Williams Wilkins Company v. United States

Full title: THE WILLIAMS WILKINS COMPANY v. THE UNITED STATES

Court: United States Court of Claims

Date published: Nov 27, 1973

487 F.2d 1345 (Fed. Cir. 1973)

Opinion

No. 73-68.

November 27, 1973.

Alan Latman, New York City, attorney of record, for plaintiff. Arthur J. Greenbaum and Cowan, Liebowitz Latman, New York City, of Counsel.

Thomas J. Byrnes, Washington, D.C., with whom was Asst. Atty. Gen. Harlington Wood, Jr., for defendant.

Irwin Karp, New York City, for The Authors League of America, Inc., amicus curiae.

Philip B. Brown, Washington, D.C., for the Assn. of Research Libraries, Medical Library Assn., American Assn. of Law Libraries, American Medical Assn., American Dental Assn., Mayo Foundation, Robert H. Ebert, M.D. (in his capacity as Dean of the Faculty of Medicine, Harvard University), The University of Michigan Medical School, The University of Rochester, School of Medicine and Dentistry, American Sociological Assn., Modern Language Assn. of America, and History of Science Society, amici curiae. Cox, Langford Brown and John P. Furman, Washington, D.C., of counsel.

Harry N. Rosenfield, Washington, D.C., for The National Education Assn. of the United States, amicus curiae.

William D. North, Chicago, Ill., for the American Library Assn., amicus curiae. Perry S. Patterson, Washington, D.C., Ronald L. Engel, James M. Amend, John A. Waters, Chicago, Ill., Thomas B. Carr, Washington, D.C., and Kirkland Ellis, Chicago, Ill., of counsel.

Charles H. Lieb, New York City, for the Assn. of American Publishers, Inc. and The Assn. of American University Presses, Incorporated, amici curiae. Paskus, Gordon Hyman, New York City, and Elizabeth Barad, New York City, of counsel.

Arthur B. Hanson, Washington, D.C., for The American Chemical Society, amicus curiae. Hanson, O'Brien, Birney, Stickle Butler, Washington, D.C., of counsel.

Davies, Hardy, Ives Lawther, New York City, for The American Institute of Physics Incorporated, amicus curiae. Robert E. Lawther, New York City, of counsel.

Robert B. Washburn, Virgil E. Woodcock, and Woodcock, Washburn, Kurtz Mackiewicz, Philadelphia, Pa., for American Society for Testing and Materials and National Council of Teachers of Mathematics, amici curiae.

Before COWEN, Chief Judge, and DAVIS, SKELTON, NICHOLS, KASHIWA, KUNZIG and BENNETT, Judges.

OPINION

DAVIS, Judge:

We confront a ground-breaking copyright infringement action under 28 U.S.C. § 1498(b), the statute consenting to infringement suits against the United States.¹ Plaintiff Williams Wilkins Company, a medical publisher, charges that the Department of Health, Education, and Welfare, through the National Institutes of Health (NIH) and the National Library of Medicine (NLM), has infringed plaintiff's copyrights in certain of its medical journals by making unauthorized photocopies of articles from those periodicals. Modern photocopying in its relation to copyright spins off troublesome problems, which have been much discussed.² Those issues have never before been mooted or determined by a court. In this case, an extensive trial was held before former Trial Judge James F. Davis who decided that the Government was liable for infringement. On review, helped by the briefs and agreements of the parties and the amici curiae, we take the other position and hold the United States free of liability in the particular situation presented by this record.

1 Prior to 1960, § 1498 provided only for patent infringement suits against the Federal Government. In that year, Congress amended the section to make the United States liable in money for copyright infringement, pursuant to Title 17 of the United States Code, the general copyright statute. This is the first copyright case to reach trial in this court.

2 We list in the Appendix, infra, several considerations to these problems.

I³

3 We borrow, with some modifications, the statement of facts from the opinion of Trial Judge James F. Davis.

Plaintiff, though a relatively small company, is a major publisher of medical journals and books. It publishes 37 journals, dealing with various medical specialties. The four journals in suit are Medicine, Journal of Immunology, Gastroenterology, and Pharmacological Reviews. Medicine is published by plaintiff for profit and for its own benefit. The other three journals are published in conjunction with specialty medical societies which, by contract, share the journals' profits with plaintiff. The articles published in the journals stem from manuscripts submitted to plaintiff (or one of the medical societies) by physicians or other scientists engaged in medical research. The journals are widely disseminated throughout the United States (and the world) in libraries, schools, physicians' offices, and the like. Annual subscription prices range from about $12 to $44; and, due to the esoteric nature of the journals' subject matter, the number of annual subscriptions is relatively small, ranging from about 3,100 ( Pharmacological Reviews) to about 7,000 ( Gastroenterology). Most of the revenue derived from the journals comes from subscription sales, though a small part comes from advertising.⁴ The journals are published with notice of copyright in plaintiff's name. The notice appears at the front of the journal and sometimes at the beginning of each article. After publication of each journal issue (usually monthly or bimonthly) and after compliance with the requisite statutory requirements, the Register of Copyrights issues to plaintiff certificates of copyright registration.

4 E. g., the November 1956 issue of Medicine has 86 pages, four of which carry commercial product advertising. The August 1965 issue of Journal of Immunology has 206 pages, nine of which carry commercial product advertising.

NIH, the Government's principal medical research organization, is a conglomerate of institutes located on a multi-acre campus at Bethesda, Maryland. Each institute is concerned with a particular medical specialty, and the institutes conduct their activities by way of both intramural research and grants-in-aid to private individuals and organizations. NIH employs over 12,000 persons — 4,000 are science professionals and 2,000 have doctoral degrees. To assist its intramural programs, NIH maintains a technical library. The library houses about 150,000 volumes, of which about 30,000 are books and the balance scientific (principally medical) journals. The library is open to the public, but is used mostly by NIH in-house research personnel. The library's budget for 1970 was $1.1 million; of this about $85,000 was for the purchase of journal materials.

The NIH library subscribes to about 3,000 different journal titles, four of which are the journals in suit. The library subscribes to two copies of each of the journals involved. As a general rule, one copy stays in the library reading room and the other copy circulates among interested NIH personnel. Demand by NIH research workers for access to plaintiff's journals (as well as other journals to which the library subscribes) is usually not met by in-house subscription copies. Consequently, as an integral part of its operation, the library runs a photocopy service for the benefit of its research staff. On request, a researcher can obtain a photocopy of an article from any of the journals in the library's collection. Usually, researchers request photocopies of articles to assist them in their on-going projects; sometimes photocopies are requested simply for background reading. The library does not monitor the reason for requests or the use to which the photocopies are put. The photocopies are not returned to the library; and the record shows that, in most instances, researchers keep them in their private files for future reference.

The library's policy is that, as a rule, only a single copy of a journal article will be made per request and each request is limited to about 40 to 50 pages, though exceptions may be, and have been, made in the case of long articles, upon approval of the Assistant Chief of the library branch. Also, as a general rule, requests for photocopying are limited to only a single article from a journal issue. Exceptions to this rule are routinely made, so long as substantially less than an entire journal is photocopied, i. e., less than about half of the journal. Coworkers can, and frequently do, request single copies of the same article and such requests are honored.

Four regularly assigned employees operate the NIH photocopy equipment. The equipment consists of microfilm cameras and Xerox copying machines. In 1970, the library photocopy budget was $86,000 and the library filled 85,744 requests for photocopies of journal articles (including plaintiff's journals), constituting about 930,000 pages. On the average, a journal article is 10 pages long, so that, in 1970, the library made about 93,000 photocopies of articles.

NLM, located on the Bethesda campus of NIH, was formerly the Armed Forces Medical Library. In 1956, Congress transferred the library from the Department of Defense to the Public Health Service (renaming it the National Library of Medicine), and declared its purpose to be "* * * to aid the dissemination and exchange of scientific and other information important to the progress of medicine and to the public health * * *." 42 U.S.C. § 275 (1970). NLM is a repository of much of the world's medical literature, in essence a "librarians' library." As part of its operation, NLM cooperates with other libraries and like research-and-education-oriented institutions (both public and private) in a so-called "interlibrary loan" program. Upon request, NLM will loan to such institutions, for a limited time, books and other materials in its collection. In the case of journals, the "loans" usually take the form of photocopies of journal articles which are supplied by NLM free of charge and on a no-return basis. NLM's "loan" policies are fashioned after the General Interlibrary Loan Code, which is a statement of self-imposed regulations to be followed by all libraries which cooperate in interlibrary loaning. The Code provides that each library, upon request for a loan of materials, shall decide whether to loan the original or provide a photoduplicate. The Code notes that photoduplication of copyrighted materials may raise copyright infringement problems, particularly with regard to "photographing whole issues of periodicals or books with current copyrights, or in making multiple copies of a publication." [Emphasis in original text.] NLM, therefore, will provide only one photocopy of a particular article, per request, and will not photocopy on any given request an entire journal issue. Each photocopy reproduced by NLM contains a statement in the margin, "This is a single photostatic copy made by the National Library of Medicine for purposes of study or research in lieu of lending the original."

In recent years NLM's stated policy has been not to fill requests for copies of articles from any of 104 journals which are included in a so-called "widely-available list." Rather, the requester is furnished a copy of the "widely-available list" and the names of the regional medical libraries which are presumed to have the journals listed. Exceptions are sometimes made to the policy, particularly if the requester has been unsuccessful in obtaining the journal elsewhere. The four journals involved in this suit are listed on the "widely-available list." A rejection on the basis of the "widely-available list" is made only if the article requested was published during the preceding 5 years, but requests from Government libraries are not refused on the basis of the "widely-available list."

Also, NLM's policy is not to honor an excessive number of requests from an individual or an institution. As a general rule, not more than 20 requests from an individual, or not more than 30 requests from an institution, within a month, will be honored. In 1968, NLM adopted the policy that no more than one article from a single journal issue, or three from a journal volume, would be copied. Prior to 1968, NLM had no express policy on copying limitations, but endeavored to prevent "excessive copying." Generally, requests for more than 50 pages of material will not be honored, though exceptions are sometimes made, particularly for Government institutions. Requests for more than one copy of a journal article are rejected, without exception. If NLM receives a request for more than one copy, a single copy will be furnished and the requester advised that it is NLM's policy to furnish only one copy.

In 1968, a representative year, NLM received about 127,000 requests for interlibrary loans. Requests were received, for the most part, from other libraries or Government agencies. However, about 12 percent of the requests came from private or commercial organizations, particularly drug companies. Some requests were for books, in which event the book itself was loaned. Most requests were for journals or journal articles; and about 120,000 of the requests were filled by photocopying single articles from journals, including plaintiff's journals. Usually, the library seeking an interlibrary loan from NLM did so at the request of one of its patrons. If the "loan" was made by photocopy, the photocopy was given to the patron who was free to dispose of it as he wished. NLM made no effort to find out the ultimate use to which the photocopies were put; and there is no evidence that borrowing libraries kept the "loan" photocopies in their permanent collections for use by other patrons.

Defendant concedes that, within the pertinent accounting period, NLM and the NIH library made at least one photocopy of each of eight articles (designated by plaintiff as the Count I-to-Count VIII articles) from one or more of the four journals in suit. These requests, as shown at the trial, were made by NIH researchers and an Army medical officer (stationed in Japan) in connection with their professional work and were used solely for those purposes. In seven of the eight counts in the petition, the article requested was more than two years old; in the eighth instance it was 21 or 22 months old.

II

We assume, for the purposes of the case, but without deciding, that plaintiff is the proper copyright owner and entitled to sue here,⁵ and we agree with plaintiff that, on that assumption, it can sue for infringement of the eight separate articles.⁶ This faces us squarely with the issue of infringement.

5 Defendant vigorously contests the publisher's claim to be the copyright "proprietor" and its right to sue in this court. The argument is that the individual authors of the articles are the owners and they have not assigned their rights to plaintiff.

6 Section 3 of the copyright statute, 17 U.S.C.A. § 3, says that, "* * * [t]he copyright upon composite works or periodicals shall give to the proprietor thereof all the rights in respect thereto which he would have if each part were individually copyrighted under this title." This means, and was intended to provide that each article in the journals is protected from infringement to the same extent as the entire issue. Advertisers Exch., Inc. v. Laufe, 29 F. Supp. 1 (W.D.Pa. 1939); King Features Syndicate v. Fleischer, 299 F. 533 (C.A. 2, 1924).

Perhaps the main reason why determination of the question is so difficult is that the text of the Copyright Act of 1909, which governs the case, does not supply, by itself, a clear or satisfactory answer. Section 1 of the Act, 17 U.S.C. § 1, declares that the copyright owner "shall have the exclusive right: (a) To print, reprint, publish, copy, and vend the copyrighted work; * * *." Read with blinders, this language might seem on its surface to be all-comprehensive — especially the term "copy" — but we are convinced, for several reasons, that "copy" is not to be taken in its full literal sweep. In this instance, as in so many others in copyright, "[T]he statute is hardly unambiguous * * * and presents problems of interpretation not solved by literal application of words as they are `normally' used * * *." DeSylva v. Ballentine, 351 U.S. 570, 573, 76 S.Ct. 974, 100 L.Ed. 1415 (1956). See, also, Fortnightly Corp. v. United Artists Television, Inc., 392 U.S. 390, 395-396, 88 S.Ct. 2084, 20 L.Ed.2d 1176 (1968).

The court-created doctrine of "fair use" (discussed in Part III, infra) is alone enough to demonstrate that Section 1 does not cover all copying (in the literal sense). Some forms of copying, at the very least of portions of a work, are universally deemed immune from liability, although the very words are reproduced in more than de minimis quantity. Furthermore, it is almost unanimously accepted that a scholar can make a handwritten copy of an entire copyrighted article for his own use, and in the era before photoduplication it was not uncommon (and not seriously questioned) that he could have his secretary make a typed copy for his personal use and files. These customary facts of copyright-life are among our givens. The issue we now have is the complex one of whether photocopying, in the form done by NIH and NLM, should be accorded the same treatment — not the ministerial lexicographic task of deciding that photoduplication necessarily involves "copying" (as of course it does in dictionary terms).

One aspect of the history and structure of the 1909 Act offers another reason for refusing to give "copying" in Section 1, as applied to these articles, its simplest "ordinary" reach. It is pointed out to us, on the basis of analysis of the copyright laws from 1790 to 1909,⁷ that the early statutes distinguished "copying" from "printing," "reprinting," and "publishing," and provided that the copyright in books is infringed by "printing," "reprinting" and "publishing," while the copyright in other works ( e. g., photographs, paintings, engraving, drawings, etc.) is infringed by "copying." Cf. Harper v. Shoppell, 26 F. 519, 520 (C.C.S.D.N.Y. 1886). The 1909 Act obliterated any such distinction in its text. It provides in § 5 a list of all classes of copyrightable subject matter (including books and periodicals), and says in § 1 that the owner of copyright shall have the exclusive right "to print, reprint, publish, copy and vend the copyrighted work." Thus, the 1909 Act, unlike the earlier statutes, does not expressly say which of the proscribed acts of § 1 apply to which classes of copyrightable subject matter of § 5. Defendant and some of the amici say that, to be consistent with the intent and purpose of earlier statutes, the "copying" proscription of § 1 should not apply to books or periodicals; rather, only the proscribed acts of "printing," "reprinting" and "publishing" control books and periodicals. The proponents of this view stress that the legislative history of the 1909 legislation does not suggest any purpose to alter the previous coverage.⁸

7 Congress enacted the first copyright statute in 1790 (Act of May 31, 1790, ch. 15, 1 Stat. 124). Thereafter, the statute was revised from time to time, notably in 1802, 1831, 1870, and 1891. In 1909, the present statute was passed (Act of March 4, 1909, ch. 320, 35 Stat. 1075) and later was codified as 17 U.S.C. (Act of July 30, 1947, 61 Stat. 652).

8 For instance, H.R. Rep. No. 2222, 60th Cong., 2d Sess. 4 (1909) states: "Subsection (a) of section 1 adopts without change the phraseology of section 4952 of the Revised Statutes, and this, with the insertion of the word `copy,' practically adopts the phraseology of the first copyright act Congress ever passed — that of 1790. Many amendments of this were suggested, but the committee felt that it was safer to retain without change the old phraseology which has been so often construed by the courts."

This is quite a serious argument. However, in view of Congress's general inclusion of the word "copy" in Section 1 and of the practice under the Act since 1909, we are not ready to accept fully this claim that infringement of periodical articles can come only through "printing," "reprinting" or "publishing." But we do believe this point — that there is a solid doubt whether and how far "copy" applies to books and journals — must be taken into account in measuring the outlines of "copying" as it involves books and articles.

Adding to this doubt that "copy" blankets such printed matter is the significant implication of a special segment of the background of the 1909 statute, a sector of history which is peripheral but revealing. The then Librarian of Congress, Herbert Putnam, was the leading public sponsor of that Act (outside of Congress itself), and was intimately involved in its preparation from at least 1906 on. While the bill was being considered in Congress, the Library's 1908 "Rules and Practice Governing the Use and Issue of Books," p. 6, specifically provided:

" Photographing. Photographing is freely permitted. The permission extends to the building itself and any of its parts, including the mural decorations. It extends to articles bearing claim of copyright, but the Library gives no assurance that the photograph may be reproduced or republished or placed on sale. These are matters to be settled with the owner of the copyright" (emphasis added).

After the 1909 Act became law, the Library continued the same provision. The 1913 version of the "Rules and Practice"⁹ added the following on "Photostat," after the above paragraph on "Photographing":

9 There was an 1911 edition, but no copy has been located.

Photo-duplicates of books, newspapers, maps, etc. can be furnished at a reasonable rate by means of the photostat, installed in the Chief Clerk's Office. Apply to the Chief Clerk for a schedule of charges.

Later editions, throughout Dr. Putnam's tenure (which ended in 1939), contained the same or comparable provisions.¹⁰ Indeed, when he left his post in 1939, he was honored by the American Council of Learned Societies because (among other things) "You have led in adapting the most modern photographic processes to the needs of the scholar, and have * * * made widely available for purposes of research copies of your collections * * *." This illuminating slice of history, covering the time of enactment and the first three decades of the 1909 Act, should not be ignored.

10 The Library's current practice is described in Part III, 3, note 16, infra.

These are the leading reasons why we cannot stop with the dictionary or "normal" definition of "copy" — nor can we extract much affirmative help from the surfacial legislative text. As for the other rights given in Section 1, "vend" is clearly irrelevant (since NIH and NLM do not sell), and the applicability to this case of "print," "reprint" and "publish" is more dubious than of "copy." The photocopy process of NIH and NLM, described in Part I, supra, does not even amount to printing or reprinting in the strict dictionary sense; and if the words be used more broadly to include all mechanical reproduction of a number of copies, they would still not cover the making of a single copy for an individual requester. If the requester himself made a photocopy of the article for his own use on a machine made available by the library, he might conceivably be "copying" but he would not be "printing" or "reprinting." The library is in the same position when responding to the demands of individual researchers acting separately.

For similar reasons there is no "publication" by the library, a concept which invokes general distribution, or at least a supplying of the material to a fairly large group.¹¹ The author of an uncopyrighted manuscript does not lose his common law rights, via publication, by giving photocopies to his friends for comment or their personal use — and publication for Section 1 purposes would seem to have about the same coverage. In any event, the hitherto uncodified principles of "fair use" apply to printing, reprinting, and publishing, as well as to copying, and therefore the collocation of general words Congress chose for Section 1 is necessarily inadequate, by itself, to decide this case.

11 To the extent that Macmillan Co. v. King, 223 F. 862 (D.Mass. 1914), may possibly suggest that "publication" can occur through simple distribution to a very small restricted group, for a special purpose, we think the opinion goes too far.

III

In the fifty-odd years since the 1909 Act, the major tool for probing what physical copying amounts to unlawful "copying" (as well as what is unlawful "printing." "reprinting" and "publishing") has been the gloss of "fair use" which the courts have put upon the words of the statute. Precisely because a determination that a use is "fair," or "unfair," depends on an evaluation of the complex of individual and varying factors bearing upon the particular use (see H.R. Rep. No. 83, 90th Cong., 1st Sess., p. 29), there has been no exact or detailed definition of the doctrine. The courts, congressional committees, and scholars have had to be content with a general listing of the main considerations — together with the example of specific instances ruled "fair" or "unfair." These overall factors are now said to be: (a) the purpose and character of the use, (b) the nature of the copyrighted work, (c) the amount and substantiality of the material used in relation to the copyrighted work as a whole, and (d) the effect of the use on a copyright owner's potential market for and value of his work.

In addition, the development of "fair use" has been influenced by some tension between the direct aim of the copyright privilege to grant the owner a right from which he can reap financial benefit and the more fundamental purpose of the protection "To promote the Progress of Science and the useful Arts." U.S. Const., art. 1, § 8. The House committee which recommended the 1909 Act said that copyright was "[n]ot primarily for the benefit of the author, but primarily for the benefit of the public." H.R. Rep. No. 2222, 60th Cong., 2d Sess., p. 7. The Supreme Court has stated that "The copyright law, like the patent statutes, makes reward to the owner a secondary consideration." Mazer v. Stein, 347 U.S. 201, 219, 74 S.Ct. 460, 471, 98 L.Ed. 630 (1954); United States v. Paramount Pictures, 334 U.S. 131, 158, 68 S.Ct. 915, 92 L.Ed. 1260 (1948). See Breyer, The Uneasy Case for Copyright: A Study of Copyright in Books, Photocopies, and Computer Programs, 84 Harv.L.Rev. 281 (1970). To serve the constitutional purpose, "`courts in passing upon particular claims of infringement must occasionally subordinate the copyright holder's interest in a maximum financial return to the greater public interest in the development of art, science and industry.' Berlin v. E. C. Publications, Inc., 329 F.2d 541, 544 (2d Cir. 1964). Whether the privilege may justifiably be applied to particular materials turns initially on the nature of the materials, e. g., whether their distribution would serve the public interest in the free dissemination of information and whether their preparation requires some use of prior materials dealing with the same subject matter. Consequently, the privilege has been applied to works in the fields of science, law, medicine, history and biography." Rosemont Enterprises, Inc. v. Random House, Inc., 366 F.2d 303, 307 (C.A. 2, 1966).

It has sometimes been suggested that the copying of an entire copyrighted work, any such work, cannot ever be "fair use," but this is an overbroad generalization, unsupported by the decisions¹² and rejected by years of accepted practice. The handwritten or typed copy of an article, for personal use, is one illustration, let alone the thousands of copies of poems, songs, or such items which have long been made by individuals, and sometimes given to lovers and others. Trial Judge James F. Davis, who considered the use now in dispute not to be "fair," nevertheless agreed that a library could supply single photocopies of entire copyrighted works to attorneys or courts for use in litigation. It is, of course, common for courts to be given photocopies of recent decisions, with the publishing company's headnotes and arrangement, and sometimes its annotations. There are other examples from everyday legal and personal life. We cannot believe, for instance, that a judge who makes and gives to a colleague a photocopy of a law review article, in one of the smaller or less available journals, which bears directly on a problem both judges are then considering in a case before them is infringing the copyright, rather than making "fair use" of his issue of that journal. Similarly with the photocopies of particular newspaper items and articles which are frequently given or sent by one friend to another.¹³ There is, in short, no inflexible rule excluding an entire copyrighted work from the area of "fair use." Instead, the extent of the copying is one important factor, but only one, to be taken into account, along with several others.

12 Leon v. Pacific Tel. Tel. Co., 91 F.2d 484, 486 (C.A. 9, 1937) and Public Affairs Associates, Inc. v. Rickover, 109 U.S.App.D.C. 128, 284 F.2d 262, 272 (C.A.D.C. 1960), vacated and remanded, 369 U.S. 111, 82 S.Ct. 580, 7 L.Ed.2d 604 (1962), which are often cited in this connection, both involved actual publication and distribution of many copies, not the simple making of a copy for individual personal or restricted use. In Wihtol v. Crow, 309 F.2d 777 (C.A. 8, 1962), 48 copies of the copyrighted song were made and distributed, and there were a number of public performances using these copies. It was as if the defendant had purchased one copy of sheet music and then duplicated it for an entire chorus.
On the other hand, New York Tribune, Inc. v. Otis Co., 39 F. Supp. 67 (S.D.N.Y. 1941), shows that copying of an entire copyrighted item is not enough, in itself, to preclude application of "fair use." Although it was already plain that an entire copyrighted item (a newspaper editorial) had been reproduced, the court ordered further proceedings to take account of other factors.

13 Verner Clapp, former Acting Librarian of Congress, has pointed out some of the uses of a photocopy for which the library copy original is unsuited (Can Copyright Law Respond to the New Technology?. 61 Law Lib. J. 387, 407 (1968)):
"I cannot submit the original conveniently in a court, in a suit of law. I cannot put the original into my filing cabinet. I can't shuffle it with notes in preparation for an address. I can't make notes on it. I can't conveniently give it to a typist. I can't use it as printer's copy. I can't send it through the mail without serious risk of loss of an original. With a photocopy I can do all these things and more, and this is the reason I want a copy."

Under these over-all standards, we have weighed the multiplicity of factors converging on the particular use of plaintiff's material made by NIH and NLM, as shown by this record. There is no prior decision which is dispositive and hardly any that can be called even close; we have had to make our own appraisal. The majority of the court has concluded that, on this record, the challenged use should be designated "fair," not "unfair." In the rest of this part of our opinion, we discuss seriatim the various considerations which merge to that conclusion. But we can help focus on what is probably the core of our evaluation by stating summarily, in advance, three propositions we shall consider at greater length: First, plaintiff has not in our view shown, and there is inadequate reason to believe, that it is being or will be harmed substantially by these specific practices of NIH and NLM; second, we are convinced that medicine and medical research will be injured by holding these particular practices to be an infringement; and, third, since the problem of accommodating the interests of science with those of the publishers (and authors) calls fundamentally for legislative solution or guidance, which has not yet been given, we should not, during the period before congressional action is forthcoming, place such a risk of harm upon science and medicine.

1. We start by emphasizing that (a) NIH and NLM are non-profit institutions, devoted solely to the advancement and dissemination of medical knowledge which they seek to further by the challenged practices, and are not attempting to profit or gain financially by the photocopying; (b) the medical researchers who have asked these libraries for the photocopies are in this particular case (and ordinarily) scientific researchers and practitioners who need the articles for personal use in their scientific work and have no purpose to reduplicate them for sale or other general distribution; and (c) the copied articles are scientific studies useful to the requesters in their work. On both sides — library and requester — scientific progress, untainted by any commercial gain from the reproduction, is the hallmark of the whole enterprise of duplication. There has been no attempt to misappropriate the work of earlier scientific writers for forbidden ends, but rather an effort to gain easier access to the material for study and research. This is important because it is settled that, in general, the law gives copying for scientific purposes a wide scope. See, e. g., Rosemont Enterprises, Inc. v. Random House, Inc., supra, 366 F.2d at 306-307; Loew's, Inc. v. Columbia Broadcasting System, Inc., 131 F. Supp. 165, 175 (S.D.Cal. 1955), aff'd, Benny v. Loew's Inc., 239 F.2d 532 (C.A. 9, 1956), aff'd by an equally divided Court, 356 U.S. 43, 78 S.Ct. 667, 2 L.Ed.2d 583 (1958); Greenbie v. Noble, 151 F. Supp. 45, 67-68 (S.D.N.Y. 1957); Thompson v. Gernsback, 94 F. Supp. 453, 454 (S.D.N.Y. 1950); Henry Holt Co. v. Liggett Myers Tobacco Co., 23 F. Supp. 302, 304 (E.D.Pa. 1938).

2. Both libraries have declared and enforced reasonably strict limitations which, to our mind, keep the duplication within appropriate confines. The details are set forth in Part I supra, and in our findings. Both institutions normally restrict copying on an individual request to a single copy of a single article of a journal issue, and to articles of less than 50 pages. Though exceptions are made, they do not appear to be excessive, unwarranted, or irrational. For instance, though on occasion one person was shown to have ordered or received more than one photocopy of the same article, the second copy was for a colleague's use or to replace an illegible or undelivered copy. Some care is also taken not to have excessive copying from one issue or one volume of the periodical. While a certain amount of duplication of articles does, of course, occur, it does not appear to be at all heavy.¹⁴ There is no showing whatever that the recipients use the libraries' photocopying process to sell the copies or distribute them broadly.

14 One survey of NIH operations shows only 4 instances of duplication in over 200 requests; at NLM, as of 1964, duplication occurred at a 10% rate in the 102 most heavily used journals (constituting one-third of total requests); if all requests were considered, the rate would be less. The Sophar Heilprin report (see Appendix), which is not friendly to library photocopying, estimates that for libraries generally the duplication rate was about 3% (p. iii).

NIH responds only to requests from its own personnel, so that its entire photoduplication system is strictly "in-house" — in the same way that a court's library may supply a judge of that court with a copy of a law journal article or a reported decision. NLM fulfills requests more generally but it has adopted the practice of not responding (outside of the Government) where the article appears in a recent (preceding 5 years) issue of a periodical on its "widely-available list". The result is that the duplication of recent issues of generally available journals is kept within the Government, and distribution to the larger medical public is limited to older, less available issues and to journals which are harder to obtain from medical libraries. It is a fair inference, supported by this record, that at the very least in the latter classes the demand has been inadequately filled by reprints and the publisher's sale of back issues. See, also, Part III, 4, infra. In those instances not covered by the "five year" policy, the impression left by the record is that, on the whole, older rather than current articles were usually requested.

Brushing aside all such breakdowns, plaintiff points to the very large number, in absolute terms, of the copies made each year by the two libraries. We do not think this decisive.¹⁵ In view of the large numbers of scientific personnel served and the great size of the libraries — NIH has over 100,000 volumes of journal materials alone, and NLM is currently binding over 18,000 journals each year — the amount of copying does not seem to us to have been excessive or disproportionate. The important factor is not the absolute amount, but the twin elements of (i) the existence and purpose of the system of limitations imposed and enforced, and (ii) the effectiveness of that system to confine the duplication for the personal use of scientific personnel who need the material for their work, with the minimum of potential abuse or harm to the copyright owner. The practices of NIH and NLM, as shown by the record, pass both of these tests, despite the large number of copies annually sent out.

15 In 1970, NIH copied 85,744 and NLM 93,746 articles.

Without necessarily accepting the full sweep of the concept that the library is nothing more than the individual requester's ministerial agent, we do agree that the NIH and NLM systems, as described in the evidence, are close kin to the current Library of Congress policy, see note 16, infra, of maintaining machines in the library buildings so that readers can do their own copying. The principal extension by NLM and NIH is to service requesters who cannot conveniently come to the building, as well as out-of-town libraries. But the personal, individual focus is still present. The reader who himself makes a copy does so for his own personal work needs, and individual work needs are likewise dominant in the reproduction programs of the two medical libraries — programs which are reasonably policed and enforced.

3. We also think it significant, in assessing the recent and current practices of the two libraries, that library photocopying, though not of course to the extent of the modern development, has been going on ever since the 1909 Act was adopted. In Part II, supra, we have set forth the practice of the Library of Congress at that time and for many years thereafter.¹⁶ In fact, photocopying seems to have been done in the Library at least from the beginning of this century. Can Copyright Law Respond to the New Technology? 61 Law. Lib. J., 387, 400 (1968) (comments of V. Clapp). In 1935 there was a so-called "gentlemen's agreement" between the National Association of Book Publishers (since defunct and the Joint Committee on Materials for Research (representing the libraries), stating in part: "A library * * * owning books or periodical volumes in which copyright still subsists may make and deliver a single photographic reproduction * * * of a part thereof to a scholar representing in writing that he desires such reproduction in lieu of loan of such publication or in place of manual transcription and solely for the purposes of research * * *." Though this understanding discountenanced photoduplication of an entire book it was regularly construed as allowing copying of articles. There have been criticisms of this pact, and we cite it, not as binding in any way on plaintiff or any other publisher, or as showing universal recognition of "single" photocopying, but as representing a very widely held view, almost 40 years ago, of what was permissible under the 1909 statute.

16 Currently, and for some time, the Library of Congress has said that copyright material will "ordinarily" not be photocopied by the Library "without the signed authorization of the copyright owner," but "[e]xceptions to this rule may be made in particular cases." The Library does, however, maintain machines which readers may themselves use for photocopying; these machines contain notices saying that "a single photocopy of copyrighted material may be made only for the purpose of study, scholarship, or research, and for no other purpose" and "the sale and/or further reproduction of any photocopied copyrighted materials is illegal."

There is other evidence that, until quite recently, library photocopying was carried on with apparent general acceptance. Witnesses in this case testified that such photocopying has been done for at least fifty years and is well-established. The National Library of Medicine Act, in 1956, by which NLM was created ( 42 U.S.C. § 275 et seq.), provided at § 276(a)(4) that the Secretary of Health, Education, and Welfare, through NLM, should "make available, through loans, photographic or other copying procedures or otherwise, such materials in the Library as he deems appropriate * * *"; and the Medical Library Assistance Act of 1965 ( 42 U.S.C. § 280b-1 et seq.) provided that grants be made to medical libraries for, among other things, "acquisition of duplication devices, facsimile equipment * * * and other equipment to facilitate the use of the resources of the library," 42 U.S.C. § 280b-7. These two pieces of legislation indicate to us that Congress knew in 1956 and 1965 of the practice of library photocopying, and assumed that it was not beyond the pale. The General Interlibrary Loan Code (revised in 1956), see Part I, supra, is a similar indication of the extent of the practice, and of the general position of the libraries (at the least) that such copying is permissible.

The fact that photocopying by libraries of entire articles was done with hardly any (and at most very minor) complaint, until about 10 or 15 years ago, goes a long way to show both that photoduplication cannot be designated as infringement per se, and that there was at least a time when photocopying, as then carried on, was "fair use." There have been, of course, considerable changes in the ease and extent of such reproduction, and these developments bear on "fair use" as of today, but the libraries can properly stand on the proposition that they photocopied articles for many years, without significant protest, and that such copying was generally accepted until the proliferation of inexpensive and improved copying machines, less than two decades ago, led to the surge in such duplication. The question then becomes whether this marked increase in volume changes a use which was generally accepted as "fair" into one which has now become "unfair."

4. There is no doubt in our minds that medical science would be seriously hurt if such library photocopying were stopped. We do not spend time and space demonstrating this proposition. It is admitted by plaintiff and conceded on all sides. See, e. g. Varmer, Photoduplication of Copyrighted Material by Libraries, Study No. 15, "Copyright Law Revision," Studies Prepared for the Subcommittee on Patents, Trademarks and Copyrights, Senate Judiciary Committee (1959), p. 49; Memorandum of General Counsel Willcox, Department of Health, Education and Welfare, June 7, 1965, Hearings before Subcommittee No. 3, Committee on the Judiciary, H. of Reps., 89th Cong., 1st Sess., on H.R. 4347, H.R. 5680, etc., "Copyright Law Revision," Part 2, 1132, 1133. The trial testimony of a number of the requesters and authors documents the point. The supply of reprints and back numbers is wholly inadequate; the evidence shows the unlikelihood of obtaining such substitutes for photocopies from publishers of medical journals or authors of journal articles, especially for articles over three years old.¹⁷ It is, moreover, wholly unrealistic to expect scientific personnel to subscribe regularly to large numbers of journals which would only occasionally contain articles of interest to them. Nor will libraries purchase extensive numbers of whole subscriptions to all medical journals on the chance that an indeterminate number of articles in an indeterminate number of issues will be requested at indeterminate times. The result of a flat proscription on library photocopying would be, we feel sure, that medical and scientific personnel would simply do without, and have to do without, many of the articles they now desire, need, and use in their work.¹⁸

17 Plaintiff itself publishes a notice to the effect that it does not attempt to keep a stock of back issues, and it refers requests for reprints to the author.

18 We think the alternative of compulsory licensing is not open to us under the present copyright statute. See, infra, Parts III, 6, and IV.

5. Plaintiff insists that it has been financially hurt by the photocopying practices of NLM and NIH, and of other libraries. The trial judge thought that it was reasonable to infer that the extensive photocopying has resulted in some loss of revenue to plaintiff and that plaintiff has lost, or failed to get, "some undetermined and indeterminable number of journal subscriptions (perhaps small)" by virtue of the photocopying. He thought that the persons requesting photocopies constituted plaintiff's market and that each photocopy user is a potential subscriber "or at least a potential source of royalty income for licensed copying."¹⁹ Studies rejecting as "fair use" the kind of photocopying involved here have also assumed, without real proof, that the journal publishers have been and will be injured. See, e. g., Project — New Technology and the Law of Copyright: Reprography and Computers, 15 U.C.L.A.L. Rev. 931 (1968); Sophar Heilprin, "The Determination of Legal Facts and Economic Guideposts with Respect to the Dissemination of Scientific and Educational Information as It Is Affected by Copyright — A Status Report" (1967).

19 It is wrong to measure the detriment to plaintiff by loss of presumed royalty income — a standard which necessarily assumes that plaintiff had a right to issue licenses. That would be true, of course, only if it were first decided that the defendant's practices did not constitute "fair use." In determining whether the company has been sufficiently hurt to cause these practices to become "unfair," one cannot assume at the start the merit of the plaintiff's position, i. e., that plaintiff had the right to license. That conclusion results only if it is first determined that the photocopying is "unfair."

The record made in this case does not sustain that assumption. Defendant made a thorough effort to try to ascertain, so far as possible, the effect of photoduplication on plaintiff's business, including the presentation of an expert witness. The unrefuted evidence shows that (a) between 1958 and 1969 annual subscriptions to the four medical journals involved increased substantially (for three of them, very much so), annual subscription sales likewise increased substantially, and total annual income also grew; (b) between 1959 and 1966, plaintiff's annual taxable income increased from $272,000 to $726,000, fell to $589,000 in 1967, and in 1968 to $451,000; (c) but the four journals in suit account for a relatively small percentage of plaintiff's total business and over the years each has been profitable (though 3 of them show losses in particular years and in all years the profits have not been large, varying from less than $1,000 to about $15,000, some of which has been shared with the sponsoring medical societies);²⁰ and (d) plaintiff's business appears to have been growing faster than the gross national product or of the rate of growth of manpower working in the field of science. Defendant's expert concluded that the photocopying shown here had not damaged plaintiff, and may actually have helped it.²¹ The record is also barren of solid evidence that photocopying has caused economic harm to any other publisher of medical journals.

20 Defendant explains the loss years and the fall-off in some subscriptions in some years as due to particular circumstances (which are spelled out) other than photocopying.

21 The trial judge referred to two instances in which subscribers cancelled subscriptions because of the availability of photocopying. Defendant is correct that both instances rest on hearsay, and in any event this small number of purported cancellations is de minimis in view of the more solid and detailed proof as to the health of plaintiff's journals and the increase in their subscription lists.

Plaintiff has never made a detailed study of the actual effect of photocopying on its business, nor has it refuted defendant's figures. It has relied for its assumption (in the words of the chairman of its board) on "general business common sense and things that you hear from subscribers, librarians and so forth." Its argument — and that of the other supporters of its position²² — is that there "must" be an effect because photocopies supplant the original articles, and if there were no photocopying those who now get the copies would necessarily buy the journals or issues. But this untested hypothesis, reminiscent of the abstract theorems beloved of the "pure" classical economics of 70 or 80 years ago, is neither obvious nor selfproving. One need not enter the semantic debate over whether the photocopy supplants the original article itself or is merely in substitution for the library's loan of the original issue to recognize, as we have already pointed out, that there are other possibilities. If photocopying were forbidden, the researchers, instead of subscribing to more journals or trying to obtain or buy back-issues or reprints (usually unavailable), might expend extra time in note-taking or waiting their turn for the library's copies of the original issues — or they might very well cut down their reading and do without much of the information they now get through NLM's and NIH's copying system. The record shows that each of the individual requesters in this case already subscribed, personally, to a number of medical journals, and it is very questionable how many more, if any, they would add. The great problems with reprints and back-issues have already been noted. In the absence of photocopying, the financial, time-wasting, and other difficulties of obtaining the material could well lead, if human experience is a guide, to a simple but drastic reduction in the use of the many articles (now sought and read) which are not absolutely crucial to the individual's work but are merely stimulating or helpful. The probable effect on scientific progress goes without saying, but for this part of our discussion the significant element is that plaintiff, as publisher and copyright owner, would not be better off. Plaintiff would merely be the dog in the manager.

22 The published literature does not reveal any careful, thorough, impartial study of this question. Often there is no attempt to ascertain the actual economic impact on the publishers and authors; when inquiry has been made of the latter, their conclusory generalizations of injury have been accepted uncritically.

Since plaintiff and those who take the same view have not attempted any hard factual study of the actual effect of photocopying, it is not surprising that others have concluded against an adverse impact. The 1962 Fry Report (George Fry Associates, "Survey of Copyrighted Material Reproduction Practices in Scientific and Technical Fields," March 1962) states that the "basic conclusion of this report is that at the present time, no significant damage occurs to the copyright holders in the scientific and technical fields although duplication of this material is widespread and is growing rapidly." In March 1965, Dan Lacy, Managing Director, American Book Publishers Council, told a House of Representatives committee: "It has been pointed out that recent technological developments have enormously increased the amount of photocopying in libraries and technology is continuing to change rapidly. Most of this photocopying, at least at present, probably consists of excerpts and probably mostly of journal articles. Most of it at present is probably undertaken in lieu of manual note taking, typing, or handwriting a copy, and in lieu of library loan rather than in lieu of buying a copy" (emphasis added). Hearings before Subcommittee No. 3, Committee on the Judiciary, H. of Reps., 89th Cong., 1st Sess., on H.R. 4347, H.R. 5680, etc., "Copyright Law Revision," Part 1, p. 120. The record in this case does not prove that the situation was any different at the time of the trial.

To us it is very important that plaintiff has failed to prove its assumption of economic detriment, in the past or potentially for the future. One of the factors always considered with respect to "fair use," see supra, is the effect of the use on the owner's potential market for the work. This record simply does not show a serious adverse impact, either on plaintiff or on medical publishers generally, from the photocopying practices of the type of NIH and NLM. In the face of this record, we cannot mechanically assume such an effect, or hold that the amount of photoduplication proved here "must" lead to financial or economic harm. This is a matter of proof and plaintiff has not transformed its hypothetical assumption, by evidence, into a proven fact.

In this connection it is worth noting that plaintiff does not have to concern itself, with respect to these journals, with authors or medical societies who are interested in a financial return. The authors, with rare exceptions, are not paid for their contributions, and those societies which share profits do not press for greater financial benefits. Indeed, some of the authors of the copied articles involved in this case testified at the trial that they favored photocopying as an aid to the advancement of science and knowledge.

6. Added to the powerful factors we have been considering is another (already suggested by the discussion in Part II, supra) — the grave uncertainty of the coverage of "copy" in Section 1 of the 1909 Act and the doubt whether it relates at all to periodicals.²³ The latitude for "fair use" is of course lessened to the extent Congress has been explicit in spelling out protection to the copyright owner. But Congress has, up to now, left the problem of photocopying untouched by express provision and only doubtfully covered to any extent by the generalizations of Section 1. The statute must, of course, "be applied to new situations not anticipated by Congress, if, fairly construed, such situations come within its intent and meaning" (Jerome H. Remick Co. v. American Automobile Accessories Co., 5 F.2d 411 (C.A. 6, 1925), cert. denied, 269 U.S. 556, 46 S.Ct. 19, 70 L.Ed. 409), but our problem is with the latter part of this quotation. That being so, we think that, in evaluating "fair use," we should give the benefit of the doubt — until Congress acts more specifically — to science and the libraries, rather than to the publisher and the owner.

23 The same is true of "print," "reprint," and "publish," as applied to the challenged practices of NLM and NIH.

While, as we have said, this record fails to show that plaintiff (or any other medical publisher) has been substantially harmed by the photocopying practices of NIH and NLM, it does show affirmatively that medical science will be hurt if such photocopying is stopped. Thus, the balance of risks is definitely on defendant's side — until Congress acts more specifically, the burden on medical science of a holding that the photocopying is an infringement would appear to be much greater than the present or foreseeable burden on plaintiff and other medical publishers of a ruling that these practices fall within "fair use."

Plaintiff's answer is that it is willing to license the libraries, on payment of a reasonable royalty, to continue photocopying as they have. Our difficulty with that response — in addition to the absence of proof that plaintiff has yet been hurt, and the twin doubts whether plaintiff has a viable license system and whether any satisfactory program can be created without legislation²⁴ — is that the 1909 Act does not provide for compulsory licensing in this field. All that a court can do is to determine the photocopying an infringement, leaving it to the owner to decide whether to license or to prohibit the practice. Plaintiff and other publishers cannot enjoin governmental libraries (because 28 U.S.C. § 1498, supra note 1, is the sole remedy), but if photocopying of this type is an infringement the owners are free under the law to seek to enjoin any and all nongovernmental libraries. A licensing system would be purely voluntary with the copyright proprietor. We consider it entirely beyond judicial power, under the 1909 Act,²⁵ to order an owner to institute such a system if he does not wish to. We think it equally outside a court's present competence to turn the determination of "fair use" on the owner's willingness to license — to hold that photocopying (without royalty payments) is not "fair use" if the owner is willing to license at reasonable rates but becomes a "fair use" if the owner is adamant and refuses all permission (or seeks to charge excessive fees).

24 Defendant and its amici strongly attack plaintiff's so-called licensing plan as nothing more than a shell. The American Library Association points out, for instance, that the Williams Wilkins license would apparently not apply to inter-library loans or to requests from persons not physically present in the library building.
There is also debate over whether a feasible ASCAP-type or clearinghouse system can be developed without legislation, and if so whether it would be desirable. See, e. g., Note, Education and Copyright Law: An Analysis of the Amended Copyright Revision Bill and Proposals for Statutory Licensing and a Clearinghouse System, 56 Va.L.Rev. 664 (1970); also published as MacLean, Education and Copyright Law: An Analysis of the Amended Copyright Revision Bill and Proposals for Statutory Licensing and a Clearinghouse System, in ASCAP, "Copyright Law Symposium, Number Twenty," 1 (1972); Breyer, The Uneasy Case for Copyright: A Study of Copyright in Books, Photocopies and Computer Programs, 84 Harv.L.Rev. 281, 33Off. (1970); Note: New Technology and the Law of Copyright: Reprography and Computers, 15 U.C.L.A.L. Rev. 939, 961ff. (1968).

25 A court's powers under the anti-trust legislation is another matter.

The truth is that this is now preeminently a problem for Congress: to decide the extent photocopying should be allowed, the questions of a compulsory license and the payments (if any) to the copyright owners, the system for collecting those payments (lump-sum, clearinghouse, etc.), the special status (if any) of scientific and educational needs. Obviously there is much to be said on all sides. The choices involve economic, social, and policy factors which are far better sifted by a legislature. The possible intermediate solutions are also of the pragmatic kind legislatures, not courts, can and should fashion. But Congress does not appear to have put its mind directly to this problem in 1909, undoubtedly because the issue was not considered pressing at that time. That statute is, unfortunately, the one we must apply, and under it we have the choice only of thumb's up or thumb's down, for the photocopying practice involved in this litigation, without any real Congressional guidance. Intermediate or compromise solutions are not within our authority.²⁶ The theme of this subpart 6 of Part III of the opinion is that, on balance and on this record, thumb's up seems to us less dangerous to the varying interests at stake during the period which remains before Congress definitively takes hold of the subject.

26 It has been suggested, however, that publishers now have the power to adopt the intermediate solution of charging more for subscriptions sold to libraries or other entities which engage regularly in photocopying.

7. The revision of the 1909 Act is now under consideration and has been for several years. The House of Representatives passed a bill in the 90th Congress (in April 1967), but the Senate has not acted.²⁷ In its report on the bill which the House adopted (H.R. Rep. No. 83, 90th Cong., 1st Sess.), the House Committee on the Judiciary discussed the existing doctrine of "fair use" at some length (pp. 29-37). We cite these comments, not as binding on us, but as the official views on the extent of "fair use" of the committee of the House of Representatives with cognizance over copyright; as such, they are and should be influential.

27 A synopsis of the revision effort (up to 1968) is set forth in Fortnightly Corp. v. United Artists Television, Inc., 392 U.S. 390, 396 n. 17, 88 S.Ct. 2084, 20 L.Ed.2d 1176 (1968).

The report makes it very clear that photocopying can be a "fair use", in proper circumstances; it negatives the notion that copying of a complete work can never be a "fair use"; and it obviously believes that the doctrine is flexible, depending upon the particular situation.²⁸ The report does not, however, express a categorical or clear view whether photocopying of the sort we have in this case is or is not a "fair use" under the doctrine as it has been developing. Rather, the committee's observations are delphic, with each side being able to quote to us one or another passage, or to argue by analogy from the specific situation (classroom teaching) considered in greatest detail in the report.

28 The report says (p. 29) that "* * * since the doctrine is an equitable rule of reason, no generally applicable definition is possible, and each case raising the question must be decided on its own facts"; that (p. 32) the committee endorses "the purpose and general scope of the judicial doctrine of fair use, as outlined earlier in this report, but there is no disposition to freeze the doctrine in the statute, especially during a period of technological change. Beyond a very broad statutory explanation of what fair use is and some of the criteria applicable to it, the courts must be free to adapt the doctrine to particular situations on a case-by-case basis; and that (p. 32) "Section 107, as revised by the committee, is intended to restate the present judicial doctrine of fair use, not to change, narrow, or enlarge it in any way."

Specifically on library photocopying the committee says (p. 36) that it does not favor a specific provision dealing with that subject, and it adds: "Unauthorized library copying, like everything else, must be judged a fair use or an infringement on the basis of all of the applicable criteria and the facts of the particular case. Despite past efforts, reasonable arrangements involving a mutual understanding of what generally constitutes acceptable library practices, and providing workable clearance and licensing conditions, have not been achieved and are overdue. The committee urges all concerned to resume their efforts to reach an accommodation under which the needs of scholarship and the rights of authors would both be respected."

We read this report, as a whole, as recognizing affirmatively that, under the existing law, library photocopying can be "fair use" in proper circumstances, and as leaving the determination of whether the particular circumstances are proper ones to an evaluation "of all the applicable criteria and the facts of the particular case." That is, of course, the overall standard we are using, and therefore we consider our approach to be consistent with that of the Committee. Although one cannot say that the report places its sanction directly on the photocopying practices now before us, neither does it suggest or intimate that they are "unfair." That question is left open. The report is nevertheless helpful because it indicates the correctness of our general approach, and also because it contradicts the concept, urged by plaintiff, that photocopying of an entire article is necessarily an infringement.

8. The last component we mention, as bearing on "fair use", is the practice in foreign countries. The copyright legislation of the United Kingdom, New Zealand, Denmark, Finland, Italy, Norway, Sweden, France, the German Federal Republic, Lichtenstein, Mexico, the Netherlands, and the U.S.S.R. have specific provisions which we think would cover the photocopying activities of NLM and NIH. Canada, India, Ireland and South Africa, while having no specific provisions permitting copying of copyrighted works for the purposes of private research and study, do provide, more generally, that fair dealing for purposes of private study or research shall not be an infringement.²⁹ These provisions in foreign countries with problems and backgrounds comparable to our own are highly persuasive that the copying done here should be considered a "fair use," not an infringement.³⁰ Where Congress has left such a large void to be filled entirely by the courts, it is appropriate for us to consider what other jurisdictions have done either by way of legislation or judicial decision.

29 The foreign laws are compiled in Copyright Laws and Treaties of the World, published by UNESCO.

30 The general report of the Committee of Experts on the Photographic Reproduction of Protected Works [a joint committee of UNESCO and the United International Bureau for the Protection of Intellectual Property (BIRPI)] recommended that libraries should have the right to provide one copy free of copyright for each user provided that such copy, in the case of a periodical, shall not be more than a single article. 4 Copyright 195, 197 (1968).

IV

Fusing these elements together, we conclude that plaintiff has failed to show that the defendant's use of the copyrighted material has been "unfair," and conversely we find that these practices have up to now been "fair." There has been no infringement. As Professor (now Mr. Justice) Kaplan observed, it is "fundamental that `use' is not the same as `infringement' [and] that use short of infringement is to be encouraged * * *." Kaplan, An Unhurried View of Copyright 57 (1967); see Fortnightly Corp. v. United Artists Television, Inc., 392 U.S. 390, 393-395, 88 S.Ct. 2084, 20 L.Ed.2d 1176 (1968).

So as not to be misunderstood, we re-emphasize four interrelated aspects of our holding. The first is that the conclusion that defendant's particular use of plaintiff's copyrighted material has been "fair" rests upon all of the elements discussed in Part III, supra, and not upon any one, or any combination less than all. We do not have to, and do not, say that any particular component would be enough, either by itself or together with some of the others. Conversely, we do not have to, and do not, say that all the elements we mention are essential to a finding of "fair use." They all happen to be present here, and it is enough for this case to rule, as we do, that at least when all co-exist in combination a "fair use" is made out.

Connected with this point is the second one that our holding is restricted to the type and context of use by NIH and NLM, as shown by this record. That is all we have before us, and we do not pass on dissimilar systems or uses of copyrighted materials by other institutions or enterprises, or in other fields, or as applied to items other than journal articles, or with other significant variables. We have nothing to say, in particular, about the possibilities of computer print-outs or other such products of the newer technology now being born. Especially since we believe, as stressed infra, that the problem of photo and mechanical reproduction calls for legislative guidance and legislative treatment, we feel a strong need to obey the canon of judicial parsimony, being stingy rather than expansive in the reach of our holding.

The third facet articulates the same general premise — our holding rests upon this record which 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. We leave untouched, because we do not have to reach them, the situations where the copyright owner is shown to be hurt or the recipients (or their interests) would not be significantly injured if the reproductions were ruled to infringe.

Finally, but not at all least, we underline again the need for Congressional treatment of the problems of photocopying. The 1909 Act gives almost nothing by way of directives, the judicial doctrine of "fair use" is amorphous and open-ended, and the courts are now precluded, both by the Act and by the nature of the judicial process, from contriving pragmatic or compromise solutions which would reflect the legislature's choices of policy and its mediation among the competing interests. The Supreme Court has pointed out that such a "job is for Congress" (Fortnightly Corp. v. United Artists Television, Inc., 392 U.S. 390, 401, 88 S.Ct. 2084, 20 L.Ed.2d 1176 (1968)), and in an earlier copyright case in which it was recognized that the owner might be morally or economically entitled to protection the Court applied "the act of Congress [as it] now stands," saying that the other "considerations properly address themselves to the legislative, and not to the judicial, branch of the Government." White-Smith Music Co. v. Apollo Co., 209 U.S. 1, 18, 28 S.Ct. 319, 52 L.Ed. 655 (1908). Hopefully, the result in the present case will be but a "holding operation" in the interim period before Congress enacts its preferred solution.

On this record and for these reasons, we hold the plaintiff not entitled to recover and dismiss the petition.

APPENDIX SOME DISCUSSIONS OF LIBRARY PHOTOCOPYING

B. Varmer, Photoduplication of Copyrighted Material by Libraries, Study No. 15, Copyright Law Revision, Studies Prepared for Senate Comm. on the Judiciary, 86th Cong., 2d Sess. (1960); G. Sophar and L. Heilprin, The Determination of Legal Facts and Economic Guide-posts with Respect to the Dissemination of Scientific and Educational Information as it is Affected by Copyright — A Status Report, Final Report, Prepared by The Committee to Investigate Copyright Problems Affecting Communication in Science and Education, Inc., for the U.S. Department of Health, Education, and Welfare, Project No. 70793 (1967); Report of the Register of Copyrights on the General Revision of the U.S. Copyright Law to the House Comm. on the Judiciary, 87th Cong., 2d Sess. at 25-26 (1961); Project — New Technology and the Law of Copyright: Reprography and Computers, 15 U.C.L.A.L. Rev. 931 (1968); V. Clapp, Copyright — A Librarian's View, Prepared for the National Advisory Commission on Libraries, Association of American Libraries (1968); Schuster and Bloch, Mechanical Copyright, Copyright Law, and the Teacher, 17 Clev.-Mar.L.Rev. 299 (1968); "Report on Single Copies" — Joint Libraries Committee on Fair Use in Photocopying, 9 Copyright Soc'y Bull. 79 (1961-62); Breyer, "The Uneasy Case for Copyright: A Study of Copyright in Books Photocopies, and Computer Programs," 84 Harv.L.Rev. 281 (1970); Note, "Statutory Copyright Protection for Books and Magazines Against Machine Copying," 39 Notre Dame Lawyer 161 (1964); Note, Education and Copyright Law: An Analysis of the Amended Copyright Revision Bill and Proposals for Statutory Licensing and a Clearinghouse System," 56 Va.L. Rev. 664 (1970); Hattery and Bush (ed.), Reprography and Copyright Law (1964).

[57] COWEN, Chief Judge (dissenting):

It is my opinion that our former Trial Judge James F. Davis fully and correctly resolved the difficult and perplexing issues presented by this case in his scholarly and well-reasoned opinion. I would therefore adopt his opinion, findings of fact, and recommended conclusions of law as a basis for a judgment by the court in favor of the plaintiff.

In its discussion of the grounds for the decision which rejected the trial judge's conclusions, the court has, in my opinion, unduly emphasized the facts that are favorable to the defendant and has given inadequate consideration to other facts which led the trial judge to reach a contrary result. For these reasons, I am incorporating in this dissent those portions of the trial judge's opinion which I think are particularly pertinent to the grounds upon which the case has been decided. In view of the court's extensive discussion of the issues and its consideration of some matters not argued to the trial judge, I am supplementing his opinion with the material that follows.

As a preface to my disagreement with the court, I think it would be helpful to point out that this is not a case involving copying of copyrighted material by a scholar or his secretary in aid of his research, nor is it a case where a teacher has reproduced such material for distribution to his class. Also, it is not a case where doctors or scientists have quoted portions of plaintiff's copyrighted articles in the course of writing other articles in the same field. We are not concerned here with a situation in which a library makes copies of ancient manuscripts or worn-out magazines in order to preserve information. What we have before us is a case of wholesale, machine copying, and distribution of copyrighted material by defendant's libraries on a scale so vast that it dwarfs the output of many small publishing companies. In order to fill requests for copies of articles in medical and scientific journals, the NIH made 86,000 Xerox copies in 1970, constituting 930,000 pages. In 1968, the NLM distributed 120,000 copies of such journal articles, totalling about 1.2 million pages. As the trial judge correctly observed, this extensive operation is not only a copying of the copyrighted articles, it is also a reprinting by modern methods and publication by a very wide distribution to requesters and users.

I

Photographic Reproduction of Plaintiff's Journal Articles Is An Abridgement of the Copyright Owner's Exclusive Right To Copy

The majority maintains there is a "solid doubt" whether and how far the word "copy" in Section 1(a) of the 1909 Copyright Act applies to books and journals. The argument continues that the infringement of periodical articles can come only through "printing," "reprinting," or "publishing." Certainly few things in the law are beyond all doubt or qualification. I think it is apparent, however, from the wording of the 1909 Act, and from the cases interpreting that Act, that Congress intended the word "copy" to apply to books and journals as well as other copyrightable materials. Section 1(a) of the 1909 Act gives the copyright proprietor the exclusive right to "print, reprint, publish, copy, and vend the copyrighted work." 17 U.S.C. § 1 (1970). It follows that copying of a substantial portion of the copyrighted work by someone other than the copyright owner would be an infringement.

I think the trial judge correctly concluded that there is nothing in the legislative history of the 1909 Act which indicates that a restrictive definition of the word "copy" was intended. A significant change in the 1909 Act was the elimination of sections 4964 and 4965 of the prior copyright statute, which it is claimed, are the source of the distinction between the copying of books and the copying of other copyrighted material.¹ By removing those two sections and by adopting the general classification of "copyrighted works" in Section 1(a) and a general listing of all copyrightable works in Section 5 of the 1909 Act, Congress obliterated the distinction, if there ever had been one, between the copying of books and the copying of other materials.²

1 These sections had been in the copyright law since 1831 and had been twice re-enacted. Act of February 3, 1831, ch. 16, §§ 6 and 7, 4 Stat. 436; Act of July 8, 1870, ch. 230, §§ 99 and 100, 16 Stat. 214; Act of March 3, 1891, ch. 565, §§ 4964 and 4965, 26 Stat. 1109.

2 The trial judge observed that it was the intent of Congress in all the copyright acts to proscribe the unauthorized duplication of copyrighted works. The words used in the various statutes were simply attempts to define the then-current means by which duplication could be effected. I believe this is a fair statement, but it is not necessary to debate the statutory history in light of the changes in the 1909 Act.

As a result of the simple clarity in the phrasing of the copyright owner's exclusive rights in the 1909 Act, it is not surprising that numerous court decisions interpreting that Act have focused on the copying of copyrighted material (including books and other items of this type) as the infringing act. See, e. g., Harold Lloyd Corp. v. Witwer, 65 F.2d 1, 16-19 (9th Cir. 1933); King Features Syndicate v. Fleischer, 299 F. 533, 535 (2d Cir. 1924).

I have not been able to find one decision since the 1909 Act which has held that the word "copy" in section 1(a) would not apply to the making of one or a number of copies of a book or other material of this type. The cases have simply not recognized the claimed distinction between copying and printing or publishing. For example, in New York Tribune, Inc. v. Otis Co., 39 F. Supp. 67 (S.D.N.Y. 1941), the court found the making of photostatic copies of plaintiff's newspaper editorial and masthead to be a "good cause of action on its face," and denied defendant's motion for summary judgment. Id. at 68. The defendant in that case had distributed the photocopies to a selected list of public officials, bankers, educators, economists and other persons. The court drew no distinction between printing, publishing, or copying. By comparison, the copying and distributing of the newspaper editorial and masthead in that case is very similar to the copying and distributing of the journal articles in the present action.³

3 For cases involving the copying of segments from a copyrighted catalog by photographic reproduction, see Hedeman Products Corp. v. Tap-Rite Prods. Corp., 228 F. Supp. 630, 633-634 (D.N.J. 1964); R. R. Donnelley Sons Co. v. Haber, 43 F. Supp. 456, 458-459 (E.D.N.Y. 1942).

Therefore, I do not think there is substantial doubt that the photocopying by defendant's libraries is a copy of the plaintiff's journal articles in violation of the copyright owner's exclusive right to copy or to multiply copies of his work under section 1(a). I can see no reason to draw a distinction between copying of "books" and copying of other materials when the distinction is expressly rejected on the face of the copyright statute, has not been observed in numerous cases applying the 1909 Act, and has no reasonable basis in light of the purposes of copyright protection.⁴

4 The fact that Dr. Putnam, the Librarian of Congress at the time of the 1909 Act, interpreted the word "copy" not to include library photoduplication is no indication that the Congress drafted the statute with this intent. The absence of any provision allowing library photoduplication in the statute or the legislative history indicates, as much as anything else, that Congress did not consider it to be exempt from the Act. The many efforts to amend the law to authorize photocopying by libraries provide a strong indication that existing law was not intended to grant this exemption to libraries. See n. 14, trial judge's opinion.

II [67] The Photocopying of Plaintiff's Copyrighted Articles Was Not Fair Use

1. Realizing the necessity for showing that the defendant's unauthorized copying of plaintiff's articles was both reasonable and insubstantial, the court relies heavily on policies which were adopted by the libraries in 1965. Although these policies were designed to limit the extent of copying that had been done in prior years, the trial judge's opinion and the findings of fact show the exceptions are routinely granted by the defendant's libraries, that there is no way to enforce most of the limitations, and that defendant is operating a reprint service which supplants the need for journal subscriptions.

In particular, the trial judge has, I think, clearly demonstrated that the claimed "single-copy-per-request" limitation is both illusory and unrealistic. He has found, and it is not disputed, that the libraries will duplicate the same article over and over again, even for the same user, within a short space of time. NLM will supply requesters photocopies of the same article, one after another, on consecutive days, even with knowledge of such facts. I find great difficulty in detecting any difference between the furnishing by defendant's libraries of ten copies of one article to one patron, which he then distributes, and giving each of ten patrons one copy of the same article. The damage to the copyright proprietor is the same in either case.

2. The law is well settled, and I believe not questioned by the court in this case, that under Section 3 of the Copyright Act, plaintiff's copyrights of the journals cover each article contained therein as fully as if each were individually copyrighted. Section 3 expressly mentions periodicals, and for the purpose of determining whether there has been infringement, each copyrightable component is to be treated as a complete work. Markham v. A. E. Borden Co., 206 F.2d 199, 201 (1st Cir. 1953).

It is undisputed that the photocopies in issue here were exact duplicates of the original articles; they were intended to be substitutes for and they served the same purpose as the original articles. They were copies of complete copyrighted works within the meaning of Sections 3 and 5 of the Copyright Act. This is the very essence of wholesale copying and, without more, defeats the defense of fair use. The rule to be applied in such a situation was stated in Leon v. Pacific Telephone Telegraph Co., 91 F.2d 484, 486 (9th Cir. 1937) as follows:

Counsel have not disclosed a single authority, nor have we been able to find one, which lends any support to the proposition that wholesale copying and publication of copyrighted material can ever be fair use.

For other cases to the same effect, see Public Affairs Associates, Inc. v. Rickover, 109 U.S.App.D.C. 128, 284 F.2d 262, 272 (1960), judgment vacated for insufficient record, 369 U.S. 111, 82 S.Ct. 580, 7 L.Ed.2d 604 (1962); Benny v. Loew's Inc., 239 F.2d 532, 536 (9th Cir. 1956), aff'd by an equally divided court sub nom., Columbia Broadcasting System, Inc. v. Loew's Inc., 356 U.S. 43, 78 S.Ct. 667, 2 L.Ed.2d 583 (1958); Holdredge v. Knight Publishing Corp., 214 F. Supp. 921, 924 (S.D.Cal. 1963). See also M. Nimmer, Nimmer on Copyright § 145 at 650-51 (1973 Ed.).

Although the majority states that the rule announced in the cases cited above is an "overbroad generalization, unsupported by the decisions and rejected by years of accepted practice," the court cites no decisions in support of its position.

3. I recognize that the doctrine of fair use permits writers of scholarly works to make reasonable use of previously copyrighted material by quotation or paraphrase, at least where the amount of copying is small and reliance on other sources is demonstrated. See, e. g., Rosemont Enterprises, Inc. v. Random House, Inc., 366 F.2d 303 (2d Cir. 1966), cert. denied, 385 U.S. 1009, 87 S. Ct. 714, 17 L.Ed.2d 546 (1967); Simms v. Stanton, 75 F. 6, 13-14 (C.C.N.D.Cal. 1896). However, I think the basic error in the court's decision is its holding that the fair use privilege usually granted to such writers should be extended to cover the massive copying and distribution operation conducted by defendant's libraries. The articles are not reproduced by the libraries to enable them to write other articles in the same field. In fact, booksellers and licensed copiers of plaintiff's journals sell copies of journal articles to the same class of users and for the same purposes as the copies reproduced by defendant's libraries.

I do not believe that anyone would contend that the ultimate use of the purchased articles by scientists, doctors, or drug companies would permit the commercial concerns mentioned to reproduce copies without plaintiff's permission. In an effort to overcome this obstacle, the majority relies in part on the nature and function of the NIH and the NLM and the fact that the articles are reproduced and distributed free of charge. I do not know of any case which holds that an unauthorized reproduction which is made without profit amounts to fair use by the infringer, and there are decisions to the contrary.⁵

5 It has been held that the copying or printing of something which has been lawfully copyrighted is an infringement "without any requirement that there be a sale or that profits be made from sale of the copies." Chappell Co., Inc. v. Costa, 45 F. Supp. 554, 556 (S.D.N.Y. 1942). In Wihtol v. Crow, 309 F.2d 777 (8th Cir. 1962), the First Methodist Church was found to be liable for a choral instructor's copying of a copyrighted song.

Moreover, as plaintiff has pointed out, almost every service provided by Government agencies is financed by appropriated funds and furnished without charge to the recipient. If Congress had intended to relieve Government agencies from liability for copyright infringement whenever the material is copied or otherwise reproduced without charge to the recipient, there would have been no need for the enactment of the 1960 Amendment, now 28 U.S.C. § 1498(b), which gives us jurisdiction of this action.

Defendant also argues that its libraries are entitled to the fair use privilege of scientists, researchers, or scholars, because the libraries act as their agent in making the photocopies at their request. This argument is so far-fetched that the majority balks at embracing it completely. It collides with reality. The libraries installed and operate the reproduction and distribution operation on their own initiative and without any kind of an agreement with the ultimate users of the copies. There is no showing that these alleged — and in the case of NLM generally unknown — principals have any say in the formulation of the policies and practices of the photocopying operation. The libraries decide, without consulting or obtaining the consent of the alleged principals, whether to loan the original of the journals or to provide photocopies. The libraries are no more than agent of the users of the material than are the venders of plaintiff's magazines and the commercial concerns which are licensed by plaintiff to make and sell copies to doctors and scientists. The essential elements of agency are wholly lacking.

4. The trial judge found that it is reasonable to infer from the evidence that the extensive unauthorized copying of plaintiff's journal has resulted in some loss of revenue and serves to diminish plaintiff's potential market for the original articles. Since the inferences made by the trial judge may reasonably be drawn from the facts and circumstances in evidence, they are presumptively as correct as his findings of fact. Bonnar v. United States, 438 F.2d 540, 542, 194 Ct.Cl. 103, 109 (1971). See also Baumgartner v. United States, 322 U.S. 665, 670, 64 S.Ct. 1240, 88 L.Ed. 1525 (1944); Penn-Texas Corp. v. Morse, 242 F.2d 243, 247 (7th Cir. 1957). Accordingly, under the standards which we employ for reviewing the findings of our trial judges, I would adopt these findings. Davis v. United States, 164 Ct.Cl. 612, 616-617 (1964); Wilson v. United States, 151 Ct.Cl. 271, 273 (1960).

Although the court states that it rejects the trial determinations as to both actual and potential damage to plaintiff, I think the opinion shows that the court's conclusion is based primarily on its finding that plaintiff failed to prove actual damages. In so doing, the majority relies heavily on evidence that the plaintiff's profits have grown faster than the gross national product and that plaintiff's annual taxable income has increased. This evidence is irrelevant to the economic effects of photocopying the journals in this case, because these periodicals account for a relatively small percent of plaintiff's total business. Moreover, the extent of plaintiff's taxable income for the years mentioned does not reflect the effect of defendant's photocopying of plaintiff's journals, and particularly the effect it will have on the prospects for continued publication in the future.

By the very nature of an action for infringement, the copyright proprietor often has a difficult burden of proving the degree of injury. It is well established, however, that proof of actual damages is not required, and the defense of fair use may be overcome where potential injury is shown. See, e. g., Henry Holt Co., Inc. v. Liggett Myers Tobacco Co., 23 F. Supp. 302, 304 (E.D.Pa. 1938). As Professor Nimmer has stated, the courts look to see whether defendant's work "tends to diminish or prejudice the potential sale of the plaintiff's work." M. Nimmer, Nimmer on Copyright § 145 at 646 (1973 ed.).

The problem posed by library photocopying of copyrighted material has long been a subject of controversy. Several studies of this problem have pointed out that extensive photocopying by libraries is unfair because of its potential damage to the copyright owner. The trial judge has quoted from the reports of several of these studies.⁶

6 See the trial judge's opinion for quotations from B. Varmer, Photoduplication of Copyrighted Material by Libraries, Study No. 15, Copyright Law Revision, Studies Prepared for the Senate Comm. on the Judiciary, 86th Cong., 2d Sess. 62-63 (1960); Report of the Register of Copyrights on the General Revision of the U.S. Copyright Law, House Committee Print, 87th Cong., 1st Sess. 25-26 (1961). M. Nimmer, Nimmer on Copyright § 145 at 653-54 (1973 ed.). See also Crossland, The Rise and Fall of Fair Use: The Protection of Literary Materials Against Copyright Infringement by New and Developing Media, 20 S.Car.L.Rev. 153, 154 (1968).

In a thorough and thoughtful discussion of the effects of reprography, prepared at the University of California, Los Angeles, and funded by the National Endowment for the Arts, it is stated:

It has long been argued that copying by hand "for the purpose of private study and review" would be a fair use. Users are now asserting that machine copying is merely a substitute for hand copying and is, therefore, a fair use. But this argument ignores the economic differences between the two types of copying. Copying by hand is extremely time consuming and costly, and is not an economic threat to authors. Viewing reprography as though it were hand copying, however, overlooks the effect of the total number of machine copies made. Few people hand copy, but millions find machine copying economical and convenient. Allowing individual users to decide that their machine copying will not injure the author and will thus be a fair use fails to take into account the true economic effect when thousands of such individual decisions are aggregated.

The problem is vividly presented by the practice of the National Library of Medicine. The Library justifies its distribution of reprographic copies of journal articles to biomedical libraries (without permission of the copyright owner) on the basis of a 1939 understanding between publishers and libraries called the "Gentlemen's Agreement." Under this agreement photocopies are permitted whenever the user would have made a hand copy himself, the rationale being that no purchases of the author's work are displaced under these circumstances. When an individual would actually copy by hand, the theory is valid; there is no sound reason to force him to do the work. But many people obtain copies from the library who would not copy by hand, and who might in fact buy a copy of the work if they were unable to receive an inexpensive machine reproduction of it. Thus, the library interprets the Gentlemen's Agreement in its favor and thereby "justifies" a substantial amount of copying. (Emphasis supplied.) Project, New Technology and the Law of Copyright: Reprography and Computers, 15 U.C.L.A.L.Rev. 931, 951 (1968).

As the majority points out, one study, made in 1962, concluded that photocopying did not result in economic damage to publishers at that time: Fry Associates, Survey of Copyrighted Material Reproduction Practices in Scientific and Technical Fields, 11 Bull.Cr.Soc. 69, 71 (1963). This study also stated:

One situation was reported during the survey in which economic damage may occur. A prominent university library in a small town with several corporate research and development centers gives excellent service on its collection. This library felt that these corporate libraries are subscribing to only the minimum number of journals. They rely on the university to supply photocopies of other material.

* * * * * *

This is the one clear-cut example disclosed during the survey of dilution of the publishers' circulation market. Id. at 119.

Indeed, this example is very nearly the situation presented by this case. Government institutions, medical schools, hospitals, research foundations, drug companies, and individual physicians are supplementing their collections, if they subscribe to any journals, by acquiring free photocopies of articles from the NLM.⁷

7 It should be noted that the Fry survey was made when photocopying was not as prominent as it is today. Even at that time, the Fry Report notes that larger publishers (who were photocopied most heavily) complained about the effects of photocopying. Fry Report at 86-87. Secondly, the Fry Report operates on the dubious assumption that in most cases the photocopy serves as a substitute for loaning the original material and does no more damage than would loaning of the original material.
In addition to the Fry Report, the majority cites a statement by Dan Lacy, Managing Director, American Book Publisher's Council, to the effect that photocopying is undertaken in lieu of manual note taking, typing, or handwriting a copy, and in lieu of a library loan rather than in lieu of buying a copy. We can hardly expect a representative of an organization of book publishers to be an expert on the problems of journal publishers. Library photocopying of books does not pose the same threat to a book publisher as photocopying of journal articles does to its publisher. Rarely are books photocopied completely. At present, there appears to be no competition for the consumer market between libraries and book publishers.

In addition to the conclusions of those who have studied the problem extensively, there are other facts and circumstances in this record which I think amply support the views of the trial judge that the system used by defendant's libraries for distributing free copies of plaintiff's journal articles attracts some potential purchasers of plaintiff's journals.

Subscription sales provide most of the revenue derived from the marketing of plaintiff's journals. It is important to remember that each of plaintiff's journals caters to and serves a limited market. Plaintiff's share of the profits from these journals has varied from less than $1,000 to about $7,000 annually.⁸ In the context of rising costs of publication, an inability to attract new customers, and the loss of even a small number of old subscribers may have a large detrimental effect on the journals. A representative of Williams Wilkins Company testified that in recent years there have been journals that have failed, and in the opinion of those at Williams Wilkins, photocopying has played a role in these failures.⁹ The majority relies on the fact that subscriptions for the four journals in this case have shown a general increase over the last five years, but two of the journals, Medicine and Pharmacological Reviews, have shown a slight decrease in subscriptions from 1968 to 1969. In addition, the Journal of Immunology showed losses in the period prior to 1961; Gastroenterology showed losses in 1967 and 1968; Pharmacological Reviews showed a loss in 1969. There is no evidence to show specifically whether any particular instance or instances of unauthorized photocopying of plaintiff's journals has or has not resulted in the loss of revenue to plaintiff. However, I think the record, as a whole, supports the determination of the trial judge that the photocopying in this case has had a tendency to diminish plaintiff's markets in the past.

8 For example, in 1968, profit from Pharmacological Reviews was $1,154.44 (on sales of about $40,000). The profit was divided, $1,039 to the American Society for Pharmacology and Experimental Therapeutics and $115.44 to plaintiff. In 1969, net income from Gastroenterology was $21,312 (on sales of about $245,000) and $11,532.35 of that amount was offset by losses the previous year, leaving a balance of $9,779.73. The balance was split between plaintiff and the American Gastroenterological Association, plaintiff getting $4,889.86.

9 Tr. at 73.

The NLM publishes a monthly indexed catalog of journal articles in medicine and related sciences entitled "Index Medicus." The index is widely distributed to medical libraries, research centers, schools, hospitals, and physicians. The catalog announces its new publications and acquisitions and thus advertises to the medical and scientific community, which constitutes plaintiff's market, that certain articles are available free of charge in the form of a photocopy.

At the present time, the NIH purchases only two subscriptions to plaintiff's journals. If nothing else, it would certainly need more than the two copies to meet the requests of the large in-house staff. Although it has been argued that the photocopies are merely a substitute for the loan of an original and does no more harm than the loan of the original material, I think this argument is fallacious. One copy of the original material could not possibly be loaned to as many requesters as the numerous photocopies, the competitive effect of which is much greater. Also, the photocopies are not required to be returned and become the property of the possessor. They can be marked, cut into segments, placed in the files, and otherwise put to uses that would be impossible with a loan of the original. While the library may look at the giving of a photocopy as a substitute for a loan, the user and would-be purchaser gets an exact copy of the original article which is a substitute for a purchased copy.

One of the new sources of income to publishers is the supplying of back issues or providing copies of such issues. When plaintiff receives a request for an out-of-print article, the customer is generally referred to the Institute of Scientific Information, which is licensed by plaintiff to make the photocopies. If the same articles can be obtained from the NLM without charge, it seems obvious that the supplying of free copies by the defendant's libraries will tend to diminish plaintiff's income from this source. NLM reproduces and supplies copies of journal articles to the patrons of other libraries. Therefore, the libraries who make the requests do not have to buy subscriptions for the use of their own patrons.

III [90] Foreign Laws Do Not Justify an Exemption From the Copyright Laws

The court relies to some extent on the copyright laws of the United Kingdom, New Zealand, Denmark, Finland, Italy, and other countries. The plaintiff says there are many differences between our copyright laws and those of other countries, and plaintiff does not agree that the defendant would be exempt from liability under the statutes of some of the countries named. However we need not delve into the details of the copyright legislation of these foreign countries. There is a shorter answer to the court's reliance on foreign laws. Unlike the legislative bodies of these countries, the Congress has not yet changed the Copyright Act of 1909 to permit the same kind of copying by the NIH and the NLM. If the time has come when the defendant's libraries should be exempted from the provisions of the Copyright Act to the extent permitted by the court's decision, the exemption should be provided by legislative action rather than by judicial legislation.

IV [92] A Judgment for Plaintiff Will Not Injure Medicine and Medical Research

The court has bottomed its decision to a very large extent on its finding, which is not disputed, that medical science would be seriously hurt if the photocopying by defendant's libraries is entirely stopped. But the court goes further and concludes that a judgment for plaintiff would lead to this result. It is not altogether clear to me how the court arrives at the second conclusion, and I think it is based on unwarranted assumptions.

The plaintiff does not propose to stop such photocopying and does not desire that result. What plaintiff seeks is a reasonable royalty for such photocopying and, in this case, a recovery of reasonable compensation for the infringement of its copyrights. Plaintiff has established a licensing system to cover various methods of reproducing its journal articles, including reproduction by photocopying. One of the licensees is a Government agency, and on several occasions plaintiff has granted requests from Government agencies and others for licenses to make multiple copies (Finding 36). In May 1967, the photocopying of plaintiff's journal articles was monitored by NLM for a 90-day period which was judged to be a representative sample. As the trial judge has shown, NLM found it would have paid plaintiff from $250 to $300 if it had granted plaintiff's request for royalty payments. The Director of NLM testified that this was, in his opinion, a surprisingly small sum. He also testified (Part III, trial judge's opinion) that the payment of a royalty to plaintiff for photocopying "has nothing to do with the operation of the library in the fulfillment of * * * [its] function. It is an economic and budgetary consideration and not a service-oriented kind of thing." This is the only direct testimony that I have found on how the payment of royalties for photocopying will affect the functions of the library, and it gives no indication or intimation that the payment of royalties to plaintiff will force NLM to cease the photocopying.

The court has laid heavy emphasis on the public interest in maintaining a free flow of information to doctors and scientists, and on the injury that might result if this flow should be stopped. However, there is another facet to the public interest question which is presented in this case. The trial judge put it well in his statement:

The issues raised by this case are but part of a larger problem which continues to plague our institutions with ever-increasing complexity — how best to reconcile, on the one hand, the rights of authors and publishers under the copyright laws with, on the other hand, the technological improvements in copying techniques and the legitimate public need for rapid dissemination of scientific and technical literature. (Part III, trial judge's opinion).

In enacting the 1909 Act, the House Committee said:

The enactment of copyright legislation by Congress under the terms of the Constitution is not based upon any natural right that the author has in his writings * * * but upon the ground that the welfare of the public will be served and progress of science and useful arts will be promoted by securing to authors for limited periods the exclusive rights to their writings. H.R. Rep. No. 2222, 60th Cong., 2d Sess. 7 (1909).

In Mazer v. Stein, 347 U.S. 201, 219, 74 S.Ct. 460, 471, 98 L.Ed. 630 (1954), the Supreme Court emphasized that the copyright protection given to authors and publishers is designed to advance public welfare, stating:

"The copyright law, like the patent statutes, makes reward to the owner a secondary consideration. * * *" However, it is "intended definitely to grant valuable, enforceable rights to authors, publishers, etc. * * *"

The economic philosophy behind the clause empowering Congress to grant patents and copyrights is the conviction that encouragement of individual effort by personal gain is the best way to advance public welfare through the talents of authors and inventors in "Science and useful Arts."

In order to promote the progress of science, not only must authors be induced to write new works, but also publishers must be induced to disseminate those works to the public. This philosophy has guided our country, with limited exceptions, since its beginning, and I am of the opinion that if there is to be a fundamental policy change in this system, such as a blanket exception for library photocopying, it is for the Congress to determine, not for the courts. The courts simply cannot draw the distinctions so obviously necessary in this area.

The court recognizes that the solution which it has undertaken to provide in this case is preeminently a problem for Congress which should decide how much photocopying should be allowed, what payments should be made to the copyright owners, and related questions. Nowhere else in its opinion is the court on more solid ground than when it declares that the "choices involve economic, social, and policy factors which are far better sifted by a legislature. The possible intermediate solutions are also of the pragmatic kind legislatures, not courts, can and should fashion." In spite of this obviously correct statement, the court has bridged the gap which the inaction of Congress has left in the Copyright Act of 1909.

I agree with the court that we have no jurisdiction to order a copyright owner to institute a licensing system if he does not wish to do so, but I think we are equally powerless to assume the congressional role by granting what amounts to a blanket exemption to defendant's libraries. Without too much difficulty, however, we can determine the amount of just compensation that is due plaintiff for the infringement of its copyrights. If that should be done, it may very well lead to a satisfactory agreement between the parties for a continuation of the photocopying by defendant upon the payment of a reasonable royalty to plaintiff.

The following portions of the trial judge's opinion are made a part of this dissent:^_

_ Certain sections of the trial judge's opinion have been omitted and the footnotes corresponding to that part of the text were also deleted.

I

Plaintiff, though a relatively small company, is a major publisher of medical journals and books. Plaintiff publishes 37 journals, dealing with various medical specialties. The four journals in suit are Medicine, Journal of Immunology, Gastroenterology, and Pharmacological Reviews. Medicine is published by plaintiff for profit and for its own benefit. The other three journals are published in conjunction with specialty medical societies which, by contract, share the journals' profits with plaintiff. The articles published in the journals stem from manuscripts submitted to plaintiff (or one of the medical societies) by physicians or other scientists engaged in medical research. The journals are widely disseminated throughout the United States (and the world) in libraries, schools, physicians' offices, and the like. Annual subscription prices range from about $12 to $44; and, due to the esoteric nature of the journals' subject matter, the number of annual subscriptions is relatively small, ranging from about 3,100 ( Pharmacological Reviews) to about 7,000 ( Gastroenterology). Most of the revenue derived from the journals comes from subscription sales, though a small part comes from advertising.^4a The journals are published with notice of copyright in plaintiff's name. The notice appears at the front of the journal and sometimes at the beginning of each article. After publication of each journal issue (usually monthly or bimonthly) and after compliance with the requisite statutory requirements, the Register of Copyrights issues to plaintiff certificates of copyright registration.

4a E. g., the November 1956 issue of Medicine has 86 pages, four of which carry commercial product advertising. The August 1965 issue of Journal of Immunology has 206 pages, nine of which carry commercial product advertising.

NIH, the Government's principal medical research organization, is a conglomerate of institutes located on a multi-acre campus at Bethesda, Maryland. Each institute is concerned with a particular medical specialty, and the institutes conduct their activities by way of both intramural research and grants-in-aid to private individuals and organizations. NIH employs over 12,000 persons — 4,000 are science professionals and 2,000 have doctoral degrees. To assist its intramural programs, NIH maintains a technical library. The library houses about 150,000 volumes, of which about 30,000 are books and the balance scientific (principally medical) journals. The library is open to the public, but is used mostly by NIH in-house research personnel. The library's budget for 1970 was $1.1 million.

The NIH library subscribes to about 3,000 different journal titles, four of which are the journals in suit. The library subscribes to two copies of each of the journals in suit. As a general rule, one copy stays in the library reading room and the other copy circulates among interested NIH personnel. Demand by NIH research workers for access to plaintiff's journals (as well as other journals to which the library subscribes) is usually not met by in-house subscription copies. Consequently, as an integral part of its operation, the library runs a photocopy service for the benefit of its research staff. On request, a researcher can obtain a photocopy of an article from any of the journals in the library's collection. Usually, researchers request photocopies of articles to assist them in their on-going projects; sometimes photocopies are requested simply for background reading. In any event, the library does not monitor the reason for requests or the use to which the photocopies are put. The photocopies are not returned to the library; and the record shows that, in most instances, researchers keep them in their private files for future reference.

Four regularly assigned employees operate the NIH photocopy equipment. The equipment consists of microfilm cameras and Xerox copying machines. In 1970, the library photocopy budget was $86,000 and the library filled 85,744 requests for photocopies of journal articles (including plaintiff's journals), constituting about 930,000 pages. On the average, a journal article is 10 pages long, so that in 1970, the library made about 93,000 photocopies of articles.

NLM is located on the Bethesda campus of NIH. NLM was formerly the Armed Forces Medical Library. In 1956, Congress transferred the library from the Department of Defense to the Public Health Service (renaming it the National Library of Medicine), and declared its purpose to be "* * * to aid the dissemination and exchange of scientific and other information important to the progress of medicine and to the public health * * *." 42 U.S.C. § 275 (1970). NLM is a repository of much of the world's medical literature. NLM is in essence a "librarians' library." As part of its operation, NLM cooperates with other libraries and like research-and-education-oriented institutions (both public and private) in a so-called "interlibrary loan" program. Upon request, NLM will loan to such institutions, for a limited time, books and other materials in its collection. In the case of journals, the "loans" usually take the form of photocopies of journal articles which are supplied by NLM free of charge and on a no-return basis. The term "loan" therefore is a euphemism when journal articles are involved. NLM's loan policies are fashioned after the General Interlibrary Loan Code, which is a statement of self-imposed regulations to be followed by all libraries which cooperate in interlibrary loaning. The Code provides that each library, upon request for a loan of materials, shall decide whether to loan the original or provide a photoduplicate. The Code notes that photoduplication of copyrighted materials may raise copyright infringement problems, particularly with regard to "photographing whole issues of periodicals or books with current copyrights, or in making multiple copies of a publication." [Emphasis in original text.] NLM, therefore, will provide only one photocopy of a particular article, per request, and will not photocopy on any given request an entire journal issue. NLM, as well as other libraries, justifies this practice on the basis of a so-called "gentlemen's agreement," written in 1935 by the National Association of Book Publishers and the Joint Committee on Materials for Research (representing the libraries), which states in part, "A library * * * owning books or periodical volumes in which copyright still subsists may make and deliver a single photographic reproduction * * * of a part thereof to a scholar representing in writing that he desires such reproduction in lieu of loan of such publication or in place of manual transcription and solely for the purposes of research * * *." [Emphasis supplied.] Each photocopy reproduced by NLM contains a statement in the margin, "This is a single photostatic copy made by the National Library of Medicine for purposes of study or research in lieu of lending the original."

In 1968, a representative year, NLM received about 127,000 requests for interlibrary loans. Requests were received, for the most part, from other libraries or Government agencies. However, about 12 percent of the requests came from private or commercial organizations, particularly drug companies. Some requests were for books, in which event the book itself was loaned. Most requests were for journals or journal articles; and about 120,000 of the requests were filled by photocopying single articles from journals, including plaintiff's journals. Usually, the library seeking an interlibrary loan from NLM did so at the request of one of its patrons. If the "loan" was made by photocopy, the photocopy was given to the patron who was free to dispose of it as he wished. NLM made no effort to find out the ultimate use to which the photocopies were put; and there is no evidence that borrowing libraries kept the "loan" photocopies in their permanent collections for use by other patrons.

Defendant concedes that within the pertinent accounting period, NLM and the NIH library made at least one photocopy of each of eight articles (designated by plaintiff as the Count I-to-Count VIII articles) from one or more of the four journals in suit. Defendant also concedes that plaintiff is the record owner of copyright registrations on the journals. That would appear to end the matter in plaintiff's favor, for § 1 of the copyright statute (17 U.S.C.) says that the copyright owner "* * * shall have the exclusive right: (a) to print, reprint, publish, copy and vend the copyrighted work * * *"; and § 3 of the statute says that, "* * * [t]he copyright upon composite works or periodicals shall give to the proprietor thereof all the rights in respect thereto which he would have if each part were individually copyrighted under this title." Simply stated, this means that each article in plaintiff's journals is protected from infringement to the same extent as the entire journal issue. Advertisers Exch., Inc. v. Laufe, 29 F. Supp. 1 (W.D.Pa. 1939); King Features Syndicate v. Fleischer, 299 F. 533 (2d Cir. 1924).^5a

5a One argument made by defendant to justify the copying of single articles from plaintiff's journals is that each article is but "part" of a journal issue, which in turn is but "part" of a journal volume; and, accordingly, defendant says, its libraries have not copied an "entire" copyrighted work. Section 3 of 17 U.S.C. fully meets that argument, for it is undisputed that plaintiff could publish and seek copyright registration on each article separately. As stated in H.R. Rep. No. 2222, 60th Cong., 2d Sess. 10 (1909):

Section 3 [of the Copyright Act] does away with the necessity of taking a copyright on the contributions of different persons included in a single publication * * *.

The noninfringement defense

Defendant contends that its acts of copying do not violate the copyright owner's exclusive right "to copy" the copyrighted work as provided by 17 U.S.C. § 1. The argument is that with respect to books and periodicals, the act of making single copies ( i. e., one copy at a time) is not, in itself, sufficient to incur liability; that the "copying," to be actionable, must include "printing" (or "reprinting") and "publishing" of multiple copies of the copyrighted work. The argument is bottomed on analysis of the copyright laws as they have evolved from 1790 to the present.^8a The early laws distinguished "copying" from "printing," "reprinting," and "publishing," and provided that the copyright in books is infringed by "printing," "reprinting" and "publishing" while the copyright in other works ( e. g., photographs, paintings, drawings, etc.) is infringed by "copying." The 1909 Copyright Act obliterated any such distinction. It provides in § 5 a list of all classes of copyrightable subject matter (including books and periodicals), and says in § 1 that the owner of copyright shall have the exclusive right "to print, reprint, publish, copy and vend the copyrighted work" [emphasis supplied]. Thus, the 1909 Act, unlike the earlier statutes, does not expressly say which of the proscribed acts of § 1 apply to which classes of copyrightable subject matter of § 5. Defendant says that to be consistent with the intent and purpose of earlier statutes, the "copying" proscription of § 1 should not apply to books or periodicals; rather, only the proscribed acts of "printing," "reprinting" and "publishing" should apply to books and periodicals.

8a Congress enacted the first copyright statute in 1790 (Act of May 31, 1790, ch. 15, 1 Stat. 124). Thereafter, the statute was revised from time to time, notably in 1802, 1831, 1870, and 1891. In 1909, the present statute was passed (Act of March 4, 1909, ch. 320, 35 Stat. 1075) and later was codified as 17 U.S.C. (Act of July 30, 1947, 61 Stat. 652).

Defendant's argument is not persuasive and, in any event, is irrelevant. It is clear from a study of all the copyright statutes from 1790 to date that what Congress has sought to do in every statute is to proscribe unauthorized duplication of copyrighted works. The words used in the various statutes to define infringing acts ( i. e., printing, reprinting, copying, etc.) were simply attempts to define the then-current means by which duplication could be effected. It is reasonable to infer that in 1909, when Congress included "copying" in the list of proscribed acts applicable to books and periodicals (as well as copyrightable subject matter in general), it did so in light of the fact that new technologies ( e. g., photography) made it possible to duplicate books and periodicals by means other than "printing" and "reprinting." The legislative history of the 1909 Act says little, one way or the other, about the matter.^9a Nevertheless, §§ 1 and 5 are plain and unambiguous on their face; and the Supreme Court held as recently as 1968, in Fortnightly Corp., supra 392 U.S. at 394, 88 S.Ct. at 2086:

9a H.R. Rep. No. 2222, 60th Cong., 2d Sess. 4 (1909) states:
Subsection (a) of section 1 adopts without change the phraseology of section 4952 of the Revised Statutes, and this, with the insertion of the word "copy," practically adopts the phraseology of the first copyright act Congress ever passed — that of 1790. Many amendments of this were suggested, but the committee felt that it was safer to retain without change the old phraseology which has been so often construed by the courts.

* * * § 1 of the [Copyright] Act enumerates several "rights" that are made "exclusive" to the holder of the copyright. If a person, without authorization from the copyright holder, puts a copyrighted work to a use within the scope of one of these "exclusive rights," he infringes the copyright. [Emphasis supplied.]

See also the 1961 Register's Report, wherein it is noted at 21-22:

* * * as several courts have observed, the right embraced in the repetitive terms of section 1(a) is the twofold right to make and publish copies.

This right is the historic basis of copyright and pertains to all categories of copyrighted works. * * * [Emphasis supplied.]

The burden, therefore, is on defendant to show that Congress intended the statute to mean something other than what it plainly says. Defendant has not carried that burden.

It is also pertinent that the courts have liberally construed the 1909 Act to take into account new technologies by which copyrighted works can be duplicated, and thus infringed. In Fortnightly Corp., supra at 395-396, 88 S.Ct. at 2087, the Court, in dealing with copyright infringement relating to television, said:

In 1909 radio itself was in its infancy, and television had not been invented. We must read the statutory language of 60 years ago in the light of drastic technological change. [Emphasis supplied.]

To the same effect is Jerome H. Remick Co. v. American Automobile Accessories Co., 5 F.2d 411 (6th Cir. 1925), cert. denied, 269 U.S. 556, 46 S.Ct. 19, 70 L.Ed. 409, which stated at 411:

* * * the statute may be applied to new situations not anticipated by Congress, if, fairly construed, such situations come within its intent and meaning. Thus it has been held both in this country and England that a photograph was a copy or infringement of a copyrighted engraving under statutes passed before the photographic process had been developed. [citations omitted] While statutes should not be stretched to apply to new situations not fairly within their scope, they should not be so narrowly construed as to permit their evasion because of changing habits due to new inventions and discoveries.

Furthermore, defendant's argument that it may "copy," short of "printing," "reprinting" and "publishing," is irrelevant under the facts of this case. NLM and the NIH library did not merely "copy" the articles in suit; they, in effect, "reprinted" and "published" them. "Printing" and "reprinting" connote making a duplicate original, whether by printing press or a more modern method of duplication. Macmillan Co. v. King, 223 F. 862 (D.Mass. 1914); M. Nimmer, Copyright § 102 (1971 ed.). "Publishing" means disseminating to others, which defendant's libraries clearly did when they distributed photocopies to requesters and users. Macmillan Co., supra; M. Nimmer, Copyright § 104 (1971 ed.).

Defendant's contention that its libraries make only "single copies" of journal articles, rather than multiple copies, is illusory and unrealistic. Admittedly, the libraries, as a general rule, make only one copy per request, usually for different users. But the record shows that the libraries duplicate particular articles over and over again, sometimes even for the same user within a short timespan. E. g., the NIH library photocopied the Count I article three times within a 3-month period, two of the times for the same requester; and it copied the Count IV and Count V articles twice within a 2-month period, albeit for different users. The record also shows that NLM will supply to requesters photocopies of the same article, one after the other, on consecutive days, even with knowledge of such facts. In short, the libraries operate comprehensive duplication systems which provide every year thousands of photocopies of articles, many of which are copies of the same article; and, in essence, the systems are a reprint service which supplants the need for journal subscriptions. The effects of this so-called "single copying" practice on plaintiff's legitimate interests as copyright owner are obvious. The Sophar and Heilprin report, at 16, puts it in terms of a colorful analogy: "Babies are still born one at a time, but the world is rapidly being overpopulated."

The "fair use" defense

Defendant contends that its copying comes under the doctrine of "fair use" of copyrighted works. "Fair use," a judicially-created doctrine, is a sort of "rule of reason" applied by the courts as a defense to copyright infringement when the accused infringing acts are deemed to be outside the legitimate scope of protection afforded copyright owners under 17 U.S.C. § 1. What constitutes "fair use" cannot been defined with precision. Much has been written about the doctrine, particularly its rationale and scope. See, e. g., A. Latman, Fair Use of Copyrighted Works, Study No. 14, Copyright Law Revision, Studies Prepared for Senate Comm. on the Judiciary, 86th Cong., 2d Sess. (1960); Comment, Copyright Fair Use — Case Law and Legislation, 1969 Duke L.J. 73; S. Cohen, Fair Use and the Law of Copyright, ASCAP Copyright Law Symposium (No. 6) 43 (1955); W. Jensen, Fair Use: As Viewed by the "User," 39 Dicta 25 (1962); L. Yankwich, What Is Fair Use?, 22 U.Chi.L.Rev. 203 (1954); Note, Fair Use: A Controversial Topic in the Latest Revision of Our Copyright Law, 34 U.Cin.L.Rev. 73 (1965); M. Nimmer, Copyright § 145 (1971 ed.); Sophar Heilprin Report at 15; R. Heedham, Tape Recording, Photocopying and Fair Use, ASCAP Copyright Law Symposium (No. 10) 75 (1959); Crossland, The Rise and Fall of Fair Use: The Protection of Literary Materials Against Copyright Infringement by New and Developing Media, 20 S.Car.L. Rev. 153 (1968). Some courts have held that the doctrine is but an application of the principle de minimis non curat lex and, as plaintiff puts it, "comes into play only when a relatively small amount of copying takes place." Principal factors considered by the courts in deciding whether a particular use of a copyrighted work is a "fair use" are (a) the purpose of the use, (b) the nature of the copyrighted work, (c) the amount and substantiality of the material used in relation to the copyrighted work as a whole, and (d) the effect of the use on a copyright owner's potential market for his work.¹⁰ While these criteria are interrelated and may vary in relative significance, the last one, i. e., the competitive character of the use, is often the most important. E. g., it has been held "fair use" to copy excerpts from literary works for purposes of criticism or review (Loew's, Inc. v. CBS, Inc., 131 F. Supp. 165 (S.D.Cal. 1955), aff'd sub nom. Benny v. Loew's, Inc., 239 F.2d 532 (9th Cir. 1956), aff'd by an equally divided Court, 356 U.S. 43, 78 S.Ct. 667, 2 L.Ed.2d 583 (1958)); or to copy portions of scholarly works (Greenbie v. Noble, supra; Holdredge v. Knight Publishing Corp., 214 F. Supp. 921 (S.D.Cal. 1963)). However, it is not "fair use" to copy substantial portions of a copyrighted work when the new work is a substitute for, and diminishes the potential market for, the original. Hill v. Whalen Martell, Inc., 220 F. 359 (S.D.N.Y. 1914); Folsom v. Marsh, 9 F.Cas. 342 (D.Mass. 1841). And it has been held that wholesale copying of a copyrighted work is never "fair use" (Leon v. Pacific Tel. Tel. Co., 91 F.2d 484 (9th Cir. 1937); Public Affairs Associates, Inc. v. Rickover, 109 U.S.App.D.C. 128, 284 F.2d 262 (1960), vacated and remanded, 369 U.S. 111, 82 S.Ct. 580, 7 L.Ed.2d 604 (1962)), even if done to further educational or artistic goals and without intent to make profit. Wihtol v. Crow, 309 F.2d 777 (8th Cir. 1962).

10 H.R. Rep. No. 83, 90th Cong., 1st Sess. (1967), which relates to revision of the copyright laws, notes that these factors are the ones used by the courts. At 29-37, there is a detailed discussion of "fair use" as applicable to photocopying for educational purposes.

Whatever may be the bounds of "fair use" as defined and applied by the courts, defendant is clearly outside those bounds. Defendant's photocopying is wholesale copying and meets none of the criteria for "fair use." The photocopies are exact duplicates of the original articles; are intended to be substitutes for, and serve the same purpose as, the original articles; and serve to diminish plaintiff's potential market for the original articles since the photocopies are made at the request of, and for the benefit of, the very persons who constitute plaintiff's market. Defendant says, nevertheless, that plaintiff has failed to show that it has been harmed by unauthorized photocopying; and that, in fact, plaintiff's journal subscriptions have increased steadily over the last decade. Plaintiff need not prove actual damages to make out its case for infringement. Macmillan Co., supra. Section 1498 of title 28 U.S.C. provides for payment of "reasonable and entire compensation * * * including the minimum statutory damages as set forth in section 101(b) of title 17, United States Code." See Brady v. Daly, 175 U.S. 148, 20 S.Ct. 162, 44 L.Ed. 109 (1899); F. W. Woolworth Co. v. Contemporary Arts, Inc., 344 U.S. 228, 73 S.Ct. 222, 97 L.Ed. 276 (1952). M. Nimmer, Copyright § 154 (1971 ed.). Moreover, damage may be inferred in this case from the fact that the photocopies are intended to supplant the original articles. While it may be difficult (if not impossible) to determine the number of subscription sales lost to photocopying, the fact remains that each photocopy user is a potential subscriber, or at least is a potential source of royalty income for licensed copying. Plaintiff has set up a licensing program to collect royalties for photocopying articles from its journals; and among the licensees have been libraries, including a Government library.¹¹ Also, there is evidence that one subscriber canceled a subscription to one of plaintiff's journals because the subscriber believed the cost of photocopying the journal had become less than the journal's annual subscription price; and another subscriber canceled a subscription, at least in part because library photocopies were available. Loss of subscription (or photocopying royalty) income is particularly acute in the medical journal field. The record shows that printing preparation costs are 50-65 percent of total cost of publication and that the number of subscriptions is relatively small. This simply means that any loss of subscription sales (or royalty income) has the effect of spreading publication costs over fewer copies, thus driving up steeply the unit cost per copy and, in turn, subscription prices. Higher subscription prices, coupled with cheap photocopying, means probable loss of subscribers, thus perpetuating a vicious cycle which can only bode ill for medical publishing.

11 There is no agreement, even among libraries and Government agencies, of what constitutes "fair use" in institutionalized photocopying. The Library of Congress will not photocopy copyrighted materials without permission of the copyright owners. Many other libraries follow the General Interlibrary Loan Code and engage in "single copy" photocopying. The U.S. Office of Education, through its Education Resources Information Center (ERIC) makes available current educational and research-related materials. ERIC will not copy copyrighted materials without permission of the copyright owner. See Sophar and Heilprin report at 39-46.

Defendant's amici fear that a decision for plaintiff will be precedent for plaintiff's seeking injunctions against non-Government libraries, pursuant to 17 U.S.C. § 101(a), thereby interfering with the free flow of technical and scientific information through library photocopying. On the basis of this record and representations made by plaintiff's personnel and counsel, that fear does not appear to be justified. Plaintiff does not seek to enjoin any photocopying of its journals. Rather, it merely seeks a reasonable royalty therefor.¹² Its licensing program would so indicate for, as far as the record shows, plaintiff will grant licenses to anyone at a reasonable royalty. No doubt, plaintiff would prefer that all of its journal users be subscribers. However, plaintiff recognizes that this is unrealistic. Some articles in its journals are in greater demand than others, and many journal users will not consider it economically justifiable to subscribe to a journal simply to get access to a few articles. Implicit in plaintiff's licensing program, therefore, is the idea that it is in the best interest of all concerned that photocopying proceed without injunction, but with payment of a reasonable fee. That would appear to be a logical and commonsense solution to the problem, not unlike the solution provided by the American Society of Composers, Authors and Publishers (ASCAP) and Broadcast Music, Inc. (BMI) in the field of music and the performing arts. For a description of how ASCAP and BMI operate in a context similar to this one, see Hearings on H.R. 4347 and other bills before Subcomm. No. 3, House Comm. on the Judiciary, 89th Cong., 1st Sess. 194, 203 (1965); Finkelstein, ASCAP as an Example of the Clearing House System in Operation, 14 Copyright Soc'y Bull. 2 (1966).

12 In his opening statement at trial, plaintiff's counsel said (emphasis supplied):

The case has nothing to do with the stopping of photocopying. The Commissioner knows that an injunction is not available in this court, nor is plaintiff, in any case, seeking to curtail this use of its articles.

Similarly, William M. Passano, plaintiff's Chairman of the Board, stated in a hearing before a Senate committee:

We feel that it is unrealistic and not in the public interest to consider restricting in any way the use of photocopying devices. They serve a useful purpose in the dissemination of knowledge. Since we, as publishers, are in that business, we certainly don't want to see the spread of knowledge curtailed.

To us the only solution to the problem is a simple system of royalty payments with a minimum of red tape. * * * [Hearings on Copyright Law Revision before the Patents, Trademarks and Copyrights Subcomm. of the Senate Comm. on the Judiciary, 90th Cong., 1st Sess. 976 (1967).

Defendant says that photocopying by NLM and the NIH library is "reasonable and customary" because it complies with a longstanding practice of libraries to supply photocopies of parts of scientific works to persons engaged in scholarly research, and is consistent with the terms of the "gentlemen's agreement," earlier noted. The "gentlemen's agreement," drafted in 1935, was the product of meetings and discussions between representatives of the book publishing industry and libraries. The representatives were interested in working out a practical accommodation of the conflict between (a) the legitimate interests of copyright owners not to have their works copied without compensation and (b) the needs of scholars and research workers for copies of parts of copyrighted works for private use in pursuit of literary or scientific investigation. The "agreement" was, in effect, a promise by the book publishers not to interfere with library photocopying under three conditions: (i) the library must warn the person for whom the photocopy is made that he is liable for any copyright infringement by misuse (presumably by making further photocopies), (ii) the photocopying must be done without profit to the library, and (iii) the amount copied must be so substantial as to constitute an infringement. The third condition is implicit in the "agreement" which says:

While the right of quotation without permission is not provided in law, the courts have recognized the right to a "fair use" of book quotations, the length of a "fair" quotation being dependent upon the type of work quoted from and the "fairness" to the author's interest. Extensive quotation is obviously inimical to the author's interest. * * * It would not be fair to the author or publisher to make possible the substitution of the photostats for the purchase of a copy of the book itself either for an individual library or for any permanent collection in a public or research library. Orders for photocopying which, by reason of their extensiveness or for any other reasons, violate this principle should not be accepted. [Emphasis supplied.]

The "gentlemen's agreement" does not have, nor has it ever had, the force of law with respect to what constitutes copyright infringement or "fair use." So far as this record shows, the "agreement" has never been involved in any judicial proceedings. Nevertheless, the "agreement" is entitled to consideration as a guide to what book publishers and libraries considered to be "reasonable and customary" photocopying practices in the year 1935. It has little significance, however, to this case. The agreement was drafted on behalf of a book publishers' organization which is now defunct and to which plaintiff never belonged. In fact, it appears that no periodical publishers were represented in the organization at the time the agreement was drafted; and, consequently, the "agreement" cannot speak for their interests or problems. See the Varmer study at 51, n. 9. Furthermore, the "agreement" was drafted at a time when photocopying was relatively expensive and cumbersome; was used relatively little as a means of duplication and dissemination; and posed no substantial threat to the potential market for copyrighted works. Beginning about 1960, photocopying changed character. The introduction to the marketplace of the office copying machine made photocopying rapid, cheap and readily available. The legitimate interests of copyright owners must, accordingly, be measured against the changed realities of technology. Professor Nimmer in his treatise Copyright capsules the point at 653:

Both classroom and library reproduction of copyrighted materials command a certain sympathy since they involve no commercial exploitation and more particularly in view of their socially useful objectives. What this overlooks is the tremendous reduction in the value of copyrighted works which must result from a consistent and pervasive application of this practice. One who creates a work for educational purposes may not suffer greatly by an occasional unauthorized reproduction. But if every school room or library may by purchasing a single copy supply a demand for numerous copies through photocopying, mimeographing or similar devices, the market for copyrighted educational materials would be almost completely obliterated. This could well discourage authors from creating works of a scientific or educational nature. If the `progress of science and useful arts' is promoted by granting copyright protection to authors, such progress may well be impeded if copyright protection is largely undercut in the name of fair use. [Emphasis supplied.]

In any event, the "gentlemen's agreement" by its own terms condemned as "not * * * fair" the making of photocopies which could serve in "substitution" for the original work, and further noted that "[o]rders for photocopying which, by reason of their extensiveness or for any other reasons" could serve as duplicates of the original copyrighted work "should not be accepted." Thus, the most that can be said for the "gentlemen's agreement" is that it supported (and probably still supports) the proposition that it is "reasonable and customary" (and thus "fair use") for a library to photocopy for a patron a part of a book, or even part of a periodical article, such as a chart, graph, table, or the like, so long as the portion copied is not practically a substitute for the entire original work. Other instances of library photocopying may also be "fair use." E.g., a library no doubt can replace damaged pages of copyrighted works in its collection with photocopies; can make a small number of photocopies for in-house administrative purposes, such as cutting up for cataloging or the like; or can supply attorneys or courts with single photocopies for use in litigation. In all those instances, and probably many more which might come to mind on reflection, the rights of the copyright owner are not materially harmed. The doctrine of "fair use" and the "gentlemen's agreement," however, cannot support wholesale copying of the kind here in suit.¹³

13 The potential pernicious effects of modern, institutionalized photocopying of copyrighted works (particularly journal articles) in the name of "fair use" is discussed at length in the Sophar and Heilprin report. The authors, at 24, characterize wholesale copying by libraries as "a non-violent form of civil disobedience."

Defendant also contends that traditionally, scholars have made handwritten copies of copyrighted works for use in research or other scholarly pursuits; that it is in the public interest that they do so because any harm to copyright owners is minimal compared to the public benefits derived therefrom; and that the photocopying here in suit is essentially a substitute for handcopying by the scholars themselves. That argument is not persuasive. In the first place, defendant concedes that its libraries photocopy substantially more material than scholars can or do copy by hand. Implicit in such concession is a recognition that laborious handcopying and rapid machine photocopying are totally different in their impact on the interests of copyright owners. Furthermore, there is no case law to support defendant's proposition that the making of a handcopy by scholars or researchers of an entire copyrighted work is permitted by the copyright laws. Certainly the statute does not expressly permit it; and no doubt the issue has never been litigated because, as a practical matter, such copying is de minimis and causes no real threat to the copyright owner's legitimate right to control duplication and discrimination of copyrighted works. The photocopying done by NLM and the NIH library, on the other hand, poses a real and substantial threat to copyright owners' legitimate interests. Professor Nimmer discusses the point succinctly, at 653-54 of his treatise, and his language can hardly be improved upon:

It may be argued that library reproduction is merely a more modern and efficient version of the time-honored practice of scholars in making handwritten copies of copyrighted works, for their own private use. In evaluating this argument several factors must be considered. In the first place, the drudgery of making handwritten copies probably means that such copies in most instances are not of the complete work, and the quantitative insignificance of the selected passages are such as generally not to amount to a substantial similarity. Secondly, there would appear to be a qualitative difference between each individual scholar performing the task of reproduction for himself, and a library or other institution performing the task on a wholesale basis for all scholars. If the latter is fair use, then must not the same be said for a non-profit publishing house that distributes to scholars unauthorized copies of scientific and educational works on a national or international basis? Finally, it is by no means clear that the underlying premise of the above argument is valid.

There is no reported case on the question of whether a single handwritten copy of all or substantially all of a protected work made for the copier's own private use is an infringement or fair use. If such a case were to arise the force of custom might impel a court to rule for the defendant on the ground of fair use. Such a result, however, could not be reconciled with the rationale for fair use suggested above since the handwritten copy would serve the same function as the protected work, and would tend to reduce the exploitation value of such work. Moreover, if such conduct is defensible then is it not equally a fair use for the copier to use his own photocopying or other duplicating device to achieve the same result? Once this is acknowledged to be fair use, the day may not be far off when no one need purchase books since by merely borrowing a copy from a library any individual will be able to make his own copy through photocopying or other reproduction devices which technological advances may soon make easily and economically available.

To the same effect is a statement in the Varmer study at 62-63:

It has long been a matter of common practice for individual scholars to make manual transcriptions of published material, though copyrighted, for their own private use, and this practice has not been challenged. Such transcription imposed its own quantitative limitations; and in the nature of the event, it would not be feasible for copyright owners to control private copying and use. But reproduction for private use takes on different dimensions when made by modern photocopying devices capable of reproducing quickly any volume of material in any number of copies, and when copies are so made to be supplied to other persons. Publisher's copies are brought for the private use of the buyer, and in some circumstances a person supplying copies to others will be competing with the publisher and diminishing his market.

Not only is such competition unfair to the publisher and copyright owner, but it may be injurious to scholarship and research. Thus, it has been pointed out that widespread photocopying of technical journals might so diminish the volume of subscriptions for the journals as to force the suspension of their publication.

Also, the 1961 Register's Report notes at 25-26:

Researches need to have available, for reference and study, the growing mass of published material in their particular fields. This is true especially, though not solely, of material published in scientific, technical, and scholarly journals. Researchers must rely on libraries for much of this material. When a published copy in a library's collections is not available for loan, which is very often the case, the researcher's need can be met by a photocopy.

On the other hand, the supplying of photocopies of any work to a substantial number of researchers may diminish the copyright owner's market for the work. Publishers of scientific, technical, and scholarly works have pointed out that their market is small; and they have expressed the fear that if many of their potential subscribers or purchasers were furnished with photocopies, they might be forced to discontinue publication.

Finally, defendant says that it is unconstitutional to construe the copyright law so as to proscribe library photocopying of scientific or technical writings because such photocopying is consonant with the constitutional purpose of copyright "to promote the progress of science." That argument misses the mark. Article I, section 8, clause 8, of the U.S. Constitution grants to Congress the "Power * * * to promote the Progress of Science * * * by securing for limited Times to Authors * * * the exclusive Right to their * * * Writings * * *." The word "Science" is used in the sense of general knowledge rather than the modern sense of physical or biological science. See Rich, Principles of Patentability, 28 Geo.Wash.L.Rev. 393, 394-97 (1960); H.R. Rep. No. 1923, 82d Cong., 2d Sess. 4 (1952); S.Rep. No. 1979, 82d Cong., 2d Sess. 3 (1952). Congress has exercised its constitutional power by enacting, and revising from time to time, copyright statutes which are the method of, and provide a system for, achieving the constitutional purpose. The system "promotes progress" by encouraging authors to write and publicly disclose their writings; by inducing publishers and entrepreneurs to invest risk capital in the dissemination of authors' writings; and by requiring other authors to create new writings, rather than plagiarize the old, all of which is in the public interest. Mazer v. Stein, 347 U.S. 201, 219, 74 S.Ct. 460, 98 L.Ed. 630 (1954), rehearing denied, 347 U.S. 949, 74 S.Ct. 637, 98 L.Ed. 1096. Congress has broad discretion under the Constitution to prescribe the conditions under which copyright will be granted, the only express restriction being that any "exclusive right" must be for a "limited time." Nothing in the present statute, its legislative history or the case law suggests that Congress intended to exempt libraries or others from liability for wholesale copying of copyrighted works, whatever be the purpose or motivation for the copying. What defendant really appears to be arguing is that the copyright law should excuse libraries from liability for the kind of photocopying here in suit. That, of course, is a matter for Congress, not the courts, to consider for it involves questions of public policy aptly suited to the legislative process. In an analogous context in Fortnightly Corp., supra, Justice Fortas noted at 408 of 392 U.S., at 2093 of 88 S.Ct.:

The task of caring for CATV is one for the Congress. Our ax, being a rule of law, must cut straight, sharp, and deep; and perhaps this is a situation that calls for the compromise of theory and for the architectural improvisation which only legislation can accomplish.

See also White-Smith Music Co., supra, where the Court noted at 18, that "considerations [of what the copyright laws should provide] properly address themselves to the legislative, and not to the judicial, branch of the government."¹⁴

14 There has been no dearth of activity to revise the 1909 Copyright Act. Some of that activity relates to library photocopying problems. See, e. g., Hearings on H.R. 4347 and other bills before Subcomm. No. 3, House Comm. on the Judiciary, 89th Cong., 1st Sess. 448, 459, 1133 (1965); S. 597, H. R. 2512, 90th Cong., 1st Sess. (1967); S. 543, 91st Cong., 1st Sess. (1969); S.Rep. No. 91-1219, 1219, 91st Cong., 2d Sess. 5 (1970); S. 644, 92d Cong., 1st Sess. (1971). For a brief history of legislative activity directed toward revision of the 1909 Copyright Act, see Fortnightly Corp., supra at 396 n. 17, 88 S.Ct. 2084; UCLA Project at 931-38.

III

Several other points raised by the parties merit comment. Defendant notes that the National Library of Medicine Act by which NLM was created ( 42 U.S.C. § 275 et seq.) provides at § 276(a)(4) that the Secretary of Health, Education, and Welfare, through NLM, shall "make available, through loans, photographic or other copying procedures or otherwise, such materials in the Library as he deems appropriate * * *"; and that the Medical Library Assistance Act of 1965 ( 42 U.S.C. § 280b-1 et seq.) provides that grants be made to medical libraries for, among other things, "acquisition of duplicating devices, facsimile equipment * * * and other equipment to facilitate the use of the resources of the library." 42 U.S.C. § 280b-7. Defendant suggests that by those statutory provisions Congress intended to exempt NLM and other grantee libraries from the copyright laws. As defendant puts it, "* * * the only reasonable interpretation [of the statutes] is that Congress knew that fair use would exempt such libraries from copyright infringement in the established use by libraries of such [photocopy] equipment." There is no merit to this. Nothing in the statutes or their legislative histories says anything about the copyright laws, and it cannot be inferred that Congress intended the statutes to be in derogation of the copyright laws, absent an express indication to the contrary.¹⁶ See generally E. Crawford, Statutory Construction § 227 (1940). No court has ever held that "fair use" applies to library wholesale photocopying; nor has there been a uniform and unchallenged policy among libraries and other institutionalized photocopiers on the bounds of "fair use." See note 11. Thus, it makes no sense to impute to Congress an intent for which there is no sound basis in judicial decision, or otherwise. The fact that the statutes authorize the libraries to make use, generally of photocopying equipment and procedures, is not controlling or even very significant. Much material in library collections is either not copyrighted or is material on which the copyright has expired; and in either event, the material is in the public domain and can be freely copied.

16 H.R. Rep. No. 941, 84th Cong., 2d Sess. (1956); S.Rep. No. 2071, 84th Cong., 2d Sess. (1956); H.R. Rep. No. 1026, 89th Cong., 1st Sess. (1965); S.Rep. No. 756, 89th Cong., 1st Sess. (1965).

Furthermore, the record shows that NLM, from the beginning, has been concerned about complying with the copyright laws and has never considered itself exempt therefrom. In 1957, NLM's Board of Regents discussed the library's photocopying practices and deemed them to create vexing copyright infringement problems. The Director of NLM was of the opinion that "sooner or later" the problems would bring "a test of the issue in the courts."

Defendant suggested at trial that payment of compensation to plaintiff for photocopying its journals would create a continuing undue and oppressive administrative and financial burden on NLM and the NIH library. Defendant has not pressed the point in its brief, perhaps because it is clear that plaintiff's right to compensation under 28 U.S.C. § 1498 (b) cannot depend on the burdens of compliance. Nevertheless, defendant's point merits comment since courts should be mindful of the practical consequences of their decisions. Based on this record, defendant's fears are not justified. Both NLM and the NIH library already have administrative procedures by which they keep detailed records of photocopying. Both libraries require that written request slips be submitted by requesters of photocopies. The slips are a permanent record of the journals and pages photocopied. It would seem a routine, albeit tedious, matter to cull from those records the information necessary to calculate a reasonable royalty on the basis of the number of articles copied, or perhaps to come up with an acceptable formula for establishing a blanket annual royalty payment. Indeed, the evidence suggests that this is so. In 1967, NLM temporarily stopped photocopying articles from plaintiff's journals, as a result of plaintiff's charge of copyright infringement and requests for a reasonable royalty. NLM was able as a practical matter, to flag all requests for photocopies from plaintiff's journals from April 27, 1967 to May 29, 1967, in order to refrain from copying them. On about May 29, 1967, photocopying was resumed and was monitored for about 90 days. Satisfied that the 90-day period was a representative sample, NLM found that it would have paid plaintiff about $250-$300 if it had acceded to plaintiff's request for royalty payment. The Director of NLM testified that, in his opinion, this was "a very small sum — surprisingly small sum." Similarly, the NIH librarian testified that payment of royalties for photocopying "has nothing to do with the operation of the library in the fulfillment of * * * [its] function. It is an economic and budgetary consideration and not a service-oriented kind of thing."

Nor does it appear that payment of royalties to other publishers will create an undue or oppressive administrative burden. The Sophar and Heilprin report notes, at 58-60, that based on a study of the photocopying practices of U.S. libraries, less than 1,000 publishers provide the material photocopied by libraries, and that about 5 percent of that number provide about 40 percent of the material copied. This simply means that nearly half of the materials photocopied emanate from about 50 publishers. No doubt, the materials photocopied by NLM and the NIH library come from an even smaller number of publishers since those libraries are highly specialized. In any event, by using modern management practices including computers and the like, it would appear that NLM and the NIH library can, with minimum disruption, cope with the necessary recordkeeping.¹⁷

17 It has been suggested that there be established a clearinghouse for access, permissions and payments for photocopying of copyrighted materials. The clearinghouse would relieve institutional copiers of the burdens of royalty distribution and might also be instrumental in setting up blanket royalty arrangements, thus relieving the institutions from most recordkeeping requirements. See, e. g., the Sophar and Heilprin report at 82. The clearinghouse concept has also been alluded to in a congressional report:

* * * Despite past efforts, reasonable arrangements involving a mutual understanding of what generally constitutes acceptable library practices, and providing workable clearance and licensing conditions, have not been achieved and are overdue. The committee urges all concerned to resume their efforts to reach an accommodation under which the needs of scholarship and the rights of authors would both be respected. [Emphasis supplied.] [H.R. Rep. No. 83, 90th Cong., 1st Sess. 36 (1967).]

And it is interesting that Sophar and Heilprin found that librarians favored, two to one, the clearinghouse approach to the problem, even though many of those in favor "indicated a desire to settle an increasingly complex matter, rather than an enthusiastic approval of the idea." Sophar and Heilprin report, at p. v of the Summary.

Postscript: The issues raised by this case are but part of a larger problem which continues to plague our institutions with ever-increasing complexity — how best to reconcile, on the one hand, the rights of authors and publishers under the copyright laws with, on the other hand, the technological improvements in copying techniques and the legitimate public need for rapid dissemination of scientific and technical literature. The conflict is real; the solution not simple. Legislative guidelines seem appropriate.¹⁸ The Sophar and Heilprin report, at pp. VIII-IX of the Summary, capsules the problem in a statement worth quoting:

18 In 1969, several bills were introduced in both the Senate and House to establish a National Commission on Libraries and Information Science. Also in 1969, H.R. 8809 was introduced to provide for a "National Science Research Data Processing and Information Retrieval System." See 1969 Register of Copyrights Annual Rep. 6. Earlier, in 1967, the Senate enacted S. 2216, 90th Cong., 1st Sess., by which there would be created a commission to study and compile data on the reproduction and use of copyrighted works. The House took no action on the bill.

From the viewpoint of the information scientist, copyright may appear as an impediment to the most efficient flow of information. It is apparently a blockage in an information system. Our early tendency was to oppose and try to limit the protection and control granted in copyright for the sake of efficiency. After careful analysis we no longer do.

There is a philosophical reason for not wanting to see copyright destroyed and there are a number of practical reasons. The philosophical reason is simply a belief that copyright is one of a number of ways in which our society expresses its belief and hope that an individual can continue his identity in a world of mass efforts by assuring the individual, his publisher or his association sufficient income from his ideas to maintain a degree of independence. The erosion of the economic value of copyright must lead to federal support of all kinds of writing and, of course, control.

The practical reasons flow from the philosophical reasons. Publishers, non-profit as well as commercial, will simply not be able to continue publishing under an eroded system. The scientific and other professional societies which, through their memberships, have done the most to develop information-handling tools and media are the ones most hurt by them. A means must be developed to assure payment to the copyright owner in return for unlimited and uncontrolled access to and duplication of the copyrighted work.

Our only concern and "vested interest" in copyright since we became interested in the problem "is to find a way to protect the `exclusive Right' of an author to his `Writings,' while permitting the advantages of modern information dissemination systems to become as useful as they may without weakening or threatening the economic urge and the need to create." We believe the two must become reconciled, not in the interests of compromise, but simply because both concepts are too valuable for either one to be permitted to severely harm or destroy the other.

KUNZIG, Judge, joins in the foregoing dissenting opinion.

[136] NICHOLS, Judge (dissenting):

I join in the Chief Judge's able dissent, and add a few words of my own. I agree with him that the photocopying of copyrighted material, as described in the findings, is not within the judge-made doctrine of fair use, and it should not be. The majority has posed a question, whose answer it triumphantly demonstrates, but it is the wrong question. The issue is not whether we should "stop" defendant's photocopying. Such a stoppage, at the behest of a publisher who refused to license on fair terms could well be unconscionable, in my opinion, but we have no such publisher before us. Plaintiff here is willing to license. If he did want to halt the photocopying, he would be in the wrong court here.

As the majority admits, we lack the power to enjoin. United States v. King, 395 U.S. 1, 89 S.Ct. 1501, 23 L.Ed.2d 52 (1969). Under 28 U.S.C. § 1498, as amended, all we can do, if we find infringement, is to award reasonable and entire compensation. The idea we are asked to "stop" the photocopying I suppose can be elaborated as follows: our decision would be stare decisis in other suits against non-government libraries in which injunctive relief is expressly authorized. 17 U.S.C. § 101(a), 112. However, the latter section authorizes injunctions on terms. There is high authority under earlier legislation that courts can refuse to enjoin copyright infringements if they deem an injunction would be unconscionable. Dun v. Lumbermen's Credit Ass'n., 209 U.S. 20, 28 S.Ct. 335, 52 L.Ed. 663 (1908). Under the 1909 Act, the Second Circuit held in National Comics Publications, Inc. v. Fawcett Publications, Inc., 198 F.2d 927 (1952), refusing to direct an injunction on remand:

* * * We think it best to leave open to the district court the question whether an injunction shall issue, since that is always a discretionary matter.

The Ninth Circuit also recognizes that same principle. Shapiro, Bernstein Co. v. 4636 S. Vermont Ave., Inc., 367 F.2d 236 (1966).

The Senate Study on Copyright Law Revision, Committee On The Judiciary Pursuant to S.Res. 240, Studies 22-25, 86th Cong.2d Sess., says at p. 127:

* * * * * *

The present law leaves it to the discretion of the court whether an injunction will be granted or denied. It has always been the rule of the courts that an injunction is an extraordinary remedy to be used only where further injury to the plaintiff is likely and the equities of the situation are on the side of injunctive relief, and the courts have denied an injunction in cases where it was thought that this remedy would be unduly harsh on the defendant.

* * * * * *

In Hecht Co. v. Bowles, 321 U.S. 321, 64 S.Ct. 587, 88 L.Ed. 754 (1944), the Supreme Court construed the wartime Emergency Price Control Act as not mandating injunctive relief, although the language of the statute was more favorable to such a construction, than that of the Copyright Act. Mr. Justice Douglas said for the Court at p. 329, 64 S.Ct. at p. 592:

* * * We are dealing here with the requirements of equity practice with a background of several hundred years of history. Only the other day we stated that "An appeal to the equity jurisdiction conferred on federal district courts is an appeal to the sound discretion which guides the determinations of courts of equity." Meredith v. Winter Haven, 320 U.S. 228, 235, 64 S.Ct. 7, 11, 88 L.Ed. 9. The historic injunctive process was designed to deter, not to punish. The essence of equity jurisdiction has been the power of the Chancellor to do equity and to mould each decree to the necessities of the particular case. Flexibility rather than rigidity has distinguished it. The qualities of mercy and practicality have made equity the instrument for nice adjustment and reconciliation between the public interest and private needs as well as between competing private claims. * * *

In view of the many persuasive reasons our majority adduces why photocopying by non-profit libraries should not be "stopped", I deem it an over-large assumption that an award by us of reasonable and entire compensation to our plaintiff would by stare decisis bind other tribunals, at the behest of other plaintiffs, to enjoin such library photocopying.

The amended § 1498 prescribes an award of "reasonable and entire compensation" which shall include the statutory minima under 17 U.S.C. § 101(b). Whether any statutory minimum is there prescribed in the case of a library photocopy of a periodical article, is a far tangent from our present inquiry, but I am satisfied, if it is, it need not be so prohibitive or punitive as to "stop" the photocopying.

Moreover, as to the question of fair use, I have difficulty regarding a use as fair, when a user benefits as extensively from the copyrighted material as this one does, yet adamantly refuses to make any contribution to defray the publisher's cost, or compensate for the author's effort and expertise, except the nominal subscription price of two copies of each periodical. Defendant's libraries, and others, have attempted to exercise a measure of self-restraint hitherto, but there is nothing in the majority decision to induce them to continue, that is not more than counterbalanced by other material that will encourage unrestricted piracy. However, hedged, the decision will be read, that a copyright holder has no rights a library is bound to respect. We are making the Dred Scott decision of copyright law.

I think the court also errs in imputing to Herbert Putnam, Librarian of Congress, an interpretation of the 1909 statute consistent with the court's. The brief for a group of amici put before us a 1908 regulation of that library which, the brief says, included this provision:

Photographing. Photographing is freely permitted. The permission extends to the building itself and any of its parts, including the mural decorations. It extends to articles bearing claim of copyright, but the Library gives no assurance that the photograph may be reproduced or republished or placed on sale. These are matters to be settled with the owner of the copyright. (Emphasis supplied.)

Assuming this is properly a matter for judicial notice, the omission to include it in the trial judge's findings (or to request inclusion) may perhaps be overlooked. Yet, as part of self-serving statements of historical fact in a brief, it avoids explanations such as an adverse party at the trial level might have furnished. The regulation possibly alluded to articles ( i.e., in common speech, short non-fiction writings) if it alluded to copyrighted printed matter. Why was not, e.g., copyrighted fiction of equal concern? The explanation that suggests itself to me as possible is that the articles mentioned are, or at least include, three-dimensional objects, or artifacts. The provision is too ambiguous in its coverage to afford any indication of how Putnam interpreted the copyright law.

The 1913 regulation, in a new section entitled Photostat, deals for the first time with reproduction of two-dimensional material. The court quotes it. It includes no caution as to copyright. The former provision as to photographing is continued unchanged. It seems a fair inference that no copyright caution was considered necessary as to two-dimensional material because the then method of photograph duplication of such material, known as photostating, was too costly, cumbersome and slow, to appear as a menace to holders of copyrights. Those of us whose memories go back to law practice in the thirties can take judicial notice that use of the method to generate copyright infringements on a major scale would have been unthinkable. A single copy of an infringing book or magazine article, produced by photostating, would have cost two dollars or so per page. If this recollection is carrying judicial notice too far, let us drop the Putnam argument altogether.

Finally, I must note the repeated alterations made in the trial judge's fact findings. Specifically:

New sentence in finding 6, that the requesters needed the articles in connection with their professional work, and used their copies solely for such purposes.

Deletions from finding 10 of part (b) explaining how copyright passes from authors to publishers.

Insertion in finding 17(b) that libraries' self-restraint policies are not abused or circumvented.

Insertion in finding 22(c), same effect.

Insertion in finding 39(b) that plaintiff's business is growing faster than the gross national product. This is irrelevant if true. Why not a comparison with the growth of the national debt? Or the total gate receipts of the National Football League?

Deletion from finding 39(d) of inference that plaintiff must have suffered some loss from photocopying and substitute statement he has failed to show substantial hurt. Trial Judge Davis also found, with record support, that at least one subscriber cancelled a subscription to one of plaintiff's journals because the cost of photocopying the journal had become less than the journal's annual subscription price. There was evidence that in another instance, a subscriber cancelled a subscription at least in part because library photocopies were available.

The relevance and effect of these changes is doubtful in light of the fact that the statutory minima under § 101(b) are apparently intended to take care of instances where a plaintiff cannot prove actual damages. Cf., Shapiro, Bernstein Co. v. 4636 S. Vermont Ave., Inc., supra.

I do not think these alterations were proper in light of the presumption that the trial judge's findings are correct. Rule 147(b). They also suggest that the court would have had difficulty reaching the conclusion it did if it had respected the findings as it should have done. If plaintiff's business is really growing faster than the gross national product or other indicia, without the court's protection, the place to take this into account is in the determination of reasonable and entire compensation.