In my post about the Author Guild’s lawsuit against HathiTrust and some of its university partners, I promise to consider further the question of why the complaint was so focused on orphan works. Why the plaintiffs put so much emphasis there, since presumably they cannot show that they or the members they represent actually hold any copyrights in those specific works? In a story for Inside Higher ed., a spokesperson for the Authors Guild maintains that the suit really isn’t about orphan works; nevertheless, I continue to think that that is exactly what it is about. Jennifer Howard from the Chronicle of Higher Education gets it right, I think, when she writes that “The plaintiffs seek to take control of those files out of the hands of libraries until Congress establishes guidelines for the use of digital libraries and orphan works.” The lawsuit seems most understandable, I maintain, as part of a strategy to dictate the terms by which the orphan works problem is “solved,” and we need to look closely at what that strategy might involve.
In his excellent post about the lawsuit, Professor James Grimmelmann makes two comments about the Authors Guild’s “grand strategy” with which I respectfully disagree. He says that the AG has switched positions and now opposes action on orphan works, and he suggests that the effect of the lawsuit will be to discourage Congress from getting into the orphan works arena at all. On the contrary, I believe that this is part of a campaign to get Congress to address orphan works by setting up a licensing scheme, similar to the one Canada uses, in which each use of an orphan work must be licensed and a royalty paid. I think the hope is to collect royalties on using orphans that will ultimately be distributed to authors (or publishers?) of similar “categories” of work. With the failure of the Google Books Settlement, I think this is the next attempt by the content industries to sell what they do not own on behalf of parties they cannot identify, and then keep the money.
One of the first bits of evidence I cite for this opinion is this short report about comments sent to the European Union, as it considers its own orphan works directive, from the International Federation of Reproduction Rights Organizations. IFFRO is the trade group for collective rights management groups, including US groups like ASCAP and the Copyright Clearance Center. In their comments, they tell the EU that “Orphan works should be administered through collective management and licensing,” the rules and conditions of which should be left in the hands of those who publish similar “categories” of works. IFRRO does not address where the money would go, but the Canadian model for orphan works gives us a clue; when royalties are paid to collecting societies in Canada, they are normally held for five years (in case a rights holder surfaces) and then become part of the regular funds of the rights organization, presumably for distribution.
What IFRRO is essentially asking the EU for is the same kind of license to exploit orphan works that Google was hoping to get in its book project settlement. That settlement seems to be completely dead, but the prospect of monetizing orphan works put dollar signs in the eyes of the Authors Guild. Now I think the AG is hoping to clear the way for Congress to establish a scheme for collecting money from users of orphan works, probably using the Copyright Clearance Center, and distributing it eventually to their members. They want to make sure that no plausible mode of using orphan works, especially a non-commercial one, gets any traction, so that their lobbyists can go to Congress and say “only you can address this problem, and we have a plan all ready for you.”
It’s really worth looking at the Canadian model, which is described in this excellent article by De Beers and Bouchard on “Canada’s ‘Orphan Works’ Regime: Unlocatable copyright owners and the Copyright Board,” The Canadian law, we should note, authorizes an administrative board, not a collective rights organization, to grant licenses upon application for particular uses. But the royalties it orders are often determined by and paid to the collective rights groups, and I strongly suspect that it is a variation on this broad scheme that the Authors Guild would like to push.
In addition to its contextualized discussion of how the Canadian scheme works, the article’s authors raise quite a few really important and challenging considerations about the specifics of any orphan works legislation. For example, how do orphan works schema address the special problems posed by “moral rights,” which are often held by a party different than the copyright (economic rights) holder? Also, exactly when can an orphan works license be issued? For exactly which works is licensing permitted, and what kinds of uses will be licensed?
The Canadian Copyright Board always requires the payment of royalties for orphan works licenses. As we look at their system, it is important to ask how these payments fit in to the incentive scheme for new creation that copyright is intended to create. Are these payments, at least quite often, simply a tax on users that do not serve any role in encouraging new works? If a similar model is proposed in the US by industry groups, especially involving the collective rights groups, we would have an even greater problem of “rent seeking.”
The most significant issue raised by the article is the relationship between licensed uses and those that benefit from a statutory exemption. In the US, that raises the issue of how a licensing scheme for orphans would fit in with fair use, which is exactly the issue the Authors Guild tries to duck by not addressing fair use in its complaint against Hathi. If we approached orphan works by creating a licensing regime, would a license be available even if a use was arguably fair use? Who would decide? And would the license thereby provide a protection against liability for a negative judgment on the fair use case, thus usurping the role of the federal courts?
As I read about the standards by which the Canadian licensing board makes it decisions about which uses to license and how to set royalties, I realized they were using criteria very similar to the fair use analysis. The good faith of the user, the commercial or non-commercial nature of the use and the character of the original work all come into play. This suggests to me that fair use — which is found in US law but not in Canada, of course — already accomplishes a lot of the work that a licensing scheme for orphan works might be expected to do, and that fair use is really the best model for dealing with orphans in the U.S.
Really the choice between a license model and fair use is an instance of the proverbial choice between security and liberty. By suing over a fair use claim about orphan works, the Authors Guild, I think, is hoping to force libraries to opt for security, and therefore pay for licenses.
The details, problems, and dangers of Canada’s solution to orphan works led me to conclude, even before this new lawsuit made me suspect that a similar, but privately administered, system was going to be pushed in the US, that a better solution under US law would lie in the remission of potential damages (which Congress considered several years ago). In short, we should treat orphan works as simply a species of fair use and define a protection against liability for good faith users of orphan works that is similar to that which we already grant to employees of educational institutions (under 17 U.S. Code section 504(c)(2)) who make good faith determinations about fair use. That is precisely the solution Hathi makes plausible, and that the Authors Guild wants desperately to suppress lest Congress figure out that a fair use solution could really works.