Professional fair use after Texaco
All professionals in the university community should be aware of an important case that could affect the way we pursue scholarship and research. On October 28, 1994, the Second Circuit Court of Appeals affirmed the decision of New York's Southern District Court that Texaco's copying of journal articles by or for its scientists was an infringement of the copyright owners' exclusive rights.1
Professional fair use: Background
Professionals, including scientists, researchers, teachers, doctors, and even lawyers, copy scholarly materials in their fields of expertise as a matter of course in the practice of their professions. Most professionals work in settings where they must share access to journals and periodicals with their colleagues either by placing themselves on distribution lists or accessing journals through the library. Only a few of these articles will be of such nature that the reader might want to recall some of the ideas expressed in them at a later time while performing research, laboratory work, or pursuing other similar activities.
Thus, occasionally the reader will make for herself, or request the librarian to make for her, a copy for future reference. The reasons for this request may include the fact that the journals need to continue to circulate or otherwise be available to others; the fact that only pertinent articles appropriately indexed or filed, not whole journals, can be kept in personal files; that many professionals prefer to make marginal notes on their copies; that the articles must sometimes be read later as time permits; and that errors that might otherwise occur in preserving the expressed ideas by note-taking or reliance on memories can be avoided by photocopying.2 Until Texaco found otherwise, professionals probably widely believed such copying was fair use under a wide range of circumstances.
In 1985, however, numerous publishers and publishing associations sued Texaco alleging that its scientists made infringing copies of articles from the plaintiffs' publications. Texaco's practice was as described above: Texaco circulated issues of plaintiffs' journals among Texaco's scientists who would make copies of those articles they wished to retain. One of Texaco's defenses was that the copies were a fair use. By stipulation of the parties, the court considered only the fair use defense since the outcome on this issue could be dispositive of other issues in the case.
Section 107: The four fair use factors
As most professionals in the university community will recognize, the copying carried out by Texaco's scientists is not unlike copying carried out in many offices and libraries and would have seemed to fit squarely within the bounds of fair use. The statute provides:
"Section 107. Limitations on Exclusive Rights: Fair Use
Notwithstanding the provisions of section 106, the fair use of a copyrighted work, including such use by reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include-
- the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
- the nature of the copyrighted work;
- the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
- the effect of the use upon the potential market for or value of the copyrighted work.
The fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors." Emphasis added.
Parts 1, 2, 3 and 4 of the statute are the "four fair use factors." The fair use test has always been difficult to apply because the results can vary with even minor changes in the facts relevant to a particular proposed use. Texaco illustrates for us how a court applies the test and weighs the factors in a particular situation.
The fair use analysis after Texaco
Texaco lost its case largely on the basis of two facts: the court described its research copying as institutional, systematic and archival in function, which the court determined weighed against a finding of fair use under the first factor; and Texaco's unauthorized copying denied the plaintiffs revenues they would have received if Texaco had asked for permission to copy, which tipped the fourth fair use factor against fair use as well. The court indicated that to the extent the Copyright Clearance Center provided Texaco various convenient and reasonably priced mechanisms for payment of fees for copying, the scope of fair use had been reduced. The Second Circuit thus affirmed the lower court's decision, "[t]hough not for precisely the same reasons."3
The first factor
The Second Circuit attempted to clarify the relationship between for-profit status and commercial or non-commercial purposes under the first factor, thus moderating one of the lower court's most troubling holdings.4 Nevertheless, it reintroduced one of the chief ambiguities raised by the lower court's distinction between for-profit and nonprofit researchers through its own differentiation between institutional and independent research. As the dissent pointed out, even university research is institutional these days, rarely being done independently of university, government or industry support. The distinction the court tries to draw will not be sustainable.
Further, the Second Circuit narrowly defined research purposes (whether institutional or independent) under the first factor. The court asserted that only copying done to accommodate laboratory work is a research purpose; it described all other copying to facilitate research as "archival" and determined that such copying weighs against a finding of fair use.5 Overall the court's discussion of the first factor lays a firm foundation for generalizing the holding to apply to nonprofit, professional copying, despite its attempt to restrict such application.6
The third factor
The court determined that copying articles from journals is copying "whole" works, rather than parts of works. Most of the works at issue in Texaco were individually copyrighted as well as collectively copyrighted (as a compilation) by the copyright proprietor. As a result, courts following Texaco will weigh the third factor against fair use for most typical research copying purposes.
The fourth factor
The court's discussion of the fourth factor further strengthens the basis for extending the holding of Texaco beyond its context, despite the court's insistence that the holding is limited to the specific facts of Texaco's copying: it strongly endorses the Copyright Clearance Center and the role it has played in the development of a viable market for permission to photocopy.
The fourth factor analysis has always been viewed as especially important. It asks the question, "What harm would the copyright holder suffer if the copying at issue were widespread?" Until recently, the nature of the harm that a publisher could show from professional copying would have been speculative (lost subscriptions). Now, with the advent of the Copyright Clearance Center and other blanket licensing arrangements, publishers can show that they are losing licensing and royalty revenues as a result of such copying.
The court notes that the case before it involved copying for which a license was readily available, thus implying that if a license were not available, the result may have been (or will be) different; however, the case gives copyright owners a tremendous incentive to register with the Copyright Clearance Center or other such licensing entity and to the extent publishers do so, they will undoubtably benefit from this holding that the scope of fair use diminishes as the ease of access to copyright works improves.7
The interplay among the factors
The fair use weighing and balancing test is not so straightforward as it might appear from reading the statute because the weight given to one factor can influence the weight given to others. For example, a nonprofit purpose under the first factor has traditionally affected the consideration of the weight of the third factor, how much one may copy, and the fourth factor, the effect of the proposed use on the market for the copyrighted work. The interplay in the Texaco decision, however, has had a surprising effect: the court's determinations under the first and third factors have markedly decreased the potential significance of a nonprofit use in the analysis of the fourth factor.
For example, in evaluating the fair use factors in the context of typical nonprofit research activities after Texaco, a court may give little weight to the nonprofit nature of university uses because the activity itself, research-related copying, can now be labeled systematic, institutional and archival and will weigh against a finding of fair use. Thus, the first factor probably will be of little benefit to the nonprofit research user who has institution, industry or government support and who receives journals or their tables of contents through departmental distribution lists. The third factor likely will weigh against the nonprofit research user who copies entire articles, while the second factor is the only one that may weigh in the user's favor; however, even the second factor can weigh against the user where research subjects and articles copied involve more creative and fanciful expression (poetry, prose, images, sounds). Thus, by the time a court visits the fourth factor, possibly two and perhaps even all three of the first three factors weigh against the user.
The interplay in this particular circumstance is a crucial element of the Texaco decision because it makes it possible for the court to consider the effect of lost revenues (the fourth factor) without appearing to violate an important rule of logic: Texaco and many of its research community supporters had faulted the lower court's opinion for employing "circular reasoning" by taking lost revenues into account in the fair use analysis. There was ample support for Texaco's charge.
First, the Williams case, which involved nonprofit library copying of articles for scientists, indicated that even if a publisher is willing to license the making of copies or there exists a way to assess and collect royalties for permission to copy, that fact alone should not be enough to defeat a fair use defense in the nonprofit setting.8 The Williams court specifically noted that it is error to "measure the detriment to plaintiff by loss of presumed royalty income" because such a measurement assumes the conclusion that the use was unfair to begin with and that the publisher had a right to issue licenses.9 Second, even though the lower court in the Texaco case determined that individual copying by Texaco's scientists did not qualify as fair use, it approved the proposition that the loss of potential royalty income should not in and of itself convert a fair use into an infringing one.10 Third, there is the strong dissenting opinion in Texaco which raises (among others) the same issue.
The majority opinion addressed the allegation of circularity in the lower court's reasoning, but, like the lower court, finessed the issue: it agreed with Texaco's criticism and ostensibly with Williams and Judge Jacob's dissent, that it would be circular to consider lost royalties and permissions fees if doing so would convert an "otherwise fair use" to an infringing one.11 In other words, the court asserts that reasoning is only circular if the court must assume its conclusion in order to reach it. If, however, the court has logically arrived at a conclusion of unfair use before considering lost royalties, that is, if the use appears unfair by the time the court reaches consideration of the fourth factor, the court asserts there would be no circularity in considering lost revenues. Thus, courts following Texaco probably will not reach the issue of whether it would be circular to consider lost revenues because Texaco has greatly reduced the possibility that there will be an "otherwise fair" research use, one that is fair after evaluation under the first three factors.
The dissent reached the circular reasoning issue because it rejected the finding under the first factor that the copying was for an unfair purpose.12 Thus, Judge Jacobs could strongly object to the majority's cursory treatment of the circular reasoning issue, stating flatly, "I do not agree at all that a reasonable and customary use becomes unfair when the copyright holder develops a way to exact an additional price for the same product."13 By the time the dissent reached the fourth factor, it was considering an otherwise fair use (two of three factors thus far were in favor of fair use) and could not logically permit lost royalties to convert that use to an unfair one. Thus, it was the interplay among the first, third and fourth factor analyses that is the key to the result.
Assuming, of course, that many publishers will register with the Copyright Clearance Center to take advantage of this court's holding, the issue for the nonprofit research community will be essentially, whether other courts will follow the reasoning of Texaco,14 its characterization of typical research copying as institutional, systematic and archival, its narrow definition of research purposes and its reliance upon the existence of the Copyright Clearance Center to justify narrowing the scope of fair use. To the extent they do, we might be prepared for the possibility that the nonprofit nature of an alleged infringer's use will have little effect upon the result.
I attempt with these recommendations to set out a path between overly conservative reaction motivated by fear of litigation and fearless assertion of the widest possible right of fair use. Thus, while I do not wish any university to become a test case to answer the questions raised by Texaco, we don't want to abandon our rights until it is clear that we no longer have them.
We are in a period of tremendous change brought about by new technologies that are creating new and exciting possibilities for electronic creation, distribution and use of research materials. In fact, it is quite possible that a case about photocopying may prove to be unimportant within a relatively short period of time. Nevertheless, and regardless of its long-term effect or whether other courts follow it, Texaco, like all cases, has raised more questions than it has answered. Thus, Texaco does not clarify the scope of professional fair use in the university environment, but it may alert us to recognize that certain activities in the nonprofit research community may, at some point in the future, be vulnerable to challenge. But Texaco should also alert us to the fact that there are new doors opening.
With this in mind, I would recommend that professionals in the university community continue to make use of the statutory fair use privilege, following traditional guidelines that suggest that copying single articles from journals and periodicals and chapters or other small parts of works is fair. When practical, consider copying abstracts or other portions rather then entire articles.
Today, however, subscribing to, circulating, and making copies of articles from journals is not the only way to access research materials. Texaco may be a call to professionals who use and professionals who provide access to research materials, our research and library community, to work together to find a balance between the traditional mode of access and newer on-line services, document delivery, and ultimately, open access. It may be that where online services are convenient and reasonably priced, they offer viable alternatives to subscribing and photocopying. Further, where they are not reasonably priced, other alternatives such as facilitating open access should be pursued, though this will involve long term change in the relationships between creators, distributors and users of scholarly information.15
We are in a period of great uncertainty, and research professionals should expect further developments, but as we learn to accommodate the realities of new technologies into the existing framework of copyright law, there will be opportunities as well as challenges within these changing relationships.
1. American Geophysical Union v. Texaco, Inc., 37 F.3d 881 (2nd Cir. 1994) (hereafter, "Texaco").
2. American Geophysical Union v. Texaco, Inc., 801 F. Supp 1, 4-5 (N.D.N.Y. 1992).
3. Texaco, at 1.
4. Thus, the appellate opinion appears to make it possible for for-profit companies to avail themselves (theoretically) of fair use; however, the Second Circuit's definition of research for all practical purposes eliminates that possibility in the research context.
5. Texaco at 5-12. As the dissent points out, however, only a tiny fraction of research actually involves laboratory experiments. Texaco, Jacobs, J., dissenting at 4-10.
6. Since the court's narrowing of the definition of "research purposes" for Texaco's scientists may be expected to serve as precedent for efforts to narrow the scope of fair use for research purposes in other fields and in the nonprofit environment as well, researchers may well wonder whether the court has not, in effect, defined research right out of the illustrative list of typical fair uses at the beginning of Section 107. How, for example, would this definition of research purposes be applied to humanities research?
7. The majority opinion appears to embrace the market failure theory as a justification for fair use. Under this theory, fair use only exists to meet the need that the market cannot efficiently provide, in this case socially beneficial research-related access to copyright works. To the extent the market can efficiently provide such access, fair use should no longer be necessary.
8. Williams & Wilkins v. United States, 487 F.2d 1345, 1359 (Ct. Cl. 1973) aff'd by an equally divided court 420 U.S. 376 (1975).
9. Williams & Wilkins, 487 F.2d at 1357, note 19. Thus, the Williams court cautions against the circular reasoning that the lower court appeared to undertake in analyzing the market effect of Texaco's copying.
10. Texaco, 802 F. Supp. at 21.
11. Texaco, at 18.
12. Texaco, Jacobs, J., dissenting, at 10.
13. Texaco, Jacobs, J. dissenting at 6-7.
14. Although the Second Circuit's holding is only binding upon lower courts in that circuit (New York, Connecticut, Vermont), the opinion will probably be important persuasion for other courts. Because New York has been the traditional home of the publishing industry, the Second Circuit's copyright decisions are especially influential.
15. For example, many commentators have suggested that it is appropriate at this time for universities to take more control over copyrights in scholarly works; for faculty to utilize electronic networks to communicate directly with their colleagues; and for university presses, libraries and computer science departments to collaborate to offer alternatives to for-profit publication of scholarly works. Where subscribing and photocopying are irreplaceable, broadly negotiated blanket license agreements may be more cost effective than transactionally based permissions fees. See Will we need fair use in the 21st century? from Filling the Pipeline and Paying the Piper, Proceedings of the Fourth Symposium on Scholarly Publishing on the Electronic Networks, Spring, 1995, and Copyright in the library: Scholarly communication.