ties to absorb, the result may be foregoing use of the material completely. (See AAUP Bulletin, 53:129 (June 1967)).

B. Library photocopying and copyright protection

Replication of copyright works is daily taking place in libraries as part of the research and educational process. At present it is primarily reflected in reprographic reproduction (reproduction by photographic methods or processes analogous to photography), and is an established and recognized practice in library administration, teaching and research.

Reprography in libraries and for educational purposes should not be confused, however, with computerized retrieval of data and information, which in its present state of development is hardly a serious threat to owners of intellectual property but which could eventually become so. Researchers, librarians and educators in the future will then become involved with new techniques of electronic document-storage and computerized information-retrieval systems just as they are presently learning about the tremendous potential of miniaturization and remote transmission of data.

Currently, the most pressing problems facing owners and users of copyrighted works lie in the reprography area as distinguished from electronic systems. Scholars, researchers and librarians, relying on the doctine of fair use, have always felt free to copy by hand the works of others for their own research and study needs. When copying machines become available, it was a simple transition for these scholars, etc., to extend their note-taking to photocopying from copyrighted material. Publishers maintain that the new machine-copiers made replication of their copyrighted materials so easy and inexpensive that their sales are being detrimentally affected to the point that if allowed to continue they will be forced out of business. As a result, the creator of information would lack the income from his ideas to maintain a degree of independence. Educators particularly object to any limitation of their right to make machinecopies on the grounds that they, like librarians, are not doing so for profit; nor for any direct or indirect commercial advantage, but rather to promote the educational process.

The traditional library position on reprography in libraries is to the effect that not only under the Fair Use doctrine, but also as a natural extension of customary library service, a library may make a single copy of copyrighted material it has purchased, for the scholarly use of any of its readers or another library, requesting such service, if done without profit. Such service, employing modern copying methods has become essential. The present demand can be satisfied without inflicting measurable damage on publishers and copyright owners. Improved copying processes will not materially affect the demand for single-copy library duplication for research purposes. Librarians also argue, no matter who is involved, whether it be the librarian, the publisher, or the creator of information, the main concern should be the public interest in access to information. Copyright protection should not be an impediment to transferring information.

C. The economics of library photocopying and the public interest

Publishers allege that although libraries are not in the business of photocopying for profit, still by doing so, they are depriving publishers of the opportunity to sell additional copies and even to maintain their current subscriptions. In the Williams & Wilkins case, however, involving a U.S. government library's unauthorized photocopying of copyrighted medical periodicals for and at the request of medical researchers and practitioners, the U.S. Court of Claims not only held this practice constituted "fair use", but that "there is inadequate reason to believe, that it (the publisher) is being or will be harmed substantially by these specific practices." Actually, this conclusion is borne out by the realization that if most of the users in libraries who photocopy copyrighted materials would be deprived of this opportunity, they would not purchase the original material. Then again, researchers, scholars and academicians rarely purchase all or even a few of the books and/or journals they use in their research. They receive complimentary copies and reprints of articles or they borrow library copies. Only if these sources fail to provide the materials sought, do they resort to photocopying. The publishers' complaint that photocopying is depriving them of profits because of lost sales is therefore not a completely valid conclusion. Many of the potential sales the publishers envision are not of the type that ordinarily occur. It appears to be, that the publishers, despite all this library photocopying, are no worse off than before.

While libraries and large industrial organizations are principally involved in replication of copyrighted materials (also there is much private and casual copying by students, faculty and others in college and university libraries) still they continue to purchase many new titles and journal subscriptions, as well as maintaining the older subscriptions. It should be also recognized in this context that these institutions cannot physically shelve more than a few copies of a journal, etc., due to lack of space and therefore would never purchase a great number of subscriptions to a journal merely because at one time there was a demand for additional copies of a given article.

In this context, we should also review the economics of publishing. It is an established fact that publishers of scientific and technical journals, publish limited editions of their issues so that they often are unable to sell additional copies on demand as early as two months after publication. They do not invest in maintaining stock of back issues of their publications, and hardly ever reprint them. Thus, depriving themselves of the opportunity to sell their back issues on demand. Still they are insisting on the payment of fees additional to the subscription price of the publications, for photocopying rights of these back issues. Then again, publishers, especially in the areas of scientific and technical reference works do market research before publishing new titles and publish them only when assured that libraries will purchase them in addition to specialists in the field. When they determine that the sale of a particular work will be limited, the list price established is increased to reflect this in order to insure a profit. Surveys have also established that as many as 80% of authors of scientific articles are more interested in dissemination of their articles than in receiving royalties. In the scientific field, it should also be noted, authors not only do not receive remuneration for their articles, but often are required to pay for the cost of having them published or absorb the cost by purchasing a stated number of reprints. It has also been noted that subjects dealt with in scientific literature and some of the other disciplines such as law have become so specialized that most researchers in those fields are interested sometimes only in one article out of the many published in a particular journal. Reprography in libraries and documentation centers appears to be the only obvious way today for researchers to have access to the many scholarly resources of their field.

Publishers complain, however, that they are bearing the economic brunt of this development. The hardware and paper used for reprography are bought and paid for by libraries, etc., why shouldn't publishers be given additional income for the right to make copies of their copyrighted works! They also add that even though scientists, etc., etc., pay for publication of their research papers, they should be interested in the survival of the scientific journals which give them an opportunity to disseminate their findings and research reports.

Librarians respond to this copyright confrontation as follows: Non-profit library institutions are not in business and have nothing to gain by photocopying for others. Their purpose is only to promote research in the sciences and humanities in the public interest. They are involved with access to knowledge and its bibliographical control so that scholars, educators, scientists, etc., can use such data in their research and in the process create new information and materials. Why should librarians, under these circumstances be caught in the middle of the conflict between owners and users of copyrighted materials, and be required to take sides? When we become concerned with technology and economics, we must realize that they are not material to the library's ulterior purpose of information dissemination.

Library institutions do have an interest in the reproduction of copyrighted materials for their own internal, nonprofit purposes. They have a vital concern in conserving copies of periodicals and of works in their collection which are out of print, under certain circumstances. They also have an interest in reproducing multilated or missing pages of works in their collection. Then again, in order to conserve their collection, they recognize the need to photocopy materials in their collection for other libraries, requesting them on inter-library loan. Ground rules should be negotiated for these purposes but not at the expense of limiting the free flow of information, and certainly not with the added cost to libraries for administering a system involving payment of fees, licensing, etc., for the benefit of owners of copyrighted works.

D. The new technology and the copyright revision bill

What position should this committee take with reference to computer technology and related copyright problems? When the integrity of a basic collection of materials, copyrighted and otherwise, compacted and stored in electronic

information-center computers, will be preserved by Xerographic printers providing facsimile reproduction by remote transmission in hard-copy form, or by video scanning of ephemeral copy on a closed-circuit TV monitor elsewhere; when the library collection will remain intact because the computer, in essence, will assume the role of a duplicating rather than a circulating library; when one copy of a book fed into such a system will service all simultaneous demands for it; when microfiche and computer print-outs will replace copyrighted hardcopy publication of research reports, as well as of scientific and technical materials currently appearing in journals, monographs and books, and when audiovisual dial-access teaching machines, operated by remote control, will provide hundreds and even thousands of students with simultaneous audio and visual access to a journal article or excerpts from a book, it is obvious that the publishers' traditional market will be affected by these developments and the copyright laws will have to respond to this "non-book" production.,

Merely on the issue as to when an infringement will occur with reference to input, storage and retrieval of a copyrighted work fed into a computer without permission of the copyright owner remains still to be resolved. Output or retrieval of the copyrighted work may be in the form of abstracts, excerpts, or the work as a whole. It may be delivered to the user in tangible form such as a photo-duplication or in ephemeral form such as the temporary projection of an image on the screen. Should the output of an information storage and retrieval system be considered a copyright infringement or derivative work if such output is an index, abstract, limited quotation or analysis of the copyrighted work? "No," reply some copyright experts, except to the extent that the output is likely to diminish the demand for a copyrighted work, because then the doctrine of fair use should govern. Some experts note that the term "copy" is a word of art construed by the courts in the U.S. to mean a copy which is "visually perceivable" and in "tangible form" and therefore when we are concerned with computer output of punch cards or tape, we are "copying". Thus, the experts cannot agree when a computer system has infringed on a copyright owner's works or for that matter to what extent. Professor Benjamin Kaplan, contends that infringement should not turn on input conversion but rather on output conversion on what is subsequently done with the stored work.

There are other copyright problems brought out by the new technology, e.g., notice of copyright and deposit, whether doctrine of fair use is applicable. The state of the art today, however, is not sufficient to warrant the acceptance of rules and regulations governing the use of copyrighted materials. When "nonbook" production will predominate, the role of the commercial publisher will probably change, especially in his relationship with authors and readers. Publishers may also decide to play a different role with reference to regional, national and international information networks. Libraries will also have to readjust their concepts of reader's services and technical operations and may even become eventually part of projected government information networks. Thus, it is possible that the new technology will change the concept of author protection and that copyright protection will be of little help to the author of scholarly works. Rather than depending on royalties, these authors and/or their publishers will sell directly to the information-system operator either as a complete sale or upon an accounting based on use. The computer could easily be programmed not only to incorporate the new "work" into the existing data but also arrange for accountability of its use in the system. Subscribers to the system will pay for its maintenance.

As a result, many other problems will arise, e.g., the amount of control government will have over these information systems, rates to be established and international agreement and treaties will have to be negotiated to reflect the needs of the system.

In light of all these possibilities, it is my thought that this Committee cannot take a position at present affecting copyright and computer based information storage and retrieval systems. We must await developments in this field to the point where we will be aware of the implications of our decisions.

I would respectfully recommend that this is a problem for the newly created National Commission on New Technological Uses of Copyrighted Works to resolve.

The long range problems arising from the effects of this new technology on copyright must first be identified by the Commission and then it should "make recommendations as to such changes in copyright law or procedures that may be necessary to assure for such purposes access to copyrighted works and to provide recognition of the rights of copyrighted owners" as it has been charged by the U.S. Congress to do.

Mr. Chairman, I appreciate the opportunity of appearing here to present the point of view of the American Association of Law Libraries.

STATEMENT OF WILLIAM M. PASSANO, CHAIRMAN OF THE BOard,
WILLIAMS & WILKINS Co.

I thoroughly enjoyed meeting with you on May 13, and I appreciate the opportunity to review with you the subscription figures for the 27 Journals published by The Williams & Wilkins Company. You may recall that a compelling reason given by the four judges of the Court of Claims for finding library photocopying of our Journals to be "fair use" was that we had not convinced them that this practice was doing harm to the financial condition of the Journals. It is true that in 1973, when the Court of Claims decision was handed down, we had no statistical proof of damage.

However, the figures now available, which compare 1973 with 1974 and which I showed you when I was in your office, do to my mind show that the library networks are, in fact, doing just what they were designed to do; namely, reducing the number of Journals which the libraries subscribe to, since the needs of library patrons can be served by obtaining photocopies of requested articles as interlibrary loans through the network systems.

You will notice that the individual subscriptions to the 27 Journals which we publish have increased nearly 17%, comparing 1974 with 1973. Foreign subscriptions of all kinds have increased approximately 13%. Furthermore, there has been a healthy increase in the number of hospital subscriptions, due primarily we believe to the ever-growing number of community hospitals. This record indicates that the Journals as a group are in demand as purveyors of scientific knowledge and are highly respected by the scientific community. Furthermore, we credit much of the increase to the effectiveness of the very substantial direct marketing efforts which we have made during the past year.

The record of individual and foreign subscriptions, however is in sharp contrast with the institutional subscriptions which in the same period of time have fallen off by 3%. I think it is safe to say that this decrease in institutional subscriptions, at a time when individual and foreign subscriptions have substantially increased, is not due to lack of popularity on the part of the Journals in question, but is because of the ease with which interlibrary loans (photocopies) are obtained through membership in library network systems, and that these photocopies can and do replace the necessity for institutions subscribing to the Journals in question. Certainly The American Chemical Society subscription figures confirm this with even larger declines.

We do not quarrel with photocopying, nor do we object to the network systems. They are effective means of efficiently disseminating scientific knowledge. We do believe, however, that those who use the Journals by photocopying them should share in their support and not leave the entire burden on the shoulders of the subscribers, the authors and the advertisers, as is the case at present. It is for this reason that Section 108(g) (2) of the proposed Copyright Bill must be retained if the scientific press is to remain viable and free from governmental subsidy and control.

It should be borne in mind that fully 65% of the cost of producing the typical scientific periodical is incurred before the first copy comes off the press. This means that only a comparatively small erosion of the subscription list can greatly affect the unit cost and therefore jeopardize the financial security of the Journal. Again, many thanks for permitting me as a member of the Proprietary Rights Committee of the Information Association to place these facts before you.

STATUS OF SUBSCRIBERS TO 27 WILLIAMS & WILKINS JOURNALS COMPARING 1973 WITH 1974

STATEMENT OF HON. DAVID MATHEWS, SECRETARY, DEPARTMENT OF HEALTH,

EDUCATION, AND WELFARE

There is now pending before your Committee H.R. 2223, a bill "For the general revision of the Copyright Law, title 17 of the United States Code, and for other purposes."

In brief, the bill as presently worded contains a provision [Subsection 108 (g)] which would severely hamper the flow of biomedical information between the National Library of Medicine and the nation's medical libraries and thereby reduce the information available to researchers and practitioners. Deletion of Subsection 108 (g) would remove this restriction. However, if deletion of this Subsection is not possible modification of the language contained therein would accomplish the same goal.

We transmit herewith a brief technical report which contains an analysis of select provisions of the bill under consideration and the effects which they might have on the programs of the National Library of Medicine, a bureau of the Department of Health, Education and Welfare.

We are advised by the Office of Management and Budget that there is no objection to the presentation of this legislative proposal from the standpoint of the Administration's program:

AN ANALYSIS OF THE POSSIBLE EFFECTS OF SECTION 108 OF H.R. 2223, GENERAL REVISION OF THE COPYRIGHT LAW

House of Representatives bill, H.R. 2223 "For the general revision of the Copyright Law, title 17 of the United States Code, and for other purposes," now pending before the Committee on The Judiciary in the House of Representatives, would provide for the first general revision of the copyright law since its passage in 1909. Section 108, "Limitations on exclusive rights: Reproduction by libraries and archives" provides that it is not an infringement of copyright for a library or archives to reproduce no more than one copy of a work for non-commercial purposes in order to preserve deteriorating materials, replace a damaged or lost copy that can not be purchased at a fair price, or provide a copy for the use of an individual library patron for scholarship and research. However, Subsection 108 (g) prohibits "the related or concerted reproduction or distribution of multiple copies or phonorecords of the same material, whether made on one occasion or over a period of time" whether intended for the use of one individual or a group. It also prohibits "the systematic reproduction or distribution of single or multiple copies" of a copyrighted work.

Subsection 108 (g) in its present form, depending on the interpretation of "systematic reproduction," could possibly make operation of the current interlibrary loan program of the National Library of Medicine and its Regional Medical Library network an infringement of copyright, thereby seriously impairing the nation's health research and scholarship.

The NLM is a "library's library" serving as the back-up source of materials requested by patrons of local medical libraries but which are at that time absent from their collections. There are many reasons for the non-availability of literature which necessitates that a local library request an interlibrary loan; among the common reasons are that the material requested are out of the local library on loan or at the bindery.

To provide more rapid dissemination of biomedical information, the Library has developed a network arrangement through which biomedical literature