SUPREME COURT OF CANADA
12 JULY 2012
(delivered the judgment of the court)
The purchase of musical works is increasingly carried out over the Internet. Some commercial Internet sites that sell music allow consumers to preview musical works before making a purchase. The issue in this case is whether those previews constitute “fair dealing” under s. 29 of the Copyright Act, R.S.C. 1985, c. C‑42.
The Society of Composers, Authors and Music Publishers of Canada (SOCAN) represents composers, authors and music publishers and administers their performing and communication rights. Its arguments before this Court were supported by the Canadian Recording Industry Association and CMRRA-SODRAC Inc.
Bell Canada, Apple Canada Inc., Rogers Communications Inc., Rogers Wireless Partnership, Shaw Cablesystems G.P. and TELUS Communications Inc. operate online music services that sell downloads of digital files of musical works. These services provide catalogues of digital audio files that allow users to identify musical works by title, album, genre and artist. The service providers also give consumers the ability to listen to free “previews” of those works before deciding which work to purchase. SOCAN seeks compensation for the provision of those previews over and beyond what would normally be paid for a music download or a CD.
A preview consists of an extract taken from the work, usually 30 to 90 seconds of a musical track. Users are able to select and listen to the preview through an online “stream”, meaning that a temporary copy of the excerpt is heard by the user in such a way that his or her computer does not store a permanent copy of the preview. The previews help users decide whether to purchase a permanent download of the work. Apple’s iTunes service, for example, allows consumers to listen to previews as many times as they want, whether or not they make a purchase or have registered with the online service provider.
As noted in the companion appeal (Rogers Communications Inc. v Society of Composers, Authors and Music Publishers of Canada,  2 S.C.R. 283), SOCAN filed proposed tariffs targeting the years 1996 to 2006 with the Copyright Board of Canada in 1995 for the determination of royalties to be paid by users when musical works are communicated to the public over the Internet.
In a decision released on October 18, 2007 (61 C.P.R. (4th) 353), the Board agreed that SOCAN was entitled to collect royalties for the downloading of musical works, but not for previews. In the Board’s view, the use of previews was not an infringement of copyright since their use was “fair dealing” for the purpose of research under s. 29 of the Copyright Act based on the factors identified by McLachlin C.J. in CCH Canadian Ltd. v Law Society of Upper Canada,  1 S.C.R. 339. Because the previews were not an infringement of copyright, no royalties were required to be paid to SOCAN for their use.
The Federal Court of Appeal upheld the Board’s decision (2010 FCA 123, 403 N.R. 57). As would I.
In Théberge v Galerie d’Art du Petit Champlain inc.,  2 S.C.R. 336, this Court noted that copyright requires “a balance between promoting the public interest in the encouragement and dissemination of works of the arts and intellect and obtaining a just reward for the creator” (para. 30).
Théberge reflected a move away from an earlier, author-centric view which focused on the exclusive right of authors and copyright owners to control how their works were used in the marketplace: see e.g. Bishop v Stevens,  2 S.C.R. 467, at pp. 478-79. Under this former framework, any benefit the public might derive from the copyright system was only “a fortunate by-product of private entitlement”: Carys J. Craig, “Locke, Labour and Limiting the Author’s Right: A Warning against a Lockean Approach to Copyright Law” (2002), 28 Queen’s L.J. 1, at pp. 14-15.
Théberge focused attention instead on the importance copyright plays in promoting the public interest, and emphasized that the dissemination of artistic works is central to developing a robustly cultured and intellectual public domain. As noted by Professor David Vaver, both protection and access must be sensitively balanced in order to achieve this goal: Intellectual Property Law: Copyright, Patents, Trade-marks (2nd ed. 2011), at p. 60.
CCH confirmed that users’ rights are an essential part of furthering the public interest objectives of the Copyright Act. One of the tools employed to achieve the proper balance between protection and access in the Act is the concept of fair dealing, which allows users to engage in some activities that might otherwise amount to copyright infringement. In order to maintain the proper balance between these interests, the fair dealing provision “must not be interpreted restrictively”: CCH, at para. 48.
CCH sets out the test for fair dealing under s. 29 of the Copyright Act:
While not specifically in issue in the case before us, ss. 29.1 and 29.2 of the Act also permit “fair dealing” for the purposes of criticism, review or news reporting.
The test for fair dealing articulated in CCH involves two steps. The first is to determine whether the dealing is for the purpose of either “research” or “private study”, the two allowable purposes listed under s. 29. The second step assesses whether the dealing is “fair”. The onus is on the person invoking “fair dealing” to satisfy both aspects of the test under CCH.
To assist in determining whether the dealing is “fair”, this Court set out the following six fairness factors for guidance: the purpose, character, and amount of the dealing; the existence of any alternatives to the dealing; the nature of the work; and the effect of the dealing on the work.
The first inquiry in this case, therefore, is whether previews are provided for the allowable purpose of “research” under the first step of the CCH fair dealing test. While CCH did not define the word “research”, it notably concluded that “‘[r]esearch’ must be given a large and liberal interpretation in order to ensure that users’ rights are not unduly constrained” (para. 51).
The Board defined “previews” as [para. 18]
a marketing tool offered by online music services, among others. A preview is an excerpt (usually 30 seconds or less) of a sound recording that can be streamed so that consumers are allowed to “preview” the recording to help them decide whether to purchase a (usually permanent) download.
Based on the evidence it heard about the purpose of previews and the way they were used by consumers, the Board concluded that previews were used “either to determine whether the track suits the user’s tastes or to verify that the track is the one the user wants to buy” (para. 101). Listening to previews helped users identify what music to purchase. Since planning the purchase of a download involved “searching [and] investigation”, the Board was of the view that previews amounted to “research” under s. 29 of the Copyright Act (para. 109).
The Federal Court of Appeal endorsed the Board’s view that listening to previews was part of planning the purchase of a download of a musical work and was therefore “for the purpose of research”, concluding [para. 20]:
.... it would not be unreasonable to give the word “research” its primary and ordinary meaning. The consumer is searching for an object of copyright that he or she desires and is attempting to locate and wishes to ensure its authenticity and quality before obtaining it .... “[L]istening to previews assists in this investigation”.
SOCAN argued that the Board and the Federal Court of Appeal misinterpreted the term “research” in two ways. It argued first that their interpretation of “research” was overly broad. Its second argument was that the purpose of “research” should have been analysed from the perspective of the online service provider and not the consumer. From this perspective, the purpose of the previews was not “research”, but to sell permanent downloads of the musical works.
SOCAN offers the definition of “research” as being “the systematic investigation into and study of materials and sources in order to establish facts and reach new conclusions” (A.F., at para. 96). Moreover, SOCAN argues, the goal of the “research” must be for the purpose of making creative works, since only uses that contribute to the creative process are in the public interest. As a result, previews do not amount to “research” since their primary purpose is not to foster creativity, but to enable users to purchase music online.
It is true that an important goal of fair dealing is to allow users to employ copyrighted works in a way that helps them engage in their own acts of authorship and creativity: Abraham Drassinower, “Taking User Rights Seriously”, in Michael Geist, ed., In the Public Interest: The Future of Canadian Copyright Law (2005), 462, at pp. 467-72. But that does not argue for permitting only creative purposes to qualify as “research” under s. 29 of the Copyright Act. To do so would ignore the fact that the dissemination of works is also one of the Act’s purposes, which means that dissemination too, with or without creativity, is in the public interest. It would also ignore that “private study”, a concept that has no intrinsic relationship with creativity, was also expressly included as an allowable purpose in s. 29. Since “research” and “private study” both qualify as fair dealing purposes under s. 29, we should not interpret the term “research” more restrictively than “private study”.
Limiting research to creative purposes would also run counter to the ordinary meaning of “research”, which can include many activities that do not demand the establishment of new facts or conclusions. It can be piecemeal, informal, exploratory, or confirmatory. It can in fact be undertaken for no purpose except personal interest. It is true that research can be for the purpose of reaching new conclusions, but this should be seen as only one, not the primary component of the definitional framework.
In urging the Court to narrow the definition of “research” as requiring the creation of something new, SOCAN relied on American jurisprudence which looks to the requirement of a “transformative” purpose before the use is seen as fair. It cited as an example United States v American Society of Composers, Authors and Publishers, 599 F.Supp.2d 415 (2009), where the New York District Court held that the use of music previews as a marketing tool to sell musical ringtones was not “transformative” in nature and therefore could not be fairly described as “criticism, comment, news reporting .... or research” under the fair use provisions in Title 17, § 107 of the U.S. Code, at pp. 424-25.
The American approach is called “fair use”. The U.S. Code provisions create an open set of purposes for fair use which include criticism, comment, news reporting, teaching, scholarship, or research: 17 U.S.C. § 107. The analysis proceeds straight to the assessment of fairness, an assessment based on factors enumerated in the Code or established by the case law. Although one of those fairness factors includes whether the use is transformative, it is not at all clear that a transformative use is “absolutely necessary” for a finding of fair use: Campbell v Acuff-Rose Music, Inc., 510 U.S. 569 (1994), at p. 579.
But even if it were a requirement under American law, this Court has previously cautioned against the automatic portability of American copyright concepts into the Canadian arena, given the “fundamental differences” in the respective legislative schemes: Compo Co. v Blue Crest Music Inc.,  1 S.C.R. 357, at p. 367. This caution has resonance in the fair dealing context.
Unlike the American approach of proceeding straight to the fairness assessment, we do not engage in the fairness analysis in Canada until we are satisfied that the dealing is for one of the allowable purposes enumerated in the Copyright Act. Under the test set out in CCH, “fairness” is not considered until the second step of the test for fair dealing: see CCH, at para. 51; see also Giuseppina D’Agostino, “Healing Fair Dealing? A Comparative Copyright Analysis of Canada’s Fair Dealing to U.K. Fair Dealing and U.S. Fair Use” (2008), 53 McGill L.J. 309, and Century 21 Canada Limited Partnership v Rogers Communications Inc., 2011 BCSC 1196, 338 D.L.R. (4th) 32, at para. 234.
In mandating a generous interpretation of the fair dealing purposes, including “research”, the Court in CCH created a relatively low threshold for the first step so that the analytical heavy-hitting is done in determining whether the dealing was fair. SOCAN’s submission that “research” be restricted to the creation of new works would conflate the allowable purpose with the fairness analysis and unduly raise the bar for entering that analysis. Moreover, its restricted definitional scope of “research” contradicts not only the Court’s admonition in CCH that “[i]n order to maintain the proper balance between the rights of a copyright owner and users’ interests, [the fair dealing exception] must not be interpreted restrictively” (para. 48), but also its direction that the term “research” be given a “large and liberal interpretation” so that in maintaining that balance, users’ rights are not unduly constrained (paras. 48 and 51).
SOCAN’s proposed definition of “research” as requiring “systematic investigation” and “new conclusions” is also at odds with its second submission about “research”, namely, that “research” be analysed from the perspective of the purpose of the online service providers, and not that of the users. But its own proposed definition shows that it sees research as a user-focused undertaking, since the investigation and creation of new conclusions are clearly done by a user, not a provider. The provider’s purpose in making the works available is therefore not the relevant perspective at the first stage of the fair dealing analysis.
This is consistent with the Court’s approach in CCH, where it described fair dealing as a “user’s right” (para. 48). In CCH, the Great Library was the provider, offering a photocopying service to lawyers requesting copies of legal materials. The Court did not focus its inquiry on the library’s perspective, but on that of the ultimate user, the lawyers, whose purpose was legal research (para. 64).
Similarly, in considering whether previews are for the purpose of “research” under the first step of CCH, the Board properly considered them from the perspective of the user or consumer’s purpose. And from that perspective, consumers used the previews for the purpose of conducting research to identify which music to purchase, purchases which trigger dissemination of musical works and compensation for their creators, both of which are outcomes the Act seeks to encourage.
The inquiry then moves to the second step, namely, determining whether the use of previews was “fair” in accordance with the CCH factors.
Whether something is “fair” is a question of fact and depends on the facts of each case: CCH, at para. 52, citing Hubbard v Vosper,  1 All E.R. 1023 (C.A.), at p. 1027. Based on all the factors, the Board concluded, properly in my view, that previews amounted to fair dealing.
The first factor identified in CCH is the purpose of the dealing, where an objective assessment is made of the “real purpose or motive” behind using the copyrighted work (para. 54).
SOCAN argued that the purpose of the previews in this case was purely commercial. This is an approach that looks at the purpose of the previews from the perspective not of the consumer, but of the service providers. I agree instead with the Board and the Federal Court of Appeal that the predominant perspective in this case is that of the ultimate users of the previews, and their purpose in using previews was to help them research and identify musical works for online purchase. While the service providers sell musical downloads, the purpose of providing previews is primarily to facilitate the research purposes of the consumers.
The Board also noted that there were reasonable safeguards in place to ensure that the users’ dealing in previews was in fact being used for this purpose: the previews were streamed, short, and often of lesser quality than the musical work itself. These safeguards prevented the previews from replacing the work while still fulfilling a research function.
SOCAN also argued that even from the perspective of the consumers, the purpose of the previews was purely commercial, since their purpose was the potential purchase of musical works. Ultimately, however, the Board’s approach is consistent with CCH’s observation that while research done for commercial reasons may be less fair than research done for non-commercial purposes (para. 54), the dealing may nonetheless be fair if there are “reasonable safeguards” in place to ensure that the works are actually being used for research (para. 66).
The second factor discussed in CCH is the character of the dealing. The Court stated that a particular dealing might be unfair if multiple copies of works are being widely distributed (para. 55). But as the Court also pointed out, if a single copy of a work is used for a specific legitimate purpose, or if the copy no longer existed after it was used, this would favour a finding of fairness (para. 55).
SOCAN’s argument was based on the fact that consumers accessed, on average, 10 times the number of previews as full-length musical works. However, no copy existed after the preview was heard. The previews were streamed, not downloaded. Users did not get a permanent copy, and once the preview was heard, the file was automatically deleted from the user’s computer. The fact that each file was automatically deleted meant that copies could not be duplicated or further disseminated by users.
The third factor identified in CCH is the amount — or quantity — of the dealing. The Board characterized the “amount” of the dealing in terms of the length of each preview compared to the length of the work, concluding that streaming a preview of about 30 seconds was a modest dealing “when compared to purchasing the whole work [approximately four minutes] for repeated listening” (para. 113).
SOCAN argued, however, that the proportion of the preview in relation to the length of the whole musical work was not the proper measure, and that the Board should have considered instead the aggregate number of previews that are streamed by consumers. Since the evidence showed that each user, on average, listened to 10 previews before purchasing a musical work for download, the overall amount of time spent listening to previews was so large that the dealing was unfair. SOCAN saw this factor as determinative in this case.
There is no doubt that the aggregate quantity of music heard through previews is significant, but SOCAN’s argument conflicts with the Court’s statement in CCH that “amount” means the “quantity of the work taken” (para. 56). 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. That, it seems to me, is consistent with the Court’s approach in CCH, where it considered the Great Library’s dealings by looking at its practices as they related to specific works requested by individual patrons, not at the total number of patrons or pages requested. The “amount of the dealing” factor should therefore be assessed by looking at how each dealing occurs on an individual level, not on the aggregate use.
Moreover, the quantification of the aggregate dissemination is already considered under the “character of the dealing” factor, which examines whether multiple copies of works are being widely distributed. Reconsidering the same “aggregate” quantity under the “amount” factor would deprive that factor of any utility in the analysis, and would erase consideration of the proportion of the excerpt to the entire work.
Further, given the ease and magnitude with which digital works are disseminated over the Internet, focusing on the “aggregate” amount of the dealing in cases involving digital works could well lead to disproportionate findings of unfairness when compared with non-digital works. If, as SOCAN urges, large-scale organized dealings are inherently unfair, most of what online service providers do with musical works would be treated as copyright infringement. This, it seems to me, potentially undermines the goal of technological neutrality, which seeks to have the Copyright Act applied in a way that operates consistently, regardless of the form of media involved, or its technological sophistication: Robertson v Thomson Corp.,  2 S.C.R. 363, at para. 49.
The fourth factor identified in CCH involves considering any alternatives to the dealing. A dealing may be less fair if there is a non-copyrighted equivalent of the work that could have been used, or if the dealing was not reasonably necessary to achieve the ultimate purpose (para. 57).
SOCAN argued that there were other methods available, like advertising, to help users identify potential music for purchase. Many of the service providers, for example, offered album artwork, textual descriptions, and user-generated album reviews. In addition, some service providers offered return policies in the event that users accidentally downloaded the wrong musical work.
But allowing returns is an expensive, technologically complicated, and market-inhibiting alternative for helping consumers identify the right music. And none of the other suggested alternatives can demonstrate to a consumer what previews can, namely, what a musical work sounds like. The Board found that “[l]istening to a preview probably is the most practical, most economical and safest way for users to ensure that they purchase what they wish” (para. 114). As a result, it concluded that short, low-quality streamed previews are reasonably necessary to help consumers research what to purchase. I agree.
The fifth factor is the nature of the work, which examines whether the work is one which should be widely disseminated. SOCAN does not dispute the desirability of the sale and dissemination of musical works, but argues that since these works are easily purchased and disseminated without the use of previews, previews are of no additional benefit to promoting further dissemination. But the fact that a musical work is widely available does not necessarily correlate to whether it is widely disseminated. Unless a potential consumer can locate and identify a work he or she wants to buy, the work will not be disseminated.
This observation is linked to the final factor: the effect of the dealing on the work and whether the dealing adversely affects or competes with the work. Because of their short duration and degraded quality, it can hardly be said that previews are in competition with downloads of the work itself. And since the effect of previews is to increase the sale and therefore the dissemination of copyrighted musical works thereby generating remuneration to their creators, it cannot be said that they have a negative impact on the work.
All of this confirms the Board’s conclusion that previews satisfy the requirements of fair dealing and that the online service providers do not infringe copyright. In so concluding, the Board properly balanced the purposes of the Act by encouraging the creation and dissemination of works while at the same time ensuring that creators are fairly rewarded. Its approach was consistent with CCH, with the interpretative principles it set out, and with its test for fair dealing under s. 29 of the Copyright Act. The conclusion, as a result, should not be disturbed.
I would dismiss the appeal with costs.
Henry Brown, Q.C., Gilles M. Daigle, Paul Spurgeon and Matthew S. Estabrooks (m/s Gowling Lafleur Henderson, Ottawa), for the appellant the Society of Composers, Authors and Music Publishers of Canada.
Glen A. Bloom (m/s Osler, Hoskin & Harcourt, Ottawa), for the appellant the Canadian Recording Industry Association.
Casey M. Chisick, Timothy Pinos and Jason Beitchman (m/s Cassels Brock & Blackwell, Toronto), for the appellant CMRRA‑SODRAC Inc.
Gerald L. Kerr‑Wilson, Ariel A. Thomas and Julia Kennedy (m/s Fasken Martineau DuMoulin, Ottawa), for the respondents Bell Canada, Rogers Communications Inc., Rogers Wireless Partnership, Shaw Cablesystems G.P. and TELUS Communications Inc.
Michael Koch (m/s Goodmans, Toronto), for the respondent Apple Canada Inc.
No one appeared for the respondents the Entertainment Software Association and the Entertainment Software Association of Canada.
all rights reserved