars,
students, and other patrons are more likely to discover and use works that they can locate

4

As noted infra, there is considerable difficulty in determining the copyright status of works published between
1923 and 1963. For purposes of the HDL and this motion, the Libraries conservatively assume that many books
published during this time period may still be protected by copyright (even though a Copyright Office report
indicated that only 7% are still in copyright). See Staff of S. Comm. on the Judiciary (Barbara Ringer), 86th Cong.,
Renewal of Copyright 31, at 220 (Comm. Print 1960).
5

Although Michigan previously developed a project called the “Orphan Works Project” (“OWP”), not a single
patron has been given access to a single work through the OWP and it is at best unclear whether or how the OWP
will continue. Michigan took great pains in formulating the OWP to protect the legitimate rights of authors and
submits that had the project proceeded, would have been lawful under—among other provisions—Sections 107,
108, and 121 of the Copyright Act. Given the suspended status of the OWP, the Libraries believe that its legality is
not ripe for judicial determination, and the Libraries’ motion on that ground remains pending before the Court. If the
Court nevertheless decides to address the OWP, Section III discusses why it (1) does not infringe a single right
under Section 106 and (2) is also a fair use.

5

through digital search” and “[s]uch increased demand for works invariably translates into
increased purchases.” (Id. ¶ 71; see also Leary Decl. ¶ 15 (“During the course of my research on
Cook, I purchased at least two books . . . that I found via HathiTrust or Google search-only
searches. Of course, had I not discovered these works through my research, I would never have
purchased them.”); Waldfogel Decl. ¶ 50 (“I conclude that a service offering searchability inside
books, like the HathiTrust does, would very likely benefit rights holders by stimulating demand
for their works.”).)
E.

The Benefits of the HDL’s Full-Text Search Functionality.

Full-text searching easily constitutes the most significant advance in library search
technology in the last five decades. (Wilkin Decl. ¶ 75.) Rather than combing through paper,
card, or electronic catalog records and attempting to discern from author, title, and subject
heading whether the work itself may be of interest, scholars can go to the HDL website and
search—but not read or copy—the actual text of copyrighted books and journals. (Id. ¶ 76.) As
Mr. Wilkin describes, the search results display bibliographic information—including title,
author, publisher, and publication date—for in-copyright books containing the search term. (Id. ¶
77.) The search results also display the page numbers on which the term is found within the book
and the number of times it appears on each page, thereby, providing some clues as to how useful
the book might be. (Id.) The text itself, however, is not made digitally available (readable or
downloadable) unless it is determined that it is in the public domain or unless the rights holder
has given permission to provide access to the work’s content. (Id. ¶ 81.)
The HathiTrust search functionality represents a dramatic step forward in the evolution of
academic search. Library search once looked like this (on the following page):

6

It can now be performed anywhere with an Internet connection yielding, in milliseconds, results
like this:

7

Absent search, the information within these resources is invisible, or nearly so. For
example, Dr. Stanley Katz, the renowned American historian, was able to discover books
supporting his research on international human rights issues that he did not find without full-text
search. (Katz Decl. ¶¶ 14-17.) Author Margaret Leary used HathiTrust search to discover
otherwise unknown books for her own book on the Michigan law graduate and donor, William
W. Cook. (Leary Decl. ¶¶ 10-13.)
The HDL also empowers scholars to perform types of inquiries that previously were all
but impossible. (Smalheiser Decl. ¶¶ 25-29.) For example, a digital research method called “text
mining” is already proving itself a powerful and important tool for scholarly research. (Id. ¶¶ 36.) The goal of text mining is to find patterns and connections from large databases of textual
material. (Id. ¶¶ 3-4.) To provide a few examples, Dr. Swanson (mentor for Dr. Smalheiser) used
text mining to propose ways to ameliorate Raynaud’s syndrome and migraine headaches. (Id. ¶
13.) Dr. Smalheiser himself used this procedure to identify a class of molecule involved in
Huntington’s Disease. (Id. ¶ 17; see also Petersen Decl. ¶ 24, Ex. W (Peter Leonard & Timothy
R. Tangherlini, Trawling in the Sea of the Great Unread at 2-3 (“Text-mining techniques . . .
contribute significantly to a scholar’s ability to understand the relationships between the works in
this collection, thereby holding out the promise that one can develop a more encompassing
understanding of a particular field.”).)
F.

The Groundbreaking Benefits of the HDL for Individuals with Print
Disabilities.

For centuries, people with a broad range of disabilities have been denied equal access to
library collections and, therefore, to the full promise of an education. (Wilkin ¶ 101.) This is
particularly true in a university, where the written record is at the heart of the scholarly pursuit.
(Id.) Given the number of works a student must review to write a typical term paper, she may

8

have to wait weeks or even months for all requested works to be converted into Braille or audio
recording formats—works that she has not yet even been able to determine will be of any use.
(Id. ¶ 102.) Even digitizing the works on a book-by-book basis can take many weeks, which
makes pursuing postsecondary education that much more difficult for a student with a print
disability.6 (Id.)
The public benefits from the Libraries’ limited uses of book scans are indisputable.
Unfindable print books can be found—yet only read in the print copy. Men and women who
cannot see or turn physical pages are afforded equal access to the cultural record. Thousands of
books that would have been lost each year are preserved. Older, out-of-print, rarely checked-out,
and public-domain books, formerly banished to offsite storage facilities, are again discovered,
purchased, and used. Authors—as well as the public—benefit.
ARGUMENT
I.

Legal Standard.
Summary judgment is appropriate where “there is no genuine dispute as to any material

fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). “Only
disputes over facts that might affect the outcome of the suit under the governing law will
properly preclude the entry of summary judgment,” Anderson v. Liberty Lobby, Inc., 477 U.S.
242, 248 (1986), and these principles “apply with equal vigor to the fair use determination,”
Maxtone-Graham v. Burtchaell, 803 F.2d 1253, 1257-58 (2d Cir. 1986).

6

In order to use the HDL’s accommodations, a person with a print disability must obtain certification from a
qualified expert who in turn informs the Michigan library the individual has a certified print disability. (Id. ¶ 105.)
Michigan explains the digital library to the patron, describes appropriate uses of the service (including warnings
about copyright infringement), and enables the patron to get secure digital access to the library. (Id.) With digital
access, a print-disabled patron can perceive the works within the HDL using adaptive technologies such as software
that translates the text into spoken words. (Id.)

9

II.

The Libraries’ Uses of Plaintiffs’ Works Are Protected Fair Uses.
A.

Fair Use and the Purpose of Copyright Law.

“The primary objective of copyright is not to reward the labor of authors, but ‘[t]o
promote the Progress of Science . . . .’” Feist Publ’ns, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340,
349 (1991) (alteration in original) (quoting U.S. Const. Art. I § 8, cl. 8). “[P]rivate motivation
must ultimately serve the cause of promoting broad public availability of literature, music, and
the other arts.” Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 156 (1975) (emphasis
added and footnote omitted).
Fair use furthers this central purpose, Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569,
575 (1994), and “is necessary to achieve the objectives of that law,” Blanch v. Koons, 467 F.3d
244, 250 (2d Cir. 2006) (internal quotation and citation omitted). For this reason, fair use “should
not be considered a bizarre, occasionally tolerated departure from the grand conception of the
copyright monopoly,” Pierre N. Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. 1105,
1110 (1990).
B.

Section 107 and the Fair Use Factors.

Congress flatly states in Section 107 that copying for the purposes of “scholarship,”
“teaching,” or “research” are examples of fair use and then directs courts to evaluate four
“factors”: (1) the purpose and character of the use; (2) the nature of the copyrighted work; (3) the
amount and substantiality of the portion used; and (4) the effect of the use upon the potential
market for or value of the copyrighted work. Blanch, 467 F.3d at 251. Courts should consider the
factors consistently with the public interest. 4 Melville B. Nimmer & David Nimmer, Nimmer on
Copyright § 13.05[B][5] (2011).

10

1.

The Purpose and Character of the Use Factor Favors Fair Use.
a.

The Libraries’ Uses Are Favored Preamble Uses and
Presumptively Fair.

The Libraries’ uses—for “teaching,” “scholarship,” and “research”—perfectly match the
codified examples of what Congress intended to be fair uses. As such, these specifically
sanctioned uses necessarily inform the meaning of “the purpose and character of the use” and all
but end the factor one analysis. Indeed, as this Court (and the Second Circuit) has held, factor
one strongly favors fair use where the use falls within the categories listed in Section 107.
Hofheinz v. Discovery Commc’ns, Inc., No. 00 Civ. 3802, 2001 WL 1111970, at *3 (S.D.N.Y.
Sept. 20, 2001) (Baer, J.); NXIVM Corp. v. Ross Inst., 364 F.3d 471, 477 (2d Cir. 2004). This key
factor, therefore, heavily favors the Libraries on its face, a conclusion buttressed by the fact that
the Libraries’ uses are for “nonprofit educational purposes,” the antithesis of the types of
commercial uses that can weigh against fair use under factor one. See 17 U.S.C. § 107(1)
(contrasting commercial uses with nonprofit educational uses).
b.

Transformative Uses Are Particularly Favored.

Fair use is more than the hallmark examples Congress named in the statute. Scholarship,
teaching, and research are “illustrative and not limitative.” Campbell, 510 U.S. at 577. A further
central inquiry under factor one is whether the use “adds something new, with a further purpose
or different character, altering the first with new expression, meaning, or message.” Id. at 579
(citing Leval, supra, at 1111). Such uses have been termed “transformative” (or “productive”).
“Transformative” does not mean that the original work must be physically altered; rather,
the issue is whether the new use is for an “entirely different purpose and meaning” irrespective
of whether changes are made to the original work. For example, in Bill Graham Archives, the
Second Circuit held that the defendants’ unauthorized use of entire (although reduced size)

11

copies of copyrighted concert posters to illustrate points in a book was “transformatively
different from the original expressive purpose” of the posters. Bill Graham Archives v. Dorling
Kindersley Ltd., 448 F.3d 605, 609 (2d Cir. 2006). This was because defendants’ use of the
poster image “to document and represent the actual occurrence” of the concerts was different
from the “dual purposes of artistic expression and promotion” of the original use. Id.; see also
Blanch, 467 F.3d at 253 (including unedited photograph in collage was for transformative
purpose of commenting on the social and aesthetic consequences of mass media); Lennon v.
Premise Media Corp., 556 F. Supp. 2d 310, 324 (S.D.N.Y. 2008) (“To be transformative, it is
not necessary that defendants alter the music or lyrics of the song. . . . Defendants’ [unaltered]
use is nonetheless transformative because they put the song to a different purpose . . . .”).7
c.

The HDL’s Search Functionality is Unquestionably
Transformative.

Using a computer to search a book for a word is not the same as reading a book. Instead
of making books readable for their expressive, aesthetic, or instructional value, the HDL finds
the pages where a search term appears. HathiTrust has never made a single line of any books
treated as in-copyright available to be read (other than for the purpose of providing access to
the blind).8 (Wilkin Decl. ¶ 16.) Copying works in order to provide discovery tools that identify
relevant works—as distinct from making the text of the works themselves accessible—
unquestionably serves a different function than that served by the original works.

7

See also the Fourth Circuit’s decision in A.V. v. iParadigms, LLC, 562 F.3d 630 (4th Cir. 2009), where there was
extensive unauthorized copying and archiving of student papers by a plagiarism detection service that compared a
student’s writing with previously-submitted papers. Id. at 634. The Fourth Circuit summarily dismissed the
plaintiffs’ argument that the use was not transformative, finding that a use “can be transformative in function or
purpose without altering or actually adding to the original work.” Id. at 639.

8

Although the HDL search results do not display a single line of text for works treated as in-copyright, the Libraries
are confident that the “snippets” provided by Google are a fair use as well.

12

As explained by Dr. Katz, “[s]ophisticated digital searching over the text of large
numbers of books has permitted me to search for specific words or names (or words or names in
relation to each other) that simply was not possible using a library card catalog, the electronic
OPAC, or other bibliographic resources.” (Katz Decl. ¶ 9.) Margaret Leary also notes: “In seven
years of research, I did not find [the] information anywhere else.” (Leary Decl. ¶ 13
(emphasis added).) In each case, the researcher located a physical copy of the book and read the
relevant portions there, not online.
The Ninth Circuit’s decision in Kelly v. Arriba Soft Corp., 336 F.3d 811, 819 (9th Cir.
2003) (which the Second Circuit panel in Bill Graham Archives cited favorably), confirms that
copying for purposes of search is a transformative use. Arriba Soft copied the plaintiff’s
photographs—and millions of other copyrighted photographs—from websites for the purpose of
facilitating search, producing exact but smaller, lower-resolution versions of the originals
(“thumbnail” images), and made the thumbnail images available as search engine results on the
defendant’s website. Arriba Soft, 336 F.3d at 815-16.
Although the defendant in Arriba Soft, like the defendant in Bill Graham Archives, made
“exact replications” of the plaintiff’s entire works, the court held that the defendant’s use was
transformative because the defendant’s images “served an entirely different function than [the
plaintiff’s] original images.” Id. at 818. In words that resonate here, the defendants’ copying was
for the purpose of “improving access to information on the internet versus artistic expression.”
Id. Here, library patrons do not get to see even a low-resolution or reduced-size copy of books;
they get to see unprotected titles of books and the page numbers where their search terms appear.
The Ninth Circuit in Perfect 10 v. Amazon.com, Inc., 508 F.3d 1146 (9th Cir. 2007),
came to the same conclusion about different (transformative) uses of exact copies when it judged

13

Google’s indiscriminate copying of Internet content to make it findable: “Although an image
may have been created originally to serve an entertainment, aesthetic, or informative function, a
search engine transforms the image into a pointer directing a user to a source of
information.” Id. at 1165 (emphasis added); see also A.V. v. iParadigms, LLC, 562 F.3d 630,
640 (4th Cir. 2009) (finding defendant’s copying and archiving of student papers “was
completely unrelated to expressive content and was instead aimed at detecting and discouraging
plagiarism”).
Like the copying in those cases—each of which was by a commercial entity—the
Libraries’ copying to enable search (but not reading) serves “an entirely different function” than
the expressive purpose of the books.9 The Libraries’ uses are unquestionably transformative and
the first fair use factor overwhelmingly supports fair use.
d.

Access to the Disabled and Preservation Also Are
Transformative Uses and Are Otherwise Fair Uses.

Making books available to the blind and disabled is transformative because the use serves
a different, underserved audience otherwise unable to access the overwhelming majority of
works in their print format. It hardly “‘supersede[s] the objects’ of the original creation,” in
Justice Story’s words, Campbell, 510 U.S. at 579 (citing Folsom v. Marsh, 9 F. Cas. 342, 348
(C.C.D. Mass. 1841) (No. 4901)), when authors themselves do not seem to care that the blind
cannot access their works, (Petersen Decl. ¶ 22, Ex. U (Cummings Dep. 56:23-57:3, May 22,
2102: Q: “So, you do not believe the print disabled should have access to those works?” A:
“No.”); ¶ 23, Ex. V (Rønning Dep. 80:21-25, May 29, 2012: Q: “[Y]ou have no understanding of
how a U.S. student . . . with a print disability would obtain access to your works.” A: “No. Why
9

Most of the works in the HDL were digitized by Google. Additional copies were made by the HathiTrust to
facilitate search, ensure preservation, provide access to the print disabled, permit research, and maintain adequate
backups for those purposes. These copies were necessary in order to make favored uses and were in no sense
commercial.

14

should I?”). Providing access to the print disabled unlocks these books and thus departs from and
transforms their original purpose and character.
The preservation of an archive of books that otherwise would be lost is equally
transformative because the purpose of preserving books for future lawful uses is
“transformatively different from the original expressive purpose” of the books. See Bill Graham
Archives, 448 F.3d at 609; cf. Field v. Google, Inc., 412 F. Supp. 2d 1106, 1118-19 (D. Nev.
2006) (finding Google’s “caching” of websites was transformative because the archival copy
allowed researchers to access information that was otherwise inaccessible, and thus served a
different purpose than the artistic function of the copyrighted works).
Even if the Court holds that copying for purposes of preservation and providing access to
the print disabled are not transformative uses, these uses are still protected fair use. While a
determination that a use is transformative is typically dispositive, the converse is not true.
Because the ultimate inquiry in each case is whether the use is beneficial to society, uses that are
not transformative but benefit the public are often held to be fair use. For example, in Sony Corp.
of America v. Universal City Studios, Inc., 464 U.S. 417 (1984), the Supreme Court held that
private copying of television broadcasts to view at a later time (i.e., “time shifting”) benefited
society and, thus, was a protected fair use even though the copyrighted works were being used
for the identical purpose. Id. at 449, 454 (noting that the noncommercial character of the use
rendered it presumptively fair); see also, e.g., Swatch Grp. Mgmt. Servs. Ltd. v. Bloomberg LP,
No. 11 Civ. 1006, 2012 WL 1759944, at *3-4, *6 (S.D.N.Y. May 17, 2012) (posting full versions
of earnings calls, while not transformative, was nonetheless a fair use because it advanced the
public interest of disseminating financial news).

15

Copying in order to permit reading by the blind and copying for preservation are in fact
mentioned by Congress as fair use. With respect to the blind: “Another special instance
illustrating the application of the fair use doctrine pertains to the making of copies or
phonorecords of works in the special forms needed for the use of blind persons.” H.R. Rep. No.
94-1476, at 73 (1976), reprinted in 1976 U.S.C.C.A.N. 5659, 5686; see also Sony, 464 U.S. at
455 n.40 (“Making a copy of a copyrighted work for the convenience of a blind person is
expressly identified by the House Committee Report as an example of fair use, with no
suggestion that anything more than a purpose to entertain or to inform need motivate the
copying.”). As for preservation, Congress noted (in connection with movies in danger of
disintegration): “The efforts of the Library of Congress, the American Film Institute, and other
organizations to rescue and preserve this irreplaceable contribution to our cultural life are to be
applauded, and the making of duplicate copies for purposes of archival preservation certainly
falls within the scope of ‘fair use.’” H.R. Rep. No. 94-1476, at 73 (1976), reprinted in 1976
U.S.C.C.A.N. 5659, 5687.
In sum, the Libraries’ uses are uses Congress named as fair; they are transformative; and
they are made by non-profit educational institutions. Factor one “strongly favors” the Libraries.
See Cambridge Univ. Press v. Becker, No. 1:08-CV-1425-ODE, 2012 WL 1835696, at *24
(N.D. Ga. May 11, 2012) (finding fair use for the vast majority of alleged infringements because
they were preamble uses and were made by a non-profit educational institution).

16

2.

The Nature of the Copyrighted Works Factor Favors the Libraries.
a.

The Majority of Works in the HDL Are Published and Out of
Print.

Copying published works is more likely to be a fair use than copying unpublished works.
Blanch, 467 F.3d at 256 (fact that plaintiff’s work was published favored the defendants).
Virtually all works allegedly infringed are published—which tilts in the Libraries’ favor.
Moreover, “[a] key, though not necessarily determinative, factor in fair use is whether or
not the work is available to the potential user. If the work is ‘out of print’ and unavailable for
purchase through normal channels, the user may have more justification for reproducing it than
in the ordinary case . . . .” S. REP. NO. 94-473, at 64 (1975) (1975 WL 370213); see also
Maxtone-Graham, 803 F.2d at 1264 n.8 (explaining that the fact plaintiff’s book was out of print
provided additional support for finding fair use). Plaintiff Authors Guild has acknowledged that
“[a]pproximately 75% of the Books in United States libraries are out-of-print and have ceased
earning any income at all for their Rightsholders,” a fact that favors fair use copying of them.
Mem. of Law in Supp. of Pls.’ Mot. For Prelim. Settlement Approval at 27, The Authors Guild,
Inc. v. Google Inc., No. 05-cv-8136 (S.D.N.Y. Oct. 28, 2008). (See also Wilkin Decl. ¶ 66 (“It is
my understanding that the vast majority of works in the corpus are now out of print (and, in fact,
for older works within the collection, have been out of print for decades).”).)
Many of these out-of-print works are also in the public domain—although it is difficult to
confirm which books are and which ones are not. Works published between 1923 and 1963
entered the public domain unless they were renewed, and the vast majority were not renewed.
For example, a 1960 Copyright Office study found that only 7% of books were renewed. See
Staff of S. Comm. on the Judiciary (Barbara Ringer), 86th Cong., Renewal of Copyright 31, at
220 (Comm. Print 1960). Since the Copyright Office records prior to January 1, 1978 are not

17

completely or reliably digitized, however, the process of confirming whether a work was
renewed requires manually checking the physical records at the Copyright Office in Washington,
D.C. for each work, an impossible task given the number of works within the HDL corpus. And
of course, even if a search confirmed that the work was renewed and still subject to copyright
protection, there is no guarantee that a subsequent search would identify the current rights holder
(which given the life-plus-70-year copyright term could be grandchildren, ex-spouses, creditors,
bankruptcy trustees, or defunct publishers).
The HDL currently contains millions of works published during the 1923-1963 time
period. If only 7% of these works are still subject to copyright protection, then well over a
million books currently being treated as in copyright are actually in the public domain.
The uncertainty about whether a book published between 1923 and 1963 is still in
copyright, coupled with the difficulty of locating the author or his/her heirs decades after
publication, has had a devastating effect on the availability of new printings of out-of-print books
during those 40 years, as reflected in the following chart from Professor Paul Heald at University
of Illinois College of Law, which shows a pronounced drop-off in the publication of older books
upon approaching the 1923 boundary (on the following page):10

10

See Rebecca J. Rosen, The Missing 20th Century: How Copyright Protection Makes Books Vanish, The Atlantic,
Mar. 30, 2012, http://www.theatlantic.com/technology/archive/2012/03/the-missing-20th-century-how-copyrightprotection-makes-books-vanish/255282/#. Professor Heald’s study involved the random selection of 2,500 fiction
books available for purchase new from Amazon. The y-axis indicates the number of books in the sample that were
first published during each decade reflected on the x-axis.

18

Fearful of expansive copyright infringement lawsuits by the tiny, all but unknowable minority
who may have renewed their copyrights, publishers simply stopped printing them, thereby
further reducing availability, a key issue for factor two.
b.

The Majority of Works in the HDL Are Factual Works.

The second factor recognizes “that some works are closer to the core of intended
copyright protection than others,” Campbell, 510 U.S. at 586, and copying factual works is more
likely to be fair use than copying creative works. Blanch, 467 F.3d at 256 (citation omitted); see
also Maxtone-Graham, 803 F.2d at 1263; Rosemont Enters., Inc. v. Random House, Inc., 366
F.2d 303, 307 (2d Cir. 1966).
As explained in Mr. Wilkin’s declaration, an analysis of the call numbers within the HDL
corpus reveals that less than 9% of the corpus consists of prose fiction, poetry, and drama. The
remainder, approximately 90% of the corpus consists of factual works such as books and
journals in the humanities, social sciences, and physical sciences. (Wilkin Decl. ¶ 67.) Since the
vast majority of works within the HDL corpus are factual, the Libraries’ fair use rights with
respect to such works are commensurately broad.

19

c.

In-Print and Fictional Works Have Limited Significance Given
the Transformative Nature of the Use.

Because the Libraries’ uses are transformative, the fact that a small minority of works
may be in print or fictional, and thus less susceptible to fair use under factor two, offers little
support for Plaintiffs’ claims. See Campbell, 510 U.S. at 586 (noting that this factor is unlikely to
“separat[e] the fair use sheep from the infringing goats” where the use is transformative); Bill
Graham Archives, 448 F.3d at 612 (noting that “the second factor may be of limited usefulness
where the creative work of art is being used for a transformative purpose”).
3.

The Libraries Copy No More of the Books Than Necessary.

The third factor focuses on whether the amount of copying was reasonable in relation to
its purpose. Sony, 464 U.S. at 449-50. “There are no absolute rules as to how much of a
copyrighted work may be copied and still be considered a fair use,” Maxtone-Graham, 803 F.2d
at 1263, and “the extent of permissible copying varies with the purpose and character of the use,”
Campbell, 510 U.S. at 586-87. The crux of the inquiry is whether “no more was taken than
necessary.” Id. at 587 (emphasis added). In some cases, it is “necessary” to copy, and even
display, entire works. See Bill Graham Archives, 448 F.3d at 613; Arriba Soft, 336 F.3d at 821.
This treatment of the third factor highlights that the proper focus of fair use is logically
on the “use” and not the copying that enabled the use. As such, courts have recognized, for
example, that “intermediate copying” of entire works is not infringing where the copying is
necessary to engage in fair use. See, e.g., Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510,
1526-27 (9th Cir. 1993) (copying entire software program for purposes of reverse engineering
was fair use); Sony Computer Entm’t, Inc. v. Connectix Corp., 203 F.3d 596, 605-06 (9th Cir.
2000) (same); Sundeman v. Seajay Soc’y, Inc., 142 F.3d 194, 206 (4th Cir. 1998) (fair use to
copy fragile manuscript so that author of critical review could study it without damaging the

20

original). “It would severely restrict scholarly pursuit, and inhibit the purposes of the Copyright
Act, if a fragile original could not be copied to facilitate literary criticism.” Id. at 206; cf. Perfect
10, 508 F.3d at 1169 (finding incidental copying to assist computer users’ access to the Internet
was a transformative fair use). Similarly, the Libraries’ digitization of entire books is justified by
their fair uses of the digital copies.
The third factor typically is concerned only with expressive uses of copied works. See
Matthew Sag, Copyright and Copy-Reliant Technology, 103 Nw. U. L. Rev. 1607, 1650 (2009)
(“The issue at the heart of the third factor is not simply what percentage of the copyright owner’s
original work has been taken, but what proportion of the work’s expressive value has been
appropriated.”). For example, in Blanch, the artist Jeff Koons scanned the plaintiff’s entire
photograph (as well as many other images from advertisements), yet in evaluating factor three
the Second Circuit emphasized that Koons ultimately had not incorporated the most expressive
aspects of the photograph into his artwork. Blanch, 467 F.3d at 257-58 (concluding that the third
factor “weighs distinctly in Koons’s favor); cf. Cambridge Univ. Press, 2012 WL 1835696, at
*43, passim (copying of works that were made available online but never accessed was de
minimis and therefore non-infringing without the need even to consider fair use).
The Libraries’ uses for search and preservation do not display any protected text from the
scanned books—not even a snippet—and therefore cannot be said to use the expressive elements
of the works. Moreover, the Libraries have copied no more than is essential for their fair uses.
The entire texts of Plaintiffs’ works must be digitized for preservation purposes. And search of
only partially-digitized texts would obviously lead to incomplete and inaccurate results. See
Arriba Soft, 336 F.3d at 821 (“If Arriba only copied part of the image, it would be more difficult
to identify it, thereby reducing the usefulness of the visual search engine.”). Imagine, for

21

example, preserving only every third chapter or searching only the first 100 pages of a book.
Factor three is at worst neutral.
4.

The Libraries’ Uses Do Not Affect the Market For or Value of
Plaintiffs’ Books.

The fourth factor asks “whether the secondary use usurps the market of the original
work.” NXIVM, 364 F.3d at 482. The “markets” at issue are “only those that creators of original
works would in general develop or license others to develop.” Campbell, 510 U.S. at 592; Bill
Graham Archives, 448 F.3d at 614 (noting that only “traditional, reasonable, or likely to be
developed markets” are relevant under the fourth factor). Because the Libraries’ uses are
noncommercial, Plaintiffs carry the burden of “showing by a preponderance of the evidence that
some meaningful likelihood of future harm exists.” Sony, 464 U.S. at 451. “[A] use that has no
demonstrable effect upon the potential market for, or the value of, the copyrighted work need not
be prohibited in order to protect the author’s incentive to create.” Id. at 450, 454 (noting that in
light of the societal benefits of time-shifting, the copyright holder needed to demonstrate a
likelihood of harm).
Plaintiffs admit they are unable to identify “any specific, quantifiable past harm, or
any documents relating to any such past harm” resulting from the Libraries’ uses of their
works. (Petersen Decl. ¶¶ 2 - 21, Exs. A - T (responses to Interrogatory No. 5) (emphasis
added).) Indeed, two Plaintiffs deposed in this action—including the Secretary of Plaintiff
Authors Guild, Pat Cummings—confessed to not even being aware of the Libraries’ uses of incopyright works within the HDL, a fact that evidences they have not suffered any demonstrable
harm from the Libraries digitization of their works. (See id. ¶ 22, Ex. U (Cummings Dep. 19:3-9,
May 22, 2012; Q: “[D]o you have any understanding of the use made by the libraries with
respect to the digitized works in the HathiTrust Corpus?” A: “I don’t know what uses they’re

22

making of it.”); ¶ 23, Ex. V (Rønning Dep. 52:8-14, May 29, 2012; Q: “[D]o you have any
understanding of the types of uses made by my clients with respect to the in copyright—the
works that are presumed to be in copyright that are included in the HathiTrust digital library?” A:
“No.”).) Instead, Plaintiffs argue they will suffer only prospective, unrealized harms.
In the first place, there is no harm. The Libraries continue to purchase books, even books
that have already been digitized. (Wilkin Decl. ¶¶ 72-74.) Last year alone, Michigan spent over
24 million dollars on library hardcopy and electronic acquisitions. (Id. ¶ 14.)
Second, were some market effect to emerge, any such harm is irrelevant to the fair use
analysis because the Libraries’ uses are transformative and thus any harm arises, if at all, in a
“transformative [i.e., unprotected] market.” See Bill Graham Archives, 448 F.3d at 614
(“Appellant asserts that it established a market for licensing its images, and in this case expressed
a willingness to license images to DK. Neither of these arguments shows impairment to a
traditional, as opposed to a transformative market.”). Plaintiffs cannot preempt markets that
currently do not exist by speculating that they might one day develop a licensing scheme for
these uses; nor can they disrupt what is otherwise a fair use, even if a market is conceivable.
If this were not the case, a plaintiff would always point to the use made by the defendant
and claim that licensing revenue would be lost if the defendant’s conduct were deemed a fair use;
a plaintiff could simply “boot-strap the specter of a fair use holding . . . as reason why the use
was not a fair use to begin with.” Hofheinz v. A & E Television Networks, 146 F. Supp. 2d 442,
449 (S.D.N.Y. 2001) (use of movie footage in television biography program a fair use). This
would “eviscerate” fair use “since every copyright infringer seeking protection of the fair use
doctrine could have potentially sought a license from the owner of the infringed work.” Hofheinz
v. Discovery Commc’ns, Inc., 2001 WL 1111970, at *6 (Baer, J.) (internal quotation and citation

23

omitted); see also Bill Graham Archives, 448 F.3d at 614 (noting that the fourth factor would
always favor the copyright holder “were a court automatically to conclude in every case that
potential licensing revenues were impermissibly impaired simply because the secondary user did
not pay a fee for the right to engage in the use”) (internal quotation and citation omitted).
Third, Plaintiffs’ alleged “prospective” harm to a “potential” market is pure conjecture, as
shown by their own evidence. Plaintiffs have not produced in discovery:
(1)

any business plans for licensing the digitization of books; or

(2)

any plans for the use books for preservation and research purposes; or

(3)

any analysis of their costs for licensing such markets or anticipated revenues; or

(4)

any communications with entities that might collectively license these rights; or

(5)

any analysis of the substantial limitations that such an entity would face in terms
of the number of works it could foreseeably license.

Plaintiffs also have not identified “any revenue or other earnings of any kind generated or
expected to be generated in whole or in part” for archiving, research, full-text search, or use by
the blind. (Petersen Decl. ¶¶ 2 -18, Exs. A – Q (responses to Request Nos. 8 -11); ¶¶ 19 – 21,
Exs. R – T (responses to Request Nos. 4 – 7 (misnumbered in Plaintiffs’ responses)).)
Plaintiffs’ speculative, generalized allegations fall short of satisfying their burden of
“showing by a preponderance of the evidence that some meaningful likelihood of future harm
exists.” Sony, 464 U.S. at 451, 454 (endorsing the District Court’s conclusion: “Harm from timeshifting is speculative and, at best, minimal.”); see also Blanch, 467 F.3d at 258 & n.9 (rejecting
speculative harm to potential market for photograph); Perfect 10, 508 F.3d at 1168 (rejecting
“hypothetical” harm to market for thumbnail images on mobile phones). And of course, the fact
that the majority of works digitized by the Libraries are out of print also supports the absence of
any adverse effect on Plaintiffs’ market. See Hofheinz v. Discovery Commc’ns, Inc., 2001 WL

24

1111970, at *6 (Baer, J.) (noting that if there were a market for the plaintiff’s film, which
was not being sold or rented, it was diminutive); Duffy v. Penguin Books USA Inc., 4 F.
Supp. 2d 268, 275 (S.D.N.Y. 1998) (noting that because the plaintiff’s book was out of print
at the time of the alleged infringement, there was no market harm).
Finally, there is no viable market likely to be developed for licensing preservation, printdisabled access, or search for the books in the HDL. This is because of the prohibitively high
transaction costs associated with seeking licenses for millions of books. Dr. Joel Waldfogel,
Frederick R. Kappel Professor of Applied Economics at the University of Minnesota,
conservatively estimates that that cost of securing permission from every possible rights holder
for books in the HDL would be $569 million—and that does not even include the cost of any
licensing fee to the rights holder. (Waldfogel ¶¶ 22-24.) And the estimate assumes each rights
holder can be identified (id. ¶ 24); of course, as noted previously, they cannot (id. ¶¶ 18-20).
As Dr. Waldfogel concludes: “The costs associated with creating and maintaining a
service like the HathiTrust would exceed any potential revenue that such a venture could earn.
Thus, I conclude that the creation and offering of a service with the functionality of the
HathiTrust, but with licensed content, is not a commercially viable endeavor (i.e., an endeavor
where the revenues could cover the costs).” (Id. ¶ 7.) This insurmountable barrier prevents any
market from forming and supports fair use. See Wendy J. Gordon, Fair Use As Market Failure,
A Structural and Economic Analysis of the Betamax Case and Its Predecessors, 82 Colum. L.
Rev. 1600, 1618, 1628 (1982) (noting that where there are prohibitively high transaction costs, a
market failure results and courts should allow fair use).
Plaintiffs appear to be relying on speculation that Congress will eventually step in and
form a market, e.g., by creating a compulsory licensing system. (See First Am. Compl. ¶ 6.) As

25

the Supreme Court noted in Sony, “it is not [the court’s] job to apply laws that have not yet been
written.” Sony, 464 U.S. at 456. The need for payment to rights holders through a compulsory
license or otherwise only becomes an issue if the Libraries’ activities are not a fair use because
fair uses require neither payment nor permission. See 17 U.S.C. § 107 (“[T]he fair use of a
copyrighted work . . . is not an infringement of copyright.”).
C.

Balancing All the Factors.

“The ultimate test of fair use . . . is whether the copyright law’s goal of promoting the
Progress of Science . . . would be better served by allowing the use than by preventing it,” Bill
Graham Archives, 448 F.3d at 608 (quotation marks omitted; ellipsis in original), but fair use
also is an “equitable rule of reason,” Sony, 464 U.S. at 448. Thus, the equities between the
parties may come into play in the fair use analysis.
Here, Plaintiffs were aware of the Libraries’ digitization at least as early as 2004 when
Google publicly announced its agreement with Michigan (among other academic libraries) to
digitize its collection—nearly seven years before this lawsuit was filed. Yet during this period in
which the Libraries were expending considerable resources digitizing their collections and
creating the infrastructure to securely store the works, and despite the fact that Plaintiff Authors
Guild in 2005 filed a lawsuit against Google based on its digitization of these materials, Plaintiffs
did not make a single objection to the Libraries. Now that Plaintiffs have filed a lawsuit against
the Libraries, one would assume that each Plaintiff would at least be familiar with the Libraries’
uses of their works, yet at least two deposed Plaintiffs—including an officer of Plaintiff Authors
Guild—are not.11 In view of Plaintiffs’ delay, the Libraries’ expenditures during the seven years
when Plaintiffs were on notice, and the cavalier attitude of at least some Plaintiffs as to the
reason for this lawsuit, the equities favor the Libraries.
11

To reduce costs, the Libraries only deposed four Plaintiffs.

26

III.

The Orphan Works Project.
As the court is aware, the Libraries have suspended any plans to make any works

available through the Orphan Works Project (OWP), and the project may never resume. (Wilkin
Decl. ¶¶ 114-16.) For that reason, as addressed in the Libraries’ pending motion, the issue is not
ripe for adjudication. But even if the Court disagrees and considers now Plaintiffs’ infringement
claims with respect to the project, there is no justiciable issue as to the infringement of any
copyright by the OWP.
This is because not one single orphan works candidate—i.e., not one single out-of-print
book whose copyright holder could not be located—has ever been distributed, displayed, or
performed to anyone. That material fact is undisputed. (Id. ¶ 116.) As a result, the Libraries are
entitled to summary judgment that the OWP has not infringed a single right under Section
106 of a single book copyright.
As for whether the hypothetical OWP is a fair use: if the OWP were ever implemented in
the manner previously contemplated (i.e., if one single out-of-print but in-copyright work whose
copyright holder could not be found despite a rigorous search were ever made available for a
student or researcher to view subject to copyright warnings and in full compliance with 17
U.S.C. § 108(e), then a federal court at that time could evaluate whether making available such
a work for the purposes of research and scholarship—Congressionally-endorsed uses—was or
was not a fair use.
Finally, should the Court decide the OWP is ripe for a fair use determination, we would
respectfully request that the parties be permitted to fully brief the issue. We would only note at

27

1100 Peachtree Street, Suite 2800
Atlanta, Georgia 30309-4530
Telephone: [phone]
Facsimile: [phone]
Email: [e-mail]
Attorneys for Defendants

29