College and Research Libraries MATT ROBERTS Copyright and Photocopying: an Experiment in Cooperation The problem of copyright and photocopying by libraries is examined, with particular reference to the arguments offered by publishers in defense of the "exclusive right" provision of the copyright law. Pro- ceeding step-by-step through these arguments, it appears rather that "fair use'' is virtually self-explanatory and that libraries not only are privileged, but are obligated to photocopy reasonable parts of copy- righted works in order to fulfill their responsibility to scholarship and to "promote the Progress of Science and useful Arts.'' IN ONE OF HIS SEARCHING essays on the problems of copyright Luther Evans writes: Copyright properly understood and wise- ly handled may be at the same time a powerful stimulus to creation and the means of opening the channels of dis- semination of thought, information and de- bate. Misunderstood, and with its true purposes lost sight of, copyright can be- come a limitation on creation and a bar- rier to free exchange and expression. Like many other products of man's genius in the realms both of science and of law, it has a capacity for good or evil depending on his understanding and the use he makes of it. 1 Librarians and publishers have been debating the problems of copyright and photoduplication for many years. Per- haps the most controversial question in the entire copyright debate has centered about the matter of "fair use" -that is, the manner in which librarians contend 1 Luther H. Evans, "Copyright and the Public In- terest," New York Public Library Bulletin, LIII (Jan- uary 1949), 4. Mr . Roberts is Chief, Circulation De- partment, John M. Olin Library, at Wash- ington University , St. Louis. 222 I they are privileged to use copyrighted materials and publishers say they are not. Controversy in this one area has led librarians and publishers along wide- ly diverging paths in search of a just solution. "Fair use" is a difficult concept to de- fine. In fact, if there is any agreement at all, it is that the term "eludes precise definition,"2 and yet many have at- tempted to define it. Fair use has been defined as "a doctrine which enables the courts to live with a law which contains a misstatement of fact so palpable that, if interpreted and enforced literally, it would involve them in absurdities con- trary to the public interest"; 3 as a "priv- ilege in others than the owner of a copy- right to use the copyrighted material in 2 U.S., Congress. 87th Congress , 1st session. Copy- right Law R evision. Report of the R egister of Copy- rights on the G eneral Revision of the U.S. Copyright Law, July 1961 (Washington, D.C ., U.S. Governm ent Printing Office, 1961), p. 24 . 3 Verner W. Clapp, " Library Photocopying and Copyright : Recent D evelopments ," in Law Library Bullet in, v. 55, February 1962, p . 12. The law to which Mr. Clapp refers is the 1909 Copyright Law. The "m isstatement of fact" refers to that part of the law which states that the owner of the copyright "shall have the exclusive right to: print, reprint, publish, copy and vend the copyrighted work." It is this statement that has led to such bitter disagreem en t b etween li- brarians and publishers. a manner without his consent ... ";4 and as a type of limited license to publish, which appears to be inherent in the na- ture of statutory copyright. 5 These and other definitions are all accurate in one sense or another, but they leave too much uncertainty as to what can or can- not be done in everyday practice. 6 Per- haps the term eludes any definition whatsoever. One purpose of this essay is to de- termine whether it is actually necessary to define "fair use" at all. It will be the contention of this paper that it is not. Were all irrelevant issues to be cleared away, and the principal arguments pro- ceed ed through systematically, it would be found that, in logic anyway, the term virtually defines itself. But before this may be done, one must accept two con- ditions. First, librarians must .agree to be less contentious about their "rights" and accept the fact that publishers, too, have rights ; and second, publishers must de- sist from their curious habit of waving the eighth commandment under librar- ians' noses·, and understand that they are only attempting to fulfill their obligation to scholarship. There are, it seems, but three basic is- sues involved in "fair use" in copyright and photocopying by libraries, and none is beyond compromise. They are: ( 1 ) Does "fair use" denote only single copies of a "reasonable part"7 for private use, and in lieu of hand transcription, or does it include the reproduction of entire 4 Horace G. Ba ll , T he L aw o f C op yright and Lit er- ary Property (New York : Banks and Co., 1944), p. 260. 5 Ralph R. Shaw, Literary Property in the United States ( n.p. : Sca rec row Press, 1950 ), p. 67 . 6 Unlike the 1909 copy right law, the proposed cop yri ght revision a t lea st mention s " fa ir u se," statin g tha t " . . . the f a ir u se of a copyrighted work is not an infringe m ent of cop yri r ht. " See: U .S. , Congress . 8 9th Congress, 1st session , A B ill f or the G en eral R evision o f the Co pyright Law, T itle 17 of the United States C od e (Washington , D.C.: U.S . Government Printin g Offic e , 1965 ), p . 17. 7 " Reasona ble p a rt' is a n o the r t erm which h as n ev- e r b een d efin ed to the satisfaction of all concern ed . L e t us say h e re, for the r e cord , that a r ea sonable p a rt w ould include on e or more pa ges , but less than a n e ntire ch a pte r or journa l arti cle . Copyright and Photocopying I 223 chapters or articles, or multiple copying for classroom ( reserve) use, in lieu of purchase?; ( 2) Does copyright and "fair use" apply only to monographs , or to both monographs and serials?; and ( 3) Does it matter if the photocopying of a reasonable part is done gratis or at a charge? Few would contend that libraries have the right to photocopy an entire copyrighted book, or a major portion thereof, because such copying would substantiate the publishers' argument that photocopying is an extension of printing, 8 in lieu of purchase. But is this kind of reproduction really a threat to publishers' interests? The immediate question must be: Why would a library want to copy an entire book, or a sub- stantial portion thereof? Reproducing a complete book, to which the library would have to add the cost of collating and binding, would b e cumbersome and economically unfeasible. If it is a matter of replacing deteriorating books, a li- brary would undoubtedly find it less ex- pensive to seek the services of an or- ganization specializing in reprinting .and/ or photocopying, and would get a b etter product. This is particularly true since all too frequently the library's copy will be in such poor condition that it cannot easily be copied , as will likely b e the situation in other libraries own- ing that particular edition. Moreover, whole book copying for preservation for the most part involves books in the pub- lic domain. 9 For these reasons, it is dif- ficult to see how publishers are injured by this kind of copying. 8 " Hou se Copyright H e arin gs: Education a nd 'Fair Use,'" in Publishers' W ee kly , v. 188, no . 9, section 1 , August 30, 1965, p. 291. 9 W . S. Buddin gton , in hi s stud y of the u se of copy- righted mate rial, reports tha t only about 10 p e r cent of materi a l copyri ghted is rene w ed for th e second twenty- eight years. See: W illiam S. Buddington, " U sing Copy - ri ghted Mate rials,'' in Sp ecial Libraries, LII ( Novem- b e r 1961 ) , 511. It is inte restin g to sp ecula te on w h y , in view of this low ren ewal r a te, publish ers a re so eager to see inc orpora ted in the prop osed cop yright law th a t the t e rm of copyri ght b e th e life of the a uthor plus fifty years. 224 I College & Research Libraries • May 1969 But there are those who contend that photocopying per se is injurious to pub- lishers, their wri~ers, and, ultimately, the public. 10 This position appears to be to- tally unacceptable. It is based on the general principle that all copying is in lieu of purchase, and this certainly is not the case. No library can expect a scholar ( or any other reader) to copy several hundred or thousand words by hand, so that he might have for future reference the gist of another man's think- ing. Single copy reproduction for pri- vate us e, and in lieu of hand transcrip· tion, is and must be associated with the precept that only the manner of expres- sion, and not ideas themselves, are pro- tected by copyright. It would seem that no reasonable man could deny the logic of this. The . publisher who contends that copying is intrinsically in lieu of pur- chase has made a serious mistake. It is abundantly clear that the alternative is not between photocopying and purchase of .another copy; it is between photo- copying and copying by hand, mem- orizing, or, what is more likely, doing without. And whereas, if libraries re- produce single copies for scholars, pub- lishers have not lost; indeed, they have gained, to the extent that new books may be forthcoming. But if libraries re- fuse to make single copies for private use, publishers have not gained even a little ; and libraries have failed in one of their principal responsibilities , which is to assist the scholar and thereby "pro- mote the Progress of Science and useful Arts." The scholar loses, and, presum- ably, scholarship suffers . The area in which publishers, and par- ticularly monograph publishers, should concentrate their efforts is in the copy- l Q R oger H. Smith, "Library Photocopying: The Stalemate Deepens," in Publ-ishers' W eekly, CLXXXIV (August 12, 1963) , 41; and, again, in B enjamin Kaplin, " Copyright, Libraries, an d the Public Inter- est," in College & R esearch Libraries, XXI (May 1960) , 215. ing of entire chapters or articles, and multiple copying, principally for class- room use, which, both in logic and in law, is clearly indefensible and dele- terious to their interests. 11 It is pub- lishing, pure and simple. Unless permis- sion is obtained in advance, there can be little justification for this kind of copying. 1 2 There is a wide range of dif- ference between one copy of ten pages for private use and ten copies of one page for public use. But one point must be crystal clear. The publisher who con- tends that there is no essential differ- ence between reproducing ten copies for ten scholars over a period of time and ten copies at one time for "public" use is mistaken. In the former, copying is purely in lieu of hand transcription; in the latter, it is in lieu of purchase. Unfortunately, there are librarians as well as publishers who fail to see the distinction. In the early days of the "fair use" controversy, librarians and publishers did agree informally to .allow single- copy reproduction for private use. The "Gentlemen's Agreement" of 1935 stipu- lated that a library could make a single copy of a copyrighted work for a scholar who stated in writing that he desired such reproduction in lieu of loan of such publication or in place of manual transcription, .and solely for the purpose of research. 1 3 The original agreement 11 Curtis G. Benjamin, " Book Publishers' Interests in Reprographic Copyright," in Library Journal, LXXXVIII (August 1963) , 2840. 12 Photocopying entire chapters or journal articles, particularly the l a tter, for inter-library loan purposes is pertinent h ere, in that it involves more than a " r ea- sonable part." While it could be argued that such copying is for private us e and in lieu of lending, it could just as easily be argued th at it is for private use but in lieu of purchase. P erhaps if specific pages to be copied could be specified, this might be resolved; but it is difficult t o specify pages b efore the " loan" is made. Nevertheless, a library might find it difficult to demonstrate that photocopying an entire journal ar- ticle for inter-library loan constitutes "fair use." This assumes, of course, that the article is copyri ght ed. 13 " The G entlem en's Agreement and the Problem of Copyright," in Journal of D ocumentary Re production, II, 29-36. was between the Joint Committee on the Reproduction of Materials for Re- search and the National Association of Book Publishers. It was later used as a basis for ALA's Materials Reproduction Code. While the agreement still serves as a guide to some libraries, faith in the Gentlemen's Agreement and strict ad- herence to the single copy principle ap- parently has fallen by the wayside.14 It fell in part because librarians con- sidered it unsatisfactory for several rea- sons, mainly because it implied an ad- mission on the part of librarians of a violation of the law which publishers were willing to overlook1 5-a license to steal, so to speak. In one sense this was an unfortunate attitude. It is possible that publishers might have continued to subscribe to the single-copy concept had librarians been willing to adhere to it in practice, if not in theory, and had they instituted an effective program to prevent multiple copying. But they did not. Today, publishers, as well as li- brarians, know all too well that an in- creasing number of libraries are pro- ducing multiple copies of chapters and journal articles for use in reserve read- ing rooms; consequently, it is not' sur- prising that many publishers are deeply concerned over the practice of photo- copying in general. Is there any distinction between copy- right of a monograph and of a periodi- cal? The present copyright law requires the author, or, in the event of transfer of right, his agent (publisher) to apply 14 Edward G. Freehafer, "Photocopying and Fair Use," in College & Researc h Libraries, XXI (May 1960), 217. 1 5 Edward G. Freehafer, "Summary Statement of Policy of the Joint Librari es Committee on Fair Use in Photocopying," in Special Libraries, LV (February 1964) p. 104 . On page 105 of this article, the com- mittee recommends that "it be library policy to fill an order for a single photocopy of any published work or any part thereof. Before making a copy of an entire work, a library sho uld make an effort by consulting standard sources to determine whether or not a copy is available through normal trade channels." Publishers have not demonstrated much enthusiasm for this pol- icy. Copyright and Photocopying I 225 for copyright, deposit two copies of the best edition of the work with the Office of Copyright (Library of Congress), and to display the copyright notice on the title page or the page following. And this is about .all that is required. This seems straightforward enough, at least for monographs, or any other form of one-time publication. But how does it apply to periodicals? Are individual ar- ticles in an issue covered? The author of a particular article may apply for .and receive copyright protection (he need send in only one copy), or he may ex- pressly assign his right to the publisher. But suppose he does neither? To those not versed in the law, section three of the 1909 act would seem at first to in- clude periodical articles. In brief, it states that "copyright ... shall protect all the copyrightable component parts of the work copyrighted." But then it goes on to say, "and all matter therein in which copyright is already subsisting. ... The 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. . . ."16 (Italics mine) . This is confusing. They are pro- tected, but apparently only if copyright already . subsists. Perhaps if the wording had been "as if," it might have been construed to mean the individual ar- ticles. Out of the confusion has come the interpretation that the copyright of an issue does not protect the individual articles, unless: · 1) rights are assigned to the publisher; or 2) the publisher or one of his employees is the author of the article. 17 16 U.S., Congress. 60th Congress, 2nd session. Con- gressional R ecord, XLIII (Washington, D.C.: U.S. Government Printing Office, 1909), 3701-3702. 17 U.S., Congress. 86th Congress, 2nd session. Copy- right Law Revision. Studies Prepared for the Subcom- mittee on Patents, Trademarks, and Copyrights of the Committee on the Judiciary United States Senate, 86th Congress, 2nd session, pursuant to S. Res. 240, studies 11-13, (Washington , D.C. : U.S. Government Printing Office, 1960) , p. 18. 2261 College & Research Libraries • May 1969 The proposed copyright law revision, unless the revision is revised, clarifies the old law, and excludes individual contributions, stating that, . . . copyright in each separate contri- bution to a collective work is distinct from copyright in the collective work as a whole, and vests initially in the author of the contribution. In the absence of an ex- press transfer . . . the owner of copyright in the collective work is presumed to have acquired only the privilege of reproducing and distributing the contribution as part of that particular collective work. . . .18 Individual articles, then, are not and will not be protected by copyright, un- less the author expressly transfers his right, and very few do so at this time.19 Some may be protected, however; and the library that copies articles freely may be infringing. Aside from this possibil- ity, if the libra1y requires copies of com- plete articles for whatever purpose, it can and should ask for permission in advance (see Appendix). Is there actually any real difference between reproducing a single photocopy for a scholar gratis and doing the same thing at a charge? Do libraries have the right to install (or have installed) coin- operated copiers? Does it matter that the library receives no income from public copiers and may even lose money in the copying it does for scholars? There seems to be little agreement among li- brarians and publishers in these admit- tedly difficult aspects of photocopying. From the publishers' side, the essen- tial point seems to be that in making a oopy for a scholar, the library is not copying for itself, but for another party; and, even if it loses money in the op- eration, the fact that it charges for the work is the real distinction. 20 The publishers' argument in essence 1 8 U.S., Congress. 89th Congress, 1st session, op. cit., p. 12. 1 9 U.S., Congress. 86th Congress, 2nd session, op. cit., p. 18. 2o Miles 0. Price, "Photocopying b y Libraries and Copyright: a Precis, " in Library Trends. VIII (January 1960) , 438. seems to be that in copying for another and for compensation, the library de- prives publishers of income which they , the publishers, would receive if the li- brary did not copy. It is again the "in lieu of purchase" argument, and, insofar as it presumes that if the library did not copy, the publisher would (i.e. , print ), it is tenuous logic. While in a sense the "for compensation" and "copying for an- other" principles are the best arguments publishers have against photocopying b:v libraries, they are valid only in an un- realistic sense, for when the logic is carried one step further-i.e., in lieu of purchase-it just does not hold up, be- cause photocopying for private use is not in lieu of purchase, but in lieu of manual transcription. The solution to this problem is for li- braries to demonstrate beyond question . that: a) single photocopies are solely in lieu of manual transcription ; and b) no profit is realized from photocopying. They have, it would appear, demon- strated the former, and they can demon- strate the latter. They must do so, be- cause there can be no justification in a library deriving profit from photocopy- ing. ·Fulfillment of these prerequisites would place libraries on safe ground in the spirit, if not the letter, of the law. When one comes to the question of coin-operated copiers, he is really in a forest of terrors , both real and imaginary. Publishers can at times be quite vehe- ment in their opposition to public cop- iers, contending that unsupervised copy- ing cannot in any sense be called "fair use ."21 One can only suppose they pre- sume all such copying to involve copy- righted materials. But librarians too can be equally adamant in the opposite di- rection.22 (One begins to suspect that Armageddon may be more appealing to 2 1 Lee C. Deighton , "Books, Not Copying Machin es, Are the Long-term Answer," in Library Journal, XC ( May 1, 1965), 2090. 2 2 Charles F. Gosnell, "The Copying Grab-bag, Ob- servations on the New Copyright Legislation ," in ALA Bulletin, LX (January 1966 ), 50. many than the market place of reason. It is probably more exciting, anyway!) From both points of view, the prob- lem here is largely, but not entirely, economic. The coin copier is valuable to the library in several respects: ( 1) in copying personal materials, e.g., lecture notes; ( 2) in relieving the library in part of the considerable staff costs in- volved in providing copies; ( 3) in pro- viding copying services when the li- brary's copying section is closed; and ( 4) in alleviating to some extent the serious problem of mutilation. The argument against public copiers is that the library has no control over the use made of them. This is a valid argument, but only if one assumes that ·they are used for copying copyrighted materials in lieu of purchase; without this assumption it is no argument at all. Let us, then, consider the possible uses that can be made of such copiers: a) they are used to copy personal materials or works now in the public domain; b) they are used to reproduce single copies of a reasonable part for private use and in lieu of hand transcription; c) they are used to copy entire chap- ters or journal articles, or to make multi- ple copies for public use, in lieu of pur- chase. Use "a" presents no problem; "b" should not be a problem; and "c" would constitute infringement. But are public copiers used for pur- poses such as in "c"? To believe that anyone is going to deposit dime after dime, or quarter after quarter, in a ma- chine in order to copy an entire chapter or article solely in lieu of purchase, or to make multiple copies for public use, calls for an imagination of some vivid- ness. It could be done, of course, and publishers are theoretically correct; but in all likelihood they are frightened b y a chimera. The third party in all this-the manu- facturers and distributors of copying de- vices-who reap, as it is said, where Copyright and Photocopying I 227 they do not sow-should be the ones to reimburse publishers for any income ac- tually lost. Unfortunately, no workable method has been devised to implement such a system. The problems involved in compensat- ing publishers for income lost, if in truth any is, because of library photocopying .are all but insoluble. 23 But the problem can doubtless be solved, and, at the same time, the entire issue of "fair use" summarized, by means of a series of questions and answers. Q. Can libraries justify photocopying in lieu of purchase? A. . .. They cannot. Q. If publishers actually do lose income, would the cost to libraries and pub- lishers alike in collecting and dis- tributing penny royalties exceed the amount collected? A. . .. Very likely. Q. If income is lost, as publishers insist, which publishers lose and how much? A. . .. No one knows. Q. Assuming that photocopies are pro- duced solely in lieu of hand trans- cription, how much copying, either by the library directly, or by means of coin-operated copiers, involves copyrighted materials? A .... No one knows. Q. Considering only the reproduction of copyrighted materials, how much in- come is lost? A .... No one knows. Q. Do publishers really lose any income at all because of photoreproduction of copyrighted materials? A .... No one knows. Q. If we assume that libraries agree to forego all photocopying, is it then reasonable to expect publishers to keep on hand an adequate number 23 Solutions have b een offered, however; some of them quite bizarre. See, for example: U.S., Congress. 76th Congress 3rd session. Congressional R ecord, LXXXVI (Jan~ary 3, 1940) 65-66; and "A Licensing System, A Proposal by the Authors League of America, Inc.," in Library Journal, XCI (February 15, 1966), 892-93 . 228 I College & Research Libraries • May 1969 of copies of each individual work, including journal articles, to satisfy demand for the life of the copyright? A .... It is not. Q. Do publishers have any obligation to provide copies on demand for the life of the copyright? A. . .. In a limited sense, they do not, .as the copyright law stipulates only that two copies of the work to be copy- righted be deposited with the Reg- ister of Copyrights, and does not re- quire that even one copy be offered for sale to the public. But in a much broader sense, they do, because pub- lishers, as well as librarians, are obli- gated to "promote the Progress of Science and useful Arts." This is no mere moralistic viewpoint, but a seri- ous and unavoidable obligation on the part of publishers-it represents the Constitutional basis of all copy- right legislation in the United States. The Constitution states that "The Congress shall have the power ... To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Invent- ors the exclusive Right to their re- spective Writings and Discoveries." (Art. 1, Sec. 1 (Par. 8) ) And while the publishers who oppose photo- copying have been quick to seize up- on the "exclusive right" concept, they have seemingly been less aware of their responsibilities concerning the promotion of science, etc. They have · taken the ideas of "progress" and "ex- clusive right" and made them their own; and, in doing so, they have failed to see that no "exclusive right" can remove from them the responsi- bility to scholarship which the Consti- tution intended and which librarians must assume. It follows then that: 1) publishers should accept fully a policy of single photocopies of a reasonable part of any copyrighted work for private use, in lieu of hand transcription; and 2) librarians, under the positive leadership of the as- sociation, should agree that, unless per- mission is obtained in advance, repro- duction of entire chapters and complete articles, and multiple copying for public use, will not be permitted. • • APPENDIX The faculty of a university frequently calls upon the library to provide multiple copies of books and periodical articles for class (reserve) use. Books represent here only a limited problem, as they can usually be obtained in sufficient quantities. In ad - dition, the use-turnover of books is usually considerably lower than that of journal articles. Periodical issues, particularly ret- rospective issues, are a problem, as they are not available very long, or long enough, after publication; and, even if they are, it is usually at a relatively high price. The library therefore finds itself in a dilemma . It must provide the necessary materials , and yet, short of photocopying, it finds it difficult to do so. The problem essentially is one of time, and to a lesser extent of cost. Admittedly, given sufficient time, the library might be able to find added copies of back issues, though as stated above at considerable cost. Also it might discover that reprints are available. Finally the library might b e able to obtain permission from publishers to make multiple copies of sp~cified ar- ticles. Often, however, there is too little time to make effective use of these al- ternatives. What then can the library do? If the solution does not lie in multiple subscriptions, because of cost; nor in spe- cific requests to copy, because of time; the library must look elsewhere. That it must do something is obvious, for if it elects to do nothing, it merely passes the problem, and solution, on to someone else. A three-way attack on the problem is possible. The library might increase the number of its subscriptions to major peri- odical titles. This need not be done solely as a solution to the reserve problem, but to provide added copies of materials which are in great demand for research. It might also subscribe to reprints (e.g., Bobbs-Mer- rill) in one or more copies. Finally, it might request permission in advance from publishers whose periodicals are considered essential for reserve use. With this final approach in mind, the author contacted the publishers of 255 pe- riodicals, requesting permission to make up to five copies of single articles for reserve use. The library agreed to pay a reasonable royalty, if required, and gave assurances that the copies would not be given away or sold, and would be destroyed when the need for them had passed. At the same time, the library took out second and even third subscriptions of frequently used jour- nals and expanded its coverage of re- prints. Replies were received from 227 of the 255 publishers contacted. Some 117 ( 45 per cent) agreed to permit multiple copy- Copyright and Photocopying I 229 ing as specified and did not demand royal- ties or attach any contingencies; forty-four ( 17 per cent) granted permission without royalty but did attach certain conditions, such as source on each copy, informing the publisher as to what was copied, send- ing one copy to the publisher, and return- ing copies when no longer needed. Nine- teen ( 7 per cent) granted permission with royalty and, in several cases, wished to be informed of what was copied. Thirty- six ( 14 per cent) reported either that their authors held copyright, or that their policy did not permit copying. The re- maining four ( 1 per cent) reported they had a reprint arrangement and therefore could not permit copying. In summary, 183 (72 per cent) extended overall per- mission, forty ( 16 per cent) refused per- mission, and thirty-two ( 12 per cent) did not reply.