Director of the Law Library &: Professor of Law, Georgetown University Law Center
I am delighted to be here to talk about copyright issues in electronic publishing. Actually, I must say that I have been looking forward to the morning because I am anxious to learn more about your concerns and to help, where I can, allay some of the fears about copyright in the electronic environment.
I am relatively used to speaking about copyright to library groups because I have a pretty good sense of what they know -- and what they don't know -- about the issues. Many of you, on the other hand, are in the business, dealing with copyright every day, and I assume you are intimately familiar -- in some ways, more familiar than I -- with the details, the problems, and the frustrations of copyright administration.
To find out what this group might want to hear about, I did a little research by reading periodical articles and by following several different discussions on the Internet. From those sources, I have tried to distill out a few concerns that I think might be of interest.
First, some writers still seemed willing to assume that the copyright system is fatally flawed and must be completely overhauled to work in the electronic environment. I will, this morning, sound a cautionary note about proceeding in that way. Second, there was concern about whether certain kinds of material can be protected when distributed electronically, and if they can be protected, how the relationship between an author and an electronic publisher should be structured. Third, there was some uncertainty about using computer technology to distribute such works as preprints, abstracts, and citation information. Fourth, how does fair use work in the electronic environment where enforcement is so very difficult. Fifth and finally if, when all is said and done, you are still not satisfied with copyright protection, how does the Act establish a framework that allows the publisher to move beyond the bounds of copyright to find other ways to protect themselves. My goal for this morning is to try to address each of those concerns at least briefly.
In an article in Publishers Weekly last summer, the author suggested that the copyright system was close to collapse and that a complete re-working of the Act might be necessary. Since I assume you all read Publishers Weekly, I thought this might be an issue of some concern.
According to the article, problems with fair use, changing concepts of authorship, and the protection of works from their creation, rather than from publication, have all introduced serious elements of uncertainty into the system. The author cites with approval a study from the Office of Technology Assessment that called for completely new concepts to replace the existing copyright system. In their study, the OTA explained that copyright worked when an intellectual work was represented in something tangible, a book, an article, or some other physical item. But the concept of a stable "work" that can be protected is meaningless, they said, in an environment where ideas are nothing more than a string of bits and bytes in cyberspace that can be transmitted instantaneously for use anywhere else in the world with or without making a physical "copy".
This author of the article has picked up on a discussion that has been going on for several years now, but many experts disagree with the conclusions of the OTA, believing instead that copyright can grow and be adapted to meet the challenges of new technology. Those on this side of the argument point out that copyright has adapted to new technologies before, to include protection for sound recordings and motion pictures, for example. They see no reason why the law cannot again expand to meet the challenges of new information technology.
At the foundation, however, the most fundamental problem with the approach of the copyright critics is that no-one has yet devised an alternative that maintains the balance between the rights of creators and the rights of users of information. They call for a new approach, but they have yet to suggest an approach that provides incentives for participants in the creative process to continue to create and write and publish.
It might interest you to know that this issue was considered soon after the OTA Report was issued by a group known as the Network Advisory Committee of the Library of Congress. That group consists of librarians, network administrators, database vendors, and electronic publishers. After a careful review of the issues, spread over two meetings a year apart, the Network Advisory Committee concluded that a complete overhaul of the copyright system was neither necessary nor desirable. I believe their conclusion reflects the careful balance of competing interests that was built into the Copyright Act and a fear that a sweeping change would create more uncertainties than it would be likely to resolve. If I were a publisher, I believe I would recognize that the existing system gives me substantial rights, rights that not only allow me to control the distribution of a work, but also creates a framework that allows me, for example through the use of license agreements, to establish limits on the actual use of a work in the electronic environment. If I were a publisher, I would not be in a great hurry to abandon a system that had given me so much control. Rather than looking for completely new models, it seems to me that the better approach is to view the new technology as an opportunity to create new markets that will generate a reasonable return to the author and publisher. More about that later.
For myself, until I hear of a plausible alternative to the copyright system, I have preferred to approach the topic cautiously, exploring with some care the interface of copyright and new technology, to understand how copyright can work in the new environment and to understand where the limits really are. The Copyright Act is more likely to be amended and changed incrementally to deal with carefully defined, narrow problems than it is to be radically overhauled, in favor of a whole new approach that, by definition, would be untested and uncertain.
With that as background, let me turn to some of the more specific concerns. Here, we will look to see how the Copyright Act is actually working. Along the way, we might identify some of the specific areas needing attention.
Before going any further, however, I must stress that it is important to keep in mind that--for the most part--the Copyright Act was written to be technologically neutral. There are a few sections that are directed at particular technologies, and some of those affect our discussion today. But most of the sections are written generally to apply to any technology in which an intellectual work might be represented, and you can get a good sense of the answer simply by asking how the same situation would be handled in the paper environment.
First, databases. Databases are a special case, because facts are not copyrightable but a compilation of facts might be. If a database or a compilation is protectable, the protection does not extend to the underlying facts, but only to the contribution made by the compiler. This is likely to include the selection of data, the arrangement of the information, and its presentation. With the growth of online databases--especially those that are automatically compiled -- many have questioned the degree to which these files may be protected.
The Supreme Court helped us with the analysis in a case dealing with the telephone book white pages--Feist v. Rural Telephone Services[1], decided March 27 of last year. Although the Court in that case resisted the temptation to give a sweeping pronouncement about databases in general, the case does provide a point of departure for the analysis of future cases.
In Feist, the Court refused to uphold a claim of copyright in the white pages, saying that they lacked sufficient "originality". As you know the Act clearly provides protection only for "original works of authorship". The Court in Feist found that originality was a "bedrock principle" of copyright. The threshold, they said was relatively low, but nonetheless, the white pages did not meet it. The Court explained that originality might be found in both the selection and arrangement of facts in a compilation or database. But they explicitly rejected the so-called "sweat of the brow" rationale that permitted protection of databases based simply on the hard work or the investment made in the compilation. This, they said, "flouted basic copyright principles" by protecting factual information without any element of creativity.
The Feist case clearly points the way to analyze future attempts to copyright databases or compilations of factual information. The Courts are now instructed "that in determining whether a fact-based work is an original work of authorship, they should focus on the manner in which the collected facts have been selected, coordinated, and arranged....the statute dictates that the principal focus should be on whether the selection, coordination, and arrangement are sufficiently original to merit protection."
No doubt, this decision is troubling to many publishers who have gone to great trouble and expense to create large scale compilations. Indeed, the article in Publishers Weekly last summer claimed that "[F]or years, a primary focus under copyright has been the right to compensation for effort." Although the courts were by no means agreed on that principle, the Supreme Court's decision in Feist makes it clear that effort or investment alone are not sufficient to justify copyright protection. A publisher who seeks to make available a compilation of facts must go further and insure that they can show some element of originality in the work, in selecting, coordinating or arranging the data.
What about regular written works? A culture of free dissemination and re-dissemination has grown up among regular users of the Internet. Why are people so cavalier about the rights of others? Do they really not care, or do they not realize what they are doing? I believe it is the latter, and that most people base their behavior on assumptions that derive from the norms of the old Copyright Act, not realizing that the law is now different.
As this group surely knows, under the old Act, protection for a work came into being when it was "published with notice". Prior to publication, a work was protected by the states under a common law doctrine known as the right of first publication. Both concepts -- "publication" and "notice" became firmly embedded in the law of Copyright. If a work was found to have been "published" without "notice" it was then in the public domain. Publication was not defined in the old Act, but it was generally held to mean when a work was "placed on sale" or "publicly distributed"[2] by the owner.
This, then, is the way many of us are used to thinking about copyright. As a result, we are likely to conclude that much of what flows over the Internet must be in the public domain. After all, except for those of us in relatively small, private electronic conferences, the works have been made available to the public and the author has not -- for the most part -- signaled any intention to protect the work by applying a notice of Copyright.
However, the new Copyright Act, as now amended, has changed both of the basic preconditions for protection. The 1976 Act provided that copyright exists from the moment of creation[3]. Protection no longer must wait for some later action such as publication; it exists from the beginning. In addition, although the 1976 Act originally had a notice requirement, when the United States joined the Berne convention in 1989, the notice requirement was dropped. Although the copyright office still encourages copyright owners to apply the notice, it is not required. Furthermore, the definition of what is protected is technologically neutral; it does not say that a work has to be printed; it only has to be "fixed in a tangible medium of expression, now known or later developed, from which [it] can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device." This section is so general that it was clearly intended to include all writings, whether in machine readable form or not.
Taken altogether, it should be apparent that an original work of authorship, created since 1989, should be presumed to be protected, whether it is printed on paper or distributed over the Internet. This is true whether or not there is a notice of copyright on the work. Prudent authors, however, would be well advised to take at least two, and better yet, three steps. First, they should make sure that the work is fixed, to a disk, to paper, or whatever. Second, they should be sure that a notice of copyright appears on all versions of the work distributed by any means. Although the notice is no longer required, it will signal the author's intentions. Finally, the author should register the work with the copyright office.
These same comments apply to electronic journals. As far as the copyright law is concerned, there is no distinction between a journal distributed electronically and one distributed in paper. Of course, it needs to be "fixed" somewhere. And again, although notice is no longer required, if a publisher wants to protect its work, it would be prudent to insure that the notice appear prominently -- at least on the first screen, and preferably and the beginning of each article -- within the journal. Moreover, if printing is permitted, the notice should appear on all printed copies.
But what about the publisher's relationship with authors? As should be apparent, the author has all the rights in his or her work from the time it is created, and the publisher acquires only such rights as the author transfers. Compounding the problem is a recent movement among scholars and universities to retain copyright for themselves. This new trend is a reaction to two things: first, the escalation of serials prices far beyond the pace of inflation, especially among for-profit serials, and second, a perception among scholars that many journals are unwilling to grant back a right to scholars to use their own work. Among scholars, these actions from some publishers have resulted in very large price increases to university libraries and an inability of an author to use or authorize the use of his or her own work.
I understand that neither of these problems has been caused by the journals represented here. Many of you operate on a break even basis, at best, and many scholarly journals have been very willing to grant back certain rights to authors. Nonetheless, if the trend among scholarly authors takes hold, you will need to be sure that you have secured sufficient rights to insure that you are able to carry out your primary functions. No doubt, you are more familiar than I with what you need to carry out your work. But I would suggest that you need to insure that you have the right to publish the work as part of the issue of the print journal and to authorize the reproduction and distribution of that journal in print form or in microform. In addition, you should obtain explicit permission for the electronic distribution of the work. By way of example, I thought I would quote from the publication agreement recently adopted for the Law Library Journal:
[Note here that there is no explicit permission granted for reprinting the journal or for issuing it in microfilm. One would have to hope that the phrase "reproducing and distributing your contribution as part of the issue of the Journal" would cover it. But it would be better if it were explicit.]
The document follows with the standard warranty that the work has not been published before, a statement that the author will credit the original publication if it is later republished, and a policy statement on copying for classroom use that allows the journal to respond to such requests. Then comes the following:
This statement clearly allows the Journal to proceed with publishing the work, both in standard print form and in a variety of electronic formats. Although you may find ways in which the statement should be refined, I commend this to you as a starting point for insuring that you have the rights to publish the work in the ways in which you want to develop.
I will not spend a lot of time on this because I believe it is one of those areas where the fact that the distribution is done electronically does not affect the outcome.
First, preprints. Although traditional print publication is still important, it is becoming increasingly common, especially in the scientific disciplines, to distribute advance copies of works electronically. Obviously, this is an extension of an even older tradition of making advance copies of works available by print means. This tradition developed because of the importance of timeliness in some fields together with the long lead times needed for regularly scholarly publication. From a copyright point of view, there is little that is different about the fact that preprints are now distributed electronically. Of course, for any such distribution, the author should signal their intention to protect their rights by applying a notice of copyright. The electronic preprint distributor should also obtain written permission to distribute the work electronically, even if it obtains no copyright. In addition, when the author does sign an agreement with a more traditional publisher, he or she should disclose that advance copies of the work were distributed over the "preprint network".
What about the distribution of abstracts and citations? I gather that several different listservs are now being used to provide indexing, abstracts, and other reference information. Again here, the fact that the information is distributed electronically does not really affect the analysis of the issues. It is interesting, however, that some of these databases are being created automatically since text analysis programs may be used to create abstracts and to extract citation information into a bibliographic database. Query, how much originality can be shown when a database is created automatically?
When I looked into the issue of indexes and abstracts, I could find no evidence of any litigation covering the production of such works. The essential issue is whether indexing and abstracting are derivative works that infringe the rights of the original publisher. Traditional indexing and online citation systems are essentially pointers to the full work. They record the author, the title, and perhaps a few subject headings to each work in question. Few would seriously contend that such references interfere with the rights of a publisher; indeed, as a system of pointers, the index actually stimulates expanded use of the original.
The situation with abstracts is somewhat less clear. Here, the substance of the work is likely to be conveyed, and the answer is likely to turn on the particular case. If the abstracts in question are essentially just a slightly expanded pointer system, then it seems likely that they would not be found to be an infringement. If, however, the abstracts were more substantial -- more like a digest--eliminating the need to go to the original, then I think a court might find it to be an infringing derivative work. Obviously, there is no bright line test, but in general, if the abstracts are sufficiently detailed to stand in the place of the original, then I would worry about publishing them without permission. If, on the other hand, they are relatively brief references to the original, there is likely to be no problem.
A separate question, of course, is whether the resulting database of citations or abstracts is, itself, copyrightable. Our earlier discussion of the Feistcase provides the proper way of thinking about such a publication. If the database exhibits no originality -- just a stream of information fed from some other source in simple alphabetic order -- then copyright is unlikely. If, however, creative elements are present in the selection of articles or the selection of journals covered, in the creation or assignment of subject headings, in the writing of abstracts, or in the way in which the entries are ordered or laid out, then copyright is possible.
As you know, under the Act, the owner of a copyrighted work is given five exclusive rights. These are:
The first right is the right to make copies. For the most part, people understand what it means to make a copy of a work. Any permanent copy on paper or in machine readable form that can be used in the same manner or in lieu of the original is a potentially infringing copy. Some have worried about whether they are making an unlawful copy when they call a work into memory or display it on the screen. In the case of software, this problem is dealt with by Section 117 of the Act which says that it is not an infringement to make a copy of the program provided that the "new copy is created as an essential step in the utilization of the computer program in conjunction with a machine and that it is used in no other manner."
Although section 117 does not apply to literary works generally, its principles are reasonable and would almost certainly be extended to the lawful use of an online literary work. If someone was using a lawful online copy of a work, they would not be prevented from reading it simply because a transient copy had to be made into RAM or onto a screen. There seems little doubt that this would be found to be a fair use. The same principle would not apply, of course, if the copy was not lawful or if the copy was used to create a permanent copy in any form.
Apart from the rights to make and sell copies, the right that is most relevant to today's discussion is the last: the right to display a work publicly. Although reading a lawful copy on a CRT is not likely to be a problem, if the copy is not authorized by the Copyright owner, the fifth right--the right to display a work publicly is likely to be implicated. The statutory definition of "display" makes it clear that a CRT or other similar device is included. Similarly, public is defined as "a place open to the public or where a substantial number of persons outside of a normal circle of a family and its acquaintances may gather." It also includes transmission to the public...whether members of the public...receive it in the same place or in separate places and at the same time or at different times. This definition is certainly broad enough to include displays in libraries and is probably broad enough to include transmissions that might be received on a personal computer in a home or office.
This right -- the right to display a work publicly -- is written in a way that is technologically neutral. But, it is clearly intended to prevent the online storage and dissemination of information without compensation to the copyright owner. This conclusion was made explicit in one of the studies done by the Copyright Office as the new Act was being written. They said:
The use of closed and open-circuit television for presenting graphic and textual material to large audiences -- could, in the near future, have drastic effects upon copyright owners' rights. Equally if not more significant for the future are the implications of information storage and retrieval devices; when linked together by communication satellites or other means, these could eventually provide libraries and individuals throughout the world with access to a single copy of a work by transmission of electronic images. It is not inconceivable that, in certain areas at least, "exhibition" may take over from "reproduction" of "copies" as the means of presenting authors' works to the public, and we are now convinced that a basic right of public exhibition should be expressly recognized in the statute."
That was written in 1965. Today, those words protect the electronic publisher, by making it plain that electronic distribution can only be made with permission -- and possible compensation -- to the copyright owner. I am not saying, of course, that electronic libraries cannot be built. I am saying that if they are built they must take into account the legitimate rights of the copyright owners.
If all this is so, where does fair use come in?
Fair use is a judicially created rule of reason that permits limited copying of protected works. The 1976 act incorporated the judicial doctrine as it had developed to that time.
...the fair use of a copyrighted work for purposes such as criticism, comment, news reporting, teaching...scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include
the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes
the nature of the copyrighted work;
the amount an substantiality of the portion used in relation to the copyrighted work; and
the effect of the use upon the potential market for or value of the copyrighted work.
Under this doctrine, each case has to be examined on its own terms against the four criteria. Some commentators have suggested that, of the four, the most important is the last, dealing with the economic impact of the copying. Although one cannot legitimately read out of the statute the other three criteria, it is clear that those situations which negatively affect the revenue stream of the copyright owner are less likely to be seen as fair use than those which do not.
The purpose of the copying has also proven to be important in deciding these cases. The SONY case (SONY Corporation of America v. Universal City Studios, Inc.[4]) created a presumption against commercial copying. That reasoning was followed in the Kinko's[5] decision which found no fair use where a store was creating anthologies of readings to sell to students. This principle was extended by the recent Texaco[6] case. Whereas previously, it was the copying itself that was commercial in character, in Texaco, the copying was internal to the company, but the fact that the researcher was in a company engaged in a commercial enterprise caused the court to find no fair use. Essentially, that court read fair use out of the statute in the for profit environment, and the decision is now being appealed.
Interestingly, the SONY case also held that the practice of time-shifting for video recording was permissible under fair use. The narrow holding of SONY permitted private home videotaping of complete works for later viewing and ultimate erasure. In reaching this conclusion, the Court emphasized the private, noncommercial character of the use. The copy was not a substitute for a purchase; it was a substitute for viewing at a particular time. This suggests that the making of private non-permanent files, even of complete works lawfully distributed over an online database, for later review and ultimate destruction might qualify for fair use provided it met the other criteria. After Texaco, however, even those principles could probably not be applied in a profit-making company.
Actually though, you and I know that in the online world people don't limit their copying to such a time-shifting function. Frequently, items are copied and forwarded to someone else on the network. Sometimes, through the use of distribution lists, such forwarding is done to hundreds or even thousands of individuals simultaneously. Because such actions are real, publishers are legitimately concerned about the enforcement of their rights in the electronic environment. How might I suggest you look at this? Because of the flexibility of the technology, it is clear that a certain amount of private copying is inevitable. On the other hand, I also believe that there are compelling reasons why publishers will disseminate their material in this way anyway.
First, it has been suggested that the cost of production and distribution of an electronic publication is substantially lower than the cost of a print publication. Although there is obviously an irreducible base below which the costs cannot fall, nonetheless, the production and distribution costs can be substantially reduced. At the other end, electronic distribution opens up substantial new markets, bypassing libraries, and reaching individual researchers directly. If an appropriate mechanism is developed for tracking and charging for the usage of a work, these new markets can potentially generate substantial revenue.
So, it seems to me, electronic distribution carries with it the risks of some abuse. But it also carries the potential for new markets and new sources of revenue. As long as the abuse is not too bad, I think most publishers will find it advantageous to move into the new market.
At base, fair use is about being reasonable. It has been said that what fair use is all about is "does the use in question deprive the copyright owner of a reasonably expected economic reward?" Under older distribution models, a certain amount of private use clearly did not interfere with the expected income stream of the publisher and thus would clearly be fair use. That may change, of course, when the pattern of distribution evolves to reach individual users directly.
My suggestion about dealing with this series of problems, then, is much the same as in the paper environment. Publishers need to develop the mechanisms for controlling access to their database and for assessing appropriate charges. No doubt, with an appropriate mechanism in place, most users would get the information by coming to the source. Publishers should probably ignore as de minimus the isolated individual who makes a single copy for himself or a colleague. On the other hand, they should not ignore large scale storage and re-distribution of electronic journals, whether done by a commercial vendor, by someone who allows anonymous FTP of full, protected works, or -- dare I say it -- even by a library acting without regard to the provisions of section 107 or 108 of the Copyright Act.
The beauty of copyright, and the reason it should not be abandoned, is that it gives the copyright owner substantial discretion concerning the making and distribution of copies. In other words, if you can reach agreement with a purchaser, you can develop terms and conditions for the distribution of the original copy that are far narrower and more precise than the general provisions of the copyright law.
Before you let someone gain access to your database, in all likelihood you will have them sign a contract. That contract will have terms and conditions about who can use the information, at what price, how many copies can be made, and so on. Moreover, the contract can have a definite term so that both parties can evaluate the costs and benefits on a regular basis. There has been a substantial amount of discussion about the development of such license agreements on the internet. Interestingly, almost everyone seems to think that licenses are an excellent way to introduce greater certainty into the situation. The one area -- other than price -- where there seems to be some concern is who is allowed access. When such licenses began to be developed, they were site-licenses, and they tended to limit usage to a particular place -- the library building, the law school, or maybe the campus. But librarians want to do what they have always done--serve their community. And to them, in the new environment, the location no longer matters. They believe students or faculty should be able to dial in from home, or even long distance, when they are on vacation.
Cognizant, however, of the publishers' concern to limit access, the librarians have proposed an alternative to the site-license. They have asked publishers to consider instead a concurrent use license. The concurrent use license would limit the number of simultaneous users to the system, but would then leave it up to the library how to control access beyond that. This seems like a reasonable proposal, intended to recognize the publishers' concern, while helping to make the library without walls a reality.
No matter how this and other issues are worked out, what is clear, is that the copyright system provides a framework, within which many rights are defined, but which leaves open the possibility of creators, publishers, libraries, and users working together to negotiate other terms whenever they deem it appropriate.
What I have tried to do here was to talk about the issues that I believe are of interest to many of you -- the basic stability of the copyright regime, the nature of publications in the electronic environment and the relations between an author and an electronic publisher, fair use, and licensing. Although I know this may have been somewhat long, I hope you found something of value in it.
[1]Feist v. Rural Telephone Service Co, Inc., 111 S.Ct. 1282, 113 L.Ed.2nd 358 (1991).
[2] 17 USC Sec 26 (1970 ed)
[3] See Sec 302
[4] 464 U.S. 417, 104 S.Ct. 774, 78 L.Ed.2d 574 (1984).
[5] Basic Books, Inc., et. al. v. Kinko's Graphics Corporation, 758 F.Supp. 1522 (S.D., N.Y. 1991).
[6] American Geophysical Union, et.al. v. TEXACO, Inc., 23 U.S.P.Q.2d 1562 (1992).