Statement before the House Committee on the Judiciary, Subcommittee on Courts, Intellectual Property, and the Internet On Hearing Title Copyright Term and Moral Rights: Forging the Future by Understanding the Present Thomas D. Sydnor II Visiting Scholar, Center for Internet, Communications and Technology Policy, American Enterprise Institute July 15, 2014 The views expressed in this testimony are those of the author alone and do not necessarily represent those of the American Enterprise Institute. 1 Chairman Coble, Ranking Member , and Members of the Subcommittee, my name is Tom Sydnor, and I thank you for asking me to testify during your historic efforts to lead what future scholars should call the Fifth General Revision of U.S. copyright law. I serve as a Visiting Scholar at the Center for Internet, Communications, and Technology Policy at the American Enterprise Institute. But I testify today in my personal capacity; my views thus reflect only my own opinions, not those of any present or past employer. I am also honored and humbled by your invitation to testify at this hearing. During my career, I have spent thousands of hours studying the testimony, bills, reports and studies that record the Fourth General Revision that produced the basis for our current copyright laws, the Copyright Act of 1976. I thus appreciate the significance of this opportunity to contribute to your efforts to lead a Fifth General Revision that produces what the Register of Copyrights has called “Our Next Great Copyright Act.” My written testimony will focus on what has long been a persistently controversial issue in copyright law—the issue of copyright term— the question of how long copyright protection should last. The Issue of Copyright Term In copyright law, the issue of copyright term has long been controversial, and for many reasons. One, in particular, best explains why I can hope that my testimony may help the Subcommittee and Congress to resolve some important and bitterly contested claims about copyright term, but I cannot expect it to resolve all disputes about copyright term by revealing the one, right answer to the question of how long the economic rights of copyright owners should last. There seems to be no one, “right” answer to the question of copyright term, and that seems to ensure that copyright term will always be a contentious issue. Simply put, the question of copyright term interacts with all of the other important – and more important – questions at issue during a General Revision of US copyright law. Consequently, the question of term cannot be isolated and resolved definitively. For example, our current patent act grants extremely broad and nearly exception-free exclusive rights to qualifying inventors that last for only 20 years. That can suggest that much broader copyrights lasting for only a few decades could adequately encourage private investment in, and production of, expressive works – but I am unaware of any broad support for making copyrights more like patents. Therefore, my testimony takes a pragmatic approach to questions relating to copyright term. It is based upon on three principles. First, during a General Revision, we should not let eternal debates about copyright term distract us from first confronting and resolving more fundamental questions. Today, those truly fundamental questions seem to relate to copyright enforcement: how, and against whom, do we want copyright owners to enforce their rights, as a practical matter, against the torrent of Internet piracy that has arisen during the past sixteen years? The potential range of answers to that question are finite, but choosing among them means deciding whether and to what extent it is desirable or practical to continue to enforce copyrights against intermediaries, (as we did in the past), or to shift its focus by requiring copyright owners to enforce their federal civil rights by suing, or threatening to sue, individual Internet users. These fundamental questions are also implicated by this hearing. Today, too many creators and investors find that the practical term of their US copyright protection is best measured in hours or days, not years or decades. Consequently, even the most principled analysis of how long copyrights should last, (if they could actually be enforced, in practice, and even by individual creators and small-and2 medium sized creative companies), can become depressingly academic and even misleading, if enforcing copyrights becomes prohibitively expensive, time-consuming, and disfavored. To be sure, we never wholly eliminated infringement and counterfeiting in the brick-and-mortar world, and we cannot expect to do so on the Internet. But as in the brick-and-mortar world, we do need Internet-enforcement systems that can, in practice, reduce infringement and counterfeiting to a dull roar. Unless we develop such systems, most existing analyses of copyright term may be unhelpful. For example, economic analyses of property-right-related questions like copyright term often presume that the rights in question can be perfectly enforced. That presumption is never entirely accurate, but if reality differs too radically from it, then analyses based on it become unhelpful or even misleading. A Fifth General Revision should thus debate copyright term, but it should not let that debate become the proverbial cart that drives the horse. Second, in the late 20th century, over a century of multinational efforts finally generated almost universally accepted de jure international norms for copyright protection. Many of those substantive norms for copyright protection are prescribed by the 1979 version of the Berne Convention for the Protection of Literary and Artistic Works and can now be enforced by trade sanctions levied through Word-Trade-Organization dispute settlement proceedings as a result of the 1994 Agreement on TradeRelated Aspects of Intellectual Property Rights, (“TRIPS” or the “TRIPS Agreement”). These Berne/TRIPS norms arose from international negotiations and agreements – and they can still be changed, if needed, through such processes. Consequently, the US should not unilaterally change its laws on copyright term, (or term-related formalities, like renewal), in ways likely to violate our WTO obligations, or let other countries deny our authors benefits they grant to their own authors. The US is now a very successful net exporter of many copyrighted works, including books, music, movies, TV shows and other audiovisual works, application software, entertainment software, and mobile “apps.” We should thus refrain from unilaterally altering our copyright laws in ways likely to violate our WTO obligations. This may seem like common sense, but too many recent “copyright-reform” proposals have ignored it. Third, before deciding whether or how copyright term should change in the future, we should first understand why federal copyright term has evolved from the 28-year maximum prescribed by the Copyright Act of 1790 into the life-of-the-author-plus-70-year term underlying our current copyright laws. Unfortunately, questions about why copyright term has changed over time have remained bitterly disputed. I hope that the first two of these principles will be relatively uncontroversial, so I want to focus on the disputed question raised by the third principle. Last month, the Center for Internet, Communication, and Technology Policy at the American Enterprise Institute, (“CICTP”) asked me to prepare a paper that investigates the question of why the term of US copyright protection has changed from 1790 to today. The preliminary results of my ongoing research follow. Understanding the Evolution of Copyright Term from 1790 to Today. US Copyright term has changed during the 224 years since the Framers of our Constitution enacted the Copyright Act of 1790. For example, assume that state common-law protection and the federal 1790 Act produced U.S. copyright protection that lasted for an average of 20 years. Today, purely federal copyright protection lasts for a term of the-life-of-the-author-plus-70 years—on average, for about 95 years. The resulting rough comparison does suggest that the average term of U.S. copyright protection has increased by 375% during the last 224 years. 3 That is a significant increase, and we do need to understand why it occurred. But the reasons for it remain bitterly contested even after the Supreme Court’s Eldred v. Ashcroft decision upheld the constitutionality of our current laws on copyright term. Nevertheless, some scholars, analysts and lobbyists still argue that copyright term has evolved, and will keep evolving, along a “Mickey Mouse curve” that will eventually lead to copyright terms that last as long as the potentially indefinite life span of a particularly valuable animated character. Predictably, other scholars and analysts, as well as creators and creative industries, vigorously deny such claims. The Subcommittee’s ongoing General-Revision efforts would benefit from an effort to assess these clashing explanations for the evolution of US copyright term. In past General Revisions, Congress has increased copyright term, decreased it, and left it unchanged. Congress increased the term of U.S. copyright protection during the First General Revision that produced the Copyright Act of 1831. It left term unchanged during the Second General Revision that produced the Copyright Act of 1870. It increased copyright term during the Third General Revision that produced the Copyright Act of 1909. During the Fourth General Revision that produced the Copyright Act of 1976, Congress increased the term of protection as to some works, and decreased it as to others. To understand why US copyright term has changed since 1790 – and whether those changes have been principled or unprincipled – one must begin with the Copyright Act of 1790, the law that first granted federal exclusive rights to authors. It is important to begin with the Copyright Act of 1790 because it remains one of the most under-appreciated documents in the history of copyright law for two reasons. First, the 1790 Act was remarkable because it was enacted in May of 1790. In 1789, the Constitution had been ratified, George Washington had been elected as the first President, and James Madison and other Framers had been elected as Members of the First Congress. The bitterly contested Ratification process thus made them the leaders of a historically untested federation of independent States that had been united by war, divided by Ratification debates and many other issues, and were, in effect, collectively bankrupt. No one could have faulted the first President and Congress had they not enacted a federal copyright law, but President Washington saw copyright protection a basic duty of any civilized nation. Consequently, in his 1790 State of the Union address to Congress, Washington urged Madison and the other Members of Congress . Six weeks later, a bill proposing copyright protection was introduced, and about three months after that, the First Congress and President enacted the Copyright Act of 1790 into law. To the Framers, protecting the copyrights of authors was a critical duty. Second, too many have wrongly dismissed the 1790 Act as just a near-clone of the first modern copyright law, Britain’s 1710 Statute of Anne. The 1790 Act did closely follow the Statute-of-Anne model in many respects – including those relating to copyright term – but it also rejected that model in a critical respect. The Statute of Anne also contained what copyright lawyers might now call a generalcompulsory-licensing provision: anyone who thought that copies of a protected work were overpriced could complain to listed governmental, ecclesiastical or university officials who could then force the copyright owner to sell copies of the work at whatever price the official set. The Framers rejected this provision of the Statute of Anne, and thus created a truly market-based copyright law. Consequently, the term-related provisions of the 1790 Act deserve study. They reveal two critical principles that the Framers used to set copyright term. 4  The Framers concluded that copyright term should last during the lifetime of a work’s author, and for a (potentially short) post-mortem-author period in which an author’s copyrights could support his or her spouse and children.  The Framers also examined and adopted international norms relating to copyright term. The term-related provisions of the 1790 Act thus closely track the term-related provisions of what was then the best international model for federal copyright protection – Britain’s 1790 Statute of Anne. During the next 224 years of federal copyright law, copyright term did change significantly – but not because subsequent Congresses later abandoned the principles that the Framers had used to calculate copyright term. To the contrary, term-calculation principles have not changed materially since 1790. Nevertheless, over time, efforts to re-apply the Framers’ principles did require copyright term to increase significantly, at least in the case of the term of federal copyright protection. Two factors drove post-1790 increases in federal copyright term. First, the Framers’ principle of providing at least life-of-the-author copyright protection repeatedly required term to increase. Over time, authors and others simply began living longer than they tended to in 1790. For example, since 1790, the average human lifespan has increased by about 100% – from about 40 years to about 80 years. The Framers’ premise of life-of-the-author copyright term then required increases in copyright term. Second, the Framers’ principle of following international norms for copyright term eventually generated a new, principled basis for calculating the term of post-mortem-author copyright protection. The term provisions of the Copyright Act of 1790 permitted copyright protection to extend beyond the death of a work’s author. In the Copyright Act of 1831, Congress provided that a renewed term of copyright protection could be claimed by an author’s descendants. The US has thus long recognized that copyright term should protect not only a work’s author during his or her lifetime, but also his or her family after the author’s death. Nevertheless, pre-1976 US copyright acts do not seem to clearly prescribe any principle for calculating the term of post-mortem-author protection. But during the 20th century, the international, Berne Convention norms did prescribe such a principle: the Berne Convention prescribes at least a life-of-the-author-plus-50-year term of protection that implements a principle of three-generation copyright protection. Under Berne, copyrights are supposed to last during the lifetimes of a work’s author and his or her children and grandchildren. This Berne-Convention principle for calculating the term of post-mortem-author copyright protection has both humanistic and economic components. Copyrights do not protect works, they protect only their expressive components. Consequently, the author of a work and descendants who knew him or her personally would tend to be uniquely well-situated to understand what a work’s author intended to express and which future uses of the work would maximize the value of the expressive components of the work that copyrights protect. The Berne Convention thus presumes that copyright protection should persist during their lifetimes. In the Copyright Act of 1976, Congress first adopted a Berne-Convention-compliant term of life-of-theauthor-plus-50 years. The US then formally joined the Berne Union in 1988, and later responded to the so-called “Rule of the Shorter Term” that had been incorporated into the Berne Convention in 1948 by enacting the Copyright Term Extension Act of 1998. 5 Collectively, these two factors thus explain why federal copyright term has increased from 1790 to 2014. They also explain why US copyright laws have consistently followed the Framers’ principle of applying changes in copyright term retroactively to both existing and newly created works. In Eldred v. Ashcroft, the Supreme Court held that why changes in copyright term have always been applied . But another factor has also been at work. In the US, legislative calculations of copyright term have always been driven by estimates of how long human authors are likely to live. Human life spans change gradually, but laws that calculate copyright terms by estimating human life spans do not – and that is another reason why changes in copyright term have been applied retroactively, to then-existing works. For example, perhaps the most significant US expansion of copyright term for published works occurred on January 1, 1978, when the Copyright Act of 1976 became effective. But that Act was intended to provide authors with three-generation, Berne-Convention compliant copyright protection – and there was no reason to conclude that authors who chose to publish their works on December 31, 1977 would likely die long before authors who published their works a day later. Consequently, the 1976 Act’s changes to copyright term applied retroactively to existing works – just as like the preceding changes in term resulting from the Copyright Act of 1790, the Copyright Act of 1831, and the Copyright Act of 1909. Predictably, Congress also later retroactively applied the extension of term effected by the Copyright Term Extension Act of 1998. Recommendations Relating to Copyright Term I hope my testimony, ongoing research, and forthcoming paper will help the Subcommittee decide when and how to prioritize questions about copyright term during a Fifth General Revision. While I do not believe that my research necessarily tells the Subcommittee exactly whether or how existing copyright laws related to term should change, I do want to conclude with some thoughts on questions about copyright term that the Subcommittee’s General Revision efforts must eventually confront. First, while three past General Revisions of US copyright laws have expanded the term of federal copyright protection, I am unaware of any present interest – by creators and creative industries, or by the public generally – in further expansion or extension of US copyright term. As a practical matter, it thus seems like a Fifth General Revision of US copyright law may focus on whether copyright term should be left unchanged. If copyright term is changed, I think that history strongly suggests that any such change should be principled, and it should not violate any Berne-Convention/TRIPS-Agreement norms. Second, the current Register of Copyrights, Ms. Maria Pallante, has proposed a change in existing laws regarding term: the Register has proposed to retain our existing term of life-of-the-author-plus-70 year copyright protection, but to condition the last twenty years of protection on the fulfillment of a formality – a renewal obligation. The Register’s proposal deserves careful consideration, and further study, by the Subcommittee, Congress and the Executive Branch. I do not think that it would violate our Berne-Convention/TRIPS-Agreement obligations. Nevertheless, I think we should further study at least three potential issues:  Will other nations, like the EU Member States, be likely to conclude that they can and should invoke the Berne Convention’s Rule of the Shorter Term against the authors of a country that conditions the last 20 years of a life-of-the-author-plus-70-year copyright term upon the fulfillment of a formality? We must further study this question in order to assess all potential costs and benefits of this proposed change. 6  Would conditioning the last 20 years of a life-of-the-author-plus-70-year copyright term upon the fulfillment of a renewal formality cause other nations to impose similar national renewal obligations? One US renewal requirement might not impose an excessive burden upon copyright owners, but many, differently configured, national renewal requirements could easily become prohibitively burdensome.  Can we articulate a principled basis for reducing the base term of formality-free copyright protection to life-of-the-author-plus-50 years? My research could suggest one. Given current, developed-world life expectancies, such a term could arguably implement two-generation copyright protection – a term of protection that would last through the lifetimes of an author and her or his children. Indeed, even were our current life-of-the-author-plus-70-year copyright term left unchanged during a Fifth General Revision, relatively small, foreseeable increases in human life expectancy might soon require us to characterize even our current approach to copyright term as providing two-generation copyright protection. In short, our approach to the issue of copyright term should continue to be principled. I hope that this testimony assists the Subcommittee’s efforts to lead a Fifth General Revision of US copyright law, and I am honored by the opportunity to participate in that historic process. 7