Access Copyright v. York University (2017): in defence of fair dealing

This is a guest post from Lisa Di Valentino originally published on her blog Fair Dealing in Education. Lisa, MLIS, JD, PhD, is Law & Public Policy Librarian at University of Massachusetts Amherst. She has a PhD from the University of Western Ontario entitled Laying the Foundation for Copyright Policy and Practice in Canadian Universities. She reports on the recent case in Canada concerning what constitutes fair dealing, involving York University and Access Copyright. We should make it clear that the situation in Canada is different from that in the UK with a different jurisdiction (albeit based on the same Commonwealth model) and a different reprographic rights organisation. However we think the parallels are interesting to consider both for those in the UK and elsewhere and we’re really appreciative of Lisa allowing us to share her analysis of this important ruling.

On July 12, 2017, the Federal Court of Canada handed down its long-awaited decision in Access Copyright’s lawsuit against York University (York), originally filed on April 13, 2013. The suit related to York’s copyright policy and whether their fair dealing guidelines accurately reflected the test set out by the Supreme Court.

Fair dealing is an “exception” to copyright infringement, where small parts of a copyrighted work can be copied for certain purposes (such as research or education), without having to seek permission or pay the copyright owner. An example of this in Canada would be copying a law journal article for legal research, or copying a few pages from a textbook for students to study from. Not every instance of copying for for research or education is fair dealing. In addition to being for an allowed purpose, the copying must also be “fair”.

In a 2006 judicial decision, CCH Canadian Ltd. v. Law Society of Upper Canada (CCH), the Supreme Court of Canada had set out a test to determine whether a reproduction can be considered “fair”, and therefore free of payment or permission. They also referred to fair dealing as a “user’s right” and an integral part of the copyright scheme, rather than merely a defence or excuse for copying.

Access Copyright (AC) is a not-for-profit organisation that represents copyright owners of textual works. AC can enter into agreements with other institutions on copyright owners’ behalf, such as “blanket licences” that allow the institution to make certain types of copies from AC’s repertoire for a yearly payment. In 2010, AC attempted to increase the yearly cost of the blanket licence, to take into account the increasing use of digital versions of texts, for example in a learning management system. Many universities objected to this increase and decided to rely on fair dealing and other “exceptions” to copyright infringement, in order to make these types of copies without payment. AC applied to the Copyright Board for a tariff, which is similar to a licence, but would apply generally and not only to those who enter into an agreement with AC. While the Copyright Board were considering AC’s application, they certified an Interim Tariff which has been in effect since.

York, and other universities, interpreted this to mean that whatever copies they made under the fair dealing right would not be subject to the Interim Tariff, so they set about instituting policies and guidelines for faculty who make copies of works for teaching purposes. York’s policy (reproduced in the court’s decision) referred to anything up to 10% of a work, or one chapter of a book, or one academic article from a journal issue (among other examples), to be fair dealing. These limits cover a great deal of what is usually used in university courses. Often these copies will be bound together in a “coursepack”, a print or digital book that is used alongside, or in place of, the traditional textbook. The blanket licence with AC that York previously used had almost identical limits.

AC brought suit against York University in Federal Court, claiming that the university’s copying policy was not representative of fair dealing as set out by the Supreme Court, and so York should have to pay all the retroactive royalties under the Interim Tariff. York countersued, seeking a declaration from the court that their policy did accurately represent fair dealing, and subsequently all the copies made under it were royalty-free.

The result was a complete loss for York on all points, and an order to pay retroactive royalties as set by the Copyright Board in its Interim Tariff. However, the trial judge’s interpretation of fair dealing is, in large part, inconsistent with the Supreme Court’s guidance that was given in the CCH case.

As the decision is likely to be appealed (and it should be, not least because it will have an effect on the dozens of other universities that use a same or similar policy), I’d like to discuss here what will we be (or should be) the main points of legal argument before the higher court. I will focus here on the fair dealing analysis (starting at para. 249) rather than the Interim Tariff, which is an issue discussed in detail by Howard Knopf and Ariel Katz.

[Michael Geist’s analysis of the York decision can be found here.]

The Supreme Court has been quite clear on several occasions that fair dealing is a right of users, necessary to maintain the balance inherent in the objectives of copyright law. They first used this language in the seminal CCH v. LSUC (CCH) case in 2004, and repeated it in a set of of copyright cases in 2012, in particular SOCAN v. Bell (Bell) and Alberta v. Access Copyright (Alberta).

Unfortunately, the Federal Court’s decision in Access Copyright v. York University does not reflect this important aspect of copyright in his analysis of the various factors of the fair dealing test.

Purpose of the dealing

Phelan J. (the judge in this case) distinguishes York from CCH in that the copying done in CCH was for others (the library made copies as requested by lawyers off-site), while the copying at York was done “to serve York’s interests and the interests of its faculty and students.” (para. 260) The significance of this distinction is not clear to me. Phelan J. may be suggesting that York’s fair dealing policy and practices are not reliable because they are focused on their own benefit rather than that of the public, or the copyright owner, or some other third party.

He subsequently claims that “It is evident that York created the Guidelines and operated under them primarily to obtain for free that which they had previously paid for.” (para. 272) However, one would think that the majority of fair dealing is done for the user’s interest in that they are copying things for their own use, and with the expectation that no payment is required. I don’t know why that would make a difference in the final analysis; there needn’t be any altruistic motivation to take advantage of one’s own rights.

Character of the dealing

Due to the “unreliability” of York’s evidence (data and expert testimony) as to the actual extent of copying and access to copies, Phelan J. determined that the character of the dealing “tends toward the unfairness end of the spectrum.” (para. 289)

Amount of the dealing

This, according to the decision, is “particularly important in this case” and “problematic” for York (para. 294). Unfortunately Phelan J. misunderstands what “amount” is referring to. In Bell, the Supreme Court explicitly cautioned that the amount to be looked at is the proportion of the work used compared to the whole, not the total amount of copies made: “Since fair dealing is a ‘user’s’ right, the ‘amount of the dealing’ factor should be assessed based on the individual use, not the amount of the dealing in the aggregate. The appropriate measure under this factor is therefore, as the Board noted, the proportion of the excerpt used in relation to the whole work.” (para. 41). However, Phelan J. asserts that “It is relevant to consider the aggregate volume of copying by by all post-secondary institutions that would be allowed if the Guidelines or similar policies were adopted.” (York, para. 301) (I would like to suggest that this factor be renamed to “Proportion of the work” or “Percentage of the work” to avoid further confusion.)

A subsequent paragraph reads: “As became apparent during the course of the trial and as is clear from the terms of the Guidelines, the permitted copying can, in fact, be 100% or such a large part of a work as to appropriate the whole (e.g. for a journal article in a periodical, a short story in an anthology, or a chapter in an edited book).” (para. 310) Phelan J. seems to be suggesting that York (or the Association of Universities and Colleges of Canada, who drafted the model policy) pulled this proportion out of thin air; however, the Supreme Court have said essentially the same thing: “It may be possible to deal fairly with a whole work…. The amount taken may also be more or less fair depending on the purpose. For example, for the purpose of research or private study, it may be essential to copy an entire academic article or an entire judicial decision.” (CCH, para. 56)

As for the qualitative aspect, or the importance of the portion compared to the whole (i.e. whether the portion copied represents the “core” of the work), Phelan J. engages in circular reasoning when he asserts that “Where a chapter from a book can stand alone and be important enough to be taken from the whole for inclusion in a course’s required reading, there is little doubt that the copied part is qualitatively significant to the work and to the author’s contribution.” (York, para. 317) This suggests that any dealing is automatically unfair to some degree because the portion copied is important enough that the copier feels the need to copy it.

Alternatives to the dealing

The Supreme Court in CCH noted that there may be alternatives such that the copying is unnecessary. They said “If there is a non-copyrighted equivalent of the work that could have been used instead of the copyrighted work, this should be considered by the court.” (para. 57) (While the court did not mention publicly-licensed and open access, these would be suitable alternatives as well.)

In Alberta the Supreme Court rejected the idea that a textbook for each student should be purchased if only a small portion is necessary to achieve the ultimate purpose (para. 31-32).

Importantly, whether the copy could have been paid for via licence (such as a blanket licence offered by AC) is explicitly not a consideration: “The availability of a licence is not relevant to deciding whether a dealing has been fair.” (CCH, para. 70). However, while Phelan J. acknowledges this in York, he goes on to suggest alternatives that are, in essence, the same as purchasing a licence: using custom book services (which themselves pay licence fees which are passed on to the customer), or purchasing individual articles from the publisher (para. 330). In CCH, the court pointed out that fair dealing is an integral part of copyright (para. 70). Obviously, paying for the copy is an “alternative” to not paying for the copy, but if it were considered an alternative to the dealing for the purposes of this factor, it would always be unfair. Fair dealing is not an infringement of copyright; it is the right to copy (parts of) a work without paying for it at all.

Despite this, Phelan J. does acknowledge that in the end, this factor slightly favours York (para. 331).

Nature of the work

Phelan J. points out that there is “significant work, research, skill, and expense” involved in bringing materials to publication, whether they are poems, journal articles, or textbooks, and that “Most of these people [professional writers] are attempting to make a living from writing and publishing.” (para. 336). This is true. However, it is not relevant to the fair dealing analysis. In CCH the Supreme Court ruled that a non-trivial “exercise of skill and judgment” is necessary for a work to be protected under copyright (para. 16) – so, for example, a collection of names in a phone book, sorted alphabetically, would not attract copyright protection, as it only involves a trivial amount of judgment. However, nowhere in CCH or the Supreme Court’s related decisions is fair dealing predicated on the amount of work or creativity or skill or expense required to publish a copyrighted work; even the magnum opus of the greatest artist in history is subject to the limitations of user rights.

By “nature of the work”, the Supreme Court are referring to whether a work is published or confidential, and whether its reproduction could lead to further dissemination, to the benefit of the public. Admittedly there is not much guidance for this factor, as Alberta and Bell do not really address it. But to interpret it to mean that “the harder an author or publisher has worked on something, and the more money that has been spent, the less fair a dealing might be” is to completely ignore the objective of copyright law, which is not only to encourage creation of works, but also their dissemination and incorporation into future works.

Effect of the dealing on the work

As Phelan J. points out, it is up to Access Copyright to provide evidence that they suffered a decline in sales that is linked to the copying taking place at York. However, he goes on to say that York’s copying does not need to be the only or dominant reason for the decline. He provides no precedent or reasoning for this other than that the Supreme Court haven’t said otherwise (para. 342). The evidence (which we do not have access to except for a summary), shows “an acceleration of the decline in the sale of works produced for the post-secondary educational market” since the introduction of the fair dealing policy (para. 347).

However, there is a chicken-and-egg element to this case. If universities have previously been contributing to sales because they were paying for things they did not have to pay for, then naturally sales will drop when they stop paying. This was even pointed out by Access Copyright — the same type of copying had been going on for many years prior to the introduction of the fair dealing policy, except it was under the terms of a blanket licence (para. 350). These terms were substantially similar to the types of copying that could be considered fair dealing after CCH and later copyright cases. Hence, universities declined to renew the licences and instituted fair dealing policies. This is not necessarily evidence that the dealing is unfair, it could equally prove that universities were simply not taking advantage of their rights in the past.

Conclusion

The sense that I get from Phelan J.’s decision is that he believes York (and by extension, other universities using the same fair dealing policy) have suddenly taken advantage of a proffered gift (fair dealing) and are running amok with it, to the ruination of publishers and authors. I believe that universities have been entitled to fair dealing all along, and that they are finally realizing that they are paying for uses that they do not have to pay for. Naturally, if a business model involves getting people to pay for what they could (are supposed to) have for free, there will be a significant drop in income once they catch on.

York’s fair dealing policy is not perfect. I have argued elsewhere (chapter 7) that these policies should provide more guidance and avoid “bright line rule-making” that obscures the flexibility of the fair dealing analysis. Neither York nor the Association of Universities and Colleges of Canada could justify the 10% threshold in this case. It’s probably too much in some cases; it could be too little in other cases.

Furthermore, the absence of safeguards appears to be a big factor in Phelan J.’s conclusion that York’s policy did not translate to actual fair dealing in practice (e.g. para. 266). It’s important for institutions to have someone that can provide guidance in interpreting the policy and make decisions in “iffy” situations. In CCH this role was taken by the reference librarian, as per the Great Library’s Access to the Law Policy, and this was noted by the court in York (“copying at a single location under the supervision and control of research librarians”, “a policy strictly applied and enforced by librarians”) (para. 262).

University librarians and copyright officers can also help faculty find alternatives to relying on fair dealing such as works that the university already pays for via publisher or database licences, public domain materials, and publicly-licensed or “open access” materials (e.g. Creative Commons). Importantly, librarians can and should encourage faculty and researchers to publish in open access journals, and contribute to open access educational materials, to further expand the pool of readily-available works.

Another claim made by AC in the lawsuit was that particular named faculty members were making excessive copies far outside what could be considered fair dealing (even by York’s own policy), perhaps under the erroneous belief that any copying done in an educational context is not infringement. This is a not uncommon belief, which underscores the need for copyright education among faculty (both research and teaching). Whatever form this education or training takes, at the very least it will make faculty aware that one the one hand there are limits to copying, even for teaching purposes, and on the other hand, they may be entitled to copy more than they think. From a risk-management perspective, this can only help to reduce the likelihood of a lawsuit, and serve as an element of a defence if one arises.