Copyright has become a battleground in the last two years. Two ends of the scholarly communication chain — librarians and publishers — see it as an important element that will allow them to maintain their roles. Many librarians see copyright as the element that will provide the way out of their budget problems. Many publishers see it as essential to maintaining sales of their publications. There has been much discussion in the last two years about what copyright is designed for, why it exists, and what power each side believes it gives to the other side. Neither side is without its bias, and very little in the way of compromise has been offered. As a non-lawyer, it is impossible for me to comment on the legal technicalities of copyright law. But it is important to discuss the practicalities of current copyright policies and hypothesize how the system might change if the new policies being discussed now became reality.

The changes being proposed are intended to shift control of dissemination of published materials from publishers to either authors or their universities, and the assumption is that the result would be an improvement. However, without the basic assurances that copyright gives — that an article will be disseminated in its true and correct form, that it will be distributed ethically and to the appropriate places and in the appropriate forms, that it will be properly valued because of its inclusion in a high-quality peer-reviewed journal — the ability of the research community to determine value would disintegrate and research output would be harmed.

Copyright Management Under the Present System

As copyright is practiced now in most traditional journals, authors are asked to transfer copyright and authorize the publisher to be the exclusive licensor of all subsidiary uses and forms of publication of the article. Any reuse of a journal article that falls beyond the guidelines of fair use must therefore be approved by the publisher. The publisher, therefore, has the following responsibilities:

• To register copyright for the journal issue, including the articles, with the Copyright Office of the Library of Congress, to pay the fee, and to deposit the required number of copies.
• To establish terms for other publishers who want to reprint an article, or a significant portion of an article, in a work of theirs.
• To establish terms under which the article may be translated and published in another work.
• To set terms for photocopying beyond that allowed by fair use (i.e. photocopying for distribution to a class).
• To solicit inclusion of abstracts and journal contents in appropriate indexing and abstracting services and (usually) send free subscriptions to the secondary publisher.
• To set terms under which the articles may be included in document delivery services.
• To establish contractual relationships for other forms of publication such as microfilm, microfiche, CD-Rom, and full-text on-line databases. This usually includes sending free subscriptions to these services.

The publisher's goals in the licensing of subsidiary rights are to disseminate the article as widely as possible in order to encourage citation by the scholarly community; to stimulate submissions and subscriptions to the journal; to encourage researchers to act responsibly when citing and reusing the work of others by enforcing copyright; to encourage students to accept the premise that scholarly work has value; and to bring in additional revenues to contribute to publication costs of the journal. To most publishers, the first goal is the most important. The last is the least important, because these transactions are generally worth small amounts and are very labor intensive.

To librarians this situation means that they must ask publishers for permission to use articles beyond that allowed by the fair use guidelines. When they order copies through interlibrary loan beyond that allowed by fair use, they must report the use and pay any associated copyright fees. Preferably these payments are made to a central agency such as the Copyright Clearance Center, but if not they must pay the publisher. When librarians buy products from secondary publishers such as University Microfilms, Information Access, CARL, or ISI, they know that copyright has been cleared and is therefore not their concern. If a library wants to produce these types of products themselves, they must clear the permissions as any secondary publisher would

To scholars, the system appears relatively easy. In most cases, these secondary products just appear magically, the permissions are cleared so their article can be more widely disseminated, and the scholars themselves are usually not bothered with the large number of small transactions associated. When a publisher is slow to answer permission requests, or charges more than an author believes is appropriate, they may become concerned and complain to the publisher. But such complaints are a tiny percentage of the total number of transactions that a publisher handles in a year. On the other side, researchers are sometimes annoyed when they must get permission to reuse another author's work in a work of their own or to use an article in a classroom. However, they are also sometimes annoyed when they see their own work reused by another researcher who they believe has not received proper permission.

There are a number of assumptions that underly the current copyright system:

• Researchers assume their work will be used by anyone who wants it and with nothing to impede reuse.
• Librarians assume that the products they purchase from secondary publishers will include the top journals their clients need and allow them appropriate access to research materials.
• Secondary publishers rely on the reputation of a journal and the imprint of the publisher to guide them to appropriate publications to license and to give them one access point for a body of research.
• Copy shops rely on publishers to be accessible and responsive when they are looking for fast permission for a professor's photocopying job.
• Publishers rely on income from the licensing of subsidiary rights — particularly for older journals — to provide some revenue to the journal.

Copyright Management Under an Author-Managed System

The organizations and individuals promoting changes in the copyright system assume that a change in policy will address the budget problems of librarians caused by increases in serials prices that outpace both the rate of the inflation and the budget increases of librarians, and that scholars will benefit from increased control of their own work Neither of these would actually prove to be the case. In fact, products that scholars and librarians depend on to guide them through the literature would be made exponentially more expensive and delayed if they continued to be available at all.

For now, we will assume that journals would continue to be published if this change came about. If authors retained copyright and gave publishers only the right to first-time publication in English, authors would have to be prepared to take over all the functions listed above that publishers now handle. As long as the copyright law was in place, photocopy shops, publishers, and librarians would feel constrained to check the availability of rights before using material. When an inquiry was received by a publisher, it would respond with a form letter saying the requestor must contact the author. Most likely the publisher would not have a current address for the author.

If the journal's policy was to permit any and all photocopying, there would still be problems. The copy shop or library would have to have a copy of the journal (instead of a copy of the article) in order to verify the existence of such a policy. Theoretically, libraries could then make as many copies as they desired from journals without having to report photocopying to the Copyright Clearance Center or to the publisher. The question then is, how many would continue to subscribe to a journal? And would that number support the journal's continued existence? The million- dollar journal that has a subscription list of one might actually become a reality. Scholars, researchers, and administrators could be irreparably damaged by the loss of outlets for peer-reviewed publication.

Assuming again that journals still existed, secondary publishers would be in a very troublesome situation. These publishers currently decide on the content of their products by determining what are the top journals in a field and approaching their publishers for a license to use the journal's material. If authors retained copyright and the exclusive right to license subsidiary rights, these publishers would have to:

• Consider each article after publication and decide whether it should be included in their CD-Rom, microfilm, on-line database, or document delivery service, i.e. make an editorial determination on an individual article basis.
• Contact each author (including multiple authors of an article) and develop a license with them individually.

The first would create a whole new portion of overhead that secondary publishers now avoid by deciding once to include everything that is published in a particular journal. The shift to decision-making after publication and individual licenses with authors would create long delays in getting the material included in the publications and distributed to the libraries. The extra cost and delay very likely would make it impossible for full-text secondary publications in any media to exist at all.

What would researchers do without these products? What would librarians do without these products? In the last year publishers have experienced cancellations of subscriptions to print journals because of the availability of the journal in a full-text CD-Rom, on-line database, document delivery service, etc. Librarians would no longer have the option to cancel and still have the full-text available to their clients in other media. Access points would still exist in the indexing and abstracting publications, but full- text would not. If librarians chose not to subscribe to the print journal, they would have to rely exclusively on interlibary loan to give their clients access to journal articles they do not receive through subscription. As the ARL /RLG Interlibrary Loan Cost Study confirmed, interlibrary loan costs are not trivial, particularly for the supplying library.

Copyright Management Under an Institution-Managed System

Proposals have been made to shift ownership of copyright from the publisher to the author's institution. Under this scenario, universities would have to be able to handle all the types of requests that publishers currently handle. Some argue that the natural place for the management of these requests is with the local university press. But aside from the fact that most universities do not have a university press, the mission of the press would change dramatically if it managed copyright for all faculty of the university. What would that do to the press's mandate for peer review and editorial board approval of all projects with the press's imprint? The value of the imprint would change. The press would also have to deal with subsidiary rights in all different fields, which is not the case now since all publishers limit the number of fields in which they publish. And as I said before, the bulk of these transactions are labor intensive and result in small amounts of revenue.

If a university did not have its own press, it would have to set up an office for handling permissions and licensing transactions for the writings of its faculty. Space, staff, benefits, equipment — it would require a definite investment. And judging from subsidiary income received by publishers under the current system, it is unlikely that the revenue received would cover the cost of administering the rights.

There would also be the inescapable problem of multi-authored articles and multiple institutional rights sharers. Would the institutions of primary authors of multi-authored articles be given the responsibility for handling the rights? If not, the rights would be very difficult to manage and not worth the time it would take for any of the authors' institutions. And there is the obvious problem of moving authors, since academics tend to reside at several institutions over the course of their career. The biggest hurdle, however, is the shift for faculty to work-for-hire status for all their writings and the concern for academic freedom that that raises.

Is Compromise Possible?

Yes, compromise is possible, and market forces do act on publishers. If publishers charge too much for subsidiary uses, market pressure from librarians, authors, and editors should eventually bring those prices in line with others. Certainly market pressure has brought other accommodations in copyright management from publishers within the last two years. Publishers are considering and implementing some of the following as responses to market pressure:

• Explicitly tell authors what rights they retain and what permission they have to reuse their own material.
• When an author wishes to retain copyright, have other options available that allow the publisher to still license subsidiary uses of the article in order to not deter dissemination of the material.
• Explain to authors what publishers do with the rights they are asking for, and what benefit there is to the author and the scholarly community from the products and subsidiary uses those rights make possible.
• Make it easy for photocopy shops and professors to get quick permission to use articles in classes, such as blanket agreements whenever possible.
• Consider routinely allowing authors to copy their own articles for use in their own classes without requesting permission and without fee to the publisher.
• Consider allowing the author's institution to make copies free for local patrons, as well as place electronic copies on reserve for local use.
• Consider reducing the length of time for exclusive rights to the publisher as long as it can be followed with nonexclusive rights to the publisher to insure the continuation of subsidiary forms of publication.
• Consider whether early electronic postings of articles on FTP sites actually hurts sales of the printed journal or whether it can be used to point to the final authoritative version in the printed journal.

These compromises and adjustments to the administration of copyright would not irreparably harm the dissemination of information, as would the more radical changes currently being discussed. We would all be hurt by such changes — not only publishers, but also librarians, university administrators, and the researchers we all exist to serve and facilitate.

Janet H. Fisher is the Associate Director for Journals Publishing at the MIT Press.

Copyright (C) 1993 by Janet H. Fisher