Lessig Testimony . . .

Stock Asylum Staff Report
March 29, 2006

(Editors note: Following is written testimony presented to the U.S House of Representatives' Judiciary Committee, Subcommittee on Courts, the Internet, and Intellectual Property about proposed orphan works legislation by the Lawrence Lessig. Footnotes have been removed.)

March 6, 2006

The Honorable Zoe Lofgren
102 Cannon House Office Building
Washington, D.C. 20515

Dear Congresswoman Lofgren:

As you’ve requested, I am happy to provide my views about the Copyright Office’s recent “Orphan Works Report” (“Report”). I recognize that this issue has been especially important to you. In large part, it has been your leadership that has helped to bring this issue to the fore. Your bill, the Public Domain Enhancement Act of 2003, was a direct response to an Internet petition that collected over 22,000 signatures, asking Congress to address the problem of orphan works. That bill then led Senators Leahy and Hatch to ask the Copyright Office to conduct the study that resulted in the “Orphan Works” report. I can therefore well understand your concern that the solutions proposed by the Copyright Office adequately address the concerns that originally motivated you.

In my view, the Copyright Office has done an excellent job validating the concerns that originally led you to act. Through an extraordinarily open set of proceedings, the Office gathered a wide range of views that addressed the costs of orphan works. The report does an excellent job in summarizing and analyzing those views. It embodies an integrity and balance that is essential for informed policy making. It will provide an important founda-tion for Congress’ work as it addresses this issue.

The most significant contribution of the Copyright Office’s Report is its implicit recognition that copyright owners have a responsibility to help make the copyright system function more efficiently. As I will describe in more detail below, the essence of its proposal is that unless a copyright owner is accessible after a “reasonably diligent search,” the remedies available to that copyright owner will be curtailed. This rule means that the copy-right owner bears some burden (the burden of maintaining accessibility) as a condition of getting the full benefits of copyright law’s protection.

This, in my view, is an extraordinarily important principle now plainly acknowledged by the Copyright Office. Every property system places some burden on the property owner to help assure that the property system functions efficiently. For some time now, some had come to view copyright law as an exception to this general principle. Building on Article 5(2) of Berne, which states that “[t]he enjoyment and exercise of these rights shall not be subject to any formality,” some had argued that Berne-obligated gov-ernments could take no steps that would condition full protection of a copyright upon the copyright owner taking affirmative steps to maintain the vitality of his right. The Copyright Office’s Report now plainly rejects that extreme view.

By recognizing that accessibility is an important value within the copyright system, and by indirectly placing upon the author some of the burden to maintain accessibility, the Report affirms that copyright owners have a role to play in making the copyright system func-tion efficiently. In my view, this is the most significant contribution of the Copyright Office’s Report. Everything else is detail.

Nonetheless, these details are also important. If we recognize the copyright owner has a role to play in maintaining accessibility, then the aim of the law should be to achieve that end (accessibility) with a minimum burden on both the copyright owner and a potential re-user. My concern with the Report is that it fails to achieve that end. Its rule is both too burdensome to the copyright owner, and too burdensome to the re-user. Or put differently, the reforms the Copyright Office proposes both go too far, and not far enough. In this brief letter, I will address both concerns.

The Recommendations Go Too Far

The Copyright Office recommends that the Copyright Act be amended to limit the remedies available to a copyright owner when the infringer:

Significantly, this limitation on remedies does not apply to old works only. It applies to all copyrighted works from the beginning of the term of copyright. The proposal thus rep-resents a fairly radical shift in the rules protecting the property right that copyright is. A “reasonably diligent search” privileges the use of a copyrighted work, even though the ordi-nary rule governing property requires permission from the property owner. The proposal thus places a burden on the copyright owner to maintain accessibility from the start, or lose full protection of his copyright.

This is a significant difference from your bill, the Public Domain Enhancement Act (“PDEA”). The PDEA targeted older works. The concern you addressed was not the gen-eral problem of securing permission to use a copyrighted work, but the particular problem of securing permission to use works whose owners were presumptively hard to find. The requirement that an owner register the work 50 years after it had been published was justi-fied because of the real difficulty in identifying the owners of old copyrights. Your registra-tion requirement would not have changed the obligations of current copyright owners at all.

In my view, your more targeted, and conservative, response makes much more sense in this context—at least when the trigger for a privileged use of a copyrighted work (“reasona-bly diligent search”) is so vaguely defined. The Copyright Office celebrates the “flexibility” of its ad hoc approach. But “flexibility” means ex ante uncertainty, and uncertainty will im-pose costs both upon the users of orphaned works (What is a reasonable search?) and upon creators (How do I assure I can be found?). From the moment a work is created, a copyright owner will thus need to assure that others can locate him with a “reasonably diligent search.” But as the proposal reaches foreign work, and unpublished work, it is not clear how such authors could easily protect their legitimate copyright interests.

This sort of burden is one of the initial justifications for the Berne Convention’s rules against formalities. When the 1908 Berlin Convention added the requirement that “[t]he enjoyment and exercise of these rights shall not be subject to any formality,” one of the main concerns was the burdens such formalities would place upon foreign authors. It is ob-viously unreasonable to require an author, as a condition of securing copyright, to formally register his work in any number of foreign jurisdictions. Those same concerns should raise doubts about the generality of the Copyright Office’s proposal: uncertainty about how rights are to be secured is an even greater burden on the proper protection of a copyrighted work than is a formality, since it is clear how to respond to the latter, but unclear how to remedy the former.

I am also concerned about the relatively indiscriminate nature of the Report’s proposal—both its failure to distinguish among kinds of copyrighted works, and its failure to distinguish works published before 1978, and works fixed from 1978 on.

The burdens of the Report’s proposal are perfectly justifiable for works published before 1978—as the law at the time required those seeking copyright protection to take minimal steps to signal that desire. Works that failed to take those steps, and that therefore are more likely to be subject to the Report’s privilege, are reasonably burdened by that privilege.

But works fixed on or after January 1, 1978, were created under a rule that imposed no formal obligations on copyright owners whatsoever. The consequence of that change is to make such work more difficult to trace. Yet it seems unfair to so vaguely burden copyright owners with a difficulty induced by Congress’ own action. At least with respect to works not amenable to simple registration or identification given current technologies— photo-graphs, for example—retroactive application of the Report’s privilege raises legitimate concerns of fairness.

Finally, I am also concerned about the effect of this proposal on unpublished work fixed after 1977. I recognize that Congress has deemed it appropriate to treat published and un-published works alike. But I am less convinced of the justice of subjecting unpublished works to an “orphan work” analysis. In our tradition, when one publishes a work to the world, the benefits of copyright protection are balanced by the reasonable burdens the rule places upon the copyright owner. I don’t see an equivalent justice in burdens imposed upon copyright owners of unpublished work. I understand that historically, copyright law found it difficult to draw a meaningful line between published and unpublished works. That diffi-culty was one reason for eliminating it in 1976. But in the context of an orphan works remedy, the weight that a distinction between published and unpublished work would have to bear is significantly less. The only consequence of a copyright owner finding his work on the “published” side of the line is that he would need to take steps to make sure he was accessible.

The Proposal Does Not Go Far Enough

In my view, the real problem of orphaned works is tied to old works. Any proposal to address that problem should therefore be triggered by the age of a work. The PDEA set 50 years as the trigger for an obligation by the copyright owner to take steps to maintain his copyright. The Copyright Office’s proposal is significantly more demanding, as the trigger for its obligation gets pulled immediately.

In light of the Copyright Office’s concern about current as well as older work, I would suggest you modify the term that triggers an obligation on the copyright owner from 50 years to 14 years—the initial term of copyright set by the First Congress. A presumptive 14 year term far exceeds the time during which the vast majority of work earns any commer-cial return at all. The danger of forfeiting a commercially valuable interest because of an orphan works remedy is therefore significantly lessened by a 14 year trigger. Thus, under this rule, any work less than 15 years old would be governed by the existing copyright rules. Any domestic work more than 14 years old would be subject to an orphan works remedy.

The question then is what that “orphan works remedy” should be. It is here that I believe the Copyright Office has not gone far enough. For again, the core of its proposal—triggering the privilege to use an “orphaned work” upon a “reasonably diligent search”—adds significant costs to the process of using, or reusing, old works. Those costs are born by both the copyright owner and the potential re-user: At least until there is extensive litigation clarifying the question, the copyright owner needs to guess about what steps are adequate to avoiding the orphan works remedy; and so too must the potential re-user guess about how much search is “reasonably diligent.”

A less burdensome rule (1) would more affirmatively specify the steps a copyright owner must take—after the 14 year term—to adequately maintain his copyright, and (2) would more clearly specify the remedies available to a copyright owner who fails to maintain properly his copyright. I describe each component below.

(1) The Responsibility to Maintain the Copyright

Background

Copyright is property. As with any property, it is supported within a property system. That system imposes certain reasonable responsibilities upon property owners, as a means to increase the value of property generally. Land owners must record their property claims. Automobile owners must register their ownership. State abandonment laws impose significant duties on property owners to keep ownership records current. And in some legal tradi-tions, even to keep a grave site, the family must periodically register the grave.

Such owner-based responsibilities are common within federal intellectual property law as well. A patent issues only upon an application; the grant then serves as a registration. After that initial grant, the patent owner must pay significant fees to the Patent Office to “maintain” his patent. And likewise, trademark owners have the responsibility to take affirmative steps to defend their mark, again in part to preserve the clarity of the trademark system.

Copyright stands out as an exception to this general rule governing property systems. While for 186 years of the American tradition of copyright, the law imposed upon copy-right owners certain responsibilities to keep title to their work clear, beginning in 1976, Congress took steps to abandon that tradition, and adopt instead the Berne rule that for-bids formalities. As I suggested at the start, some had viewed Berne as an absolute bar on owner-based obligations. Again, the great virtue of the Copyright Office Report is that it rejects that extreme view.

Nonetheless, there continues to be confusion about the Berne rule about formalities. In my view, that rule serves an important and obviously sensible objective—to protect foreign works against burdens that would be, within a global copyright system, unreasonable. It does not stand for the principle that copyright owners, alone among property owner, bear no responsibility to help make the copyright system function efficiently.

The Berne Rule Against Formalities

The rule against formalities limits the ability of a government to impose upon owners of foreign works obligations as a condition of the “enjoyment and exercise of” the copy-right. It is not a limitation on the ability of a nation to impose obligations on owners of domestic works. Thus, United States, for example, imposes burdens upon “United States works” that it does not impose upon foreign works. The law requires that to bring an action for infringement of a “United States work,” that work must be registered. This “condition” on the “exercise of” a copyright is not a violation of Berne because the obligation is not imposed on foreign works.

It was this understanding precisely that led you in the Public Domain Enhancement Act to impose your 50 year registration requirement on domestic works only. Once the value of such a system was obvious, other nations could impose similar requirements on their own domestic works. Eventually, an international registry could be built out of local registrations. I continue to agree with you that this strategy is the simplest way to address the orphan works problem: A maintenance requirement applied to domestic works only, supplemented perhaps by an equitable power to cure any failure to properly maintain a work.

The Copyright Office Report rejects any solution that distinguishes between foreign and domestic works. Its reasons are not, in my view, convincing. First, the Report argues that such a distinction would leave too many foreign works inaccessible. That, however, assumes that foreign governments wouldn’t also enact an orphan works remedy. If solving the orphan works problem is sensible for the United States, there’s no reason to expect it won’t be sensible for nations as well.

Second, the report asserts the distinction would add complexity. No doubt it would add some burden. But so too does the Report’s remedy add some burden to the copyright system. Between a clear prescription of steps taken to maintain a copyright, and an inherently vague “reasonably diligent search” standard, in my view, your system would be much less of a burden on copyright owners and re-users.

Third, and most importantly, the Report suggests such a solution would “discriminate against United States copyright owners.” This criticism, in my view, is simply miscon-ceived.

The problem that the Berne rule addressed was the obviously irrational situation in which every country required every copyright owner to register in that country in order to gain the benefits of copyright protection. Such a rule would cripple the protection of copyright internationally, since each copyright owner would need to register in more than a hundred jurisdictions. But Berne does not limit the opportunity for countries to implement a plainly rational alternative: That each nation require of its own copyright owners that they register locally, and that those registries be available internationally.

Your proposal would be a first step to this international regime. As nations recognized the cost of the orphan works problem, the solution you proposed would become increas-ingly salient. Local registries that could be coordinated internationally would minimize the burden on a copyright owner to maintain his work, while enabling re-users from around the world to identify the holders of particular copyrights.

Thus, following the structure of your original bill, a better orphan works remedy would be for Congress to require that 14 years after a domestic work was published, a copyright owner must take steps “to maintain” his or her copyright. Those steps would be specified through rule making by the Copyright Office. The Copyright Office would be free to modify those requirements in light of current technological and market conditions.

Thus, for example, the Copyright Office might determine that registering within a registry maintained by the government was the necessary step to “maintain” the copyright. I would not favor that requirement, but it is certainly a reasonable one. Alternatively—and I believe this is likely to be the better solution—the Copyright Office could develop a set of minimum protocols for private copyright registries, and require that a copyright owner of a domestic work register his work with one of many competing registries. Such protocols would assure that searches could be made across copyright registries. They would encourage innovation in the development of new registry techniques.

In this way, a copyright registry could function analogously to the Internet’s “domain name system” (DNS). As you know, to maintain a domain name, the owner must pay a fee for each year the domain name is held. That fee is paid to one of many DNS registrars. These registrars feed the necessary information to a central registry. That registry is then publicly available to resolve DNS addresses.

An analogous copyright registry would function very similarly. Once a protocol was set for registering copyrighted works, competing registries could offer copyright owners regis-try services. Some would no doubt specialize in particular types of copyrighted material— photographs, or published books, for example. Those specialized registries could provide added services that would enhance the value of the registration. But each would comply with interoperability protocols that would enable the data collected by each to be universally accessible. The result would be a very efficient system for identifying the owners of copyrighted works who wish to maintain full copyright remedies.

The advantage to this private registry system is the advantage of competition within a market for services. The government would play an important role in creating the demand for these services, and facilitating a protocol to assure interoperability. But once it had established the protocol, the investment necessary to build this registry, and the innovation necessary to keep it up to date, would come without the government. And as you know, there are already private entities that are working to build just such a registry system.

The Copyright Office report is skeptical of the value of a registry, as were the majority of comments submitted to the Copyright Office. But none of those comments considered a registry as I’ve described here. For example, the concerns about the burden of registration under the 1909 Act reflected the burden of the requirement of an immediate registration, and the burden of a costly technology for registration. This proposal would require no im-mediate registration when a work is copyrighted, and it would take advantage of digital technologies that have radically reduced the transaction costs of registration.

Likewise, the report identified the inequity of the 1909 system of formalities, calling that system a “trap for the unwary.” As the report concluded “it is likely that the man-datory registration requirements in the proposed systems would contain similar traps.” But there are many obvious ways to avoid the “traps” and inequities of the old system of formalities. For example, the law could include an equitable procedure to cure any failure to register properly. More particularly, and as I describe more below, it could protect the owner’s interest by requiring the payment of a royalty for any commercially exploited work that had failed to register in a timely fashion.

In short, there are many ways the burdens of a registration system—both technical and equitable—could be mitigated. And as no one is promoting a system that is as harsh as the system established by the 1909 Act, it is unclear why anyone would expect it “likely” that a newly created registration requirement would recreate the same “traps.”

(2) Limitation in Remedies

Once an orphan works remedy identifies the steps the copyright owner must take to preserve the full protection of his right, it must then specify the consequences of failing properly to maintain a copyright. The Copyright Office’s Report recommends that the failure to maintain a copyright properly (by being accessible to a “reasonable diligent search”) means a limitation in the remedies that a copyright owner would be entitled to un-der the Copyright Act. Your proposal was both clearer and most consequential—the failure properly to register would mean the copyrighted work would pass into the public domain.

Your solution is plainly the simpler one. It draws a clear line between works governed by copyright, and works not. And it balances Congress’ recent practice of extending the term of copyright.

I continue to believe that your solution is the right one. But if you followed the alterna-tive of the Copyright Office (and the recommendation of Congresswoman Bono to amend the Public Domain Enhancement Act)—by limiting remedies as the consequence for failing adequately to maintain a copyright—then I believe the Report’s particular limitations are much more complicated than is necessary.

Part of the reason for this complexity relates again to the problem I identified at first—that the Report’s recommendations apply to all copyrighted work, regardless of its age. If the focus of Orphan Work reform were on old works only, then the pressure to preserve traditional remedies would be reduced, and the resulting system could be much simpler.

Thus in my view, a better solution would be to specify a very minimal royalty rate for any commercial use of a work that has not been properly registered. That rate would be set by rule, and would differ depending upon the use. The rule could either require that funds collected under this provision be held in trust for a limited period of time, or deposited in the Copyright Office. After that period elapsed, the funds would return to the re-user.

The advantages of this system are many. First, by liquidating precisely the exposure a re-user faces, it becomes possible for a business to build that cost into their cost of produc-tion. Second, by shifting funds into a trust, the system could create an incentive to locate lost copyright owners, and help them reclaim their rights. Third, by setting the rates in advance, the system would avoid the uncertainty that the Report’s “reasonable compensation” standard would produce. A clear rule would again facilitate better protection for the under-lying works.

A Proposed Alternative

The PDEA embodied most of the virtues that I have identified here. It would have produced a clear line between the public domain and copyrighted works. It would have tar-geted its reform on old works. But in light of the findings of the Copyright Office, I would suggest that you consider a modified version of your proposal. That modified proposal would require: