· The creation of copyright works for educational uses is a complex and substantial endeavour, particular in a country as large, diverse and sparsely populated as Canada. Careful thought should be given before adopting fair dealing rules that threaten the health of educational publishing.
· “Fairness” is an eye-of-the-beholder concept that must be revisited on the facts of every new case. Only six years after the landmark CCH case on “fair dealing”, this issue is returning to the Supreme Court for a second review!
· By exposing all copying with any connection to “education” to a vague fairness assessment, Canada risks violating its international treaty commitments. Such a broad and ambiguous phrasing would be unparalleled among the world’s major trading partners.
· The government’s stated desire to promote new technologies in the educational setting would be far better achieved by implementing targeted exceptions that will not be vulnerable to abuse.
French translation is published here, Jan. 26, 2011.
In mid-December, Professor Michael Geist published a blog entry entitled “Clearing Up the Copyright Confusion: Fair Dealing and Bill C-32”, a comment on the government’s proposed addition of another user right in the form of “fair dealing for the purpose of … education”. He concludes that this reform would not affect the balance between creators of educational materials and the users of such materials.
Of all of the critically important amendments proposed in Bill C-32, this reform may have the largest measurable impact. Professor Geist himself has noted that educational institutions and students currently spend more than a billion dollars a year on books and hundreds of millions of dollars on access to educational databases. A representative of the rightsholder collective Access Copyright testified before a Parliamentary committee that Canada’s education sector alone reproduces more than half a billion pages of text for use in classrooms. Currently, authors are compensated for much of this copying because only a percentage of these copies is made for allowable “research” or “private study” purposes. Under the new regime, any educational purpose would be an allowable purpose, leaving it up to courts to decide whether or not the dealing is fair. While estimating the potential future losses to publishers resulting from the change will never be an exact science, projections raised in committee have ranged between $40-million per year (Bloc MPs Carole Lavallée and Serge Cardin) and $60-million per year (Access Copyright). Due to concerns about the breadth of this reform, both the Liberal Party and the Bloc Québecois have suggested that they will seek to amend this provision.
It must also be remembered that the educational marketplace in Canada is an intricate ecosystem with characteristics that are not present in larger marketplaces such as the United States. While certain popular texts will receive wide circulation in Canada, many other texts will be aimed at far smaller marketplaces. Any legislative rule must be calibrated to avoid disincentives for authors and publishers to publish new educational works for these smaller marketplaces, such as those aimed at cultural and linguistic minorities. With so much at stake, the importance of correctly evaluating the policy implications of this particular provision cannot be overstated.
Unfortunately, I do not have the same confidence as Professor Geist that the effects of the proposed amendment will promote new technologies while minimally impairing the interests of authors and publishers. In particular, I am convinced that Professor Geist significantly understates the potential impact of the proposed new educational fair dealing exception, in part because he places too much faith in the second “fairness” stage of the fair dealing test to govern behaviour in the marketplace. Further, Professor Geist’s survey of other laws omits a principled analysis of whether the proposed exception risks putting Canada offside its international treaty commitments, in particular the three-step test enshrined by TRIPS and the Berne Convention for the Protection of Literary and Artistic Works.
In responding to Professor Geist, I offer the disclaimer that although this blog entry is my personal opinion and was prepared in my private time, part of my understanding has been developed through retainers as counsel for a number of rightsholder organizations in this field. (As counsel, I have also advised numerous institutional users of copyright on their fair dealing rights, and appreciate that users have real concerns with respect to managing use of copyright works in a complex technological landscape.)
As a final introductory point, I should note that it is not always easy to respond to Professor Geist, primarily due to his a tendency to make bold pronouncements without explanation or support. A representative example is Professor Geist’s statement that “Educational institutions and students spend over a billion dollars each year on books and hundreds of millions of dollars on licencing [sic] for access to databases. That will not change with the inclusion of education within fair dealing.” Professor Geist does not provide a source or a rationale for this critically important prediction. Furthermore, he tells a different story elsewhere in testimony before a Parliamentary Committee, where he admitted that the fair dealing for education reform would result in lost revenues for authors and publishers:
Mr. Pablo Rodriguez (Honoré-Mercier, Lib.):
Regarding the education example, don’t you think that the exemption, as it is now, would cost revenues to some of the creators, writers, or producers?
Prof. Michael Geist:
Right. As I mentioned in my opening remarks, I think that any copying that takes place, including under the new exception for education, must still be fair. It would be disingenuous to argue that there is going to be no copying that’s currently compensated for that might now fall within fair dealing, but by definition any copying that does indeed qualify through the court’s analysis is fair. [Emphasis added.]
Given such disparities in message, and given Professor Geist’s ad hominem complaint that rightsholders have made “misleading claims about potential losses, inaccurate comments on copyright and Internet materials, and dubious arguments about the compliance of the reforms under international copyright law”, it has been necessary to respond at length to the main contentions in Professor Geist’s blog entries, articles and testimony.
2) Why Is Educational Publishing Important?
Canada is not the first country to step to the brink of considering creating broad exceptions to copyright to reduce public expenditures for education. While Professor Geist tells the story of the United States and Israel to promote his desired result, the experience of the United Kingdom explains why passing such an exception may lead to unanticipated and undesirable results.
In the United Kingdom, the case for not treating educational publishing differently than other sectors of the publishing industry was made by the Whitford Report as long ago as 1977. Decades before the age of the Internet, the makers of the Report recognized that there was an ongoing friction between the educator’s desire to promote “new methods of teaching” and the rights of publishers and authors. The Whitford Committee’s conclusion, which is worth quoting at length, was that governments should resist the temptation to save on education funding by draining the resources of publishers and authors, on the grounds that it was a self-defeating exercise:
The needs of education in modern society are changing. In the new teaching situation […] it is no longer considered appropriate for text books to be issued to each pupil. We were told that new methods of teaching and learning require the use of a wide range of teaching material to meet the particular needs of a student or group of students. This concept of ‘resource-based learning’, whether applied to independent or class activity, was said to require the availability of a diversity of material, extracted from a great variety of sources.
In view of the growth of reprography as a problem in the educational field since the time of the Gregory Committee, we have considered first the question whether there should be any express exceptions in favour of educational establishments at all. We feel that the fact that “education” is a good cause is not in itself a reason for depriving copyright owners of remuneration. Nobody suggests that the makers of notebooks, compasses and rulers should supply these products to educational establishments free of charge. Although the types of material used in such places today are very different from the text books of the past and indeed are much more diverse, education is still in a large measure dependent upon the work of authors, artists and composers. Education is equally dependent upon the work of the publishers who first produced the material which the authorities want to copy for educational purposes.
(Report of the Committee to consider the Law on Copyright and Designs (London: Her Majesty’s Stationery Office, 1977) at paras. 254-56, 268-69)
In evaluating the same issues recently in a copyright dispute between educators and authors, the UK Copyright Tribunal affirmed the Whitford Report, and warned that crossing this line could destroy a longstanding symbiotic relationship between publishers and the education system:
In declining to create a wide generalised defence for educational establishments the legislature has struck a balance between the interests of copyright owners on the one hand, and the interests of education and scholarship on the other. A healthy publishing industry is important in general, but of particular importance to those in education. Wholesale exemption from the copyright laws for educational establishments would be damaging to the publishing industry, and in consequence damaging to education. On the other hand the publishing industry depends on academic authors for much of its raw material. If education is burdened too heavily with copyright restrictions, teaching and scholarship is discouraged, to the disadvantage of the public interest in general, and the publishing industry in particular. It is a good example of a symbiotic relationship.
(Universities UK Ltd v Copyright Licensing Agency Ltd  E.M.L.R. 35)
As I discuss in the final section of this comment, this symbiotic relationship for Canada might be better achieved by implementing a targeted series of exceptions to promote concrete goals, a path that would minimally impair the educational publishing sector while promoting the interests of educators in developing new means of reaching out to students.
3) Is There a Stable Set of Rules to Govern Fair Dealing in Canadian Law?
By far the most important assumption underlying Professor Geist’s blog entry is that the new reform will have little or no effect on authors’ rights because the right is limited only to those dealings that are “fair”. In his view, authors will be protected because judges are “required to assess the fairness of any use – including education – before it was treated as fair dealing”. Since the reform respecting “education” affects only the threshold “allowable purposes” portion of the fair dealing test, nothing will change. (See also Professor Geist’s September blog on this point.)
The stability of rules established by the courts has also been a prominent talking point of the government. In testimony before the Parliamentary committee charged with examining Bill C-32, Industry Minister Tony Clement stated:
I want to give some comfort to this committee on education, because that’s another issue that’s going to come up, the fact that we’ve added education to the fair dealing concept. Again, I’m conscious that I’m at a copyright hearing, so I will attribute my remarks to Professor Geist, who educated me that fair dealing is not free dealing, and there’s a big difference between the two.
Fair dealing means that the work must be for a non-commercial purpose, that the original material was lawfully acquired, and that the use of that original material must not harm the market for that material. That’s a very different concept from just saying, because we’ve added education to fair dealing, all the rules are gone. That’s not true: the rules are still in place and they seek to create that balance. Again, I want to give you that assurance.
This endorsement requires a few clarifications. Critically, the three “rules” stated in the second paragraph of the statement above do not reflect the state of the law in Canada. Rather, in the landmark decision CCH Canadian Ltd. v. Law Society of Upper Canada, 2004 SCC 13, a unanimous Supreme Court found (i) a commercial, for-profit purpose can qualify as a fair dealing (para. 51), and that (ii) the effect of the dealing on the market of the copyright owner is neither the only factor nor the most important factor that a court must consider in deciding if the dealing is fair (para. 59). In that decision, the Supreme Court made no finding that the original material must be lawfully acquired, nor have any other Canadian courts. Last, the Supreme Court found that there is no set test for fair dealing: whether a dealing is fair is a question of fact and will depend on the facts of each case (paras. 52-53).
These points aside, Professor Geist’s justifications for the reform also rely on the proposition that fairness “rules” established by the courts will avoid disruption to this marketplace. With the greatest of respect, this is a faith proposition that appears unsound in light of experiences in both Canada and the United States.
The first problem with relying on fairness alone is the notorious instability of the CCH fairness test. This instability was highlighted by the December 23 decision of the Supreme Court to hear SOCAN’s appeal of a fair dealing decision that found that “fair dealing for the purpose of research” could include previews of songs sold for commercial profit. SOCAN’s leave application argued that the Federal Court of Appeal has divided considerably on how to treat the test for fairness in Canadian law, and a number of commentators have predicted that the SOCAN appeal will be accompanied by an appeal from another Federal Court of Appeal decision affirming that classroom use of educational materials was not fair dealing for the purpose of research or private study.
Accordingly, the “assurance” and “comfort” that are offered by Minister Clement and echoed by Professor Geist are not evident in the case law. Since the fairness test is an open-ended, case-by-case discretionary analysis, it does not provide the hard rules they describe. Rather, the fairness test is notoriously soft and subjective, an eye-of-the-beholder test that will be left in almost every case to the judgment of the copier. This uncertainty was why the well-known U.S. reformist Lawrence Lessig described fair use as an “astonishingly bad” system amounting to little more than “the right to hire a lawyer”. (Lawrence Lessig, Free Culture, at 187.)
The Canadian counterbalance to the soft fairness test has always been to maintain a hard “allowable purposes” test at the threshold of the fair dealing right. If a person cannot plausibly say his or her copy is made for the purpose of research, private study, criticism, review, or news reporting, the inquiry is over. The copy is simply not a fair dealing. This is a sharp contrast to the United States, where users who seek to make a copy for free may offer any soft “fairness” justification and (effectively) dare rightsholders to sue them.
The need for the stability of the threshold test is precisely why the government should consider carefully before using an unprecedentedly expansive term at the hard “allowable purposes” threshold. As Professor Geist states at point 10 of his blog entry, the Supreme Court has construed “education” in broad terms to include “more informal training initiatives, aimed at teaching necessary life skills or providing information toward a practical end”. If this language were applied to the fair dealing test, the critical threshold test would be eradicated in respect of any person “providing information toward a practical end”. All that would be left is the soft fairness test and the user standing alone at the copy machine.
4) Is the Proposed Reform Too Ambiguous?
At point 2 of his blog entry, Professor Geist takes on the contention of the writer Douglas Arthur Brown in his committee testimony that “no one at this point really knows what fair dealing for the purpose of education really means – except more copying without compensation to creators, and more costly lawsuits.” Rather than confront the merits of Mr. Brown’s argument, Professor Geist dismisses it altogether by quoting from an Access Copyright leave memorandum to the Supreme Court. This bait-and-switch tactic allows Professor Geist to gloss over a dangerous challenge to his central claim that the fairness test alone provides certainty to creators.
Professor Geist implies repeatedly that stability will emerge because courts and tribunals can distinguish between fair dealings and unfair dealings. Fair dealings will remain free, and unfair dealings will be addressed appropriately, with a royalty in the form of damages or tariff monies. Thanks to this stability, institutions and other significant users will develop policies that will accurately instruct individuals what is fair and what is not, and these policies will be routinely followed.
From a practicing lawyer’s perspective, this aspiration appears hopelessly naïve, specifically because it is impossible to constructively model behaviour on a rule whose very nature is ambiguous.
There are three primary ambiguities within the proposed user right. The first is what is meant by “education”. The second is what is meant by “fair”. The third ambiguity involves all institutions who have a role in copying works for others: even if it is proved that certain dealings of those other persons were unfair, the institution may nonetheless claim immunity on the basis that its policies and practices were “education-based and fair”. All practicing lawyers know that ambiguities will (i) be interpreted in self-interested ways by adverse parties; and (ii) cost unimaginable amounts of time and money to clarify.
If the law is enacted in its current form, the first ambiguity will result in a vast series of users claiming that their copies were made for “educational” purposes, whether they are within the (likely intended) category of provincially funded teachers copying a substantial part of a work for classroom analysis or the (likely unintended) category of yoga instructors. Inevitably these users will refer to the Supreme Court’s decisions in CCH and in Vancouver Society of Immigrant and Visible Minority Women v. M.N.R. to support even the most far-fetched of claims.
The second ambiguity allows any user of a copyright work with an educational connection to calculate whether a given dealing is fair or not. If “fair” is free and “unfair” bears significant financial and administrative costs, it is only natural for any given user of a copyright work to conclude that his use falls on the fair side of the line, a temptation that will only increase when it is observed that it is pragmatically impossible for copyright owners to target anyone beyond the largest institutional users.
The second ambiguity also threatens inconsistent results whenever a case makes it to a court or tribunal, where lawyers will exercise their considerable ingenuity and research to argue what is fair and what is not. This is now a wide-open battle in Canada given that the Supreme Court has held that the effect of a dealing on the marketplace for the original work is not the primary consideration for fairness, a significant departure from the norm in other jurisdictions. (By contrast, see the U.S. cases Harper & Row, Publishers, Inc. v. Nation Enters., 471 U.S. 539, 566 (1985) and Bouchat v. Baltimore Ravens LP, (4th Cir. Sept. 2, 2010) at 13, calling the effect on the market the “single most important element of fair use”.) In the United States, where the only consideration is whether a use is fair, scholars have shown that courts have ruled with tremendous inconsistency in this area of the law, with the result that outcomes become nearly impossible to predict at the outset. The obvious consequence of such uncertainty is that the law fails to serve as a guide to either rights holders or users, at a great societal cost.
CCH also leaves behind a third real ambiguity in terms of the circumstances in which an institution will be held responsible for copies made by others acting under its authority or benefitting from its facilities and services. This is because the Supreme Court created a generalized defence to infringement when it held: “Persons or institutions relying on the s. 29 fair dealing exception need only prove that their own dealings with copyrighted works were for the purpose of research or private study and were fair. They may do this either by showing that their own practices and policies were research-based and fair, or by showing that all individual dealings with the materials were in fact research-based and fair.” (para. 63). In CCH, this meant that the Great Library was not held responsible for any unfair dealings by its patrons, but could rely on the absolute defence that it had “fair” policies and practices.
If the amendment is passed, it would appear that educational institutions may be able to rely on the same concept in terms of educational copying. Given such an opportunity, it would only be prudent for institutions to immediately adopt “fair” policies and practices as shields against infringement lawsuits or copyright tariffs. Institutions can raise these shields whether or not the persons making the actual copies are acting fairly, with rational rightsholders thinking very hard before spending significant amounts of money (if they have the money to spend) to take a shot at convincing a court that a given “fair” policy is only a mask for an unfair practice.
The prospect for ambiguities leading to a series of expensive court battles is not just theoretical. As I have mentioned, this year the Supreme Court will consider a SOCAN judicial review on the fair dealing purpose of “research”, which the Copyright Board and the Federal Court of Appeal deemed applicable to commercial services providing 30-second digital previews of music to potential purchasers. SOCAN will argue that the concept of “research” has been stretched far beyond its natural bounds; the music services will argue that “research” is a neutral and flexible concept that can be fairly applied to such activities; the only certainty is that this case will have cost millions of dollars to litigate by the time of its resolution. Both sides will refer extensively to the Supreme Court’s prior decision in CCH, the 2004 case that was also fought in respect of “fair dealing for the purpose of research”, but which has left behind many uncertainties with respect to the bounds of “research” and the nature of “fairness”.
The fact that the Supreme Court will now be dealing twice with the same tiny clause in the Copyright Act proves the cost and danger of this particular combination of ambiguities. In the United States, the complexities of “fair use” have led to four separate visits to the Supreme Court, an outcome derided by the leading copyright scholar David Nimmer as “a system whose upshot would be the same … had Congress instituted a dartboard rather than the particular four fair use factors embodied in the Copyright Act” (David Nimmer, “‘Fairest of them All’ and Other Fairy Tales of Fair Use” (2003) 66 Law and Contemporary Problems 263 at 280). Other policy impacts of such ambiguities are detailed at length in Sookman & Glover, “Why Canada Should Not Adopt “Fair Use” (2009) Osgoode Hall Rev.L.Pol’y 139, which concludes that a system that is too open-ended risks damage to copyright owners and users alike.
5) Is the Breadth of the Term “Education” Mirrored Elsewhere?
Rightsholders concerned by the proposed reforms have pointed out that the word “education” can mean many different things to many different people, a quality rarely desired in a term bearing such legal weight. As a consequence, Parliament should consider whether it is wise to introduce into Canadian legislation a term that may apply narrowly to classroom instruction or broadly to any formal or informal activity related to learning.
Although he concedes that the Supreme Court has construed the term “education” in very broad terms in the Vancouver Society of Immigrant and Visible Minority Women v. M.N.R. case, Professor Geist contends that the change proposed in Bill C-32 would be “evolutionary”, not “revolutionary”. Yet he is unable to point to a single legislative system across the entire world that has gone as far as adopting a fair dealing right for the purpose of “education” rather than narrower educational activities.
The closest outliers Professor Geist can cite are Israel, which approves as an allowable purpose the specific activities “instruction and examination by an educational institution”, and the United States, which considers “teaching (including multiple copies for classroom use)”, “scholarship”, and “research” to be acceptable purposes.
A close look at Israel’s Copyright Act demonstrates that its educational exception is far narrower than the proposed reforms in Canada. Section 19 of the Act does not broadly apply to all educational activities, but rather identifies the activities legislators consider fair (“instruction”, “examination”). Such copies cannot be made by anyone, but are available only to an “educational institution”, whose definition is controlled by the Minister of Education under s. 67 of the Israeli Copyright Act).
The language used in the U.S. Copyright Act is also narrower, not only because “teaching”, “scholarship”, and “research” are more closely defined aspects of education, but also because the courts of the United States have generally resisted calls to broaden these terms. Cases such as American Medical Colleges v. Mikaelian, 734 F. 2d 3 (3d. Cir. 1984) and Rubin v. Boston Magazine Co., 645 F. 2d 80 (1st Cir. 1981) have interpreted teaching, scholarship, and research in a common-sense fashion by looking to whether the “students” named as the beneficiary of the right received a degree, became qualified or certified as a result of the course, or used the course as a prerequisite for further education and training. By contrast, it is not clear that any lower court in Canada could do the same given the Supreme Court’s statement in CCH that fair dealing is “a user right… that must not be interpreted restrictively”, and given its broad construal of the term “education” in Vancouver Society of Immigrant and Visible Minority Women v. M.N.R.
The United States is also a cautionary example in the sense that it continues to produce extensive litigation on fairness issues. While Professor Geist states broadly in the current blog entry that “jurisdictions with even broader fair dealing or fair use provisions rarely experience significant litigation between publishers/authors and educational institutions”, and cites the “far more litigious U.S.” as an example, a look at the tangled docket in the highly contested test case launched by publishers against Georgia State University shows otherwise. The Georgia State saga suggest that the U.S. “fair use” doctrine continues to raise difficult questions for decision, at great cost to rightsholders and users alike, with the parties largely locked at a standstill following two years of litigation, and commentators leery that the case will ever provide true guidance for acceptable activities in an academic setting.
Perhaps wary of the U.S. model, the bulk of other legislators worldwide have proceeded with more caution, confining many educational exceptions to targeted exceptions (as in the European Union). A characteristic approach is that of New Zealand, whose newly amended Copyright Act provides at s. 44A that educational establishments will not infringe copyright if they store a copy of “a work that is made available on a website or other electronic retrieval system” only if the material: (a) is stored for an educational purpose; (b) displayed under a separate frame or identifier; (c) identifies the author (if known) and the source of the work; (d) identifies the educational establishment’s name and the date when the work was stored; and (e) available only to authenticated users. While provisions like New Zealand’s take more time and care to craft, they provide exact guidance to publishers and users alike without imposing undue technological barriers. Similarly, copyright legislation in Australia (see Parts VA, VB) and the United Kingdom (see ss. 32-36) prefers to take a targeted incremental approach to educational exceptions.
6) Will the Proposed Reform Satisfy Canada’s Three-Step Test Obligations?
As Canada is a member of Berne and TRIPS, and is currently engaged in free-trade negotiations with the European Union that will touch on intellectual property matters, it must ask itself whether its policies fall in line with international standards and practices. Professor Geist’s blog analysis makes only a bare analysis of this issue, concluding that because no trade action has yet been filed against the narrower Israeli and U.S. regimes, the likelihood of an international challenge to Canada’s law is “incredibly remote”.
A full response to Professor Geist requires a review of Canada’s international treaty obligations. Under the Berne-TRIPS “three-step test”, an author’s exclusive rights such as the right to reproduce or the right to communicate a work are the norm. By contrast, user rights must be carefully calibrated to ensure they target a specific good while minimally impairing the author’s right. The specific expression of the three-step test is as follows:
“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.”
“Members shall 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.”
An example of how this balance is maintained elsewhere is set out in the European Copyright Directive, which explains why the exclusive right of the author should not be lightly disturbed:
“Any harmonisation of copyright and related rights must take as a basis a high level of protection, since such rights are crucial to intellectual creation. Their protection helps to ensure the maintenance and development of creativity in the interests of authors, performers, producers, consumers, culture, industry and the public at large. Intellectual property has therefore been recognised as an integral part of property.”
In asking whether the C-32 fair dealing exception is “evolutionary” rather than “revolutionary”, Canada should ask itself whether incorporating such an exception into domestic legislation could form the basis for a trade challenge under the WTO regime.
Though Professor Geist does not mention it, there is existing international case law demonstrating that ambitious domestic legislation can entangle countries in lengthy and expensive international trade disputes. In 1999, the European Community initiated a trade challenge to an exception imposed by the United States in its Fairness in Music Licensing Act of 1998. This Act created an exception to copyright that permitted a wide variety of eating and drinking establishments to publicly perform copyrighted musical works without the authorization of the owners of those works. The EC contended that the exception was too broadly targeted to survive the three-step test.
In 2000, a WTO panel found (full decision; summary) that by introducing the exception into domestic law, the United States had violated Articles 11bis(1)(iii) and 11(1)(ii) of the Berne Convention, thereby also violating Article 9.1 of the TRIPS Agreement. This led to a settlement arrangement between the U.S. and EC that involved the payment of millions of dollars into a rights-holders’ fund. As a further consequence of this dispute, the U.S. is also required to file ongoing status reports with the WTO respecting its efforts to remedy the violation of international trade law.
Of particular interest to Canada is the Panel’s finding that because the United States’ “business exemption” covered a major category of establishment critical to rightsholders, it could not meet any of the stages of the three-step test set out in TRIPS and Berne. Like the proposed reform in Canada, the “business exemption” cut out too broad a swath of the rightsholders’ market, and did not provide for a compulsory licence in the targeted area.
Canada must therefore ask whether its use of the term “education” as a fair dealing purpose risks violating the three-step test as well by reason of a similar broadness. Even with a backstop “fairness” test, there are a number of reasons that Canada’s reform could meet the same fate as the U.S. “business exemption”.
First, taken literally, the contention that “fair dealing” is not “free dealing” is not true. Any fair dealing right permits many free dealings, as it grants an absolute defence to an exclusive right, thus allowing users to make a copy or copies of a work without payment to the owner of the copyright. The result of any fair dealing is – at least potentially – a lost royalty. Any government creating such a user right places itself in the Berne-TRIPS “three-step test” red zone: it has carved out an exception to the core exclusive rights, transferring wealth from the creator of a work to the users of a work. If the scope of “education” is interpreted broadly to include any fair dealing involving learning, the amount of the intended transfer of wealth could be profound.
Second, the Supreme Court’s expression of the CCH “fairness” factors is a unique amalgamation of Commonwealth and U.S. law that does not appear to have been constructed with an eye on Canada’s international treaty obligations. For example, in the United States, the Supreme Court in Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U.S. 539 at 566 (1985) stated that “the effect of the use upon the potential market for or value of the copyrighted work … is undoubtedly the single most important element of fair use”. If applied appropriately by U.S. courts, this statement arguably goes a long way to satisfying the second and third stages of the three-step test. By contrast, in CCH, the Supreme Court of Canada stated that the effect on the market “is neither the only factor nor the most important factor that a court must consider in deciding if the dealing is fair”. Accordingly, the other five factors described by Professor Geist, which do not map easily onto the three-step test, would appear to have an equal or even more important role in deciding whether a given dealing is fair. It would thus appear that certain dealings may be deemed permissible in Canadian law even if they have a disruptive effect on the normal exploitation of the work or the legitimate interests of the author. Such a result would risk disharmony between Canada’s domestic standards and its international obligations.
Third, one of the six factors cited in CCH at the second “fairness” stage of the test is the purpose of the dealing. A dealing identified as “fair” at the first stage will attain a presumptive advantage at the second stage. A broad fair dealing exception is arguably double-counted, raising the risk that it will apply beyond the “certain special cases” required by Berne and TRIPS.
Fourth, it must be remembered that the first “threshold” stage of the fair dealing test in Canada plays a critical role in preserving rightsholder interests. Whenever a user of a copyright work cannot plausibly fit the purpose of his or her activities into one of the acceptable purposes of research, criticism, review, or news reporting, and cannot otherwise take advantage of a targeted exception under the Canadian Act, that user must seek authorization to copy a substantial part of a given work. This provides certainty to both users and rightsholders, and cuts down on “Hail Mary” defences in infringement actions. As recent cases on the meaning of the acceptable purpose of “research” demonstrate, a term that can be interpreted in several ways can give rise to mischief, as a plaintiff will always construe it narrowly and a defendant will always construe it broadly, giving rise to expensive litigation and (potentially) inconsistent decision-making by courts that leave no one certain of where the law stands.
Last, a host of respected neutral observers have concluded that an expansive education exception could not pass the three-step test, particularly if it is a pure exception as opposed to a remunerative licence. In World Copyright Law, 3d ed. (London: Sweet & Maxwell, 2008) at 529, the respected expert J.A.L. Sterling expresses the opinion that:
The reference to ‘certain special cases’ is intended to indicate the general limitations and exceptions to the reproduction right (for instance, a limitation which provides that ‘reproduction of any work may take place for any purpose connected with education’) would not be permissible.
The esteemed Australian expert Professor Sam Ricketson has expressed the same essential conclusions. In a study conducted for WIPO, he concludes that any exception involving libraries, archives or educational institutions needs to be clearly specified, with defined limits. In these circumstances, he concludes that a paid statutory licence for such works might pass the three-step test (see p. 76). Similarly, in a text co-authored with the well-known American expert Professor Jane Ginsberg, International Copyright & Neighbouring Rights: The Berne Convention and Beyond, 2d ed. (Oxford: Oxford University Press, 2006), Professor Ricketson concludes at paragraph 13.36 (pages 782-783) that an exception for teaching purposes could not pass the three-step test, and that a broad student-based “copying for instructional purposes” might be justifiable under Berne-TRIPS only where it was made subject to appropriate limitations, and subject to compensation for authors.
Given these conclusions, the broad full defence envisioned by Bill C-32 would appear to violate Canada’s international treaty commitments.
7) Are the Reforms Truly a “Middle Ground “?
In his blog entry, Professor Geist suggests that the new fair dealing provisions in Bill C-32 are a “middle ground” because they add two categories requested by creator groups and one requested by user groups.
With respect, this seems to be a disingenuous argument. All fair dealing rights serve as exceptions to an exclusive right. They benefit persons who use a pre-existing work. In the case of parody and satire, a fair dealing right will allow one creator to use the work of a predecessor creator to his or her benefit without paying that predecessor. It is a special kind of user right focused on a class of creators.
But the fatal error is to treat these three rights as indistinguishable units with indistinguishable impacts. The parody and satire rights are transformative rights, in the sense that they do not permit free consumption of a work, but rather the use of an old work in creating a new work. As such, the new works will not necessarily sap the market for the original works, but may in some cases lead to works that revive or enhance interest in the originals, leading to royalties that would not otherwise have been reaped. These dealings are, arguably, a win-win for creators and users.
The educational fair dealing right is, by contrast, a consumptive right. It does not result in the creation of new works, but results in royalty-free uses that transfer wealth directly from rightsholders to users. The impact goes in only one direction, and that direction may be profound indeed, judging from the $40 million to $60 million estimates raised in testimony at Committee.
Accordingly, when the true impacts of these exceptions are measured, it would appear that the Bill as drafted would result in a considerable loss to creators. Considered critically, Professor Geist’s account of an unchanged future is inconsistent with the story set out in his own recent blog entry, where he suggests that there is a “growing sentiment on Canadian campuses that it is time for post-secondary education to decline the coverage by walking away from Access Copyright” and its proposed post-secondary tariff.
How can this be done? Professor Geist suggests that institutions, professors and students can decline coverage “by relying on site database licences that are already in place, open access licences to scholarly research, and fair dealing.” [Emphasis added.] Even prior to the passage of a “fair dealing for education” exception, Professor Geist has catalogued (here, here and here) a series of schools that are acting on this suggestion. Clearly, if fair dealing is expanded beyond its current bounds to all educational activities that are “fair” in nature, it would dramatically increase the ability of institutions to take a tough stance on liability, including by opting out of such tariffs and challenging rightsholders to sue for failure to pay. Forced to rely on the unfairness stage of the test alone, publishers will face an extreme uphill battle.
8) Do the Reforms Support or Suppress New Technologies?
In his blog entry, Professor Geist argues that the addition of education as an acceptable purpose “will open the door to the use of new technologies in the classroom without fear of potential liability”. He cites the example of a colleague who uses a photograph from court documents as part of his classroom discussion as an example of how the fair dealing right would enhance the educational experience.
While Professor Geist’s example seems eminently reasonable, Professor Geist omits any consideration of how the proposed fair dealing right would affect publishers’ multi-million dollar investments to support classroom technologies that are meant to transform the educational experience into a collaborative, multi-media endeavour. For example, publishers have invested significantly in CourseSmart, a North America-wide electronic repository of thousands of the most commonly used textbooks from a variety of higher education publishers. Instructors can use these resources to find and review the information they need to choose their textbooks and course materials, and students can purchase online access to these textbooks at a considerable discount from the print editions. Other promising new environments are being developed via partnerships between publishers and third-party course management operators such as SymText, eConcordia, and WebCT/Blackboard. These systems involve an online proprietary virtual learning environment system that is sold to educational institutions and used in e-learning. Content for these environments is provided via licence agreements with publishers.
Publishers are also making significant investments in creating original digital content, either as an online-access supplement to existing print textbooks, or as standalone material specifically dedicated to the educational sectors. These investments require a considerable reworking of traditional publishing operations, including the establishment of specialized digital departments with dedicated employees.
Parliament should consider very carefully whether the fair dealing right would undercut these developing technologies and services by offering a free alternative, thus sending a perverse signal to publishers and technology innovators not to invest in these new delivery systems. The result would be exactly contrary to the government’s stated goal of “enriching the educational experience and facilitating use of the latest technologies”.
9) Did the UK Gowers Review Support These Kinds of Reforms?
As part of Professor Geist’s middle ground argument, he suggests by way of a reference to the well-respected Gowers Review of Intellectual Property that fair uses of copyright are good for the economy and will not damage the interests of copyright owners. He quotes the fair dealing section of the Gowers Review as stating that “fair uses of copyright can create economic value without damaging the interest of copyright owners”.
A close look at the Gowers Review shows that it was not commenting on any fair dealing right that remotely resembled the education right proposed in Bill C-32, but rather on transformative parody/satire rights that would enable works to be created that are not “not necessarily substitutes for the original work”. This passage did not comment on any fair dealing right that remotely resembled the education right proposed in Bill C-32, and cannot be taken as approving a far-sweeping consumptive fair dealing right. Indeed, the Gowers Report specifically recommended that educational issues be dealt with by the way of incremental amendments to the existing targeted educational exceptions in the Copyright, Designs and Patents Act 1988. While the UK government has continued to monitor the U.S. fair use system, no government report has recommended that it be adopted.
10) How Can We Best Accommodate Both Rightsholder and User Interests?
While fights over ambiguities in fair dealing disputes provide for fascinating case law, it must be asked whether they are healthy for the copyright system. If the government’s wish is to boost new technologies in the educational setting, a far less risky means of supplementing the current fair dealing rights would be to create targeted exceptions meant to achieve specific goals. This is the way Canada has done it in the past with its exceptions for educational institutions, libraries, archives and museums. This is also the way it is done in the European Union, with whom Canada is in significant trade talks, and in the legislation of major Commonwealth countries such as Australia (see Parts VA, VB), New Zealand (see ss. 44-49), and the United Kingdom (see ss. 32-36).
In framing such exceptions, Parliament should recognize that laws are used by people primarily, and lawyers secondarily. Whatever the government means specifically by the currently proposed phrase “for the purpose of education”, it must recognize that any new legislative pronouncement will find its way into guidelines, paraphrases, and urban legend, some of which will be in the neighbourhood of a correct interpretation, and some of which will not. Both the correct and the incorrect interpretations will result in the making of copies by individuals and institutions. The broader the key terms, the greater the chance that there will be copies made that Parliament never intended to permit. This is precisely why New Zealand drafted its recent amendment to be specific about the educational activities it was blessing, and the conditions required to obtain the benefits of the exception, but neutral about the exact delivery mechanisms used to perform such activities. Such an approach is far more likely to satisfy the three-step test than the current broad-brush approach in Bill C-32.
To achieve the stated purposes of the government through this more refined legislative mechanism, Canada would have to commit to freshen up these targeted exceptions when they grow less relevant. One way to ensure a continued up-to-date approach to these kinds of exceptions might be to push the exact mechanics of the target exceptions into regulation, as Canada has done with its management of the record-keeping process for its existing education, libraries, archives and museums exceptions, as well as its management of the retransmission regime. This would allay the concern of many rightsholders and many users about the slow pace of copyright reform in Canada, including in respect of the existing educational exceptions in the Copyright Act.
*A copy of this article is posted in PDF format here.
*Dan Glover is an associate in the Intellectual Property Group at McCarthy Tétrault LLP. This comment is written in Mr. Glover’s personal capacity.