The counter-reformation movement is presently at the stage of a whispering campaign, in which ministries in countries are told that fair use (and by extension possible liberal fair dealing provisions) violate the "three-step" test. And who wants to violate the three-step after all? The appeal by counter-reformation forces to external and abstract concepts like the three-step test is a time-worn tactic: when you can't win on the merits, shift the debate elsewhere to grounds on which you think you can win. Given that few ministry officials are experts in copyright law, much less arcana like the three-step test, these appeals -- made by those who claim to be such experts -- can be effective. They shouldn't be. National governments should make policy decisions based on the merits of the proposals, free from such scare tactics. The three-step test is not a bar to a single proposal of which I am aware.
The biggest of the three-step scare tactics is that Section 107 of the U.S. Copyright Act is incompatible with the test. Baloney. WIPO and European copyright experts testified before the U.S. Congress during the hearings on U.S. adherence to Berne, hearings that spanned four years: 1985, 1986, 1987, and 1988: there was no lack of time or opportunity to raise any concerns. Congress even went to Geneva and convened a round table discussion there on November 25 and 26, 1987 with WIPO and European copyright experts, the sole purpose of which was to determine which parts of U.S. law needed to be amended to permit Berne adherence. Not once at this round table or during four years of hearings were the words "fair use" ever raised by a foreign expert who appeared before Congress nor did any domestic witness (of whom there were many dozens) consider there to be a potential problem. (A transcript of the round table is reproduced in the House Hearings: "Berne Convention Implementation Act of 1987, Serial No. 50, 100th Congress, 1st & 2d sessions 1135- 1213(1987, 1988)). I can say from direct experience of having been involved in these efforts at the Copyright Office that I never heard a single European expert claim there was a compatibility issue with fair use.
In addition to Director General Arpad Bogsch who was the lead witness in the very first congressional hearing (before the Senate on May 16, 1985), the following foreign experts participated in the round table: Mr. Shahid Alikhan, Mr. Gyorgy Boytha, Mr. Jean-Louis Comte, Mrs. Mialgros De Corral, Professor Dr. Dittrich, Mr. Mayer Gabay, Mr. Roland Grossbacher, Professor Gunnar Karnell, Mr. Jukka Liedes, Mrs. Margaret Moller, Dr. Werner Rumphorst, Mr. Victor Tarnofksy, Professor Dr. Dirk W.F. Verkade, and Mr. Jean-Alexis Ziegler. If any of these illustrious experts had any thoughts on fair use, they were not expressed publicly.
The following testimony, given by then WIPO Director General Arpad Bogsch (he was DG for 25 years, from 1973 to 1997) to the U.S. Senate may surprise many:
The only real difference -- and that has nothing to do with the level of protection -- that makes the U.S. law incompatible with the Berne Convention consists in the notice and registration requirements. One can solve that in two ways: make compliance with those two requirements voluntary rather than mandatory -- for any work, or make compliance with those requirements voluntary only for foreign works that would have to be protected under the Berne Convention.
May 16, 1985 hearing, page 10).
When Dr. Bogsch was asked about the compatibility of U.S. limitations and exceptions, the only one he thought might be a problem was the Section 116 jukebox license, and even here, he testified, "I would say it could be regarded as a matter of not sufficient importance from an economic viewpoint, so that nobody would be very excited about it, even if you don't change your law." (page 14). Nor can one say Dr. Bogsch was simply unfamiliar with U.S. law. Although he was a native of Hungary, he had worked in the U.S. Copyright Office, obtained an American law degree, and became a U.S. citizen. (See an obituary of him here).
So why now, 23 years after Dr. Bogsch's testimony, is the issue of fair use compatibility being raised with national ministries? If there really was a problem, why wasn't the issue taken to a WTO panel rather than the pathetically minor amendment to Section 110(5) that was taken to a WTO panel ? Why wasn't the obvious lack of moral rights in the U.S. objected to? Why wasn't the U.S. failure to provide retroactive protection to foreign works as required by the Berne Convention complained about, a failure that existed from 1988 to 1995? Why single out fair use 23 years after Dr. Bogsch testified that the only incompatibilty issues were notice and registration? The answer is obvious.
The three-step itself is under attack too, and is being cast as a rigid limitation on limitations, rather than as a recognition that copyright rights only go so far. There is an excellent book on the three-step test, Martin Seftleben's "Copyright, Limitations, and the Three-Step Test: Analysis of the Three-Step Test in International and EC Copyright Law," published by Kluwer Law International in 2004 and available here. Mr. Senftleben was a student of the great Dutch copyright scholar Bernt Hugenholtz who just co-authored a major piece on the issue, available here. In addition to the counter-reformation revisionism about the compatibility of Section 107, there is counter-reformation revisionism about the three-step test.
The test is found in Article 9(2) of the 1971 Paris text of the Berne Convention, and reads:
It shall be a matter for legislation in the countries of the Union to permit the reproduction of such works in certain special cases, provided that such reproduction does not conflict with a normal exploitation of the work and does not unreasonably prejudice the legitimate interests of the author.
The three steps are (1) certain special cases which (2) do not conflict with a normal exploitation of the work, and (3) do not unreasonably prejudice the legitimate interests of the author. Counter-reformation forces act as if Article 9(2) is a centerpiece of the Convention, something essential to the goals of ensuring that authors' rights aren't eroded by those thieves and pirates who insist on using the author's work without permission. Not true: the three-step test is a very recent addition to the Berne Convention, which was founded in 1888. The provision was proposed in the 1967 Stockholm protocol and then included in the 1971 Paris text, only 14 years before WIPO General Bogsch testified before Congress on U.S. adherence to the Convention. As Mr. Senftleben points out in his insightful chapter 3 ("The Contextual Background to the Three-Step Test"), the introduction of the three-step test was part of a deal, the deal being providing formal recognition to the reproduction right: "The feasibility of the plan to attain the formal recognition of a general right of reproduction, however, depended on whether or not the Conference would succeed in finding a satisfactory formulation for possible limitations." (page 47).
That formula depended on recognition of a vast variety of existing national limitations on rights; this issue raised what Senftleben calls the "dualism inherent in the three-step test" (pages 81-82). It is this dualism that the counter-reformation forces wish to obliterate. As Senftleben points out
Article 9(2) BC was distilled from typical features of the described extensive list of limitations that existed in 1967. The 1965 Committee of Governmental experts unequivocally took the view that in the course of the preparatory work for the Stockholm conference that 'the main difficulty was to find a formula which would allow of exceptions, bearing in mind the exceptions already in many domestic laws.
What were those exceptions? A study group found 14 common limitations: republication of public speeches; quotations; copying of school books and chresomathies; copying of newspaper articles; reporting current events; ephemeral recordings; copying for private use; reproduction by copying in libraries; reproduction on special characters for the use of the blind; copying of sound recordings of literary works for the use of the blind; copying of the text of songs; taking pictures of sculptures on permanent display in public places; use of artistic works used as background in films and TV programs; and reproductions in the interests of public safety. (Senftleben p. 48).
Current efforts to use the three-step test as a sword rather than a shield might well deny some of these uses as qualifying under the three-step test. The phenomenon of the three-step as a sword should be understood as part of a larger picture, in which copyright has become an absolute property right, what Blackstone (an advocate of copyright as a natural right) hyperbolically called an individual's "sole and despotic dominion ... over the external things in the world, in total exclusion of the right of any individual in the universe." This very recent phenomenon has led to what Professor Neil Netanel has rightly described as “spawn[ing] a vicious cycle. It reinforces the view that copyright owners are intrinsically entitled to control and reap the full value of each and every use of copyright-protected expression, thus smoothing the way to still further expansion.” Copyright's Paradox p. 55.
The attacks on fair use and the transmogrification of the three-step test are an essential part of this cycle: it is not enough to have vast rights: corporate content owners see a need to eliminate any limitations on those rights too, and one must hand it to them for their Karl Rove strategy of going after your opponent's strong points, rather than his weak ones. But unlike John Kerry, we need not and should stand by and let limitations and exceptions be swift-boated. Mr. Sentfleben points out that the origins of the three-step lie not only in the necessity of preserving existing limitations as a quid pro quo for the formal recognition of the reproduction right, but that the formulation chosen for the three-step was not an enumerated list of such existing uses, but rather abstract criteria. As Mr. Senftleben points out: "A comparison of the various observations made by the members countries elicits the specific quality of the abstract formula...: due to its openness, it gains the capacity to encompass a wide range of exceptions and forms a proper basis for the reconciliation of contrary opinions." (page 51). That sounds a lot like fair use.
Openness and toleration of diverse approaches have been an important hallmark of the Berne Convention, given its desire to include both civil law and common law countries, as well as developed countries and lesser developed countries. That same openness and toleration of diverse approaches -- so under attack today in the new TRIPS trade world of IP -- is what led Dr. Bogsch to approve of U.S. adherence to the Berne Convention, adherence where fair use was so obviously compatible with the letter, spirit, and history of the three-test step that not a single WIPO official or foreign copyright expert ever mentioned it in the four years U.S. adherence was being debated and eagerly sought. In a response to a question from Senator Mathias about the attitude of other Berne members, Dr. Bogsch replied, "They ardently desire it. They would welcome it without qualification." (May 16, 1985 p. 12). Without question they desired adherence in compliance with Berne's requirements, but the U.S., like any country, was entitled to rely on the opinion of the Director General of the organization administering the treaty as to what steps were necessary to comply. And Dr. Bogsch gave his opinion, an opinion in which the only limitation mentioned was the jukebox compulsory license, a limitation Dr. Bogsch though was of no moment. There was no mention of fair use, because no one thought fair use was incompatible with the three-step test.
The current counter-reformation movement should therefore be seen as the ahistorical effort it is. National governments are free to craft laws that serve their own needs and policies, including liberal limitations and exceptions.
Postscript Added on April 6, 2008
The great Israeli scholar Orit Fischman kindly pointed out to me that in 1996, a number of countries, as part of the TRIPs process, asked each other questions about their laws. The United States was asked many questions on a wide variety of topics, but only three dealt with fair use, and even then the questions were about specific applications. The EU asked about parody, Australia asked about a case on speed-up kits and a case about photocopying, and New Zealand asked about commercial research and scholarly uses. Here is the link to the entire document. It is worth pointing out that no country has filed an objection since the U.S. gave its answers. Here are the questions and the answers, with the answers in italics.REPLIES TO QUESTIONS POSED BY THE EUROPEAN COMMUNITIES AND THEIR MEMBER STATES
1. Under 17 U.S.C. § 107, the United States permits "fair use" of a copyrighted work. Please explain how the fair use doctrine, as it has been broadly applied and interpreted by US courts, particularly in connection with a "parody" that diminishes the value of a work, is consistent with TRIPS Article 13, given the obligation to "confine limitations or exceptions to exclusive rights to certain special cases which do not conflict with a normal exploitation of the work and do not unreasonably prejudice the legitimate interests of the right holder".
The fair use doctrine of US copyright law embodies essentially the same goals as Article 13 of TRIPS, and is applied and interpreted in a way entirely congruent with the standards set forth in that Article. Fair use was his to rically developed as a safety valve to the exclusive rights granted by copyright, permitting limited and reasonable uses without permission or payment - precisely those types of uses which do not interfere with the copyright owner's normal exploitation of the work or unreasonably prejudice his or her rights. The classic example of fair use is the quotation in a book review of portions of the book being reviewed. The copyright owner ordinarily does not prepare reviews of his own work, and the impact on his interests will ordinarily be minimal. If, on the other hand, so much material is quoted that the review will substitute for purchases of the book in the marketplace, the use will not be considered fair.
The articulation of this concept that was developed by the courts over the years and codified in the Copyright Act in 1976 (17 U.S.C. § 107) identifies four fac to rs that courts must examine in determining whether a use is fair: (1) the purpose and character of the use, including whether such use is of commercial nature or is for nonprofit educational purposes; (2) the nature of the copyrighted work; (3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and (4) the effect of the use upon the potential market for or value of the copyrighted work. The balancing of the four fac to rs represents a means of evaluating whether in fact the use in question conflicts with a normal exploitation of the work and unreasonably prejudices the copyright owner's interests. The Supreme Court has stated that the fourth fac to r, which specifically focuses on the impact on potential market exploitation of the work, is the most important. See Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 566 (1985). Other factors may be considered as well, allowing a flexible and sensitive calibration of the impact of the particular use under the circumstances.
In applying the fair use doctrine, the courts have consistently refused to excuse uses that go to o far and interfere with the copyright owner's normal markets for the work. See, e.g., Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U.S. 539 (1985); American Geophysical Union v. Texaco, Inc., 37 F.3d 881 (2d Cir. 1994); Pacific & Southern Co. v. Duncan, 744 F.2d 1490 (11th Cir. 1984); Basic Books, Inc. v. Kinko's Graphic Corp., 758 F. Supp. 1522 (S.D.N.Y. 1991). Those uses that have been permitted have generally involved productive and non-commercial uses of limited amounts of material from the copyrighted work. See, e.g., Arica Institute, Inc. v. Palmer, 970 F.2d 1067 (2d Cir. 1992); Wright v. Warner Books, Inc., 953 F.2d 731 (2d Cir. 1991); Triangle Pub., Inc. v. Knight-Ridder Newspapers, Inc., 626 F.2d 1171 (5th Cir. 1980).
In response to the specific question about parody, it is important to understand that not all parodies qualify as fair use under US law. Rather, each parody is dealt with on a case-by-case basis, applying the factors set out in Section 107 in a manner that takes in to account the nature of parody. Thus, the US Supreme Court has held that the purpose and character of parody involves both criticism and humour, giving it social and cultural value, and that a parody, in order to be effective, must necessarily take enough of the copyrighted work to make the object of the parody recognizable. Campbell v. Acuff-Rose Music, Inc., 114 S.Ct. 1164 (1994). Not all humorous works that draw on pre-existing works will enjoy the benefit of this more lenient application of the fair use factors, however; the defendant's work must qualify as a true parody, targeting and commenting on the copyrighted work rather than simply using it as a vehicle for unrelated comment. Id. at 1172.
In examining the fourth fair use factor, the effect on the actual and potential market for or value of the copyrighted work, the Supreme Court held that any negative impact caused solely by the critical aspect of a parody could not be considered. Id. at 1178. This analysis is consistent with TRIPS Article 13, since any such lessening of public interest in the copyrighted work does not constitute "unreasonable prejudice" to "legitimate interests" of the copyright owner - rather, it is equivalent to the lessening of public interest that may result from a scathing review. This type of harm should not be actionable by a copyright owner; otherwise only flattering commentary or benign parody is likely to be permissible. In contrast, harm to the market for other adaptations of the copyrighted work will clearly weigh against a finding of fair use. Id.
[Follow-up question ]
The US answer equates the likely effect of a critical parody of a work with that of a scathing review. The answer also suggests that a much greater part of the copyright work will be reproduced in the parody than in the review. Unlike a review, therefore, exploitation of the parody could apparently replace any desire to exploit the copyright work.
How does the US justify this possibility given the requirement in TRIPS Article 13 that limitations and exceptions to exclusive rights should not conflict with a normal exploitation of the work?
If exploitation of a parody is of a nature that it replaces any desire to exploit the copyright work, the fourth fair use factor applied by the courts will weigh strongly against the defendant, and the use will not be considered fair. As noted in our original answer to this question, the Supreme Court of the United States has specifically directed the lower courts to consider the effect of the parody on the market for other adaptations of the work, in applying the fair use factors. As a result, the right holder's ability to control the exploitation of copies or adaptations of the work is protected.
REPLIES TO QUESTIONS POSED BY AUSTRALIA
1. Please explain, having regard to the decisions in Sega Enterprises v Accolade, Inc. and Prince to n University Press v Michigan Document Services Inc. and any similar cases whether and how the US law of fair use complies with Article 9(2) of the Berne Convention and Article 13 of TRIPS.
In the United States' written responses to the first set of questions posed by the European Commission, we explained why the doctrine of fair use complies with Berne Article 9(2) and TRIPS Article 13. Please see our answer to the Commission's question 1
As to the two cases referred to in Australia's question :
In Sega Enterprises v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992), the court applied the doctrine of fair use to the specific facts of the case before it. As explained in our discussion of fair use generally, courts are required to balance several fac to rs in light of all the facts and circumstances. The Sega court did exactly that. This narrow holding based on the particular facts in the case is a defensible judgement for the purposes of the exceptions permitted under the TRIPS Agreement
The decision in the other case, Princeton Univ. Press v. Michigan Document Servs., 74 F.3d 1512 (6th Cir. 1996), has been vacated. Princeton Univ. Press v. Michigan Document Servs., reh'g en banc granted, vacated, stay granted 74 F.3d 1528 (6th Cir. 1996).
REPLIES TO QUESTIONS POSED BY NEW ZEALAND
1. In respect to the reply to the European Communities and their Member States on fair use (their question 1), please clarify how practice, especially case law, has affected the use of works in respect to teaching, research, and scholarship. In particular, how open ended is the multiple copying right for teaching, does research encompass any commercial research, and what is encompassed by scholarship?
It is important to understand that the purposes listed in the first sentence of Section 107, which include teaching, scholarship, or research, do not automatically qualify a use as a fair use, but are simply examples of the types of purposes that are likely to qualify as fair. See Twin Peaks Prods., Inc. v. Publications Int'l, Ltd., 996 F.2d 1366, 1374 (2d Cir. 1993). Each use must still be examined in light of the four statutory factors found in Section 107. Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 127 L. Ed. 2d 500, 514-515 (1994).
Accordingly, extensive multiple copying for the purposes of teaching has been held to fall outside the scope of the fair use doctrine and to constitute an infringement of copyright in the works copied. See Basic Books, Inc. v. Kinko's Graphic Corp., 758 F. Supp. 1522 (S.D.N.Y. 1991), noted in our original answer to the question from the European Communities. Guidelines as to the permissible amount and timing of such copying, negotiated by the educational and publishing sectors, are set out in the legislative history to the Copyright Act. H.R. Rep. No. 94-1476, 94th Cong., 2d Sess. 68-70 (1976).
Additionally, a leading case held that the copying of journal articles by a commercial researcher did not qualify as fair use. See American Geophysical Union v. Texaco, Inc., 37 F.3d 881 (2d Cir. 1994), noted in our original answer to the question from the European Communities
Finally, the label of "scholarship" in itself will not resolve the fair use issue, and there is no established definition of the term. If a particular use is designed to advance knowledge, this should be an aspect of the purpose and character of the use that weighs in favour of a fair use determination.