PETER B. HIRTLE, Cornell University
58th president of the Society of American Archivists
N.B.: President Peter B. Hirtle presented an abbreviated version of this address at the opening plenary session of the 67th annual meeting of the Society of American Archivists in Los Angeles on Aug. 21, 2003.
In April of this year, the Washington Post reported that the papers of Bob Woodward and Carl Bernstein, the Post reporters who uncovered most of the Watergate scandal, had been sold to the Harry Ransom Humanities Research Center at the University of Texas for $5 million. The article in the Post noted that other collections had been sold for similar large amounts, including Susan Sontag's papers for $1.1 million, Francis Crick's papers for $1.3 million, the Zapruder film of the Kennedy assassination for $16 million, and some of Winston Churchill's papers for $18.4 million. The paper could have cited additional purchases, such as the $18 million that the U.S. government paid for Richard Nixon's papers in 2000, or the current offering by Sotheby's of some of Martin Luther King's papers that had been offered to the Library of Congress for $20 million and have been appraised for $30 million.
That archival records can have monetary value is not a surprise to any practicing archivist. We are custodians of records that are of incalculable value in the study of our history, culture, and values—and that can also be sold on eBay or in an auction house for substantial sums. I have heard it estimated that at my own institution, Cornell University, the library is the single largest asset that the university owns. Since Cornell has among its seventy million manuscripts items such as a copy of the Gettysburg Address in Lincoln's hand and some of James Joyce's manuscripts for Ulysses, it would be easy to imagine that most of the value in the library resides in the archival and manuscript collections.
In short, archives and manuscript repositories control and manage assets worth hundreds of billions, if not trillions, of dollars. And yet, archives are traditionally underfunded. Many archives have become "land poor": we have millions in capital assets, but lack money to hire staff, maintain the facility, or pay the utility bills.
Given the need for funds and the understandable (and applaudable) reluctance to sell assets, it is not surprising that many archives are seeking to derive revenues from their control over archival materials. The sale of reproductions of archival materials and the licensing of material for commercial use are becoming ever more important as possible sources of income for archives.
Archives, in seeking to draw revenue from the content that they own, are following in the footsteps of museums, which have long used licensing programs to augment their income. For some museums, licensing of the art in their collections has become a vital source of revenue that supports educational and curatorial programs, and managing intellectual property a major administrative task.
Efforts to digitize archival holdings have also contributed to an increased awareness of their possible economic value. The belief that some day digital content will generate substantial sums of money is growing. Lesley Ellen Harris, for example, a lawyer who writes about intellectual property issues for museums and libraries, has identified digital property as the "currency of the 21st century." Most archivists are aware of commercial digital photo collections such as those at Corbis, Getty Images, and the New York Daily News and wonder if they could generate similar revenues from their holdings. Clearly, the recognition that archival holdings are economic assets is increasing. We are aware that even the most insignificant scrap of paper in our archives is likely to bring something on eBay if we were to sell it, and many of us are hoping that through digitization we might be able to convert our fixed capital assets into liquid cash.
I believe that the economic exploitation of archival holdings can be implemented in a manner that supports the general archival mission of making the resources of the past accessible to the future. This desirable outcome is not preordained, however. A real risk exists that the desire to raise revenues through the commercial exploitation of archival holdings could conflict with our core principles and values.
This morning, therefore, I would like to discuss with you the nature of our ownership of archival holdings and the implications it may have for the exploitation of works. We often hear archivists talk about protecting "our stuff," but what is the real nature of archival ownership of documentary materials?
As many of you know, I am very interested in issues of intellectual property. Some of the concepts of intellectual property law, and especially the notion of the public domain, can inform our understanding of what it means to own archival records. I want to convince you that ownership of public domain materials is not like the ownership of a car or a house. We do not really own some aspects of the material in our archives—the public does—and we are doing a public disservice if we attempt to exert a monopoly-like control over that material. Thus, the complaints that one frequently sees on the Archives & Archivists listserv about the use of "our stuff" or the frequent requests for advice on watermarking or encryption technologies to control the use of digital images are misplaced.
To respect the public's interest in archival materials, we will need to carefully craft our licensing schemes. If archivists are careful not to compromise our own fundamental values while seeking to generate revenue, and if we respect the public's interest in public domain material, I think that we will be able to avoid much of the criticism that many museums have received over their jealous attempts to monopolize and control their collections.
Intellectual Property Law and Ownership
Let's think, then, about what intellectual property law can tell us about the nature of our ownership of archival records. First, copyright law draws a firm distinction between the physical ownership of an item and the copyright in that item. Just because you have a letter, or a photograph, or film, does not mean that you own the copyright in that work. My experience has been that most archivists know and understand this, though I am told that some first-year law students have trouble with this concept.
There are at a minimum four possible scenarios regarding the ownership of the physical and intellectual property found in archives. First, it could be that the archives owns neither the physical item nor the intellectual property of the item, as when an item is placed on deposit. In such cases, what an archives can do with an object is dependent upon the terms of the deposit agreement and is likely to be severely limited.
What if an archives actually owns a physical item, either via transfer from a parent institution, or through donation, purchase, or some other mechanism? Who can own the intellectual property? There are three possible scenarios when an archives has physical title to an object: the archives can also own the intellectual property in the piece, third parties can own the intellectual property, or the work is in the public domain. Depending upon which theorist you read, copyright in a work in the public domain belongs either to the public or to no one.
Why does it matter who owns the copyright in a work? Federal copyright confers on the copyright law some important rights. Among the rights given to the copyright owner are the exclusive rights to reproduce a work; to distribute copies of that work to others, either by sale or by lending; to prepare derivative works based on the original copyrighted works; and, in more limited cases, to display or perform a work.
Obviously, copyright law has tremendous implications on the operations of archives. We "display" photographs from our collections to users; we make copies of records for patrons; we offer to distribute copies by mail upon request; and we sometimes transcribe, edit, or publish works in our archives. When it comes to the digital realm, we do all this in a way that is easy for others to detect and observe.
Archivists who wish to generate revenue from their holdings will need to take copyright law into account. If an archives owns both the physical item and the intellectual property in the item, then it is free to exploit the material in any fashion consistent with its mission. An archives that owns both the physical manifestation and the intellectual property rights is no different than a publishing house, or a software company, or a movie studio, or any other business that licenses and sells the intellectual property it owns as part of its business.
What if a third party owns the copyright in works that are physically owned by an archives? Whether an archives can easily generate revenues from such materials is much less clear. Since reproduction and distribution are the exclusive rights of the copyright owner, you most likely will need the permission of the copyright owner to make and sell reproductions. It may also be possible to make reproductions by using one of the exceptions to the exclusive rights of the copyright owner found in the law, especially if the copyright owner is unknown or cannot be found. In almost all cases, however, any reproductions or distributions must be made without any direct or indirect commercial advantage. Rather than generating revenue for your repository, copying for the purpose of making revenue could in theory make your reproduction of the copyrighted work illegal and thus make your institution liable for substantial civil and criminal penalties.
The Public's Interest in Copyrighted Collections
There is one important exception to general exclusive rights of copyright owners. It is, I believe, the most unqualified exception in copyright law. Normally, the limitations on the rights of copyright owners, such as the "fair use," have limitations on the amount of a copyrighted work that may be copied or the amount of time one may retain the copy. Other limitations on the exclusive rights of copyright owners are restricted by format; different rules apply for text, audiovisual materials, and works of art. One exception, however, places no limits on the amount that can be copied, nor the time the copy can be kept, nor the format of material to be copied. And it is available only to repositories with unpublished materials. That exception allows archives and libraries to copy entire unpublished works for the purpose of preservation or for deposit for research use in another library or archives.
Why did Congress accede to the request for this exception submitted to them by Julian Boyd on behalf of SAA? No indication in the historical notes accompanies the bill explaining why, in the case of unpublished materials, Congress was so willing to ignore the interests of the copyright owner. The parallel section for published materials contains no such provision.
The answer seems to be that Congress recognized that the public's interest in having ready access to unpublished archival materials outweighed the rights of the copyright owner. The public's ability to be able to read unique unpublished material is so important that the interests of the copyright owner must give way.
The Public's Interest in the Public Domain
Public interest is clearest in the case of items in the public domain. Many of the materials in our archives are in the public domain. On the first of this year, every unpublished work by an author who died before 1933 and every unpublished work by an anonymous author or a corporate author that was created before 1883 entered the public domain, joining all the works of the federal government, which are automatically part of the public domain. Works in the public domain have no copyright, and the exclusive rights of the copyright owner, including the rights to make reproductions and distribute and display works, belong to everyone.
Some, including Congresswoman Mary Bono, have called for the abolition of the public domain. They do not understand why copyrights do not continue into perpetuity. The existence of the public domain is fundamental to the American system of copyright. Copyright is in effect a balance. The public grants to the creator of a work a limited monopoly to exploit that work as a reward for the act of creation. Without some type of monopoly grant, all works that are created would be in the public domain. The fear is that if this were to happen, authors would stop producing. In return for the limited grant of monopoly, the public receives new works of authorship, including books, movies, music, and art, that they may read, study, and learn from. When the monopoly term expires, the public is then free to use and reuse the once-copyrighted work. Disney's movie Pinocchio, for example, was released less than one year after American copyrights in Carlo Collodi's book expired, and many of Disney's other most famous movies, including Snow White and the Seven Dwarfs, Cinderella, The Hunchback of Notre Dame, and Alice in Wonderland, are based on works in the public domain. A former copyright owner can no longer control when or how a work is reproduced, nor object to any use the public may wish to make of the work.
The addition of a large percentage of the nation's archival holdings to the public domain is of unquestionable benefit to both the users of such material and the archival repositories that hold it. Such repositories can now contemplate ways of fully exploiting their holdings without having to worry about whether they are infringing on the rights of the copyright owner.
The Emergence of Quasi-Copyright Control over the Public Domain
Many repositories would like to maintain a kind of quasi-copyright-like control over the further use of materials in their holdings, comparable to the monopoly granted to the copyright owner. One strategy that many museums and some archives use to exert such quasi-copyright control is based on their ownership of the physical manifestation of a once-copyrighted work. Kathleen Butler discusses this in an article that I would admire even if it was not entitled "Keeping the World Safe from Naked-Chicks-in-Art Refrigerator Magnets: The Plot to Control Art Images in the Public Domain through Copyrights in Photographic and Digital Reproductions." She writes, "Object owners, by controlling physical access to the objects, have the opportunity and power to govern how reproductions of those object are made, used, and licensed." For example, in museums, photography may be forbidden entirely, or the use of tripods might be banned so that no publication-quality photographs can be taken.
According to art historian Robert Baron, frequently the principles that guide how museums allow reproductions of public domain materials to be used do not arise from the educational and public mission of the institution, but instead are products of the museum's desire to make income from for-profit publishers or to protect its reputation by hoarding its collection. "In these cases," Baron concludes, "the museums are prisons and the pictures are prisoners serving to bolster the self image of the museum."
In addition to using their control over the conditions of access to unique physical items to control subsequent use of those items, some museums (and some archives) try to enforce a monopoly on reproductions of the unique public domain items in their collections. Reproductions are made available to researchers only if they sign agreements that limit what can be done with those reproductions. In an online environment, users are often required to "click through" an agreement that regulates the use of images and documents that would otherwise be in the public domain. In one typical, though particularly thorough, online licensing agreement users must agree to thirty-eight conditions before they can view the site. They must agree not to use the material found on the site for personal or financial gain without permission, even if the work is in the public domain. The material may not be distributed or duplicated, rights normally reserved to the copyright owner, without the permission of the institution that holds the original material. Copies of the digital documents on the site are available for purchase, but again permission is needed for commercial use, publication, manipulation, display, or distribution. In the words of Robert Baron, the documents have become "prisoners."
I would argue the existence of at least four categories of reasons—legal, principled, ethical, and practical—why efforts to use our physical control of public domain documents to impose a quasi-copyright control over them are doomed to failure.
Legal Arguments against Control of Public Domain Material
There are a number of legal reasons why the license agreements that govern access to public domain objects and the use of reproductions of those objects may be suspect. First, there is a growing legal recognition that items can be part of our common cultural heritage, and as such cannot belong to any individual or organization. A part of them belongs to everyone. As Joseph Sax notes in his very thoughtful volume, Playing Darts with a Rembrandt: Public and Private Rights in Cultural Treasures, "some objects are constituent of a community, and that ordinary private dominion over them insufficiently accounts for the community's rightful stake in them." Sax, for example, argues that because great works of art belong in part to the public, private collectors should be required to periodically loan these works to public institutions so that the public may have free access to them.
We are all familiar with zoning laws that impose upon the unfettered interests of private property owners. Historic preservation laws may limit how a building can be altered, and in places like California, some property owners must allow the public access to the beach, even if it means crossing private property. In effect, the public has a share in the ownership of the historic house or the beachfront property. The same may be true of the documents in our archives. They belong to the institution, but they also belong to the community, and the community may have the right of public access to those works.
Second, if the public owns the copyright in public domain works (as opposed to no one owning those copyrights), then the public may have a legal right to gain access to and copy public domain items. In the interesting case of Reid v. CCNV, the court ordered that the Community for Creative Non-Violence (CCNV) must grant John Reid access to a sculpture so that he could make a copy of it. Reid had made the sculpture for CCNV, but retained copyright in the work. The court said that it did the copyright owner no good to own the copyright if he or she could not have access to a unique original item to be able to make a reproduction of it. It could be argued that a logical extension of this ruling would be that the public should be free to make reproductions of public domain works in institutions regardless of any access restrictions since, in effect, the copyright in the work belongs to the public.
Third, some legal scholars have suggested that efforts to replicate copyright protections in contract law are suspect. The section in the copyright law on "preemption" states that with regard to any of the "rights that are equivalent to any of the exclusive rights within the general scope of copyright," federal copyright law takes precedence over any state laws. The issue, as yet untested in the courts, is whether repositories can use contract law to re-establish the exclusive rights of the copyright owner via state contract law once federal copyright protection has expired.
Lastly, many repositories place copyright notices on their reproductions of public domain works. In Bridgeman v. Corel, however, the court held that exact photographic copies of public domain works are not themselves copyrighted because they are not original. Making an exact photographic copy may require great skill and effort, but that alone is not enough to warrant copyright protection. Placing a specious copyright notice on these reproductions may actually place a host institution at risk. Including a copyright notice that one knows is false on a work is a criminal offense punishable by a $2500 fine. To date no archives has been charged with this offense, but in general it is sound institutional practice to obey the law.
Archival Principles and the Control of Public Domain Material
Laws about access to cultural heritage objects, the public's apparent right to reproduce items whose copyright it owns, preemption, and strictures against fraudulent copyright notice are all legal reasons why attempts to control access to public domain works in our possession may be invalid. These reasons share at root a sound principle based on the purpose of copyright ownership. The public, remember, grants to authors a time-limited monopoly to exploit their creations, after which the work is supposed to belong to the public. Museums (and archives) that seek perpetual control over the use of a work are in effect saying that act of stewardship of a work is more important than the act of creation. Society need only grant creators a temporally limited monopoly, but the interests of the repository need to be protected forever. Do we really believe that ownership is more worthy of reward than creation?
Let's not forget as well the lofty purposes for which archives exist. Our task is to take care of the objects placed in our care to the best of our abilities. We have the responsibility to pass them along in good condition to our successors. We also have the responsibility to distribute knowledge about our holdings to the public. We make the historical record available to society to maintain, as John Fleckner taught us, a just society, to protect the rights of citizens, to serve as a check on tyrannical government, and to ensure to us—individually and collectively—the ownership of our history. What will happen to these ideals if archives become only assets ripe for personal or institutional exploitation?
I have been told that a proposal was floated recently at one of the very best research universities that access to materials in the archives and manuscript departments should be restricted to faculty and students of the institution. While common practice fifty years ago, this sort of thinking has until now been mostly rejected. Administrators at this university viewed the archives as a valuable university asset that should be exploited to the benefit of the university and not as a public good whose stewardship the university had assumed. The proposed commodification of the archives reversed almost fifty years of progress in archival principles.
Ethical and Practical Issues when Controlling Use
In addition to the legal and principled reasons to avoid claiming too much control over public domain works is the ethical obligation we may have to donors. When I was a manuscript curator, I spent much of my time trying to convince people to donate personal archives, along with the copyright in those archives, to the repository where I worked. The reasons I used were that such a donation would support scholarship and research, and would be in the public interest. Never did I discuss the potential ongoing monetary value of the donation, nor indicate that we sought the papers to be able to generate revenue for ourselves from them. Such concerns might not matter with material that has been purchased, and the archives may have more freedom to exploit such material. But with material that has been donated, the conditions and terms of donation should inform any decision about marketing reproductions of materials from the donation.
Lastly, on a practical level, efforts to monopolize public domain reproductions are likely to fail. Even if, for example, a license that forbids further reproduction of a reproduction made from a public domain work were binding, the contract is only between the institution and the original user who requested a reproduction. If a copy of the reproduction should fall into the hands of a third party, they may be able to reproduce it with impunity; all the repository can do is bring action against the individual who made the unauthorized reproduction (assuming he or she can be found). Some institutions have sought to use watermarking, encryption, and other technological measures to limit use of digitized resources. These solutions are imperfect, expensive, and may still require legal action on behalf of the repository. And they can be easily subverted if the repository sells hard copy reproductions of works that can be easily digitized on an inexpensive scanner.
The conclusion we must draw is inescapable. Efforts to try to monopolize our holdings and generate revenue by exploiting our physical ownership of public domain works should not succeed. Such efforts make a mockery of the copyright balance between the interests of the copyright creator and the public. They ignore the public's ownership interest in our holdings, may be legally unenforceable, and, depending upon the implementation, may actually be criminal.
Yet it is undeniable that, to be good stewards of the public's documents, archives need more revenue. How can we keep from being "land poor"? How can archives generate the revenues they need if they cannot exert a quasi-copyright monopoly right over their holdings?
Responsible Revenue Generation
I have two suggestions to offer as to how archives can responsibly generate revenue from their holdings. First, when making copies of public domain material, I would encourage all repositories to charge whatever the market will bear and their mission (and implicit agreements with donors) will tolerate. You should not impose restrictions on further use of those reproductions, and so someone might elect to compete with you by offering copies of your reproductions for less than you charge. But again, I would remind you that "our stuff" is also "their stuff" when it is in the public domain, and that the only real right you have is the right to sell reproductions of your documents. For that first reproduction you should feel comfortable charging what you want.
Bernard Reilly of the Center for Research Libraries in Chicago has provoked my second suggestion. Reilly has proposed that libraries and museums should be more like Enron. I would suggest that archives, too, should be more like Enron, but for slightly different reasons than he has suggested. And no, I am not talking about shady accounting, stock manipulations, or the illegal destruction of records as the path archives should follow to deal with their budget woes.
Think about Enron for a moment. What business was Enron in? It was one of the world's largest energy and commodities companies—but it did not own any oil wells, or power plants, or other sources of energy. Initially begun as a pipeline company, Enron grew tremendously after it stopped trading in actual material assets and instead began to trade in the rights to resources and later in the value of future and as yet undiscovered resources. By the time Enron collapsed, it was less an energy company than an investment association trading in energy futures, bandwidth, and even the weather. Other people owned the assets; Enron grew rich marketing and using its information about those physical assets.
I hope the parallel with archives is obvious. We do not entirely own the physical public domain items that are found in our repositories; while they are private property, they are at the same time public goods in which the public has a stake. How can we make money from objects that we do not entirely own and cannot monopolize? The answer lies in marketing, like Enron, not the physical assets we have in our collection, but the information that we have about our records. The reason people should order reproductions of public domain works from an archives's Web site is not because we have used license agreements to try to restrict future use of these works, and thus maintain monopolistic control over the content, but because we can offer information and services that the user cannot find anywhere else. It might be complete and accurate metadata, or the provision of context, a guarantee of authenticity, an assurance that the resource will be available at a future date, or a system that makes it easy to locate, order, and receive works that are of interest.
In an interesting recent article in the Communications of the Association of Computing Machinery, Alan Karp, a research scientist at Hewlett-Packard, suggests that the content industries should be more like the pornography industry. Pornographic Web sites, which normally require a subscription to access, are among the most visited Web sites in spite of the fact that the individual images are easily copied and are often available for free somewhere else. Pornographic Web sites, he suggests, have succeeded because they have provided added value. They have slick interfaces, good performance because of their substantial investment in infrastructure, and easily accessible metadata tied to user needs. "Adding value," Karp notes, "gives customers a reason to buy from you instead of directly from your suppliers."
I have one example of the role of value-added services in marketing. The Cornell University Library recently created a Web site to commemorate the seventh million volume added to the collection. The book was Alexander Gardner's Photographic Sketch Book of the War, published in Washington, D.C., in 1865 and 1866. One of the pages in the site is devoted to plate 23, "President Lincoln on the Battle-field of Antietam." It includes information on the context for the photograph, identifies the photographer and the photographic processes used, and offers other valuable information. No information is given on how to order a copy of the photograph, but an astute researcher who examines the rest of the site might be able to learn that a printout of unspecified size could be had for as little as $20.
A copy of the same photograph appears in the New York Times online store. Here the photograph is identified as "Lincoln and Troops at Antietem (sic), 1862," and Alexander Gardner is credited as the photographer. There is no context for the picture, no indication that it comes from a printed volume, and no information on the original photographic process. What it does have is an easy-to-use order form: prints run from $195 to $495 unframed, and $340 to $745 framed.
Why, you might ask, would anyone pay these prices when a fairly high quality digital copy is available on the Cornell Web site, at the Library of Congress's American Memory Web site, and at several other easily identified sites on the World Wide Web? Print copies at much lower prices are available from all of these sites as well.
The answer is that the New York Times provides a value-added service that the historical repositories do not—easy searching, easy ordering, online examples of the frames that can be ordered with the prints, and so on. The service it provides, and not an artificial monopoly over the content that it is providing, is the Times's business model. As a recent study of business models for the development of digital cultural content noted, for cultural heritage institutions to deliver online digital cultural content successfully in the future, they will need to develop a more commercial approach to the marketing of the products, especially in the area of understanding user needs and demands. An added benefit is that all of the value-added services that an archives markets to its users do belong to the archives and can be protected in a manner that public domain documents cannot be.
Enron, pornographic web sites, and the New York Times are the models that archival asset management systems of the future should emulate. The real assets in an archives are not the holdings, but the skills, talents, knowledge, and abilities of its trained archival staff. It is these archival assets that archival repositories must promote.
This paper is dedicated to the memory of John C. Coolidge, my classmate and friend of over thirty years. Four days before I presented the paper in Los Angeles, I had the honor of speaking at his funeral. We continued to share a love of history even as John became a lawyer and I became an archivist. John also maintained throughout his too-brief life a commitment to protecting the rights of the less fortunate. This essay would be the stronger if it could have benefited from his perspective.
© 2003 Peter B. Hirtle. This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike License.