Also available in PDF format.
Testimony of Gigi B. Sohn
President, Public Knowledge
Before the
U.S. House of Representatives
Committee on the Judiciary
Subcommittee on Courts, the Internet, and Intellectual Property
Hearing on:
H.R. 4279, the Prioritizing Resources and Organization for Intellectual Property Act of 2007
December 13, 2007
Chairman Berman, Ranking Member Coble, and members of the Subcommittee, my name is Gigi B. Sohn. I am the president of Public Knowledge, a nonprofit public interest organization dedicated to defending user rights in the emerging digital culture. I want to thank you for inviting me to speak today on H.R. 4279, the Prioritizing Resources and Organization for Intellectual Property Act of 2007, and the opportunity to discuss the need for reform in copyright law.1
Introduction and Summary
While we agree that enforcement of intellectual property laws is essential to encouraging creativity, certain provisions in the proposed Act risk undermining this essential goal by threatening ordinary consumers with an overbroad and inapposite enforcement regime.
The Act has undergone many changes over the course of the last few months, and I want to thank the Subcommittee for working hard to maintain open channels of communication as the bill has evolved through its various drafts. While this Act is a lengthy and wide-ranging piece of legislation, I would like to focus today on three particular provisions that remain problematic to Public Knowledge and to the many companies, trade associations, and public interest groups with which we work.
First, the bill would disaggregate the parts of a compilation or derivative work for the purpose of calculating damages, multiplying severalfold the already-onerous statutory damages associated with copyright infringement. Eliminating the current method of calculating these damages worsens the already evident problem of disproportionate penalties for infringement. Such a provision stands in stark contrast to the House’s work to reduce the ill effects of excessive damages in patent law.
Second, the bill significantly expands the forfeiture provisions attached to four different types of IP violations, applying the exact same sets of standards to each. This expansion risks even further upending rational copyright remedies and ignores the significant differences in the various subject matters of copyright, trademark, and anti- bootlegging laws.
Third, the bill eliminates the requirement that copyrights be registered before criminal copyright enforcement proceeds. Copyright registration serves an essential purpose in giving the public notice of a work’s copyright status and the proper holder of copyright. Without a vibrant copyright registry, subsequent users of a work are often hard-pressed to locate the original copyright owner to obtain permission to use the work, leading to “orphan works” that can no longer be exhibited, reproduced, or seen. Reducing the incentives for creators and authors to register their works can only exacerbate this problem.
In summary, these provisions of the bill merely increase penalties and remove safeguards against disproportionate awards. They represent a step away from a rational, realistic copyright regime—one that can allow a law last overhauled before the invention of the VCR to adapt itself to a world of Tivo and YouTube. It’s no secret that the law needs to better reflect technological and social reality. Otherwise, we subject millions of people to the threat of unrealistically harsh penalties. These issues ought to be addressed more directly, before we simply raise those penalties again.
Disaggregation of Damages
Section 104 of the bill proposes to significantly increase damages by allowing each part of a compilation or a derivative work to be counted as a separate work for the purposes of infringement. This is precisely what current law prohibits. The current language states that for statutory damages purposes, “all the parts of a compilation or derivative work constitute one work.”2
The major change proposed by the bill would greatly escalate statutory damages beyond their currently bloated state. Though statutory damages are necessary to allow for action when actual damages are insufficient deterrence or difficult to measure, current levels, where non-willful infringers must pay a minimum of $750, and may pay up to $30,000 per violation, and willful infringers up to $150,000, are already stretching the bounds of reason.
A forthcoming article by Utah law professor John Tehranian discusses, among other things, the disproportionate nature of damages. Cataloguing the ordinary activities of a hypothetical person—forwarding emails, passing out news articles, reciting a poem, singing “Happy Birthday”—Tehranian finds that these mundane acts of a single day can subject his imaginary person to $12.45 million in damages,3 all without a single act of P2P file sharing or other commonly recognized “bad acts.”
In another example, much recent news coverage has been devoted to the case of Jammie Thomas, a single mother in Minnesota who, found liable for sharing 24 songs on a P2P network, was fined $222,000, or $9,250 per song.4 The fact that this award is far below the maximum penalty for willful infringement, or even innocent infringement, is less a sign of clemency on the jury’s part, and more of a clear indication that statutory damages exceed any value rationally tied either to the actual injury or any effective deterrent value. Instead, such penalties merely distort the ongoing copyright debate by forcing faster and larger settlements, rather than bringing important legal questions before the courts.
The bill heightens these problems by allowing, for instance, an infringed album of ten songs to multiply tenfold the damages awarded against the infringer. A copy of a scholarly journal, if found to be infringing, could be counted several times over, with a new $150,000 penalty for each article. A series of 24 photographs reprinted from the same collection would suddenly risk a multi-million dollar penalty. In the Internet context, a highly litigious website or magazine owner could assert a separate infringement for every separate photograph in a magazine, or every separate image and block of text on a website.
Increasing damages through disaggregation will also have a major chilling effect upon legitimate uses of copyrighted works, while providing little additional deterrent effect on willful commercial pirates. Recently, twenty five different intellectual property law scholars have expressed their concern over this particular provision, noting, for example, that a documentary filmmaker’s potential liability for using period music from a single album could increase from $150,000 to $450,000, or that a poetry reviewer excerpting from different poems in a published book could face an increase in exposure from $150,000 to $750,000.5 Whether or not such values might be awarded by a court, the fear of such damages will drive creators away from making fair uses.
The dangers of aggregating statutory damages in this way were anticipated in the language of the existing Copyright Act, which explicitly prevents such a calculation. This decision was deliberately made during discussion of what was to become the 1976 Act to prevent awarding multiple instances of damages for a single act of infringement, such as when an infringement of a work might infringe both the first edition and the current one.6 The Copyright Office, in its comments on the legislation, continually noted that compilations and derivative works should be counted as one work for the purposes of calculating damages.7
The Second Circuit has recognized the general danger of multiplied statutory damages as well. In a 2003 decision, the court noted that aggregating large numbers of statutory damages risks distorting the purpose of statutory damages, instead “creat[ing] a potentially enormous aggregate recovery for plaintiffs, and thus an in terrorem effect on defendants, which may induce unfair settlements.”8 The court went on to note that these multipliers risk running afoul of constitutional due process.9
Regardless of the constitutional limits on damages, and the courts’ deference to Congress on these matters, Congress should endeavor to ensure that statutory damages are reasonable. The House, and this Subcommittee, have already recognized that multiplying intellectual property damages risks warping the risks and incentives of litigation. For instance, H.R. 1908, passed by the House this September, calls on the courts to apportion patent damages relative to the actual harm done.10 In the patent context, the House has recognized that multiplied damages can threaten innovation by encouraging more frivolous claims and distorted settlements. The disaggregation proposal in the bill takes a step backward in the copyright realm even as patent law takes a step toward sanity.
Proponents of the disaggregation provision might point to judicial discretion as a safeguard against such ruinous damages. However, judges can only exercise their discretion in those cases that are resolved in their courts. The majority of copyright cases will settle under the cloud of expanded damages long before a judge will have any say in the rationality of the award, or the negotiations leading up to it.
Of the many problems present in copyright law and its damages provisions, the accounting of works in a compilation is not the most pressing. If anything, counting these works as separate is a step away from a rational copyright damages policy.
Forfeiture Provisions
The expansion of forfeiture provisions in the bill also risks creating disproportionate penalties. In these cases, the bill casts too broad a net, offering up for forfeiture materials and devices that may have only a fleeting connection to the offense. The proposed changes also remove the important safeguard of judicial discretion for copyright and audio bootlegging offenses. Furthermore, adding civil forfeiture provisions increases the risk that innocent parties will forfeit their property.
The bill applies essentially the same forfeiture language for four different areas of intellectual property, whether it be counterfeit goods, copyright infringement, bootlegging live music performances, or recording in a movie theater. The criminal provisions require forfeiture of not only the infringing goods, but “any property used, or intended to be used, to commit or substantially facilitate the commission of an offense.” In contrast, existing law casts a more narrowly tailored net. Not only do the existing statutes more specifically target the infringing materials and devices, the current laws also tailor the scope of the forfeiture provisions more to the nature of the offenses.
The bill characterizes its changes to the law as a “harmonization” of the various realms of IP law. However, as Professor Yochai Benkler has noted, harmony isn’t created when everyone is singing the same note.11 By the same token, harmonization need not require a one-size-fits-all approach. This becomes apparent when these forfeiture provisions are applied to certain types of activities. For instance, requiring the forfeiture of devices used in infringement makes more sense in commercial counterfeiting cases, where removing expensive, dedicated manufacturing equipment is a proportional punishment for large-scale infringements and an effective way of preventing further violations. This is less true when a family’s general purpose personal computer is used to illegally download music. The general purpose nature of the machine makes its forfeiture a much less appropriate penalty. Should a teenager’s illegal downloads deprive him or his siblings of Internet access or word processing? Furthermore, removing the computer fails to create a further deterrent to a determined infringer. On top of the already-astronomical monetary damages, a thousand dollars more will not affect a rational cost-benefit calculation.
Under current law, a court has the discretion to order the forfeiture of devices merely intended to be used in copyright infringement.12 The proposed bill removes this discretion, mandating a wider range of property be forfeited. A second home computer in an infringer’s house should not be marked for forfeiture based upon the fact that it contains a CD burner.
The expanded forfeiture provisions also remove judicial discretion from the audio bootlegging statute. Currently, 18 U.S.C. 2319A(b) specifically draws a line between the infringing goods, such as the tapes themselves, and the equipment used to make or reproduce them. While the actually infringing articles must be forfeited, the statute explicitly requires a court to consider the “nature, scope, and proportionality” of the equipment’s use before deciding whether to order its forfeiture. In contrast, the proposed bill eliminates this crucial distinction, and thus a safeguard sorely needed in other areas of IP enforcement. While, as I’ve noted above, judicial discretion is not a cure-all for potentially onerous penalties, denying judges the ability to make fact-specific determinations helps no one.
Although these changes are being made in the name of harmonization with the counterfeiting statute, the current counterfeiting provisions were put in place just two years ago, when the justification was the need for harmonization with copyright law.13 The proposed bill also creates a new class of civil forfeitures across these four different areas of law, again requiring the forfeiture of any property “used, or intended to be used, to commit or facilitate” an offense. Creating extensive civil forfeiture provisions runs the risk of unintended consequences, given the lower burden of proof for civil forfeiture. There is a long history of civil forfeiture being over-applied in other contexts, and this history ought to be examined thoroughly before extending it to a completely new realm.14
I do want to comment on one major improvement that the bill has over its companion bill in the Senate, S. 2317. The Senate bill furthers the unwarranted expansion of the forfeiture provisions by requiring the forfeiture of equipment used in circumventing copy protection mechanisms, increasing the penalties for violations of a law with a long and storied history of overbreadth and misapplication. I thank the Subcommittee for its foresight in removing that provision as being disproportionate and inapposite, and similarly, I look forward to the other forfeiture provisions being more narrowly tailored to fit the offenses they seek to remedy.
Registration Requirements and Orphan Works
Section 102 of the proposed Act alters existing law to imply that a copyrighted work need not be registered before the government pursues criminal enforcement. Not only is this provision unnecessary, it further erodes the incentives for copyright holders to register their works, adding to the problem of orphan works.
Orphan works are works whose copyright owners cannot be found. This means that permission cannot be granted or even asked for by subsequent users of the works. Lacking a known owner, these works are locked away from public distribution, display, or further reproduction, and therefore lost to the public. Any who attempt to distribute the works without permission risk massive liability if the owner emerges later, with damages of up to $150,000 per infringement. These considerations have prevented families from reprinting heirloom photographs because the professional photo finisher cannot identify the original photographer.15 Libraries, archives, and museums are left unable to collect and display works with unknown authors.16 Software developers are left unable to improve upon copyrighted programs because a dissolved company has left no clear indication as to the ownership of any copyright in the program.17
In its 2006 report on orphan works, the Copyright Office noted that one of the contributors to this body of unusable works was the fact that, after the 1976 Copyright Act, works no longer needed to be registered with the Copyright Office to receive copyright protection.18 Registration of copyrights helped to prevent this buildup of valuable yet unusable works by providing a central resource by which subsequent users can find copyright owners. Now-defunct renewal provisions also ensured that copyright owners maintained a point of contact for permissions requests after a certain period of time. These requirements reflected part of the copyright bargain—that in exchange for a government-granted monopoly rights over a creative work, the author also makes himself or a representative available to review requests for licenses.
But in the absence of registration and renewal requirements, some incentive must be present for copyright owners to register their works and maintain a point of contact. Current law accounts for this by allowing civil or criminal enforcement only after a work has been registered, though registration is allowed after the allegedly infringing conduct has occurred.19 This feature of our copyright law helps to maintain the vitality of the copyright registry and stem the tide of orphan works. In opposition to this vital mission, and with insufficient justification, the proposed bill eliminates the need for a work to be registered before criminal enforcement can proceed.
It’s unclear what common harm the provision is meant to avert. The vast majority of pirated works are commercially produced and duly registered, precisely because their producers are aware that they might be infringed. If too few works are being registered, removing incentives for registration is not the answer.
Instead, maintaining the registration system as is helps align the interests of copyright owners in protecting their works with the public’s interest in being able to seek those authors’ permission to use their works. By doing so, we can help to prevent valuable follow-on users from having to choose between letting a work slip into obscurity and facing millions of dollars in damages for reprinting a copy of a dozen orphaned works.
Conclusion
I want to end today by calling attention to an assumption that seems to underlie all too many efforts to improve IP enforcement. The assumption is that ever-higher penalties will somehow create a deterrent effect currently absent in today’s laws. However, of all the changes that need to be made to IP law, increasing the severity of the penalties is one of the least necessary, and quite possibly the most counterproductive.
We should be clear that several problems confront IP law, and specifically copyright law, and these several problems have distinct causes and distinct solutions. It doesn’t help to combat piracy if in doing so both commercial pirates and ordinary home consumers are subjected to the same harsh penalties. When the mere act of forwarding your email or updating your blog can infringe copyright, it makes more sense to have the law comport with reality before increasing the sanctions that accompany infringement.
While a complete review and overhaul of copyright law might be an aspirational ideal, a set of more modest reforms can provide some relief for the immediate future. At a talk at Boston University this past October, I proposed six steps that would help to narrow the gap between copyright law and norms.20
First, fair use needs to be preserved in the digital age. While the current statute explicitly recognizes some of the most storied types of fair use—teaching, research, commentary, criticism, and news reporting—newer uses, consistently found to be fair by the courts, are still facing challenges by the litigious. Thus, incidental, transformative, and non-commercial personal uses should be added to this non-exhaustive list. Furthermore, the anti-circumvention provisions of the DMCA should not be used as an end-run around fair use. Circumventing technological protection measures for lawful uses of the work must be allowed if fair use is to remain relevant for digital media. H.R. 1201, introduced earlier this year by Representatives Boucher, Doolittle, and Lofgren, takes an important step towards this goal.
Second, limits on secondary liability should be dealt with statutorily. The standard set out in 1984 by the Supreme Court in Sony should be codified—technologies should be protected from any secondary liability if they are capable of substantial non-infringing uses. In addition, secondary liability should be based upon actual damages suffered by the copyright holder, not upon the inflated statutory damages discussed today.
Third, there must be protections against copyright abuse. There should be clearer penalties when copyright holders recklessly send DMCA takedown notices. There should be an affirmative cause of action where plaintiffs abuse copyright as a means to stifle free speech. And overstating copyright holders’ rights should be considered unfair and deceptive trade practices by the Federal Trade Commission.21
Fourth, the process of licensing works, especially music, must be made rational. Congress should also revisit its earlier attempts to simplify clearing the composition right, and should also resolve the problem created by the Copyright Royalty Board’s recent decision to massively increase the royalty rates paid by Internet radio. The extreme disparity between high payments by Internet radio, and no payments at all by broadcast radio, must also be resolved. Likewise, the line between performances and distributions should be made clear. Simply adding a recording function to a radio, whether digital or analog, should not require an additional distribution license from the copyright holder.
Fifth, the problem of orphan works must be addressed. The Copyright Office has proposed that a user unable to find a copyright owner after a reasonable search should be liable only for “reasonable compensation,” and not immense statutory damages.22 One implementation of this proposal, H.R. 5439, was introduced in the last Congress by Representative Smith. The thoroughness of this reasonable search can be improved by creating competitive visual registries, which not only would help users make use of orphaned works, but also help to reunite those orphans with their creators.23
Sixth, technical and legal restrictions on the use of copyrighted works must be clearly and plainly disclosed to consumers. If consumers are to receive the products they pay for, and the rights to use those goods they normally expect, any use of technology to remove some of those uses should be communicated to them before a purchase is made. It should also be clear to consumers if altering digital locks on the products they buy will subject them to legal liability. As Professor Pamela Samuelson of U.C. Berkeley’s Boalt Hall Law School has noted, concealing these limitations on the media that consumers buy is an unfair and deceptive trade practice, and should be recognized as such by the FTC.
Each of these proposals directly addresses a situation where an ordinary consumer might face the already-draconian sanctions of copyright law. If the problems that create the disconnect between the law and the reality of copyright use aren’t tackled first, increasing the severity of those sanctions further does very little good.
Thank you for inviting me to testify. I look forward to answering your questions.
1 I would like to thank Public Knowledge Staff Attorney Sherwin Siy for his assistance in preparing this testimony.
2 17 U.S.C. 504(c)(1).
3 John Tehranian, Infringement Nation: Copyright Reform and the Law/Norm Gap 2007 UTAH L. REV. (forthcoming 2007) (online draft at 543-46), available at http://www.turnergreen.com/publications/Tehranian_Infringement_Nation.pdf.
4 Jeff Leeds, Labels Win Suit Against Song Sharer, N.Y.TIMES, Oct. 5, 2007, http://www.nytimes.com/2007/10/05/business/media/05music.html.
5 Letter from 25 Intellectual Property Professors to the Honorable John Conyers, Jr., Chairman, Committee on the Judiciary, U.S. House of Representatives (Nov. 13, 2007) available at http://www.publicknowledge.org/pdf/professors-letter-20071113.pdf.
6 See WILLIAM PATRY, PATRY ON COPYRIGHT § 22:156 (2007) (citing Copyright Law Revision Part 4: Further Discussions and Comments of the Preliminary Draft for Revised U.S. Copyright Law 139-140.).
7 PATRY ON COPYRIGHT § 22:157 (citing Copyright Law Revision Part 5: 1964 Revision Bill with Discussion and Comments 203-204.).
8 Parker v. Time Warner Entertainment Co., 331 F.3d 13, 22 (2d. Cir. 2003).
9 Parker, 331 F.3d at 22. See also Fitzgerald Pub. Co. v. Baylor Pub. Co., 670 F. Supp. 1133, 1140 (E.D.N.Y. 1987) (“[Copyright statutory] damages should bear some relation to the actual damages suffered”). The Supreme Court has recognized that statutory damages can violate constitutional due process when they are “so severe and oppressive as to be wholly disproportioned to the offense.” Zomba Enters. v. Panorama Records, 491 F.3d 574 (6th Cir. 2007) (paraphrasing St. Louis, I.M. & S. Ry. Co. v. Williams, 251 U.S. 63, 66-67 (1919).). In the related area of punitive damages, the Court has held that damages greater than four times the amount of harm done were constitutionally suspect, and that few awards of more than ten times the amount of damage could satisfy due process. .” BMW v. Gore, 517 U.S. 559, 581 (1996) (quoting Pacific Mut. Life Ins. Co. v. Haslip, 499 U.S. 1 (1991); State Farm Mut.Auto Ins.Co. v. Campbell, 538 U.S. 408, 425 (2003).
10 Patent Reform Act of 2007, H.R. 1908, 110th Cong. § 5(a) (2007).
11 Yochai Benkler, Professor, Harvard Law School, Welcome Address to the Second Access to Knowledge Conference (Apr. 27, 2007).
12 17 U.S.C. § 509.
13 See H.R. Rep. No. 109-68 7 (2005) (“This section would amend current law to require the forfeiture of any property derived, directly or indirectly, from the proceeds of the violation as well as any property used, or intended to be used in relation to the offense. This is intended to provide forfeiture and destruction provisions similar to those already enjoyed by copyright and trade secret holders.”).
14 See Douglas O. Linder, Evil in the American Justice System: Case 2: Zero Tolerance and Asset Forfeiture, http://www.law.umkc.edu/faculty/projects/ftrials/evil/evilP14.html (noting forfeiture seizures in the drug context that included a fishing boat confiscated when a crew member was, unbeknownst to the owner, in possession of 1.7 grams of marijuana).
15 United States Copyright Office, Report on Orphan Works 24-25, January 2006, at http://www.copyright.gov/orphan/orphan-report-full.pdf.
16 Id.
17 Id. at 28-29.
18 United States Copyright Office, Report on Orphan Works 41-43, January 2006, at http://www.copyright.gov/orphan/orphan-report-full.pdf.
19 Contrary to statements from the Department of Justice, the change proposed by the bill is not a mere clarification. The language of 17 U.S.C. § 411 is crystal clear: “no action for infringement of the copyright in any United States work shall be instituted without preregistration or registration of the copyright claim.”
20 Gigi B. Sohn, Six Steps to Copyright Sanity: Reforming a Pre-VCR Law for a YouTube World, Presented to the New Media and Marketplace of Ideas Conference, Boston University College of Communication, (Oct. 26, 2007) available at http://www.publicknowledge.org/node/1244.
21 Though it has declined to act on a complaint on this issue so far, the FTC has just this month recognized the importance of addressing consumer rights to content in the near future: “Widespread use of inaccurate copyright warnings could contribute to consumers’ misunderstanding of the statutory protections available to them under the Copyright Act. Further, if consumers routinely confront exaggerated and inaccurate copyright warnings they may tend to disregard them altogether…” Letter from Mary K. Engle, Associate Director for Advertising Practices, Federal Trade Commission, to Edward J. Black and Matthew Schruers, Computer and Communications Industry Association (Dec. 6, 2007).
22 MARYBETH PETERS, REPORT ON ORPHAN WORKS: A REPORT OF THE REGISTER OF COPYRIGHTS (2006), available at http://www.copyright.gov/orphan/orphan-report-full.pdf.
23 See Public Knowledge, Competitive Visual Registries for Copyrighted Works, http://www.publicknowledge.org/pdf/pk-visualregistry-memo-20070129.pdf.
