The Authors Guild et al v. Google Inc.

Filing 239

BRIEF Amicus Curiae. Document filed by New York Law School, Institute for Information Law and Policy.(Grimmelmann, James)

The Author's Guild et al v. Google Inc.

Doc. 239

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ------------------------------------------------------X THE AUTHORS GUILD, et al. : Plaintiffs, v. GOOGLE, INC., : : : CIV. NO: 05-CV-8136 (DC)

Defendant. : ------------------------------------------------------X

BRIEF OF AMICUS CURIAE

JAMES GRIMMELMANN 57 Worth Street New York, New York [phone] and KORNSTEIN VEISZ WEXLER & POLLARD, LLP 757 Third Avenue New York, New York 10017 [phone] Attorneys for Amicus Curiae, Institute for Information Law and Policy, New York Law School

Dockets.Justia.com

Table of Contents Table of Authorities .................................................................................................................... iii-v Statement of Interest of Amicus ..................................................................................................... 1 Introduction and Summary of Argument........................................................................................ 1 Argument ....................................................................................................................................... 1 I. THE CURRENT UNAVAILABILITY OF ORPHAN WORK BOOKS HARMS THE GOALS OF THE COPYRIGHT ACT AND THE CONSTITUTION.......... 1 A. B. C. II. Copyright Seeks to Maximize Public Access to Creative Works by Giving Authors the Incentive of Exclusive Rights ................................ 2 Millions of Books Are Orphan Works........................................................ 2 Orphan Works Are Failures of the Copyright System................................ 4

THE SETTLEMENT BOTH SERVES AND THREATENS THE PUBLIC INTEREST ............................................................. 5 A. B. The Settlement Serves the Public Interest by Making Many Orphan Work Books Available........................................... 6 The Settlement Threatens the Public Interest by Creating a Dangerous Concentration of Power in Google and the Book Rights Registry ............. 8 1. 2. 3. The Settlement Gives Google Exclusive Access to Orphan Work Books....................................................... 8 The Structure of the Settlement Raises Serious Antitrust Concerns................................................ 10 This Concentration of Power Threatens the Goals of the Copyright Act and the Public Interest ................................ 13

III.

THE SETTLEMENT INAPPROPRIATELY ATTEMPTS TO SOLVE A LEGISLATIVE PROBLEM THROUGH A CLASS ACTION SETTLEMENT ................................................................... 14 A. B. The Settlement Raises Grave Doubts About the Suitability of a Class Action........................................................................................................ 15 The Orphan Works Problem Is Inherently Legislative ............................. 17 1. Large and Diverse Constituencies ................................................ 17

i

2. IV.

A Problem for the Legislative Branch .......................................... 18

THIS COURT SHOULD SOLICIT ADDITIONAL VIEWS BEFORE RULING ON THE MOTION TO APPROVE THE SETTLEMENT.................. 20 A. B. C. D. The Procedural Posture of the Case Has Deprived this Court of Essential Perspectives .......................................... 20 This Court Should Solicit the Views of the Copyright Office, the Department of Justice, and the Federal Trade Commission ............... 21 This Court Should Appoint Counsel to Represent the Subclass of Orphan Work Book Copyright Owners........................... 22 This Court Should Scrutinize Closely Provisions of the Settlement Agreement That Could Give Google and the Named Plaintiffs Open-Ended Powers to Alter Its Terms.................. 23

Conclusion .................................................................................................................................... 25

ii

Table of Authorities Cases Amchem Products, Inc. v. Windsor, 521 U.S. 591 (1997).................................................................................................................. 16 Ariz. v. Maricopa County Medical Soc., 457 U.S. 332 (1982).................................................................................................................. 10 Associated Press v. United States, 326 U.S. 1 (1945)........................................................................................................................ 8 Bailey v. Great Lakes Canning, 908 F.2d 38 (6th Cir. 1990) ...................................................................................................... 20 Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., 441 U.S. 1 (1979)................................................................................................................ 11, 12 Cablevision Systems Dev. Co. v. Motion Picture Ass'n of Am., 836 F.2d 599 (D.C. Cir. 1988) .................................................................................................. 21 Campbell v. Acuff-Rose Music, 510 U.S. 569 (1994).................................................................................................................. 18 CCC Information Services, Inc. v. Maclean Hunter Market Reports, Inc., 44 F.3d 61 (2d Cir. 1994) ........................................................................................................... 2 Durrett v. Housing Authority of Providence, 896 F.2d 600 (1st Cir. 1990)..................................................................................................... 20 Eldred v. Ashcroft, 537 U.S. 186 (2003)................................................................................................................ 1, 2 Emerson v. Davies, 8 F. Cas. 615 (No 4,436) (CCD Mass. 1845) ........................................................................... 18 Holmes v. Continental Can Co., 706 F.2d 1144 (11th Cir. 1983) ................................................................................................ 15 In re Literary Works in Elec. Databases Copyright. Litig., 509 F.3d 116 (2d Cir. 2007), cert. granted sub nom ............................................................................................................... 10 Mazer v. Stein, 347 U.S. 201 (1954).................................................................................................................... 2 iii

Medimmune, Inc. v. Genetech, Inc., 549 U.S. 118 (2007).................................................................................................................. 10 N.Y. Mercantile Exch. Inc. v. IntercontinentalExchange, Inc., 497 F.3d 109 (2d Cir. 2007)........................................................................................................ 2 NYNEX Corp. v. Discon, 525 U.S. 128 (1998).................................................................................................................. 10 Ortiz v. Fibreboard Corp., 527 U.S. 815 (1999).................................................................................................................. 22 Reed Elsevier, Inc. v. Muchnick, 129 S.Ct. 1523 (Mem.)(2009)................................................................................................... 10 United States v. Am. Soc'y Composers, Authors & Publishers, 2001 WL 1589999 (S.D.N.Y. 2001)......................................................................................... 12 United States v. Broad. Music. Inc., 1994 WL 901652 (S.D.N.Y. 1994)........................................................................................... 12 United States v. Socony-Vacuum Oil Co., 310 U.S. 150 (1940).................................................................................................................. 10 Statutes and Other Authorities 1 Annals of Cong. 972 (Joseph Gales ed. 1834) ............................................................................. 8 15 U.S.C. § 1................................................................................................................................. 10 17 U.S. C. § 302(a) ......................................................................................................................... 5 17 U.S. C. § 302(c) ......................................................................................................................... 5 17 U.S. C. § 505.............................................................................................................................. 4 17 U.S.C. § 106(4) ........................................................................................................................ 12 17 U.S.C. § 504(c) .......................................................................................................................... 4 Am. Lib. Ass'n, Intellectual Freedom and Censorship Q & A, http://www.ala.org/ala/aboutala/offices/oif/basics/intellectual.cfm ......................................... 13 Fed. R. Civ. P. 23(b)(3.................................................................................................................. 15 Fed. R. Civ. P. 23(e) ............................................................................................................... 15, 16 iv

Greco, Albert N., The Book Publishing Industry (2d ed. 2005) ............................................................................................................................. 18 Manual for Complex Litigation (Fourth) § 21.631..................................................................................................................................... 24 Netanel, Neil Weinstock,Copyright and a Democratic Civil Society, 106 Yale L.J. 283 (1996) ............................................................................................................ 8 Orphan Works Act of 2008, H.R. 5889 110th Cong. (2008) ................................................................................................. 19 Picker, Randal C. , The Google Book Search Settlement: A New Orphan-Works Monopoly? 20 (University of Chicago Law School John M. Olin Law & Economics Working Paper No. 462 (2d series)) ................................................................................................................... 11, 17 Report on Orphan Works 1, U.S. Copyright Office (2006) ............................................................................................. 3, 4, 5 Rich, Motoko, Google Hopes to Open a Trove of Little-Seen Books, N.Y. Times, Jan. 5, 2009, at B1.................................................................................................. 3 Richards, Neil, Intellectual Privacy, 87 Tex. L. Rev. 387 (2008)....................................................................................................... 14 Rose, Mark, Authors and Owners: The Invention of Copyright 47 ............................................... 9 Shawn Bentley Orphan Works Act of 2008, S. 2913 110th Cong. (2008)...................................................................................................... 19 Schultz, Jason, The Myth of the 1976 Copyright "Chaos" Theory", http://www.lessig.org/blog/archives/jasonfinal.pdf .................................................................... 3 Sprigman, Christopher, Reform(al)izing Copyright, 57 Stan. L. Rev. 485 (2004) ........................................................................................................ 3 Wright, Miller, and Kane, Federal Practice and Procedure § 1797.3 ............................................................................... 15

v

Statement of Interest of Amicus Amicus Institute for Information Law and Policy is New York Law School's home for the study of law, technology, and civil liberties. Its mission is to understand the interplay of law and technology and influence their development to serve democratic values in the digital age. It aims to prepare students to enter the legal profession, to extend human knowledge, and to harness new informational tools to the goals of social justice. The Institute is led by the faculty of New York Law School and its work is carried out collaboratively by faculty, students, and staff. New York Law School, a 501(c)(3) nonprofit corporation, is the second-oldest independent law school in the United States. Introduction and Summary of Argument The proposed Settlement Agreement offers substantial public benefits by making books more broadly available to the public while compensating copyright owners where they can be found. Unfortunately, it is well known that a substantial fraction of copyright owners in books cannot be located. The present proposed Settlement Agreement ("Settlement") may threaten the public interest by treating the silence of these absent class members as consent to a reallocation of rights that creates a dangerous concentration of power in Google and the Registry. In light of this danger, this Court has a duty to make a searching inquiry into the effects of the Settlement and to ensure that it is fully advised on all relevant points of view. Argument I. THE CURRENT UNAVAILABILITY OF ORPHAN WORK BOOKS HARMS THE GOALS OF THE COPYRIGHT ACT AND THE CONSTITUTION "[C]opyright's purpose is to promote the creation and publication of free expression." Eldred v. Ashcroft, 537 U.S. 186, 219 (2003) (emphasis in original). Unfortunately the copyright

owners for millions of books are either unknown or unfindable. Copyright is neither providing an incentive for these books' creators nor promoting their publication. A. Copyright Seeks to Maximize Public Access to Creative Works by Giving Authors the Incentive of Exclusive Rights

Copyright's importance is recognized in the Constitution, which empowers Congress "To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries," Art. I, § 8, cl. 8. "The economic philosophy behind the [Copyright Clause] is the conviction that encouragement of individual effort by personal gain is the best way to advance public welfare through the talents of authors and inventors . . . ." Mazer v. Stein, 347 U.S. 201, 219 (1954). "Without such incentives authors `might direct their energies elsewhere, depriving the public of their creations and impeding the advance of learning.'" N.Y. Mercantile Exch. Inc. v. IntercontinentalExchange, Inc., 497 F.3d 109, 118 (2d Cir. 2007) (quoting CCC Information Services, Inc. v. Maclean Hunter Market Reports, Inc., 44 F.3d 61, 66 (2d Cir. 1994)). Copyright thus has two closely intertwined goals. On the one hand, it seeks to give authors "rewards commensurate with the services rendered." Mazer, 347 U.S. at 219. On the other, it seeks to maximize the quantity and quality of creative works actually made available to the public. "The two ends are not mutually exclusive; copyright law serves public ends by providing individuals with an incentive to pursue private ones." Eldred, 537 U.S. at 212 n.18. Serving the Constitutional purpose of the "Progress of Science and useful Arts" requires both that authors be incentivized to create and that works, once created, reach the public. B. Millions of Books Are Orphan Works

Unfortunately, many books are "orphan works," books that are under copyright but whose copyright owners "cannot be identified and located by someone who wishes to make use 2

of the work in a manner that requires permission of the copyright owner." U.S. Copyright Office, Report on Orphan Works 1 (2006). Some books are orphaned when their copyright owners move or the copyrights change hands and the public record no longer provides enough information to know whom to ask for permission to use the copyrighted material. Id. at 26­29. Changes in ownership upon the death of an author or the dissolution of a publisher are particularly likely to cause trouble; the new copyright owners may not even realize that they are copyright owners. Id. at 28­29. It is impossible to know with certainty precisely how many books are orphan works, since it is impossible to know with certainty that a particular book is an orphan. A copyright owner could always emerge at any time. Still, some estimates are possible. As books age, the fraction worth keeping in print drops precipitously. One study found that only 2.3 percent of books published in the United States between 1927 and 1946 are still in print. See Jason Schultz, The Myth of the 1976 Copyright "Chaos" Theory", available at http://www.lessig.org/blog/archives/jasonfinal.pdf. Out of the seven million books that Google had scanned when the Settlement was proposed, about five million are not commercially available. (Another one million are in the public domain.) See Motoko Rich, Google Hopes to Open a Trove of Little-Seen Books, N.Y. Times, Jan. 5, 2009, at B1. Books that have been out of print for decades have a substantial chance of being orphans. Indeed, multiple studies of renewal rates under the 1909 Copyright Act found that copyright owners renewed the copyrights in only ten to twenty percent of works. See Christopher Sprigman, Reform(al)izing Copyright, 57 Stan. L. Rev. 485, 519­21 (2004) (summarizing studies). Unrenewed works were either already orphans or were considered by their owners as not worth keeping under copyright. The Settlement also suggests that a majority

3

of out-of-print books may be orphans. It sets aside $45 million for cash payments to class members whose books were digitized, at a rate of $60 per book, Settlement § 2.1(b), suggesting that the parties estimate making cash payments for only about 750,000 scanned books. The copyright owners of the remaining four million or more books would then presumptively be orphan owners, if they could not be found by a $7 million notice program and did not claim their cash payments. C. Orphan Works Are Failures of the Copyright System

When a copyrighted work becomes an orphan, both goals of copyright are frustrated. A copyright owner who is unfindable is a fortiori unpayable. Even where there is substantial public demand for an orphaned book, its author thus realizes no "rewards" for "render[ing]" the "service" of writing it, and its publisher has no useful incentive to print a new edition. At the same time, the public is deprived of effective access to the book. Even when a user--a new publisher, for example--would like to pay for the right to make a particular use, she "cannot reduce the risk of copyright liability for such use, because there is always a possibility, however remote, that a copyright owner could bring an infringement action after that use has begun." Report on Orphan Works at 15. This problem is particularly severe for institutions, such as libraries and museums, whose mission is to preserve and make available large archives of historical works. Id. at 37­38. If a library were to make available to its patrons electronic copies of the long-out-of-print books in its collection, it would face ruinous potential copyright liability. Even if any individual book was unlikely to be the subject of a lawsuit by a disgruntled copyright owner, the chance would be substantial that some book would be. Features of the Copyright Act compound the severity of the orphan works problem. Copyright owners may recover statutory damages of up to $150,000 per work infringed, 17 U.S.C. § 504(c), and attorneys fees, id. § 505, strongly deterring even minor and socially 4

beneficial infringements of orphan works. Copyright now endures for the life of the author plus 70 years, id. § 302(a), or 95 years from first publication for works made for hire, id. § 302(c). Once a book goes out of print and is no longer actively making money, it is still likely to be subject to copyright for many decades--decades in which it runs a heightened risk of being orphaned. (Indeed, a term that extends past the life of the author explicitly guarantees that a copyright will see at least one change in ownership.) The 1976 Copyright Act, as revised to comply with the Berne Convention, also eliminates formalities that previously helped to keep copyright records current and made it easier to locate copyright owners, including registration as a prerequisite to copyright protection, renewal with the Copyright Office after 28 years, and required notice of copyright on the work itself. Report on Orphan Works at 41-44. II. THE SETTLEMENT BOTH SERVES AND THREATENS THE PUBLIC INTEREST While some provisions of the Settlement would create substantial public benefits, others are dangerous and perhaps illegal. The Settlement would adjust the rights of members of the Settlement Class to authorize Google to make orphan work books available to the public while collecting revenue on behalf of orphan work book copyright owners. This authorization would further the goals of the Copyright Act by making these books available to the public again while also providing fair compensation to copyright owners and respecting their rights to control the use of their works. It would also, however, threaten the public interest by giving Google and the Registry substantially exclusive control over the distribution of orphan work books. This exclusive control raises serious antitrust concerns and imperils intellectual freedom.

5

A.

The Settlement Serves the Public Interest by Making Many Orphan Work Books Available

The initial lawsuits as filed challenged only Google's right to scan books, to create an index based on them, and to display to users short excerpts in search results, all activities for which Google had a strong fair use claim. The Settlement, however, goes much further. It authorizes Google to make available the full text of books, for individual purchase, institutional subscription, and through whatever other business models Google and the Registry agree to. These full-text access programs go to the heart of the exclusive rights granted by copyright. Without this Court's authorization, such commercial exploitation of the complete contents of books for consumptive uses by readers would be obvious infringement unprotected by fair use. The Settlement's authorizations are particularly important for orphan work books. The Settlement sets a default for books that are not Commercially Available (i.e. out of print, see Settlement §§ 1.28, 3.2(d)(i)­(ii), Attachment A § 3.1) that Google is allowed to make Access Uses, including, crucially placing the books in the Consumer Purchase and Institutional Subscription programs. Id. § 3.2(b). But, of course, almost all orphan work books are out of print, and many out-of-print books are orphan works. Thus, by default, perhaps millions of orphan work books will be part of Google's book-sales programs. Although the Settlement allows copyright owners to opt out, id. § 10.2(b), to request Removal of their books entirely, id. § 3.5(a), or to Exclude their books from particular uses, id. § 3.5(b), all of these options require action by the copyright owner. Orphan work book copyright owners, however, can be expected not to take any actions at all. In effect, Google will receive perpetual immunity from copyright infringement liability for the uses it makes of orphan work books as specified in the Settlement. Google will be allowed to sell these books, so long as it sends 63% of the revenues to the Registry. Settlement 6

§§ 1.87, 4.5(a)(i), and so long as it has not heard from the individual copyright owners. At any time, a previously orphan work book copyright owner can appear and assert his or her rights. Settlement § 3.2(e). Google may not thereafter sell the book without permission, but is not liable for its past actions. The copyright owner is thus limited to the 63% revenue share as her principal remedy for what would have (but for the Settlement) been a blatant copyright infringement by Google. In its broad outlines, this limitation of remedies would be consistent with the Copyright Act's goals. On the one hand, it would enable Google to provide public access to the complete texts of millions of books that would otherwise have been mostly unavailable. The public would therefore benefit from an enormous expansion in the set of books available to anyone with a computer. On the other hand, it is probably impossible to do much better at compensating the owners of orphan work books than the Settlement does. If 63% of the revenues is a fair portion for book copyright owners actively managing their works, this fact creates at least a presumption that it is also a fair portion to collect on behalf of orphan work book owners. The Settlement also does not substantially prejudice orphan work book owners' rights to take active control of the commercial exploitation of their books should they ever reappear. Thus, the Settlement will substantially increase public access to books without doing much more to impair incentives for authors than the mere fact of the orphan works problem itself has already done. Indeed, by creating a profitable new market for books, the Settlement can be expected to increase authors' and publishers' incentives to supply the public with creative works, and to help encourage at least some orphan work book copyright owners to reappear. By creating this framework of selling access while collecting revenues on copyright owners' behalf, the Settlement advances the public interest in a well-functioning copyright system.

7

B.

The Settlement Threatens the Public Interest by Creating a Dangerous Concentration of Power in Google and the Book Rights Registry

Other provisions of the Settlement, however, are less benign. Critically, while the Settlement provides that Google will be able to sell copies of books, it is structured in a way that will effectively prevent any competitor from doing the same. For the same legal reason--blatant and unexcused copyright infringement--that Google could not legally make these uses absent the Settlement, competitors will be unable to make them. The Settlement authorizes Google, and Google only, to sell these books. This exclusivity may be, in and of itself, a violation of the antitrust laws. Moreover, such exclusive control over the distribution of copyrighted works runs counter to the goals of the Copyright Act and to the public interest in the uncensored exchange of diverse ideas. 1. The Settlement Gives Google Exclusive Access to Orphan Work Books

Concentrated power is antithetical to the goals of copyright. "In adopting the Constitution's Copyright Clause and enacting the first federal copyright statute, the Framers were animated by the belief that copyright's support for the diffusion of knowledge is `essential to the preservation of a free Constitution.'" Neil Weinstock Netanel, Copyright and a Democratic Civil Society, 106 Yale L.J. 283, 289 (1996), quoting 1 Annals of Cong. 972 (Joseph Gales ed. 1834). "[The First Amendment] rests on the assumption that the widest possible dissemination of information from diverse and antagonistic sources is essential to the welfare of the public." Associated Press v. United States, 326 U.S. 1, 20 (1945). Copyright advances that diversity by enabling the widest possible range of authors to express themselves to the public, free from state control or the whims of private patrons. Netanel at 352­56. The original English copyright act, the Statute of Anne, was enacted to limit the power of the control over publishing enjoyed by the Stationers' Company; "Parliament was concerned about stationers' monopolies [and] with the 8

near monopoly that the Stationers' Company itself had held on the book trade by virtue of its charter." Mark Rose, Authors and Owners: The Invention of Copyright 47. American copyright law reflects this heritage in its hostility to monopolies and its preference for decentralization. The Settlement, however, threatens to establish monopoly control over digital publishing by creating a reconstituted Stationers' Guild made up of Google and the Registry. The Settlement would give them exclusive digital access to millions of orphan work books. Precisely because millions of books whose copyrights would be licensed to Google via the Settlement are orphan works, no competitor will ever be able to obtain the necessary permissions to make competing uses of them. For the same reason, Google itself could never, under any circumstances, have privately negotiated the permissions the Settlement would grant it. Only the fortuitous coincidences that the plaintiffs filed the present lawsuit in the form of a class action, and were then willing to negotiate away the rights of millions of orphan class members who could be counted on not to object, would purport to give Google those rights. From now until the expiration of the copyrights associated with class members (many of which will still be in force a century from now), Google will have a legally enforced monopoly on their commercial exploitation. Nor could any potential competitor of Google's--whether commercial or dedicated to the preservation of our literary heritage--expect to travel the same road to legality as Google. Any other party wishing to acquire the same permissions Google will enjoy would need to negotiate a similar class-action settlement. To do that, however, it would need to be sued for copyright infringement, on a class basis, with class counsel willing to negotiate similar terms. None of these three conditions would be within the competitor's control. Any potential declaratory judgment lawsuit would need to be filed against a defendant class with millions of members,

9

posing insurmountable jurisdictional obstacles and ripeness issues. See In re Literary Works in Elec. Databases Copyright. Litig., 509 F.3d 116 (2d Cir. 2007) (rejecting copyright class action settlement on basis that some members of settlement class could not have sued individually for infringement), cert. granted sub nom. Reed Elsevier, Inc. v. Muchnick, 129 S.Ct. 1523 (Mem.)(2009); Medimmune, Inc. v. Genetech, Inc., 549 U.S. 118, 128 (2007) (stating that an "imminent threat" is required to satisfy the Article III case-or-controversy requirement). Similarly, while orphan works legislation would be desirable and could potentially put Google's competitors on an equally footing with it, the possibility that Congress might change the law does not relieve this Court of its duty to rule in accordance with the law as it currently stands. On the current state of the law, Google's proposed actions would be inarguably illegal but for the Settlement. Any competitor wishing to do the same but without access to the Settlement's protections would necessarily be a massive infringer of orphan work book copyrights. 2. The Structure of the Settlement Raises Serious Antitrust Concerns

This exclusivity matters for several reasons, the first and most important of which is that the Settlement may require the parties to act in violation of the antitrust laws. Section 1 of the Sherman Act prohibits "[e]very contract . . . in restraint of trade." 15 U.S.C. § 1. Horizontal price-fixing by competitors is one of the few per se antitrust violations. See NYNEX Corp. v. Discon, 525 U.S. 128, 133 (1998). Once horizontal price-fixing is proven, "no showing of socalled competitive abuses or evils which those agreements were designed to eliminate or alleviate may be interposed as a defense." Ariz. v. Maricopa County Medical Soc., 457 U.S. 332, 345 (1982), quoting United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 218 (1940).

10

Here, the Settlement Controlled Pricing option for the Consumer Purchase program allows copyright owners to delegate to Google the prices to be charged for their books. Google is directed to "maximize revenue for each Rightsholder," Settlement § 4.2(b)(ii)). It is true that this language can be interpreted to mean that Google must set prices as copyright owners in a competitive market would, lowering prices so that the books compete effectively with each other. But the language can also be interpreted to mean that Google must set prices such that "each Rightsholder" does better than she would in a competitive market, by setting all prices high simultaneously. Under this latter interpretation, Settlement Controlled Pricing would strongly resemble a coordinated price-fixing cartel. See Randal C. Picker, The Google Book Search Settlement: A New Orphan-Works Monopoly? 20 (University of Chicago Law School John M. Olin Law & Economics Working Paper No. 462 (2d series)). Because the Settlement makes Settlement Controlled Pricing the default, orphan works would form an enormous group of would-be competitors acting in concert to set their sales prices "in restraint of trade." Implementation of the Settlement might therefore require a per se violation of the Sherman Act. Readers would be forced to pay higher prices, making the book market less competitive and reducing the incentive for authors to write better books. The Settlement's treatment of the Institutional Subscription and the Registry is also troublesome. Here, the Settlement requires that Google and the Registry negotiate terms for blanket subscription licenses to libraries, universities, and other institutions. Each subscribing institution would pay one set fee for access to the complete collection. While the Supreme Court has held that blanket licenses to copyrighted works can be legal under the antitrust laws, they are nonetheless subject to rule of reason scrutiny. See Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., 441 U.S. 1, 16­24 (1979).

11

The closest analogy to the Registry's role under the Settlement would be the performing rights organizations (PROs): ASCAP, BMI, and SESAC. These organizations administer the public performance rights of musical work copyright owners, 17 U.S.C. § 106(4). Composers license their compositions to a PRO on a nonexclusive basis, the PRO offers blanket licenses to its entire catalog whose price is based on the size and nature of the licensee, and the PRO divides the revenues (less administrative fees) among the composers in proportion to the uses made of their compositions. That structure exactly parallels the Registry's role: obtain non-exclusive licenses from authors, offer blanket licenses, and pay authors in proportion to uses. However, the two largest PROs are both subject to antitrust consent decrees. See United States v. Am. Soc'y Composers, Authors & Publishers, 2001 WL 1589999 (S.D.N.Y. 2001); United States. v. Broad. Music. Inc., 1994 WL 901652 (S.D.N.Y. 1994). ASCAP agreed to one in 1941, and BMI in 1966. Both consent decrees require that the PROs offer uniform blanket licenses to all comers, and allow dissatisfied licensees to petition the supervising District Court to set the license fee when the licensee and the PRO cannot agree on terms. The Supreme Court placed substantial weight on the existence and terms of the consent decrees when dismissing an antitrust challenge to BMI's blanket license. Broadcast Music Inc., 441 U.S. at 13. In the Court's words, "But it cannot be ignored that the Federal Executive and Judiciary have carefully scrutinized ASCAP and the challenged conduct, have imposed restrictions on various of ASCAP's practices, and, by the terms of the decree, stand ready to provide further consideration, supervision, and perhaps invalidation of asserted anticompetitive practices." Id. The Registry is not so constrained. It can and must deal with Google on much more favorable terms than with Google's competitors, See, e.g., Settlement § 4.7 (authorizing Registry to negotiate terms of new revenue models with Google). Only Google can initiate arbitration

12

with the Registry in case of disagreements. See Settlement § 9.1(a) (requiring arbitration only for "disputes between and among Google, Rightsholders, the Registry, and Participating Libraries"). The Department of Justice and the courts do not "stand ready" to respond if the Registry oversteps. The executive branch has not "closely scrutinized" the Registry and offered an implicit endorsement of its legality of the sort that might be entitled to some deference from this Court. Without its own equivalent to the analysis, supervision, and consent decrees that have defanged ASCAP and BMI, the Registry could pose a significant antitrust threat. 3. This Concentration of Power Threatens the Goals of the Copyright Act and the Public Interest

The Settlement could also be harmful to the cultural diversity that copyright strives for. A vibrant and competitive media sector is a critical protection for intellectual freedom, "the right of every individual to both seek and receive information from all points of view without restriction." Am. Lib. Ass'n, Intellectual Freedom and Censorship Q & A, available at http://www.ala.org/ala/aboutala/offices/oif/basics/intellectual.cfm. The marketplace of ideas cannot thrive unless the marketplace for books is also open and free. A single dominant distribution platform for digital books would stand in stark contrast to the 50,000 publishers active today in the United States. A single bottleneck on the digital distribution of books could impinge on intellectual freedom. Under the Settlement, Google specifically reserves to itself the right to exclude any book in its sole discretion for "editorial reasons." Settlement § 3.7(e). If the Registry chooses not to engage an alternative partner--which it is not required to do--the book will become unavailable to potential readers. Google thus will have substantially unfettered power to censor any book in the collection. The Settlement also severely threatens reader privacy. As Neil Richards explains, 13

Consider a list of book purchases, library records, Web sites read, or the log of a search engine: these records reveal our interests and often our aspirations or fantasies. When such records are not kept in confidence, but are instead available for access by the government, what is at stake is not merely our privacy in general, but the intellectual privacy necessary for us to engage in the freedom of thought that enables the exercise of our First Amendment rights. Neil Richards, Intellectual Privacy, 87 Tex. L. Rev. 387, 439 (2008). Google will have access not just to readers' "book purchases" and "library records" but to their precise reading choices, page by page, and minute by minute. If they make annotations using the features the Settlement allows, Settlement § 3.10(c)(5), Google will also have access to their marginal annotations. This highly private information should be subject to careful controls limiting its reuse and its sharing with third parties, but the Settlement contains none. Indeed, the one gesture it makes towards reader privacy is actually worse than nothing: Fully Participating Libraries must "keep track of and report[] all such uses of [their Library Digital Copies] to the Registry." Settlement § 7.2(b)(vii). Approving the Settlement in its current form poses great risks to the public interest in open and uninhibited exchange of ideas. III. THE SETTLEMENT INAPPROPRIATELY ATTEMPTS TO SOLVE A LEGISLATIVE PROBLEM THROUGH A CLASS ACTION SETTLEMENT A court in a copyright lawsuit should give close attention to any proposed judgment that would compromise the core exclusive rights granted by copyright. That close attention is doubly important where, as here, the lawsuit is framed as a class action and the court is responsible not only for the public interest but the rights of absent class members. It is triply important where, as here, the proposed settlement would affect broadly-held rights in a prototypically legislative fashion. The procedural context of this lawsuit therefore requires the Court to give particularly close scrutiny to those portions of the Settlement that would affect orphan work books.

14

A.

The Settlement Raises Grave Doubts About the Suitability of a Class Action

All of these issues would be serious enough if they were present in an ordinary settlement agreement between named parties. This Court, however, is being asked to place its approval on a class-action settlement that affects the rights of millions of parties not before the Court. While class-wide settlements are an expected and encouraged part of class actions, see Fed. R. Civ. P. 23(e), they require close scrutiny and the burden is on the proponents of a settlement to show that it is fair, reasonable, and adequate. Holmes v. Continental Can Co., 706 F.2d 1144, 1147 (11th Cir. 1983). Here, multiple features of the Settlement should heighten the Court's skepticism. First, the Settlement Class contains millions of members who are especially unlikely to appear to represent themselves: owners of orphan work books. If they could appear and demonstrate their copyright interest, then by definition they would not be owners of orphan works. While this inability to represent themselves makes the class action form potentially "superior to other available methods," Fed. R. Civ. P. 23(b)(3), it creates a heightened risk that their interests will not be adequately represented in the class action. The orphan works problem creates this large subclass that is all but guaranteed not to have its voice heard in the settlement negotiations, or in opt-out requests or objections to the Settlement. Second, the Settlement waives future claims by members of the Settlement Class. "The settlement of future claims has been viewed as an area where there is a need for increased judicial scrutiny . . . ." Wright, Miller, and Kane, Federal Practice and Procedure § 1797.3. As the Supreme Court stated when rejecting certification of a class of plaintiffs exposed to asbestos, "Many persons in the exposure-only category [i.e. those with future claims] . . . may not even know of their exposure, or realize the extent of the harm they may incur. Even if they fully appreciate the significance of class notice, those without current afflictions may not have the information or foresight needed to decide, intelligently, whether to stay in or opt out." Amchem 15

Products, Inc. v. Windsor, 521 U.S. 591, 628 (1997). Here, members of the subclass of orphan work book copyright owners are, by their very nature as orphan work owners, unable to "decide, intelligently, whether to stay in or opt out." Indeed, the orphan work book copyright owners are even less able to recognize and defend their interests than the future claimants in Amchem. The Settlement does not merely compromise future claims for past conduct, as in Amchem. Instead, it would release Google from liability for its future conduct: scanning books it has not yet scanned, and selling copies it has not yet sold. It is thus particularly inappropriate for this Court to waive the future claims of the members of the orphan work book copyright owner subclass without a searching examination to ensure that their rights are adequately protected. Third, Settlement Class Counsel have not fairly and adequately represented the interests of all members of the Settlement Class. A court hearing a motion for approval of a class action settlement must determine that the proposal is "fair, reasonable, and adequate" to class members. Fed. R. Civ P. 23(e)(1)(C). Here, however, the Proposed Settlement is structured to disadvantage the orphan work book copyright owner members of the Settlement Class for the benefit of active members. Any money paid to the Registry that is unclaimed after five years-- as the money for many orphan works will be--will be reallocated away from the copyright owners whose works were responsible for it. Some of that money will be reallocated to other copyright owners who have registered with the Registry. Settlement § 6.3(a)(i), attachment C § 1.1(e). This is a direct conflict between the interests of orphan work book copyright owners and the interests of the named plaintiffs who negotiated the Settlement. Fourth, the Settlement would also, as a practical matter, be preclusive of the rights of many non-parties to the case. The Settlement sets forth, in substantial detail, the obligations libraries will be compelled to accept in exchange for releases from the settling parties.

16

Settlement art. VII, attachments B-1, B-2, B-3. It imposes specific terms on which Google will offer book purchases to individuals and subscriptions to institutions. Neither readers nor libraries, however, are parties to the lawsuit. The notice of class action settlement was not addressed to them, nor have they been invited to participate in defending their rights. As described above, the Settlement might also effectively foreclose many of Google's competitors from competing fairly with it in selling access to digital books. The need for pre-approval input from competitors and antitrust enforcers is especially important where, as here, the parties to the settlement might later argue that it creates immunity from antitrust scrutiny under the NoerrPennington doctrine. See Picker at 30­32. B. The Orphan Works Problem Is Inherently Legislative

The political branches of government, rather than the judicial branch, are best positioned to respond to the orphan works problem, which bears all the hallmarks of a traditionally legislative issue. The scale of the problem requires settling the rights and responsibilities of millions of parties. Those parties belong to many constituencies with varying interests and perspectives. The political branches have recognized these facts and Congress has been actively considering orphan works legislation. 1. Large and Diverse Constituencies

Not only are there vast numbers of orphan work books, but there are large and diverse constituencies with interests in them. The book industry is split between authors and publishers, who have distinct interests in books. Both communities are internally heterogeneous. Some authors, including many popular novelists, have primarily economic motivations; others, such as academics, write not to maximize their income but to spread their ideas as widely as possible. The publishing industry is home both a concentrated core of large publishers (such as the plaintiffs in the Publisher lawsuit) and to more than 50,000 smaller houses. See generally Albert 17

N. Greco, The Book Publishing Industry (2d ed. 2005). These authors and publishers include both United States nationals and many foreigners relying on the United States' adherence to the terms of international copyright treaties, as the international notice program recognizes. Nor are copyright owners the only ones with an interest in orphan works. Copyright policy favors making books available to the reading public, which has an obvious and substantial interest in the terms on which books are made available to it. Future authors themselves have an interest as readers in having access (both physical and legal) to the raw materials for their own creativity: "`Every book in literature, science and art, borrows, and must necessarily borrow, and use much which was well known and used before.'" Campbell v. Acuff-Rose Music, 510 U.S. 569, 575 (1994), quoting Emerson v. Davies, 8 F. Cas. 615, 619 (No 4,436) (CCD Mass. 1845). Libraries and archives have an institutional mission to preserve and provide the full catalog of books, even and perhaps especially those that have become orphans. Google and other entities that want to improve access to books need to know the terms on which those books will be available to them. Everyone connected to the copyright system--which is to say, essentially everyone--has a reason to care about the orphan works problem. 2. A Problem for the Legislative Branch

Problems that affect large segments of the public, and that affect different constituencies in complex, cross-cutting ways, are ideally suited to the deliberative legislative process. Congress possesses the institutional competence to hear from and weigh the interests of different segments of society. The political branches have the deliberative capacity and technical expertise to sort through the complicated dilemmas of the orphan works problem. They also have the procedural and electoral accountability required to make their decisions--which will inevitably better suit some constituencies than others--democratically legitimate.

18

Recognizing their institutional competence and responding to the urgency of the orphan works problem, the political branches have been crafting solutions to it. In 2005, Members of Congress requested that the Copyright Office examine the orphan works problem. See Letter from Orrin G. Hatch and Patrick Leahy, United States Senators, to Marybeth Peters, Register of Copyrights, Jan. 5, 2005; Letter from Lamar Smith, United States Representative, to Marybeth Peters, Register of Copyrights, Jan. 7, 2005; Letter from Howard Berman, United States Representative, to Marybeth Peters, Register of Copyrights, Jan. 10, 2005. The Copyright Office conducted a year-long study. It received 721 written comments and 146 reply comments, held three public roundtables and seventeen informal meetings, and produced a 127-page report that included proposed legislative reforms. Corresponding legislation was introduced in Congress, see Orphan Works Act of 2008, H.R. 5889 110th Cong. (2008); Shawn Bentley Orphan Works Act of 2008, S. 2913 110th Cong. (2008), and passed the Senate. The orphan works problem is the subject of active attention from the political branches best equipped to respond to it. Significantly, the Settlement goes well beyond the scope of the orphan works legislation considered by Congress in the previous Session. Both the Senate and House versions would have required a "diligent effort" to search for the copyright owner. Under the Settlement, however, Google and the Registry are not required to make any individual or ongoing efforts to locate copyright owners. Congress's hesitation to enact even a much more limited version of orphan works legislation should counsel this Court to proceed with caution. It may be that the Settlement's deal of public access in exchange for escrowed revenue is a fair and broadly applicable solution to the orphan works problem. If so, Congress is in a better institutional position than this Court to consider the risks and tradeoffs such a sweeping revision of the Copyright Act would entail.

19

IV.

THIS COURT SHOULD SOLICIT ADDITIONAL VIEWS BEFORE RULING ON THE MOTION TO APPROVE THE SETTLEMENT These concerns suggest that the Settlement as it currently stands may be harmful to the

public interest and contrary to law. Still, the Settlement offers significant enough public benefits in the form of increased access to orphan works that this Court should not reject it out of hand. Instead, the Court should seek counsel some of the many groups who are not presently before the Court but who will be affected by the Settlement or have special expertise in the issues it raises. By proceeding with caution, this Court may be able to craft a response that modifies the Settlement sufficiently to save it. A. The Procedural Posture of the Case Has Deprived this Court of Essential Perspectives

A court is required to consider the effect of a proposed settlement on the public interest. Bailey v. Great Lakes Canning, 908 F.2d 38, 42 (6th Cir. 1990). In addition to ensuring that a proffered settlement is not itself contrary to law, a court must look to its effects on interested non-parties and on the public at large. See, e.g. Durrett v. Housing Authority of Providence, 896 F.2d 600, 604 (1st Cir. 1990). Here, where the Settlement would reshape the publishing industry, affect the rights of millions of copyright owners, and affect the terms on which libraries and readers would obtain access to books, those effects cannot be analyzed based solely on the limited perspectives presented to the Court by the Plaintiffs and Defendant. Moreover, because counsel for Plaintiffs did not seek class certification before opening settlement negotiations, the Court has never previously had an opportunity to consider the interests of class members. Instead of allowing this Court to consider the appropriateness of the proposed class in a properly adversarial setting, the motion for approval of the Settlement provides this Court only with the Plaintiffs and Defendant's joint argument in favor of approval. 20

Here, a substantial portion of the Settlement Class consists of the owners of copyright in orphan works. By definition, they are not capable of representing their own interests in litigation and settlement; the same reasons it is impossible to find them to ask permission to use their works also make it impossible to find them to solicit their approval of a settlement. This Court, therefore, cannot rely on the normal objection process to provide it with a full understanding of the interests of the Settlement Class. B. This Court Should Solicit the Views of the Copyright Office, the Department of Justice, and the Federal Trade Commission

The first source of additional counsel this Court should seek is those agencies of the federal government with special expertise over the subject matter of the lawsuit: the Copyright Office, the Antitrust Division of the Department of Justice, and the Federal Trade Commission. The Copyright Office of the Library of Congress is the arm of the federal government with unique knowledge about copyright law and policy. See, e.g., Cablevision Systems Dev. Co. v. Motion Picture Ass'n of Am., 836 F.2d 599, 608 (D.C. Cir. 1988) ("The Copyright Office certainly has greater expertise in such matters than do the federal courts . . . ."). Its Report on Orphan Works gives it a particularly informed perspective on the orphan works problem. The Antitrust Division of the Department of Justice has authority to prosecute violations of the antitrust laws, and has already notified this Court that it is investigating this matter. The Federal Trade Commission has authority both to prevent anticompetitive mergers and to protect consumers from unfair terms in product licenses, both concerns that are present here. The fact that the orphan works problem has a legislative character legislative is not an absolute bar to this Court's consideration of the issues raised by the Settlement. It does, however, imply that this Court, in the prudential exercise of its authority, should work carefully to avoid frustrating Congress's goals in the Copyright Act and the Executive and Legislative 21

Branch's carefully determined policies for enforcement of the antitrust laws. Due respect for these coordinate branches of government counsels considering their views before the Court takes on the characteristically legislative task of determining rights in relation to orphan work books. C. This Court Should Appoint Counsel to Represent the Subclass of Orphan Work Book Copyright Owners

This Court should also ensure that the Settlement serves the public interest by appointing separate counsel to represent the subclass of orphan work book copyright owners. Current Class Counsel are unable to fairly and adequately represent the interests of that subclass. There is a fundamental and irreconcilable conflict between the interests of present copyright owners and the interests of absent ones. The provisions on unclaimed funds amply illustrate how those copyright owners capable of actively managing their copyrights profit at the expense of orphan work book copyright owners. The conflict is exacerbated because the Settlement purports to waive future claims on behalf of orphan work subclass members. The Supreme Court called it "obvious" that "a class divided between holders of present and future claims (some of the latter involving no physical injury and attributable to claimants not yet born) requires division into homogeneous subclasses under Rule 23(c)(4)(B), with separate representation to eliminate conflicting interests of counsel." Ortiz v. Fibreboard Corp., 527 U.S. 815, 856 (1999). Separate counsel, charged specifically with representing the interests of orphan works subclass members, would ensure this Court that the settlement does not misuse orphan works for the private gain of Google and active copyright owners. Appointing separate counsel would also reveal how the interests of orphan works owners diverge from the simplistic model presented to this Court by the parties. Class members should not be deemed to wish to settle on terms that would constitute a violation of antitrust law. Nor should their consent to a settlement that could place them at the mercy of a Google monopoly 22

over digital scanned book distribution be presumed. They also have interests that go beyond simple economic maximization of their revenues. All authors have expressive and reputational interests in being read that may justify foregoing some possible revenues (this motivation is particularly strong with academic authors, for example). All authors are also readers and will benefit from increased access to books; it may be that the settlement should reflect that interest with a greater emphasis on access and a decreased emphasis on revenue. Such questions cannot be effectively answered without someone to speak for the orphans; the Court should appoint counsel to ensure that someone will. D. This Court Should Scrutinize Closely Provisions of the Settlement Agreement That Could Give Google and the Named Plaintiffs Open-Ended Powers to Alter Its Terms

Several terms of the Settlement are susceptible to an interpretation that would give Google and the named plaintiffs post-approval powers to effectively rewrite it to their advantage. This Court should require explanations of those terms from the settling parties, require the parties to disclaim possible interpretations that would give the settling parties open-ended powers to modify the settlement unilaterally, and enforce those disclaimers as part of its retained jurisdiction over the implementation of the settlement. Three features of the Settlement, in particular, require close scrutiny: First, the Settlement contains a secret termination clause: Google, the Author Sub-Class, and the Publisher Sub-Class each will have the right but not the obligation to terminate this Settlement Agreement if the withdrawal conditions set forth in the Supplemental Agreement Regarding Right to Terminate between Plaintiffs and Google have been met. Any decision by Google, the Author Sub-Class or the Publisher Sub-Class to terminate this Settlement Agreement pursuant to this Article XVI (Right to Terminate Agreement) will be in accordance with the procedures set forth in the Supplemental Agreement Regarding Right to Terminate. The Supplemental Agreement Regarding Right to Terminate is confidential between Plaintiffs and Google, and will not be filed with the Court except as provided therein. 23

Settlement Art. XVI (emphasis added). This secret side agreement gives Google and the named plaintiffs the right to revoke the Settlement under undisclosed conditions and using undisclosed procedures. There is no guarantee that those secret conditions do not include terms unfairly favoring the named plaintiffs or other terms not in the interests of class members or the public. It is unfair for the settling parties to ask class members to bind themselves to a settlement agreement if they themselves will not themselves be bound. Class members cannot make an informed decision to stay in or opt out of the class, and this Court cannot make an informed evaluation of whether the Settlement is "fair, reasonable, and adequate," unless the contents of the Supplemental Agreement Regarding Right to Terminate are disclosed. Second, the Settlement's provisions allowing copyright owners to remove themselves from Display Uses and Revenue Models are de facto unavailable to orphan work book copyright owners. Thus, if the terms of the Settlement unduly favor Google, active copyright owners may opt-out, Remove, or Exclude their books en masse, leaving only orphan owners in the new, unfair programs. Other settlements have dealt with this type of danger by including opt-out thresholds, "in which a defendant conditions its agreement on a limit on the number or value of opt outs." Manual for Complex Litigation (Fourth) § 21.631. The Settlement may require thresholds for both de jure opt-outs, in which copyright owners remove themselves from the Settlement Class, and de facto opt-outs, in which copyright owners Remove or Exclude their books while remaining in the settlement class. Both forms of opt-out pose substantially the same risks of prejudicially disparate treatment to orphan work book copyright owners. Third, as noted above, the definition of the Settlement Controlled Price conceals dangerous ambiguities. This Court should insist that the Settling Parties stipulate that this clause is intended to and will be interpreted consistently with the requirements of class-action and

24

antitrust law. They should agree that "optimal such price for each Book" (emphasis added) requires Google to maximize revenue for books individually, mimicking the behavior of sellers in a competitive market. Conclusion For the reasons given, the Court should solicit the views of the United States Department of Justice, the Federal Trade Commission, and the Copyright Office on the Settlement; it should appoint counsel to represent the subclass of orphan work book copyright owners; and it should require the Settling Parties to disclaim interpretations of the Settlement that could be prejudicial to the interests of absent class members and the public. The Settlement offers great benefits and poses substantial risks. This Court has an obligation to separate the two. Dated: New York, New York September 3, 2009 Respectfully submitted, By:_/s/ James Grimmelmann____ James Grimmelmann (admitted pro hac vice) [e-mail] 57 Worth St. New York, NY 10013 [phone] - and KORNSTEIN VEISZ WEXLER & POLLARD, LLP Daniel J. Kornstein (DK-3264) [e-mail] Mikaela A. McDermott (MM-6833) [e-mail]

25

757 Third Avenue, 18th Floor New York, NY 10017 [phone] Attorneys for Amicus Curiae, Institute for Information Law and Policy, New York Law School

26