Opposing Copyright Extension

Legislative Materials (104th Congress)

Written Testimony
H.R. 989
A Bill to Amend Title 17, United States Code
With Respect to the Duration of Copyright, and for Other Purposes

Return to the Opposing Copyright Extension Home Page

Click here for the changes that  H.R. 989 would have effected

Professor of Law
Arizona State University
H.R. 989
A Bill to Amend Title 17, United States Code
With Respect to the Duration of Copyright, and for Other Purposes
Rayburn House Office Building
Washington, D.C.
July 13, 1995

The proposed legislation (H.R. 989) would extend the term of copyright protection for all copyrights, including copyrights on existing works, by 20 years: For individual authors, the copyright term would extend for 70 years after the death of the author, while corporate authors would have a term of protection of 95 years. Unpublished or anonymous works would be protected for a period of 120 years after their creation. The legislation would also extend the copyright in works that may be as old as our Republic or even older but that were never published prior to 1978 (when these works were first brought into the federal copyright system). Initially, these copyrights would be extended by another 10 years (to the year 2013), and if the copyright owners publish the works prior to 2013, copyrights in these already ancient works would continue in force until the year 2047.

We believe that enactment of this legislation would impose substantial costs on the United States general public without supplying any public benefit. It would provide a windfall to the heirs and assignees of authors long since deceased, at the expense of the general public, and impair the ability of living authors to build on the cultural legacy of the past. In following a European model of regulation and rigidity, it would hinder overall United States competitiveness in international markets, where the United States is currently at its most powerful. We therefore conclude that it would be a mistake to extend any of the copyright terms of protection.


Various reasons have been offered in support of the extension proposal: Some say that the extension is necessary as an incentive for the creation of works. Some argue that the current period for individual authors--50 years after the death of the author--was intended to provide an income stream for two generations of descendants and that the longer human life span now requires a longer copyright term. Some maintain that we should adopt an extended term because the countries of the European Union have done so, in order to "harmonize" our law with theirs. Some claim that the longer copyright term is necessary to prevent royalty inequality between United States and European copyright owners.

None of these arguments take into consideration the costs to the United States public of an extended copyright term. Moreover, the arguments are either demonstrably false or at best without foundation in empirical data. If incentives were the issue, there would be no need to extend the copyrights on existing works, even if one were to accept the dubious proposition that the extra 20 years provide an incentive for the creation of new works. If we were worried about two generations of individual descendants, we should prohibit the first generation from selling the copyright outright, and we would have no need to extend the term for corporate authors. If we believe in harmonization, it is in any event not achieved under the proposed legislation nor does supposed royalty inequality provide a basis for extending the term. The discussion below shows the failure of these arguments in detail. It also shows that the costs to the United States general public vastly exceed even the gains to those relatively few copyright owners who would benefit from the extension and that the general public itself would receive no compensating benefits.

Once the errors in the arguments for increasing the term have been exposed, the real reason for the legislation becomes clear: The maintenance of royalty revenues from those relatively few works from the 1920's and 1930's that continue to have significant economic value today. The continued payment of these royalties is a wealth transfer from the United States public to current owners of these copyrights. These copyright owners are in most cases large companies and in any case may not be descendants of the original authors whose works created the revenue streams that started flowing many years ago. To our knowledge, no one has made a study of just how great this wealth transfer would be, although it is clearly large enough to generate fervent support for the proposed legislation by performing rights societies, film studios, and other copyright owners in economically valuable works whose copyrights are otherwise due to expire in the next few years.

The works about to enter the public domain, absent this legislation, were created in 1920. At that time and for many years thereafter, society's "bargain" with the actual authors was a period of exclusive rights under copyright for a maximum of 56 years. Those authors produced and published their works with the understanding that the works would enter the public domain 56 years later. Yet, notwithstanding that bargain, the period was extended by 19 years in 1976 to 75 years, as were the terms of all copyrights acquired after 1920. Now, 19 years later, these same copyright owners have returned seeking yet another extension to continue the wealth transfer for another 20 years, without supplying any evidence, or even any arguments, that the public will benefit.

This wealth transfer from the United States general public to copyright owners is, moreover, only a part--probably a small part--of the total cost that we and coming generations will bear if the extension is adopted. It is important to remember that the extension would apply to foreign as well as United States works. Therefore, in order to maintain a flow of revenue to the owners of United States copyrights, the general public will continue to pay on foreign copyrights from the 1920's whose terms must also be extended. No one has shown that there will even be a net international inflow of royalties from the works at issue.

Even worse, to maintain the royalty revenues on those few works from this period that have continued economic viability, the copyrights must be extended on all works. This includes letters, manuscripts, forgotten films and music, out-of-print books, and much more, all potential sources on which current authors and scholars can base new works. Copyrights can and usually do have very complicated multiple ownership so many years after an author's death. The transaction costs of negotiating for use can be prohibitively high, even for works that no longer have economic value. None of the arguments for extension take into consideration the loss to both revenue and culture represented by the absence of new popular works that are not created because underlying works that would have served as a foundation remain under the control of a copyright owner. By definition, this loss can never be known, but that makes it no less real or substantial.

The creation of new works is dependent on a rich and vibrant public domain. Without good reason to expect a substantial compensating public benefit, we should not risk tying the hands of current creative authors and making them less competitive in domestic and international markets just to supply a financial windfall to owners of copyrights in works created long ago. Just as Santa Claus and the Easter Bunny are part of the public domain that anyone can use every Christmas and Easter season, so eventually should Mickey Mouse and Bugs Bunny also join our freely available cultural heritage. That is a crucial part of the copyright "bargain" that the public made at the time these works were created.

We recommend that the proposed legislation be rejected. The issue is certainly an important one, but the legislation is premature at best where there has been no empirical demonstration of a public benefit and no thorough exploration of alternative approaches.


Both Congress and the courts have uniformly treated United States copyright law as an instrument for promoting progress in science and the arts to provide the general public with more, and more desirable, creative works:

United States copyright tradition is in this respect philosophically different from that of many other countries that treat intellectual property as natural rights of individual creators. Under our system, Congress need not recognize intellectual property rights at all, but if it does, the purpose must be to promote innovation in science and the useful arts.

Our system of copyright protection is delicately balanced. We recognize exclusive rights in creators so that consumers have available an optimal number and quality of works but want those rights to be no stronger than necessary to achieve this goal.(2) We do not recognize new intellectual property rights, or strengthen old ones, simply because it appears that a worthy person may benefit; rather, we do so only for a public purpose and where it appears that there will be a public benefit. The current statutory foundation of copyright protection, the Copyright Act of 1976, is itself the product of lengthy debate and represents innumerable compromises that seek to achieve the proper balance between private returns to authors and public benefit, including a broad public domain that permits current authors to build on the cultural heritage from those who have come before them.

We are aware of no effort by the proponents of this extension legislation to show that the public benefits from its enactment would outweigh the costs. Indeed, they have demonstrated no public benefit whatsoever and have barely attempted to do so. Yet, the public cost in the form of a diminished public domain is obvious.(3) As we demonstrate below, this public cost is not offset by any increased incentive to create new works, nor does international trade in intellectual property rights fill the gap between public costs and public benefits.

Europe, whose copyright law is based more on a natural rights tradition, has recently moved to a life + 70 regime for individual authors and a 70-year period of protection for corporate authors. That should not cause us to change our underlying intellectual property philosophy. Nor does it provide a reason for avoiding the careful cost/benefit analysis called for by that philosophy. The United States joined the Berne Convention for many good reasons, one of which was to become an influential leader in world intellectual property policy. Our underlying policy has served us well, as shown by our dominant position in the worldwide markets, particularly for music, movies, and computer software. Rather than following Europe we might better seek to persuade Europeans that our approach to intellectual property rules both rewards creativity and promotes economic efficiency.

In the following sections we consider in some detail the arguments put forward in support of the extension. We first show the very real and substantial costs to the public that would result from adoption of this legislation--costs that are ignored by the arguments of its proponents. We then go on to show that the arguments in favor are either logically fallacious or unsupported by any plausible evidence.


While the asserted public benefits of an extended copyright protection period range from speculative to nonexistent, two identifiable costs are real and substantial: The first is the economic transfer payment to copyright owners during the period of the extension from consumers or other producers who would otherwise have free use of works. The second is the cost to the public of works that are not produced because of the diminished public domain.

Economic Costs and Transfers

The direct economic costs of a 20-year-longer period of protection, although difficult to calculate precisely, includes higher cost to the consuming public for works that would otherwise be in the public domain. That these costs are substantial is shown by the very claims of the proponents of this legislation that they will miss out on the European windfall if we do not extend our term to that of Europe. This windfall does not arise out of whole cloth. Rather, it is ultimately paid by consumers, that is, by the general public. And if Europeans will be paying for the right to use United States works in Europe, the United States public will be paying for the right to use both United States and European works here at home, increasing the windfall to copyright owners at the expense of United States consumers.

In the legislative history of the Copyright Act of 1976, it was argued that the general public received no substantial benefit from a shorter term of protection, because the cost for works in the public domain was frequently not significantly lower than that for works still under copyright.(4) Even without the fervor of the special interest protagonists of this legislation, however, economic theory tells us that the price to the public for popular works must, through competition, decrease to the marginal cost of producing the work if there are no exclusive rights. If the work is under copyright, the marginal cost of production would have to include the royalty owing to the copyright owner, even if there is general licensing to competing producers of the work. Moreover, if there is no general licensing of a copyright-protected work, the price can be expected to be set at the level that maximizes the return of the copyright owner, which is invariably higher than the marginal cost of production. Consequently, any claim that the public pays the same for public domain works as for protected works is implausible, at least in general.(5) Educational and scientific uses would also seem to be large markets for public domain works. At a time of rising educational costs we should inquire into the effect on our schools of a reduced public domain due to an extended protection period. Something more than anecdotal evidence should be presented before we accept the claim that the consuming public will not incur higher costs from the longer period.

Cost of a Diminished Public Domain

An even more important cost to the public is that paid in desirable works that are not created because of the continuing copyright in underlying works:

While primary control over the work, including the rights to refuse publication or republication and to create derivative works, properly remains in the author who has created it, giving such control to distant descendants of the author can deprive the public of creative new works based on the copyright-protected work. Artistic freedom to make creative derivative works based on public domain works is a significant public benefit, as shown by musical plays like Les Miserables, Jesus Christ Superstar, and West Side Story, as well as satires like Rosencrantz and Guildenstern are Dead and even literary classics like James Joyce's Ulysses. Although these might not necessarily be considered infringing derivative works even if the underlying work were under copyright, or might be excused by the fair use doctrine if otherwise infringing, their authors must necessarily take a cautious approach if a license is unavailable. When copyright subsists long after an author's death and there is no provision for compulsory licensing, the creation of derivative works that closely track a substantial part of the underlying work can be absolutely prohibited by copyright owners who have no creative relationship with the work at all. Authors of histories and biographies can also be inhibited from presenting independent analyses of earlier authors and their works by descendants who, for whatever personal reason, use copyright to prevent the publication of portions of protected works.

An important cost paid by the public when the copyright term is lengthened, therefore, is contraction of the public domain. The public domain is the source from which authors draw and have always drawn.(7) The more we tie up past works in ownership rights that do not convey a public benefit through greater incentive for the creation of new works, the more we restrict the ability of current creators to build on and expand the cultural contributions of their forebears. The public therefore has a strong interest in maintaining a rich public domain. Nobody knows how many creative works are not produced because of the inability of new authors to negotiate a license with current copyright holders, but there is at least anecdotal evidence that the number is not insubstantial.(8) Unless evidence is provided that a life + 70 regime would provide a significant added incentive for the creation of desirable works, the effect of an extension may well be a net reduction in the creation of new works.

This point may be highlighted by the rapid developments now occurring in digital technologies and multimedia modes of storing, presenting, manipulating, and transmitting works of authorship. Many multimedia works take small pieces of existing works and transform them into radically different combinations of images and sounds for both educational and entertainment purposes. The existing protection period, coupled with termination rights, may well be distorting or inhibiting the creation of valuable multimedia works because of the transaction costs involved in negotiating the number of licenses required. Ultimately, the rapid changes in the intellectual property environment for creating and disseminating works may necessitate a reassessment by the international community of the underlying intellectual property rules. In the meantime, extending the protection period can only exacerbate this problem. The United States should be leading the world toward a coherent intellectual property policy for the digital age and not simply following what takes place in Europe.

Incentives for the Creation of Works

It does not follow that a longer term automatically drives creative authors to work harder or longer to produce works that can be enjoyed by the public. Indeed, there is necessarily a type of diminishing return associated with an ever-longer protection period, because the benefit to the author must be discounted to present value. As Macaulay observed over 150 years ago:


Thus, while an additional year of protection has little or no incentive effect at the time of a work's creation, the costs are immediate and substantial if the extension is to apply to existing works, as provided in the proposed legislation.

The copyright industries are by their nature very risky, and no one in these industries makes financial decisions based on even 50-year, let alone 70-year, projections. Moreover, under the United States Copyright Act, most transfers of copyright by an individual author may be terminated 35 years after the grant.(10) The existence of these inalienable termination rights in individual United States authors makes it even more unlikely that anyone would pay more to exploit a work under the extended term than would be paid under the current life + 50 period.(11) The extension, therefore, holds little promise of financial benefit to individual authors.

The absence of any additional incentive for corporate authors from the extension of the copyright period to 95 years is also easily seen. Consider an assured $1,000 per year stream of income. At a discount rate of 10%, the present value of such a stream for 75 years is $10,992, while the present value of a 95-year stream is $10,999, a difference of less than 0.1%. Even at a 5% discount rate, the present values are only $20,485 and $20,806, respectively, a difference of about 1.5%. And these minuscule present value differences are for guaranteed streams of income. When risk is factored into the analysis, the present value of a 75-year stream and that of a 95-year stream must be considered essentially identical. The chance that a given copyright will still have nontrivial economic value 75 years after the work is created is very small--only a tiny fraction of all works retain economic value for such a long time. No company will take the "extra" 20 years into consideration in making a present decision to invest in the creation of a new work. In fact, an ongoing successful company like Disney is more likely to be spurred to the creation of new works like The Lion King or The Little Mermaid because it realizes that some of its "old reliable" moneymakers, like Mickey Mouse, are about to enter the public domain.

It is therefore extremely unlikely that an additional 20 years of protection tacked onto the end of a copyright protection period that is already very long will act as an incentive to any current author to work harder or longer to create works he or she (or it) would not have produced in any event. What is certain, however, is that such an extension of the copyright term would seriously hinder the creative activities of future as well as current authors. Consequently, the only reasonable conclusion is that the increased term would impose a heavy cost on the public--in the form of higher royalties and an impoverished public domain--without any countervailing public benefit in the form of increased authorship incentives.

Indeed, if incentives to production were the basis for the proposed extension, there would be no point in applying it to copyrights in existing works. These works, by definition, have already been produced. Yet, if the extension were purely prospective (i.e., applicable only to new works), we could be certain that support for it would wither rapidly. Thus, the real issue is the continued protection of old works--not those that will enter the public domain 50 (or 70) years from now but rather those due to enter the public domain today. These works were originally published in 1920 (works published before 1978 have a flat 75-year copyright rather than the current life + 50 for individual authors). At that time, the law afforded a maximum of 56 years of copyright protection. This period was expanded to 75 years in 1976, and now the descendants and assignees of these authors want yet another 20 years. The very small portion of these works that have retained economic value have been producing royalties for a full 75 years. In order to continue the royalty stream for those few copyright owners, the extension means that all works published after 1920 will remain outside the public domain for an extra 20 years. As a result, current authors who wish to make use of any work from this period, such as historians or biographers, will need to engage in complex negotiations to be able to do so. Faced with the complexities of tracking down and obtaining permission from all those who by now may have a partial interest in the copyright, a hapless historian will be tempted to pick a subject that poses fewer obstacles and annoyances.

Copyright in Works Never Published Prior to 1978

Until the effective date of the Copyright Act of 1976, works that had never been published were protected under the various state copyright statutes. Only published works were governed by the federal statute. However, the 1976 Act preempted state protection for unpublished as well as published works and, as a quid pro quo for the loss of perpetual state copyright protection, recognized a copyright in these previously unpublished works until the year 2003. As an incentive to publication of these works, the current law also extends their copyrights until the year 2027, provided they are published prior to 2003. The proposed legislation would extend these periods by 10 and 20 years, respectively, so that a previously unpublished work will be protected until 2013 and, if published prior thereto, it will remain under copyright until the year 2047.

An example is the recently discovered fragment from a draft of Mark Twain's Huckleberry Finn. The copyright on the published novel was registered in 1884, renewed by Twain's daughter in 1912, and expired in 1940. Even if a life + 70 system had been in place at the time of the work's creation, the copyright would have expired in 1980, along with everything else Mark Twain wrote (because he died in 1910). Because this story of Huckleberry Finn and Jim in the cave has now been published, however, current law recognizes the copyright until 2027. Under the proposed extension, the copyright on this story, already over 110 years old, will continue until the year 2047.

We are not aware of any arguments in support of these particular extensions of the copyright period of protection. In contrast to the Mark Twain fragment, most of these works have only scholarly value, because if they were readily available and had economic value, they would already have been published. Moreover, many of these works are truly ancient--letters and diaries from the founding fathers, for example--and constitute a vital source of original material for historians, biographers, and other scholars.

Obviously, the normal copyright incentive to creative authorship is not involved here. This is simply an incentive to current owners of copyrights in very old works to find the works and publish them so that they will be accessible to everyone. By the year 2003 we will already have afforded the very distant descendants of the authors of these works 25 years of protection, plus the possibility of 50 years of protection if they find and publish the works. Twenty-five years is enough time for these owners to accomplish the ministerial tasks. These unpublished works should be allowed to go into the public domain in 2003, so that others will then have an incentive to find and publish them.

Finally, even as to such of these works that are published prior to 2003, we can think of no argument, whether founded in natural law or otherwise, to support extending their term of protection until 2047. Fifty years of copyright protection for such old works, in favor of people who have no creative relationship with the works at all, is more than enough.

Support for Two Generations of Descendants

It is also argued that the copyright protection period was initially designed to provide a source of income to two generations of descendants of creative authors. Given the longer life spans of today, the argument goes, a longer term is necessary to achieve this goal.

Far from requiring longer copyright terms to compensate for longer life expectancies, these actuarial changes could be an argument for keeping the current term of life + 50, or perhaps even reducing it, because the longer life expectancy of the author automatically brings about a longer period of copyright protection. A longer overall life expectancy, moreover, does not in itself imply that the second generation loses anything in comparison with earlier eras. The crucial age for the second generation is not the absolute number of years grandchildren may be expected to live but rather the number of years they survive after the author's (i.e., their grandparent's) death. The copyright period is measured from the death of the author, and if grandchildren are living longer, so too are authors themselves. Certainly no one has provided data to show that grandchildren of today have significantly longer life expectancies than today's grandparents, let alone 20 years longer. Consequently, we should expect the current cohort of authorial grandchildren to remain alive for roughly the same length of time after their grandparents' deaths as at other times in this century.

Second, protection of two generations of descendants is not the inevitable result of a longer protection period. The copyright in a work that has been exploited and become popular will often have been transferred by the author or her descendants. Any termination rights with respect to the work will have already been exercised before the descendants in question here ever come into the copyright picture.(12) It is very likely that the copyright will have been retransferred after any termination before the current life + 50 year period has expired. Unless these transfers provide for a continuing royalty, there will be no royalties for the author's descendants who are alive thereafter. Moreover, even if the transferee is under obligation to pay a continuing royalty, it cannot be assumed that the royalty stream will accrue to distant relatives of the original author, such as great-grandchildren. The royalty may well be transferred outside the family, by will or otherwise, by earlier descendants. If sustenance to two generations of authorial descendants is really the goal, we should be considering prohibitions on transfers and/or stronger termination rights rather than a longer term of protection.

Third, even the "natural law" argument on behalf of such distant descendants of authors is very weak. These equitable claims to a continued income stream obviously diminish with increasing temporal distance of descendants from the creative author. More important, while one can understand the desire of authors to provide a substantial estate to their immediate offspring, one must question the economic efficiency of a system that, as a matter of policy, seeks to grant an easy flow of income to a group of people the majority of whom the actual author may never have known. The descendants themselves would probably be better off, and certainly the general public would be better off, if they were to engage in some productive activity. United States copyright policy is not and has never been designed as a welfare system. It is therefore not entirely flippant to say to these distant descendants of creative authors who died 50 years ago what many now say to current welfare recipients: "Get a job!"

Fourth, while the Directive in the European Union mentions protection for two generations of descendants as one of twenty-seven "Whereas" grounds for the extension in Europe,(13) it has never been recognized as a goal of United States copyright law. Indeed, today's longer life expectancies were offered as a basis for the recent substantial extension of the copyright term in 1976, from 56 years to life + 50 years, without any mention of a "two generation" goal.(14) Surely life expectancies have not increased since 1976 to justify an additional 20 years of protection on this ground. Going to our current life + 50 system was necessary in order for the United States to join the Berne Convention, and one could at least make a coherent argument that the benefits of joining Berne might outweigh the costs of the diminished public domain resulting from the longer copyright. The "two generation" argument, however, is devoid of any relationship to a public benefit. We therefore question whether such a claim comports with basic United States copyright principles and the social bargain that places works in the public domain after the copyright has expired.

Finally, even if "two generations of descendants" were a valid basis for extending the copyright term for works of individual authorship, it provides no justification whatsoever for extending the term for corporate authors from 75 to 95 years.

We conclude that the "two generation of descendants" argument is invalid on its face, advocates economic inefficiency, fails to comport with basic United States copyright principles, and is applicable at best to the term for individual authors. It cannot serve as a basis for the diminished public domain that the extension would effect.

"Harmonization" with European Law

The European Union has now directed its members to adopt a life + 70 term of copyright duration. Possibly because of the European natural rights tradition, neither the proposal in Europe nor its adoption was based on a careful analysis of the public costs and benefits of extending the term. Nevertheless, some argue that we must do the same to "protect" United States copyright owners, against whom the "rule of the shorter term" may be used to provide a shorter period of protection in Europe for United States works (life + 50) than is given to European works (life + 70). They also argue that harmonization of the worldwide term of protection is a desirable goal in its own right and that failure to adopt the European term will have an adverse effect on the United States balance of international trade. We first consider the general harmonization goal and, in the next sections, take up the question of the supposed "prejudice" United States copyright owners and the balance of trade would suffer in Europe were we not to follow the European example.

Harmonization of worldwide economic regulations can often be useful, especially if differences in legal rules create transaction costs that inhibit otherwise beneficial exchanges. In some cases harmonization can be beneficial even if the uniform rule is in some sense less than ideal. Thus, a uniform first-to-file rule for patents might make sense even if we believe that a first-to-invent rule is better in the abstract, because otherwise United States inventors--the very people whom we are hoping to encourage through the offer of a patent monopoly--might find it too burdensome to seek international protection. In that case the uniform rule goes to the very existence of the patent and not simply an extension of the duration of protection. We need not, however, seek uniformity for its own sake, if it means compromising other important principles. If the United States determines that works should belong to the public domain after life + 50 years, no transaction cost problem is posed to United States authors by the longer period in Europe. The ultimate owners of their copyrights will, of course, be able to exploit them for a shorter period, in both Europe and the United States, but that is the result of our policy choice to make the works freely available and not because of the absence of harmonization.

In addition, even if harmonization is desirable, the question remains, who should harmonize with whom? Although doubts were expressed about the constitutionality of a life + 50 year period of protection at the time the Copyright Act of 1976 was adopted,(15) that standard could then accurately be denominated international(16) and was in any event necessary if we were ever to join Berne. Life + 70 years is not an international standard today, notwithstanding recent actions in the European Union, nor will it become one without United States support. It was not even the standard in Europe until the European Council of Ministers directed that its member states adopt a uniform term of protection equal to the longest of any of its members. If the cost/benefit analysis required by our copyright tradition does not justify changing the social policy balances we have drawn, we might better use our influence to encourage the rest of the world to remain with our standard, and Europe to return to it, rather than follow a decision in Europe that was made without consideration of the factors we have always deemed crucial to the analysis.

Moreover, the proposed legislation is not really aimed at harmonizing United States and European law. It would, for example, extend the copyright period for corporate "authors" to 95 years (or 120 years if the work is unpublished). The European Union, by contrast, now offers corporate authors, for countries recognizing corporate "authorship," 70 years of protection, which is less than the 75 years we currently offer such authors. Consider also the works of Sir Arthur Conan Doyle, who died in 1930 and whose works have since 1981 been in the public domain in England (and Europe). Because works first published before 1978 have a 75-year period of protection rather than the current life + 50 term, those works of Conan Doyle published in the 1920's remain under United States copyright. Thus, production in this country of public domain collections of his entire works is prohibited, although Europeans may do so freely. Because his last work was apparently published in 1927,(17) it is scheduled to go into the United States public domain at the end of the year 2002. The extension would continue this "disharmony" until the year 2022.

There are many other features of copyright law that are not "harmonized" even within Europe, let alone between Europe and the United States, including moral rights and the important United States concept of fair use. "Harmonization" is therefore not in itself a valid ground for extending any of our current copyright protection terms.

Unequal Treatment of United States Copyright Owners

In addition to lengthening the copyright term for individuals to life + 70 years, the European Union has adopted the "rule of the shorter term," under which works are protected only for the shorter of the European term or the term in the country in which the work originates. Therefore, it is true that retaining our current term of protection would deny some United States copyright owners (mainly companies rather than individuals) the financial benefit of this European windfall. But the mere fact that the European Union has adopted a bad idea does not mean that the United States should follow suit. France might elect in the future, for example, to give the works of Voltaire or Victor Hugo perpetual copyright protection, but that would be no reason for us to do the same with Mark Twain or Emily Dickinson. The European copyright tradition, as we have noted, differs in important ways from that of the United States, primarily by treating copyright as a kind of natural entitlement rather than a source of public benefit. The European approach may on balance tend to discourage, rather than promote, new artistic creativity. We should not, therefore, assume that a policy giving a few United States firms and individuals an added financial windfall from works created long ago necessarily is one that promotes our long-term competitiveness in the production of new works.

This extension proposal is perhaps an occasion to consider the special character of United States copyright and the features that distinguish our law from its continental counterparts. The constitutional concept of a limited term of copyright protection is based on the notion that we want works to enter the public domain and become part of the common cultural heritage. It is worth noting that in this century United States cultural productivity and international market share has been much greater than that of Europe. The genius of the American system is that it balances public and private rights in such a way as to provide a rich collective source on which to base new and valuable productions. This makes us wealthier not only culturally but in a hard-nosed economic sense as well.

We must ask whether we really wish to remake our cultural industries in the image of Europe. This is not, in fact, a conflict between Europe and the United States. The real conflict, in both Europe and the United States, is between the interest of the public in a richer public domain and the desires of copyright owners (who may or may not be relatives of authors) to control economic exploitation of the copyright-protected works that remain in their hands. That Europe has resolved the conflict one way does not mean that we should blindly follow suit.

The arguments for maintaining a rich public domain in the United States are not diminished by the withdrawal of works from the public domain in Europe, or even by the partial withdrawal of only "European" works. If Europe protects "its" copyright owners for a life + 70 year period, its public domain is reduced, and the European general public suffers a net loss. The United States public, however, as opposed to individual copyright owners, is not harmed by the absence of protection in Europe 50 years after the death of a United States author. Conversely, the public will pay a real cost, both as consumers and as potential creators of new works, to the extent the public domain is further reduced by the longer protection period.

It should be borne in mind that we are no longer talking about authors, whether European or American, of the works that would remain protected for the extra 20 years. Those authors will have been dead for 50 years. We are talking about current authors, however, who create new and valuable works based on the public domain. If the underlying work is unprotected in Europe as well as in the United States, those new United States derivative work creators, as authors, will reap the kind of economic benefits in both jurisdictions for which copyright is indisputably designed. There is real cultural value in allowing works to become part of the common heritage, so that other creative authors have the chance to build on those common elements.

In this context, therefore, the notion of international "harmonization" simply obfuscates the real issue: There is no tension here between Europe and the United States. The tension, rather, is between the heirs and assignees of copyrights in old works versus the interests of today's general public in lower prices and a greater supply of new works. Europe has resolved the tension in favor of the owners of old copyrights. We should rather favor the general public.

The Balance of Payments

We have conceded that certain United States copyright owners will receive royalty payments from European users for a shorter period than will European copyright owners from European users, if the United States does not follow Europe in extending the copyright term. It does not follow, however, that this will have any net negative effect on the United States balance of trade, even in the short term and much less over the longer term.

Increasing the term in the United States means not simply that European users will pay longer. It also means that United States users will pay longer, and not just to United States copyright owners but also to owners worldwide. Works that are about to enter the public domain were created in 1920, and while Europeans may take more of our current works than we take of theirs, that is not necessarily true of works from the 1920's and 1930's. Our use of European works of classical music and plays as well as art works from this era may outweigh the use Europeans make of United States works from the same period. Short term balance-of-trade analysis therefore requires an investigation of whether our use of such works that would remain protected under the proposed extension would cost more than we would receive in return.

Moreover, a shorter term of protection in the United States will encourage rather than discourage the production of new works for worldwide markets. We must recall that the public domain is the source of many of our finest and most popular works. The United States market is itself so large that, with both European and United States works in the public domain here 50 years after the author's death, it alone serves as a strong creation incentive. If the new work is based on a United States work that is also unprotected in Europe, that new work should be a part of the continuing United States export engine in the world market. Even if the new work is based on a European work that remains under protection in Europe, popularity of the work in the United States will necessarily result in a license (to use the underlying work) in Europe, again with a net export gain to the United States.

The argument that United States copyright owners will unfairly "lose" royalty revenues from Europe is therefore both wrong and incomplete. It is wrong because it is not unfair that a work enter the public domain 50 years after the death of its author. It is incomplete because it does not consider that the royalties in question will be paid not just by Europeans but also by Americans, and not just to United States copyright owners but also to copyright owners worldwide. Additional revenues to a few owners of old copyrights is not a public benefit justifying adoption of the legislation, and this remains true even though some part of those revenues would be paid by Europeans. The extension represents, rather, a heavy public cost, both in additional royalties paid by the United States public and in the loss of creative new works that will not be produced because the exclusive rights of copyright remain in full force on works that cost/benefit analysis would clearly place in the public domain.


The proposed legislation extending all copyright terms by 20 years is a bad idea for all but a few copyright owners. None of the current copyright terms of protection should be extended.


1. . Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 156 (1975)(footnotes omitted).

2. 1 P. Goldstein, Copyright § 1.1, at 6-7.

3. . The proponents of the extension could at least have considered less drastic means of achieving their asserted goals. They might have proposed, for example, a "no injunction" regime 50 years after the author's death, which could provide a continuing royalty to the owners of copyrights in economically valuable works (at the expense of the public) but would at least permit current and future authors to use all old works, 50 years after their authors' deaths, in creating new ones. The proponents might also have considered a reversion of all rights in the extended term to the descendants of the individuals who created the work, whether in a work-for-hire situation or not. Or they might have suggested at least prospective limitation of the work-for-hire term to 70 years, in the interest of harmonizing our law with that of Europe. The law professors who have signed this testimony are not in agreement about whether any such limitations might temper their objections to the bill. The absence of any sign that measures of this type have even been considered, however, shows that the proponents of the extension have not concerned themselves with the public cost of their proposal. Congress, as representative of all the people and not just the special interests whose voices are loudest, must seek to maintain an appropriate balance by very carefully weighing the costs against the purported benefits.

4. . H.R. Rep. No. 94-1476, 94th Cong., 2d Sess. 133 (1976).

5. . Of course, the market for many public domain works may often be small, with the result that competition is thin, or even nonexistent. This can allow, say, a book publisher to charge a price for a republished public domain work that is consistent with prices for similar types of books that are under copyright. Given the thin market, such a price may be necessary for this publisher even to cover production costs (including a normal return). This does not mean that the public domain status is irrelevant, because if a royalty were required in addition, such a book might not be republished at all.

It may also be that the works in question are not public domain works but rather derivative works based on public domain works. A new derivative work is, of course, itself copyright protected and can be expected to sell at the same price that the public pays for other protected works in that category. In this case, continued copyright protection for the underlying work may require sharing of the profits generated by the new work, with no economic benefit to the public in the form of a lower net price. As there is also no net economic cost to the public, however, the economic effect of lengthening the protection period requires identification of the parties sharing the monopoly. One of those parties is, by hypothesis, the new author, whose creativity has resulted in the new derivative work. The other will be the owner of the copyright in the underlying work, who may or may not be distant descendants of the original author. In this case, true concern for authors would seem to favor not lengthening the protection period.

Finally, as discussed below, when the underlying work remains under copyright, the real cost to the public may come from those new derivative works that are not created because of the new author's inability to negotiate permission from whoever owns the copyright 50 years after the original author's death.

6. . Robert W. Kastenmeier & Michael J. Remington, The Semiconductor Chip Protection Act of 1984: A Swamp or Firm Ground?, 70 Minn. L. Rev. 417, 459 (1985); see also Peter Jaszi, When Works Collide: Derivative Motion Pictures, Underlying Rights, and the Public Interest, 28 U.C.L.A. L. Rev. 715, 804-05 (1981).

7. . See generally Jessica Litman, The Public Domain, 39 Emory L.J. 965 (1990); David Lange, Recognizing the Public Domain, 44 L. & Contemp. Probs. 147 (1981). For an argument that copyright is also intended to accommodate users' rights, see L. Ray Patterson & Stanley W. Lindberg, The Nature of Copyright (1991), which includes a Foreword by former Congressman Kastenmeier.

8. . Nearly 50 years ago Professor Chafee pointed to examples in which the veto power of copyright in an author's descendants deprived the public of valuable works. Chafee, Reflections on the Law of Copyright: II, 45 Colum. L. Rev. 719 (1945). There have been press reports of refusals by the estate of Lorenz Hart of permission to use Hart's lyrics to any biographer who mentions Hart's homosexuality and of censorship by the husband of Sylvia Plath of the work of serious biographers who wish to quote her poetry. Professor Jaszi has provided examples of derivative-work films whose continued distribution has been limited or even suspended because of conflicts with the owner of the copyright in the underlying work. Peter Jaszi, supra note 6, at 739-40.

9. . 8 Macaulay, Works (Trevelyan ed. 1879) 199, quoted in Chafee, Reflections on the Law of Copyright: II, 45 Colum. L. Rev. 719 (1945), requoted in R. Gorman & J. Ginsburg, Copyright for the Nineties 307 (4th ed. 1993).

10. . 17 U.S.C.A. § 203.

11. . No human author can possibly receive anything more in exchange for terminable rights in his or her work under a life + 70 regime than under the current life + 50 regime. The reason, quite simply, is that no purchaser of copyright rights will pay anything for the "extra" 20 years of the term, because those supposed extra years can be freely terminated, along with whatever remains of the current period, before they ever begin. An exception is the right to continued exploitation of derivative works, which cannot be terminated. Even in this case, however, the maximum "extra" value to the transferring author is the present value difference between a 50-year and a 70-year protection period. Even for guaranteed income streams, this difference is around 5.4% (at an assumed 5% discount rate). That is, a guaranteed income stream of $1,000 per year for 50 years has a present value of $19,256 while the same stream for 70 years has a present value of $20,343. The purchaser of the derivative work right, however, will not be willing to pay anything close to this difference in present value, because of the overwhelmingly high risk that the derivative work created pursuant to the purchased right will have an economic life, like most works, far less than even the 50 years now afforded.

12. . Termination rights accrue 35 years after a grant by an author and expire 40 years thereafter. Because the extra 20 years that would be added by the extension to the protection period begin 50 years after the author's death, all termination rights with respect to any authorial transfer will either have been exercised or have expired.

13. . Council Directive 93/98/EEC (Oct. 29, 1993).

14. . H.R. Rep. No. 94-1476, 94th Cong., 2d Sess. 133-34 (1976).

15. . E.g., 14 Omnibus Copyright Revision Legislative History, House Hearings 1975 (Part 1) 133-34, 141-42 (testimony of Irwin Goldbloom, Deputy Assistant Attorney General, Civil Division, Department of Justice). Some believe that special constitutional problems arise from an extension of the period of protection for works already under copyright, because it recaptures from the public domain works that should be freely available under the "bargain" made at the time the work was created and offers no countervailing public benefit. They argue that the constitutional term "limited times" must be interpreted in terms of the constitutional goal to promote the progress of science and the useful arts.

16. . E.g., id. at 108 (testimony of Barbara Ringer, Register of Copyrights); id. at 120 (testimony of Joel W. Biller, Secretary for Commercial Affairs and Business Activities, Department of State).

17. . The Adventure of the Veiled Lodger was published on January 22, 1927, and The Adventure of Shoscombe Old Place was published on March 5, 1927. Robert Burt de Waal, The World Biography of Sherlock Holmes and Dr. Watson 13, 23 (1974). This same source lists other Conan Doyle stories as having been published in 1921, 1922, 1923, and three each in 1924 and 1926.

The undersigned are all university professors who regularly teach or conduct legal research in the fields of copyright or intellectual property.
Howard B. Abrams,
University of Detroit Mercy School of Law
Martin J. Adelman
Wayne State University Law School
Howard C. Anawalt
Santa Clara University School of Law
Stephen R. Barnett
University of California at Berkeley School of Law
Margreth Barrett
University of California Hastings College of the Law
Mary Sarah Bilder
Boston College Law School
Dan L. Burk
Seton Hall School of Law
Amy B. Cohen
Western New England College School of Law
Kenneth D. Crews
Indiana University School of Law - Indianapolis
Robert C. Denicola
University of Nebraska-Lincoln College of Law
Jay Dratler, Jr.
University of Hawaii William S. Richardson School of Law
Rochelle C. Dreyfuss
New York University School of Law
Rebecca Eisenberg
University of Michigan Law School
John G. Fleming
University of California at Berkeley School of Law
Laura N. Gasaway
University of North Carolina School of Law
Dean M. Hashimoto
Boston College Law School
Paul J. Heald
University of Georgia School of Law
Peter A. Jaszi
American University, Washington College of Law
Mary Brandt Jensen
University of Mississippi School of Law
Beryl R. Jones
Brooklyn Law School
Dennis S. Karjala
Arizona State University College of Law
John A. Kidwell
University of Wisconsin Law School
Edmund W. Kitch
University of Virginia School of Law
Robert A. Kreiss
University of Dayton School of Law
Roberta Rosenthal Kwall
DePaul University College of Law
William M. Landes
University of Chicago Law School
David L. Lange
Duke University School of Law
Marshall Leaffer
University of Toledo College of Law
Mark Lemley
University of Texas School of Law
Jessica Litman
Wayne State University Law School
Peter S. Menell
University of California at Berkeley School of Law
Robert L. Oakley
Georgetown University Law Center
Harvey Perlman
University of Nebraska College of Law
L. Ray Patterson
University of Georgia School of Law
Leo J. Raskind
Brooklyn Law School
David A. Rice
Rutgers-Newark School of Law
Pamela Samuelson
University of Pittsburgh School of Law
David J. Seipp
Boston University School of Law
David E. Shipley
University of Kentucky College of Law
Robert E. Suggs
University of Maryland School of Law
Eugene Volokh
University of California at Los Angeles School of Law
Lloyd L. Weinreb
Harvard University Law School
Sarah K. Wiant
Washington & Lee University School of Law
Alfred C. Yen
Boston College Law School
Diane L. Zimmerman
New York University School of Law
The undersigned is in agreement with the conclusions of this Written Testimony for substantially the reasons given.
Wendy J. Gordon
Boston University School of Law