|
Copyright in the LibraryInterlibrary Loan |
Section 108(g) attempts to balance the interests of publishers and libraries regarding interlibrary loan arrangements. During the years preceding amendment of the copyright law in 1976, publishers lobbied hard for the inclusion of the strongly worded clauses in sections (1) and (2) in the bill before Congress.1 Libraries, fairly certain that such language would prevent normal and customary interlibrary loan arrangements, lobbied hard for the addition of language to clarify that the systematic copying prohibited by the statute was not that which was customary so long as it did not substitute for subscriptions or purchases of the affected works.2
Later, the members of the National Commission on New Technological Uses of Copyrighted Works ("CONTU") negotiated guidelines that described what amounts of copying would be "such aggregate quantities as to substitute for a subscription to or purchase of such work," a critical measurement in the balance. The CONTU Guidelines appear to have been much more successful than their Classroom counterparts 3 in fixing realistic or reasonable limitations in that the library community seems comfortable with the limitations.
As we near a more fully electronic library environment, however, new technology appears once again likely to upset the balance that was achieved in an earlier era. Interlibrary loan has matured into document delivery service, and publishers appear to be quite nervous that libraries are, in fact, competing with commercial service providers who pay royalties on each and every copy made and do not enjoy the benefit of the "suggestion of five." In fact, participants in the recent Conference on Fair Use (CONFU) who were charged with negotiating guidelines for interlibrary loan in the electronic environment were only able to agree that they could not agree on practically anything.
The library may be either a requestor or responder in the interlibrary loan context ("IL"). As requestor, we are responsible for compliance with copyright law and, where applicable, the CONTU Guidelines. As responder, we are only responsible to ask whether the requestor has so complied. Further, the CONTU Guidelines only apply to certain materials (articles from periodicals published fewer than five (5) years before the date of the request and small parts of other works) and do not define aggregate quantities of copying that would substitute for subscriptions to materials that are older than five (5) years or whole works under Section 108(e).
As requesting party, the library must comply with Sections 108(e) & (g):
As requesting party, the library must comply with Sections 108(e) & (g) and the CONTU Guidelines:
As requesting party, the library must comply with Sections 108(d) & (g) and the CONTU Guidelines:
Most observers believe that the supplying library in an IL transaction does not have responsibility for compliance with copyright law. Further, where the request is for an entire book and not for a part of a periodical or other work covered under Section 108(d), the supplying library has no responsibility under the CONTU Guidelines for requiring the statement of compliance with the Guidelines.
It nevertheless appears to be standard practice to require a statement of compliance with copyright law from the requesting library. Asking for and receiving assurance that the requesting library has complied with copyright law may have the effect of providing the supplying library with some "insurance" against the possibility that the supplying library could somehow be held responsible for engaging in related or concerted copying. Section 108(g) allows libraries to copy in accordance with the provisions of Section 108 so long as the library has no "awareness" or "substantial reason to believe" that it might be engaging in related or concerted reproduction or distribution of multiple copies of the same material. Courts have on occasion held that persons are responsible for "knowing" that which a reasonable inquiry might have revealed. If such an interpretation were followed in this context, a library that asks for and receives an assurance that the requesting library has complied with copyright law would be better able to say that it had no awareness than a library that did not ask.
As responding party, the library will not have responsibility for compliance with copyright law, but should nevertheless request a statement of such compliance from the requesting library, as explained above. Further, a request for articles or small parts of other works is covered by the CONTU Guidelines so the library should require a statement of compliance with the Guidelines as well.
Many libraries have begun to offer document delivery service, though in general there may not be a bright line between IL transactions and document delivery anymore. In the case of large public institutions like ours where libraries recognize a patronage that is, in essence, statewide, the application of Section 108 rights can be quite sweeping. Does document delivery by its very nature substitute for subscriptions or is it merely the natural expansion into electronic media of IL practices that are well-established and accepted? If IL practices and Section 108's affirmation of such practices clearly contemplate photocopies, why should the medium make a difference? Questions about the medium are not perhaps the most difficult; what about requests that come to the library from librarians or others in for-profit businesses, law firms, investigator's offices and commercial research entities? What about requests from commercial document delivery services? Is the requestor's declaration that he or it has complied with copyright law or the CONTU Guidelines adequate?
These questions evidence an approach to the analysis of document delivery that attempts to understand the practice in the context of "making copies," an activity that is closely circumscribed and essentially an infringement, except for our narrow exemptions about which both libraries and copyright owners may be very uncomfortable. In truth, document delivery is poised to take center stage as the basic mechanism for distribution in the 21st Century (see discussion below on Database Licenses and Interlibrary Loan). The exceptions provided by Sections 107 and 108 are about to swallow the rule. Sections 107 and 108 were never intended to be the main vehicles for the distribution of knowledge and information but they appear to be headed towards that function. Perhaps the idea of making copies and focusing on those copies and trying to control and count those copies is impeding our collective recognition that a fundamental change is taking place in the nature of distribution whereby the copy will not have the same meaning, significance and importance that it used to have. When the simple proscription, "Don't make a copy" has the power to prevent all use of a document in so far as even glimpsing it requires that one and perhaps many copies be made, then a "copy" is not what it used to be.
Document delivery will continue to raise issues related to whether it is essentially an abuse of Section 108 in that it is, by its very nature, related and concerted reproduction and distribution under section 108(g). Our libraries must be rigorous in complying with the requirements of copyright law and the CONTU Guidelines, where they apply, but we cannot be sure whether that will be enough.
Section 108 was modified in 1998 by provisions in the Digital Millennium Copyright Act ("DMCA") and the Sony Bono Copyright Term Extension Act ("SBCTEA"). Although neither of these Acts changed subsection (g), the part of Section 108 that specifically authorizes IL activities, they did affect IL indirectly. For more detailed information about these changes, please see Complying with the Digital Millennium Copyright Act and Sonny Bono Copyright Term Extension Act: Making Digital Copies in the Library. The changes themselves are shown on 17 U.S.C 108 Redlined to Show DMCA Changes and Clean Copy. It would be advisable to be able to refer to the latter document during the rest of the discussion in this section.
The DMCA and SBCTEA changes affect IL as follows:
The new notice requirement requires libraries to reproduce any copyright notice that appears on the work, and if there is none, to include a notice similar to those libraries have used before the law changed.
The limited exemption from SBCTEA's 20 year term extension applies only in the last 20 years of protection for published works and the new subsection (h) that implements the exemption details the conditions for exercising it. It appears to provide rights unrestrained by the part of subsection (g) that prohibits systematic copying that substitutes for subscription to or purchase of a work (with the CONTU guidelines establishing what that means), because that section's proscription only applies to copies made pursuant to subsection (d) (parts of a work copied in response to a patron's request). Subsection (h) appears to provide rights free of all of the other restrictions of the Section, except the restrictions in the first subsection that establish qualifications for relying on Section 108 rights.
Thus, it might seem that this new right is broad and valuable, but one might question how often today patrons or libraries request copies of works published 76 years ago. As a percentage of all requests, they probably are quite limited. Next year we will add another year's worth of works to this number: works 76 or 77 years old. And so it will go until we have 20 years worth of works that will qualify for this exemption, but always, they will be works that are very old. Eventually, the equation will relate the libraries' exemption to the number of years after the author's death, that is, it will apply to works by authors who have been dead for 51 through 70 years. If I were handy with numbers, I could establish the year that will happen. See Lolly Gasaway's chart online for more information about when works pass into the public domain.
There are no recent cases on interlibrary loan; however, two recent cases about authors' rights indirectly could affect interlibrary loan practices, as well as other library practices involving licenses to make copies or access and use electronic works generally.
The two cases are Tasini v. The New York Times Co., ___ F.3d ___ (2d Cir. 1999) ("Tasini") and Ryan v. Carl Corporation, 23 F.Supp. 2d 1146 (N.D. Cal. 1998) ("Ryan"). In each case freelance authors who wrote articles for periodical publications asserted that the defendants in the lawsuits had ignored the authors' copyright ownership interests.
In Tasini, newspaper and magazine publishers sold back issues and current issues of their publications containing the authors' articles to online and CD-ROM database companies who put them into electronic databases that users could search for individual articles. These companies did not compensate the authors for the use of their articles in this way. The companies claimed that they had acquired the rights to do this contractually, and that even if they did not have contractual rights, they were authorized by Section 201(c) of the Copyright Act to create the electronic databases without permission from the authors. Section 201 authorizes them as holders of copyright in collective works, to reproduce and distribute the articles as part of a collective work or a revision of the collective work. 4
The companies argued that databases were just such revisions. At first they lost on the contractual arguments and won on the statutory arguments, but the case was appealed. On appeal they lost on the statutory argument too. The court said that in the absence of a contract with an author that clearly gives them the rights, publishers do not have the right under Section 201(c) to place the entire contents of their collective works in cumulated electronic databases, or to license others to do this for them, or to permit users to access individual articles from such comprehensive databases.
The defendant company in Ryan (Carl's UnCover business) isn't a publisher and didn't have a database of articles. UnCover had a database of titles its customers could search to identify articles of interest. UnCover would simply go to a library and get a photocopy of whatever article a customer requested and send it to the customer. (Sound familiar?) UnCover paid a copyright fee to the publisher or to the Copyright Clearance Center ("CCC") after copying the article. UnCover did not ask permission before making a copy and it did not, in most cases, pay any fees directly to authors or even attempt to identify whether authors were copyright owners. This was the basis for the lawsuit. The freelance authors alleged that they owned the copyrights in the articles in question, that they had not transferred the rights to their publishers, that Section 201(c) did not authorize their publishers to grant permission to UnCover to make photocopies of these articles and therefore, any permission to be obtained and any moneys to be remitted had to be obtained from and remitted to them, not their publishers.
Ryan's authors won that part of their case addressing the issue of whether Section 201(c) authorizes publishers to grant permission to photocopy individual articles. According to Ryan, Section 201(c) says that for-profit document delivery requires permission from and payment to any author who owns copyright in his or her article.
The facts the Ryan authors allege certainly imply that their publishers and the CCC were not properly authorized by the authors to grant permission or receive royalties. But the authors did not sue their publishers and the CCC. They only sued UnCover. UnCover's activities look an awful lot like interlibrary loan, although they were done for a profit. Should we, like UnCover, rely on publishers and the CCC to have the rights they say they do when they give us permission to make copies for IL transactions exceeding the suggestion of five?
Why did the freelance authors sue UnCover? Why didn't they sue CCC? What might keep them from suing libraries? I think the answer has something to do with bad faith and UnCover's apparent willful ignorance of authors' rights. Those of us who rely on others' representations that they have the rights to authorize our activities must hope that good faith makes a difference because Ryan shows us that as a technical matter, asking permission from or paying royalties to the wrong person is infringement.
So, good faith. Does CCC have it? I think it does. CCC handles royalties differently from the way UnCover handled them at the time the lawsuit was filed. It does not presume to have permission and remit fees to publishers only, but gets prior permission by contract with the copyright owner and pays royalties directly to authors in any case where it has been advised that the authors are the copyright owners. CCC actively recruits authors and tries to make it as easy as possible for them to register. The largest part of CCC's licensed works, and the ones most often requested, are works where the copyright typically is owned by publishers. On the other hand, CCC is cognizant of those areas where this is not the norm, as evidenced by its formula for distribution of money it receives from foreign rights collectives, where titles sometimes are not identified. CCC also obtains from each rightsholder a warranty that the rightsholder has the authority to grant CCC the rights granted in the contract. But then, the problem is that these warranties are not always accurate because publishers are not always aware of or right about what rights they have and don't have.
Today, CCC and UnCover are "partners" with the National Writers' Union in the effort to make getting permission from authors easier. The climate that brought about the Ryan case seems to have changed dramatically, probably at least in part because of the case.
Are libraries that perform similar functions to those performed by UnCover likely to be targets for a lawsuit? Do we risk liability by relying on the CCC and publishers to tell us that it's ok to make a copy of an article? In all likelihood, the risk is quite small for several reasons: libraries do not make a profit on IL and they don't bear any ill will towards authors attempting to make the publishing industry recognize their claims to a share of the new revenues coming from electronic uses of their articles.
What about the impact of Tasini? Can libraries rely on publishers who license databases of electronic articles to have the rights they claim to have (the rights to create and license the database) and to share royalties with authors who still own copyright in their contributions to the publishers' collective works? Need libraries worry about such things? I think we are much more likely to see some increase in the price of databases to accommodate the required "split" with authors than to see lawsuits against library licensees of those databases. Further, there may be only a small percentage of material in some databases in which the publisher does not own copyright. Remember, these cases are about the rights of freelance authors and the issues raised would not apply to cases where an author has assigned his or her rights to a publisher or is an employee of the publisher.
Libraries increasingly license access to electronic works they previously acquired exclusively in analog form. It has been the rule for some time that most licensors of electronic databases did not permit IL or document delivery from their digital materials. Recently there has been some shift in this almost uniform armoring against traditional modes of sharing. Ultimately, however, newer forms of licensing could greatly reduce the need for and practicality of IL and document delivery.
For general information about acquiring materials by contract, please see Acquisition Under Contract, another article in this series. Here I would like to discuss the future of IL by exploring a recent straw-poll about the need for IL in the electronic context, conducted among the subscribers to the Liblicense-l listserv, Elsevier's IL provision in its Elsevier Science license agreements and Elsevier's PEAK licensing arrangement that permitted article-level access to thousands of Elsevier journals for fixed fees.
The following is from the Liblicense Website describing the results of a straw poll:
"On July 13th, after an animated thread on this list re: ILL and whether librarians should accept licenses that do NOT permit it, we conducted a straw poll that 137 of you (just under 10% of the list's readers) responded to. The responses to the ILL question appear below, with a breakdown by category of respondent (we've broken US out from non-US to see if there are any clearly US-centric views among librarians, but this proves not to be the case.).
As you will see, 5 librarians out of 123 think that ILL is not an effective concept to carry into the electronic environment; 118 believe some provision needs to be made for electronic "ILL." That's over 95%.
More interesting are the reasons librarians advocate for this strongly held view. A number of respondents characterize this accommodation of electronic ILL as possibly an interim need (without speculating just how long an interim period might be) until the time when easy and affordable by-the-piece delivery from publishers might be an effective substitute. "
Perhaps Elsevier Science's IL license provision meets the "interim" need in the electronic environment, perhaps not. Might Elsevier's PEAK database model have been an effective substitute for IL?
In June 1998, Elsevier Science announced that its licensees would be permitted to use electronic copies of articles to make analog copies that could then be delivered by IL to other libraries. While this seemed to be a giant step forward from its earlier position, unfortunately Elsevier also required that the library that provides the copy must keep records and pay Elsevier if it sends more than 5 copies from the same journal to the same library, regardless of the age of the articles.
This requirement directly conflicts with the CONTU guidelines which explicitly acknowledge that the suggestion of 5 only applies to journal articles 5 years old or newer and require the borrowing library to keep records (and, logically, remit copyright royalties) for copies to which the suggestion of 5 applies. Thus, many who might desire to take advantage of this option may not because it "would add considerable costs to an already cumbersome and expensive operation" and "will cause confusion because the requesting library will undoubtedly continue to report the copy requests [as] this function is built into all relevant software and is a routine part of all procedures[,] and pay the CCC..." David Goodman on Thu, 11 Feb 99, email sent to liblicense-l listserv.
Perhaps this problematic provision will be or has been modified by Elsevier. In that case, the ability to utilize an electronic copy to create a paper copy to send to another library certainly eliminates some of the burden of responding to IL requests. Perhaps such a provision may become more widespread.
This project was an experimental research proposal funded and administered by the University of Michigan. The research model provided participants (10 University libraries in 1999) the option to order e-subscriptions to specified Elsevier journals for an additional fee over the cost of the print version, order and pre-pay for a set number of articles from any of the journals in the PEAK database for about $4.50 each, or order articles from the database as needed for a higher per-article fee.
The research team characterized its innovative pricing model (pre-paying for a set number of articles) as "a way to prepay (at a discount) for interlibrary loan requests" with the added advantage that all of a participant's users would have access to any article once it was paid for. 5 Indeed, a preliminary report from Vanderbilt indicates that a large part (55%) of the usage it observed during its participation was requests for articles from journals to which it did not subscribe previously, articles it would have obtained through traditional IL. 6
Project pricing was steeply discounted to encourage participation and further structured to discourage print cancellations (to encourage Elsevier to participate). But, even assuming that a realistic price for a pre-paid allotment of articles would be the same as or similar to the price a library would pay for its current selection of subscriptions, might the new pricing model still profoundly affect IL? With this model the library obtains exactly what its patrons want for the same price that it otherwise would have paid to obtain some fraction of what they wanted (and a lot of what they don't want), with the balance obtained through IL at additional cost. In fact, how would such a pricing model affect document delivery, the CCC and everything to do with distribution and use of journal articles if anything like this were to become widespread?
Suppose a University pays for 10,000 journal articles at some fixed price per article. Under this model, the publisher should not care whether the library provides the articles to its own faculty and students, or to another library's patrons. But why would a library wish to give away part of its allotment? Even more importantly, why would a library ask another library to do that if any library can acquire any article it needs from the source for a reasonable fee? Will journal article IL just quietly disappear if the PEAK model were to become widespread? Is this likely? In our lifetimes?
1 Section 108. Limitations on exclusive rights: Reproduction by libraries and archives
...
(g) The rights of reproduction and distribution under this section extend to the isolated and unrelated reproduction or distribution of a single copy or phonorecord of the same material on separate occasions, but do not extend to cases where the library or archives, or its employee--
(1) is a ware or has substantial reason to believe that it is engaging in the related or concerted reproduction or distribution of multiple copies or phonorecords of the same material, whether made on one occasion or over a period of time, and whether intended for aggregate use by one or more individuals or for separate use by the individual members of a group or
(2) engages in the systematic reproduction of distribution of single or multiple copies or phonorecords of materials described in subsection (d): ...
2 Section 108(g) proviso (the last clause in Section 108(g)(2):
Provided, That nothing in this clause prevents a library or archives from participating in interlibrary loan arrangements that do not have, as their purpose or effect, that the library or archives receiving such copies or phonorecords for distribution does so in such aggregate quantities as to substitute for a subscription to or purchase of such work.
3 See discussion in Reserve Room Operations, Print Copies.
4 Section 201 Ownership of Copyright
...
(c) Contributions to Collective Works -- Copyright in each separate contribution 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 express transfer of the copyright or of any rights under it, 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, any revision of that collective work, and any later collective work in the same series.
5 A Report on the PEAK Experiment: Usage and Economic Behavior, Jeffrey K. MacKie-Mason, Juan F. Riveros, Maria S. Bonn, Wendy P. Lougee, D-Lib Magazine, July/August 1999.
6 Project PEAK: Vanderbilt's Experience with Articles on Demand, John Haar, Presented at NASIG, June 1999. See also, Pricing and Bundling Electronic Iinformation Goods: Field Evidence, Jeffrie K. MacKie-Mason, Juan F. Riveros, and Robert S. Gazzale, Presented August 25, 1999; System Design, User Cost and Electronic Usage of Journals, Robert S. Gazzale and Jeffrie K. MacKie-Mason, March 22, 2000; and Peak and Elsevier Science, Karen Hunter, March 23, 2000.
Fair Use (Section 107): |
Library Reproduction (Section 108): |
Other: |
|
Top
| Search
Crash Course in Copyright | Intellectual
Property Section | Office of General
Counsel