some ten men and women sitting around a table, that they read an article on Milton's poetry that appeared ten years ago in Publications of the Modern Language Association, and if two of them over the next week were to go to that college's library and look at that article and decide that they wanted to take copies back to their dormitory for further study, we don't see how there is any practical way in which a library can prevent that kind of reproduction of a single copy on separate occasions, and we don't think they should have to. And yet, the Senate Committee report on S. 1361 (S. Rept. 93-983) cites such an instance. Section 108 (g) (2) says that the rights of reproduction and distribution do not extend to a library which "engages in the systematic reproduction or distribution of single or multiple copies or phonorecords of material described in subsection (d)." The materials referred to in (d) are journal articles or small portions of other copyrighted works.

The question immediately arises as to what constitutes systematic reproduction. To the extent that we are able to puzzle it out, it appears to have been aimed at practices of the kind which were upheld as fair use by the Court of Claims in the Williams & Wilkins case. In listening to my publisher and author friends, the preeminent example which they give of systematic reproduction has always been the Regional Medical Library System, with the National Library of Medicine at its apex. Those practices of the National Library of Medicine were, of course, upheld by the Court of Claims in Williams & Wilkins in a decision which was affirmed this year by the United States Supreme Court.

Now, how does the Regional Medical Library system really work? Well, it starts off with the user, who discovers that he needs access to some particular information, often found in an article in a professional journal in the biomedical field. He usually starts off by going to the library in the hospital with which his practice is affiliated, and may find it there. If it is one of the most important journals, the hospital may well have it. But, since there are thousands of journals in the medical and health sciences field, the chances are that the hospital library may not have this, particularly if it is older material. The request would then go to one of the eleven Regional Medical Libraries over the country which are supported by Congress, and from there as a last resort to the top of the pyramid which is the National Library of Medicine and which now has over 25,000 different journals, the biggest medical collection in the world. It is obviously not pos sible for the smaller hospital library, or sometimes even the Regional Medical Library, to have a sizable portion of this vast amount of material, so some kind of access, such as photocopying, must be relied upon to get the information to the doctor or the other health professional when urgently needed. This kind of organization of access to scientific and technical knowledge seems to us to be the intelligent way of doing things. It should be noted also that the Regional Medical Libraries are not only striving to augment their collections as rapidly as possible but likewise are urging the smaller hospital libraries to upgrade theirs, thus providing all along the line an ever-increasing number of subscriptions with accompanying increased financial gain for the publishers. Mrs. Joan Titley Adams, of the Medical Library Association, who is with us here today, can provide for any of the Committee members who are interested further details about this highly significant work in the medical and health fields.

Another large and highly important type of system for which this systematic reproduction poses problems is that of the county and multi-county library systems throughout the whole country. These libraries came into being largely through the opportunity provided by the federal Library Services and Construction Act. This was and still is an effort to bring books and other library materials to the millions of people, often in rural areas, who had not heretofore had library service available. To get counties to join together, vote the necessary taxes, agree on a common governing board, and gain consensus on the sites for a central library and for the smaller satellite libraries in the system is a difficult task. It is often made possible only by the promise to the citizens of much broader areas of information which will be made available to them not only from their small but growing collection in each neighborhood, but also through loans from the central library and through it from larger collections elsewhere. In this, some copying of periodical articles is occasionally involved, but it does not result in fewer subscriptions-in fact, before the founding of many of these libraries there were no periodical subscriptions at all in the area.

Because interlibrary loan is one of the vital elements in this concept which has been so mutually beneficial to all, it is urgent that no restrictions be imposed

which would diminish the effectiveness of the program. Such a diminution, if it occurred, would be as much against the interest of the publishers as against the citizens the libraries serve. Let me give you an illustration from my home state of Oklahoma which I know well. A few years ago, the Western Plains Library System was established consisting of four counties in Western Oklahoma. At the time of its organization, there was a single library in each of two counties. The other two had no libraries. Now there are seven libraries in the four counties and two bookmobiles are operating regularly. At the beginning the two original libraries subscribed to 20 periodicals between them. The seven libraries now subscribe to over 300. The combined annual book budget of the two original libraries was under $2500. The annual book budget for the seven is now $42,000. In addition, they have encouraged school libraries to develop collections of periodicals and books and are now promoting with success the creation of home collections of books and periodicals. This tremendous increase in acquisition of materials has obviously benefited the publishers of materials as well as the citizens the libraries serve.

This kind of multi-county library is now found in every State in the Union, and over the two decades the Library Services and Construction Act has been in existence millions of dollars of federal money and matching local funds have been expended for this kind of service. The importance of this activity was recognized in the Senate report last summer accompanying S. 1361 (S. Rept. 93-983) in the portion discussing systematic reproduction by saying, "The photocopying needs of such operations as multi-county regional systems must be met," but no provision was made in the law to specifically provide for these needs. Section 108 (g) (2) would prohibit their copying activity and do much mischief indeed.

It was also pointed out to our publisher friends that many systems are not organized for the purpose of copying materials of any kind. For example, one of the large "systems" is SOLINET, an acronym for Southeastern Library Network. This is a group of about 100 libraries in the Southeastern States devoted solely to providing centralized cataloging and catalog card preparation and distribution to member libraries. Other systems have the purpose of encouraging the building of better library collections and the bringing to the area more journals, sets and bibliographies not now represented in the areas. To say that a library merely because it happens to belong to such a "system" is prohibited from photocopying where if it did not belong, it would be permitted to do so, seems to us farfetched indeed.

We are also concerned with section 108(h) which would limit the rights otherwise granted under section 108 by excluding a musical work, pictorial, graphic and other audiovisual works. These exclusions are illogical. The need of the scholar doing research in music for a copy of a portion of a score is as legitimate and proper as that of the scholar doing any other kind of research. Likewise, the copying of one map from an atlas or a page of diagrams and plans from a technical journal may be just as important as any other kind of material for research.

It seems to us that libraries ought to be encouraged to collect and preserve all of the forms in which knowledge is published and distributed, and that it should be possible for users of libraries to have access for their study and scholarship to all of these forms, not just some of them. If a student of the cinema asks a library to make a copy for him of a few selected frames of some famous motion picture which is being studied, so that he may consider at his leisure a certain key point which is made in an article he is reading, we think the library ought to be able to do that.

Mrs. Susan Sommer of the Music Library Association is with us today and can provide further information about the problems posed by this section of the bill in relation to music. Dr. Frank McKenna, of the Special Libraries Association, is also here and can discuss the problems in relation to atlas or other graphic materials in books and periodicals.

In reporting S. 1361 last July, the Senate Judiciary Committee recommended that "representatives of authors, book, and periodical publishers and other owners of coprighted material meet with the library community to formulate photocopying guidelines to assist library patrons and employees." And concerning library photocopying practices not authorized by the reported bill, the Committee recommended "that workable clearance and licensing procedures be developed." In response to this request by the Senate Judiciary Committee, representatives

of the different views on this subject were convened in November 1974 by invitation of the Register of Coyprights and the Chairman of the National Commission on Libraries and Information Science. The resulting "Conference on Resolution of Copyright Issues" established a smaller working group to carry out preliminary discussions. The working group and several subcommittees have since met on frequent occasions to consider and prepare papers on a variety of technical and procedural matters.

There are, of course, different views of the significance of the work performed to date by the Conference and its working group. The work has focused upon the mechanics and the feasibility of possible mechanisms for collecting payments for photocopying of copyrighted materials. It must be emphasized, however, that there has been no agreement as to whether such a payment mechanism is acceptable to libraries even if it is workable, and also I may say no seemingly workable mechanism has yet been advanced in that it still appears it would take dollars to collect dimes. There has also been no agreement as to the categories of publications to which such a mechanism should be applied and no change in the position of libraries that their current photocopying practices are entirely lawful and within the fair use holding of the Williams & Wilkins case, and should not in any respect be treated as infringing rights of the copyright proprietor in the provisions of any new legislation.

The publishers will probably tell you that they, too, are for photocopying but they want money for it without any outlay or trouble on their part. I should like to point out some reasons why licensing and payment of royalties by libraries for the photocopying they do is not justified. First, many publishers already have variable pricing for journals; that is, they charge a considerably higher price for the same journal for a library subscription than for an individual subscription. These prices to libraries often run quite high-subscriptions of $100 to $300 per year are not uncommon; a few run $1,000 or more; and the $50 to $100 price is quite commonplace in the scientific field. These higher subscription prices to libraries presumably are designed in many cases to include charges for anticipated copying. Some journal publishers have received substantial federal assistance in modernizing their editorial and manufacturing procedures. Other journals, and also some of those just mentioned, have already had major contributions of public funds in the nature of per-page charges, usually in the range of $50 to $100 per printed page paid by the author or by a federal grant which is financing his work. The author is usually not paid by the publisher for his work in writing the article but the library or the institution where the author is located often spends a sizable amount for interlibrary loan postage and handling to aid him in preparation of his article which the periodical then receives without cost. As an example, my own small library spent during this past year over $100 on interlibrary loan expense for books to enable a professor to write an article for an historical journal, but the journal did not pay him anything for the article.

In light of these contributions which the libraries and the public already make to the publication of these works, it seems unreasonable for journal publishers to demand still further payment from libraries, and eventually the public, for the occasional photocopying of individual articles for library users. It seems even more unreasonable in view of the fact that by making the information concerned available to those with current, specific needs for it, library photocopying fosters the basic purpose of the authors of such articles. But when it is also noted that there is no evidence that the libraries' policies have caused publishers any harm whatsoever and may actually increase their subscriptions, it is clear that such demands are completely unjustified and the public interest requires that they be rejected by Congress.

For the reasons we have advanced above, we urge that sections 108(g) (1) and (2) and (h) be deleted from the bill. This would also be in accord with the Williams & Wilkins decision and would permit libraries to continue the long established library service of providing a single photocopy of a single article or excerpt from a copyrighted periodical or book for a patron's use without incurring liability for copyright royalties.

It has been a pleasure to appear before you today, Mr. Chairman, and I assure you that we are ready to be of assistance in any way we can toward a satisfactory resolution of this very difficult but important problem.

[The following prepared statements and correspondence were received for the record:]

STATEMENT OF JOHN P. McDONALD, EXECUTIVE DIRECTOR, ASSOCIATION OF RESEARCH LIBRARIES

The Association of Research Libraries, an organization of the principal university and research libraries in the country, believes that the copyright revision bill ultimately approved by the Courts, Civil Liberties and the Administration of Justice Subcommittee must include provisions which will ensure that the customary, long-established library service of providing a single photocopy of a single article or excerpt from a copyrighted periodical or book for a patron's private use may be continued without incurring liability for copyright royalties. The bill adopted by the Senate last year, and reintroduced in the 94th Congress as H.R. 2223, gives explicit recognition to and protection for library photocopying. However, that bill also incorporates provisions in Section 108 (g) which encumber and confuse the expressly recognized right to an extent that would severely hamper libraries' service to the public and exclude practices which are presently lawful. It is imperative that the bill be amended to restore to libraries and the public the rights which they presently enjoy to make limited photocopies of copyrighted works. Section 108 (g) (2) should be removed from the bill because: 1. It restricts practices which are reasonable, customary and lawful under the decision in the Williams & Wilkins case. 2. Copyright owners (e.g., publishers) have advanced no evidence showing that such practices in any way injure their economic interests, much less evidence that it is in the public interest to forbid them. 3. The cost of the libraries and ultimately the public of prohibiting or imposing a royalty requirement upon such practices will be extremely high. 4. The primary purpose of the authors of the sorts of articles most frequently copied is to disseminate the results of their research, not to earn royalties.

At issue is the making, whether at the request of a patron or at the request of another library, of single copies of copyrighted matter for the private use of a scholar or other reader. Such copies may be of articles from law reviews, medical journals or scientific or technical periodicals, or they may be passages from other published works. They are made in response to individual requests for single copies, although more than one individual may request a copy of a particular part of a work in a library's collection. In providing this service, a library may make a copy from a work located on its premises, or in the case of a work not in its own collection, it may request the copy from another library, just as it might obtain the original work itself on an inter-library loan for a patron who wished to borrow it. The right to make a single copy for personal use is important to a wide variety of scholars and other library users, from the high school student who wants a copy of an article in a issue of a magazine for a debate or science project to the physician requiring the material for research work or patient care. In the overwhelming number of cases it is the only way in which a researcher can obtain a copy of an article from a issue of a periodical for reference.

Both libraries and the public have traditionally considered the making of such copies to be a natural and necessary part of libraries' services to their users. It is simply one way in which published material is made accessible. Such copies have been made by photographic and other reprographic techniques since before the enactment of the 1909 Copyright Act. No court has ever held that these traditional practices result in liability under the copyright law, and in the test case brought by publishers, Williams & Wilkins v. United States, the U.S. Court of Claims held that the libraries' practices were a fair use of the published materials. That holding was recently affirmed by the United States Supreme Court.

It is vital that the copyright revision bill recognize the right of libraries to make single photocopies of works for the private use of patrons without incurring liability under the copyright law. Although Williams & Wilkins is authority that traditional library photocopying is within the doctrine of "fair use", express statutory treatment is necessary to remove the threat of suit against libraries arising from varying judicial interpretations of what is or is not "fair use". Failure to include such provisions would abandon this area of major public interest to judicial “legislation", and could lead to further costly litigation.

Section 108 of H.R. 2223 extends the necessary recognition, but limitations written into its provisions, principally in Subsection 108 (g)—and particularly clause (2) of that subsection-seriously erode the rights which it intends to recognize. Clause (g) (2) excludes from the library photocopying permitted under Section 108 any instance of “systematic reproduction and distribution”. Because this restriction was written into the bill by the Senate Patents, Trademarks and

Copyrights Subcommittee at the last minute (after public hearings had been held) and is only vaguely and confusingly explained in the committee report, it is impossible to determine exactly what it means. Such cursory explanation of the limitation as was offered by publishing interests before this Subcommittee seems to confuse it with "related or concerted" reproduction-which is separately treated by Section 108 (g) (1) of the bill-and merely disguises the real import of the restriction. It appears, however, to be potentially applicable whenever a library makes a photocopy of an article or other portion of a published work in the context of a "system". There are, of course, many such systems of libraries, from city or county branch library systems to the university with branch campuses to regional library consortia. When it applies, Section 108(g) (2) would reach the making of a single copy for a single requester, of any part, however small, of a copyrighted work. It is precisely the right to make such copies which Section 108 was intended to confirm.

The Senate Judiciary Committee report states that systematic reproduction or distribution within the intent of Subsection 108(g) (2) occurs “when a library makes copies of such materials available to other libraries or groups of users under formal or informal arrangements whose purpose or effect is to have the reproducing library serve as their source of such material." An example which seems to fit this description would be arrangements under which the Legislative Research Service of the Library of Congress provides copies of materials, such as articles from economic or business periodicals, at the request of Members of Congress. An example listed by the Senate committee's report is the case in which a branch of a library system obtains at a user's request a copy of an article which it does not carry in its own collection. The example most frequently cited by publishers is the regional medical library system, by which local hospital and medical school libraries have access upon request to seldom-read and highly specialized periodicals carried by regional medical libraries or the National Library of Medicine. Each of these examples involves practices which are traditional and obviously reasonable. Just such photocopying practices of the NLM were at issue in the Williams & Wilkins case and were held to be lawful.

The sole rationale offered for the new restrictions is an assertion that they are necessary in order to prevent present and potential subscribers from relying on library photocopying machines in the place of journal subscriptions. That assertion is simply and clearly not valid. The argument has a certain surface plausibility, but in spite of the many opportunities presented to them, notably in the Williams & Wilkins case and most recently at the hearings before this Subcommittee, publishing interests have never offered any evidence to substantiate their claims of damage or to show that their fears for the future have any basis in fact within the context of the limited library photocopying which would be recognized under subsections (a) through (f) of Section 108. In Williams & Wilkins such an inference of injury mistakenly indulged in by the presiding Commissioner was overruled by the Court of Claims which held instead that "this record .. fails to show a significant detriment to plaintiff but does demonstrate injury to medical and scientific research if photocopying of this kind is held unlawful." The publishers' reference to practices by the University Microfilms subsidiary of the Xerox Corporation has no relevance to library photocopying, and the firm's profit-making (and royalty-paying) enterprise geared to the reprinting of multiple copies of articles and entire journal issues would clearly be outside the provisions of Section 108 in the absence of Subsection 108 (g).

What evidence is available strongly suggests that traditional library photocopying does not injure publishers, and in many instances may actually increase subscriptions. It is reasonable to suppose that libraries which have frequent requests for particular works will purchase those works, if they are available, to better serve their users and avoid the delay and administrative expense necessarily involved in inter-library loan transactions. Results of ARL's examination of one inter-library loan network showed a very low rate of coincidence among requests. Rarely was the same article requested by the libraries. It also revealed that 54 percent of all requests were for foreign periodicals and domestic pub