Fair Use and “Market Effects”: Which Potential Markets Count?
I’m very grateful to Bill and Derek for inviting me to join their blog! There’s a bit of a learning curve with WordPress, I’m finding, but I’m gradually getting the hang of it. As Derek mentioned, I’m preparing to join the faculty of the University of Cincinnati College of Law, and I thought I’d begin by throwing out a topic that is causing me some headaches as I prepare the syllabus for my fall Introduction to Intellectual Property course.
The fourth factor under the federal fair use statute requires courts to consider “the effect of the use upon the potential market for or value of the copyrighted work” when deciding whether a particular use is fair. This factor is easy enough to understand when well-developed markets for the copyrighted work already exist. What about when there are no such markets, though? To put it differently: what should courts do where an infringing act affects only a possible future market that might or might not come into existence (depending, frequently, on the outcome of the litigation)? Which potential markets “count” for purposes of determining whether a given use is fair?
There are lots of ways to answer this question. We might say:
- Harmful effects on a market for the opportunity to discover a copyrighted work count against a finding of fair use, irrespective of the effect on any market for the work itself; or
- Harmful effects on a market for derivative works, which doesn’t exist at the time of the infringement but which actually comes into existence during litigation, count against a finding of fair use; or
- Harmful effects on hypothesized markets other than the market for the work itself don’t count against a finding of fair use, because the copyright holder isn’t entitled to exclusive rights in markets other than that for its own work.
What makes this a hard subject to synthesize (and to prepare to teach) is that, just within the last six months or so, courts in high-profile cases have said all three of the above things about fair use and harm to potential markets.
Case #1 is BMG Music v. Gonzalez, in which the Seventh Circuit found that a user who downloaded songs from a peer-to-peer network for the purpose of “sampling” them, intending to purchase the songs the user enjoyed, could not avail herself of the fair use defense. Analogizing to radio broadcasts, the court reasoned that “there is also a market in ways to introduce potential consumers to music.” The (negative) effects on that market, rather than the (positive) effects on the market for sales of the music itself, justified the rejection of the user’s fair use defense.
Case #2 is Perfect 10 v. Google, in which a district court said that Google’s display of scaled-down thumbnail versions of Perfect 10’s copyrighted photos in its Google Image Search results page wasn’t fair use, reaching the opposite conclusion from the Ninth Circuit in Kelly v. Arriba Soft on similar facts. Perfect 10’s claim succeeded where Kelly’s failed, the court reasoned, because during the course of the litigation Perfect 10 had contrived to create a “market” for the sale of thumbnails through licensing them to a single cell-phone provider. (And wouldn’t you like to know who paid whom in that exchange!)
Case #3 is Bill Graham Archives v. Dorling Kindersley Ltd., noted by Professor Lessig here, which is sort of a Kelly v. Arriba Soft for the offline world (the alleged infringement involved scaled-down concert posters reproduced in a book, rather than scaled-down digital photos reproduced on a web site). Upholding a fair use defense, the Second Circuit remarked, as to the fourth fair use factor, that ‘a copyright holder cannot prevent others from entering fair use markets merely “by developing or licensing a market for parody, news reporting, educational or other transformative uses of its own creative work.’” This language seems to forbid, at a minimum, the “bootstrapping” post hoc market creation of Perfect 10 v. Google, and possibly also BMG Music v. Gonzalez’s sleight-of-hand, which attempts to shift the focus away from effects on the market for the copyrighted work itself.
Is there any articulable principle to be drawn from these cases that explains which potential market effects count against a finding of fair use and which do not? Can the cases be reconciled, or have the courts just left fair use law in this area a mess? I don’t have an answer here (although I suspect my sympathies are pretty clear), but would love to hear what others think. After all, it’s only a few short months now until I have to try to actually teach this topic to a roomful of skeptical law students. 
Filed under: Copyright, Education & Copyright, Law School
Not reconcilable – is my view. To be sure the fair use doctrine conceals the complex normative and practical issues that the juridical framework is being used to address. As many have observed, some issues are not readily amenable to juridification. Have done a short article that deals with the Perfect 10/Kelly Arriba point (note the argument pursued by the claimant in the latter case and also judicial notice taken of the spider). What is interesting about Perfect 10 is that due to the amenability of the complex issues for judicial solution – the result actually creates some wiggle room for parties to produce a negotiated outcome. This neatly shows the instrumentality of the fair use defence – it is dealing with trade-offs as well normative and practical issues.
[...] The outcome of the law suit came as some sort of surprise: Case #2 is Perfect 10 v. Google, in which a district court said that Google’s display of scaled-down thumbnail versions of Perfect 10’s copyrighted photos in its Google Image Search results page wasn’t fair use, reaching the opposite conclusion from the Ninth Circuit in Kelly v. Arriba Soft on similar facts. Perfect 10’s claim succeeded where Kelly’s failed, the court reasoned, because during the course of the litigation Perfect 10 had contrived to create a “market” for the sale of thumbnails through licensing them to a single cell-phone provider. (And wouldn’t you like to know who paid whom in that exchange!) from blogs.law.harvard.edu [...]
[...] The court cites and distinguishes the Second Circuit’s recent Bill Graham Archives case (discussed by Tim here). When that decision came out, some observers thought it would be a very significant and useful case, but I was more skeptical. The fact that the court felt compelled to deal with the case here – even though, since it’s from a different geographical circuit, it is not controlling – provides an early indication that I was wrong and that the Bill Graham case may come to be seen as a major precedent. [...]
[...] So, not only would settlements lock in Google as the super-dominant player as Pasquale says (and as many techies have begun to fear anyway), it also would short-circuit the movement of the law to a reasonable accommodation of search technology. To be sure, that movement is slow, indirect, and sometimes fumbling, but it is happening. The Perfect 10 appeal (which Tim discussed here) is one place where the issues will be teed up fairly soon. Yet given the risks inherent in waiting around for courts and then possibly Congress, I would not be surprised if Google indeed chose to settle. And then the litigation would disappear and Congress would go back to its many other worries. [...]
This is a very enlightening round-up of cases. My humble effort to “bring order” to this subject is here:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=850704
and includes cites to the seminal work in this field from Loren, Africa, and Duhl.
As for my take: I have a fear that courts are going to “Cut the gordian knot” by adopting a misunderstanding of Gordon’s transactions costs approach…i.e., they will say that any use that can be paid for, should be paid for. That’s why I’ve focused my fair use scholarship away from the fourth factor and towards the first factor.
[...] Some people with whom we spoke fretted, quite understandably, about the trend toward rightsholders slicing and dicing content into smaller bits and offering licenses for every conceivable tiny use. As Tim Armstrong explained a few months ago, the mere existence of a market for a particular use of content has been seen by some courts as severely undermining a fair use defense. Educators thus forced to pay constantly for microchunks of content that ought to be covered by fair use might soon be priced out of those markets — a perverse effect of the “any conceivable market counts” approach to fair use. On the other hand, there certainly are times when fair use just does not apply, and educators would like to have an easier time clearing rights. Our case study about the Database of Recorded American Music is an example of such a situation: its sponsors wanted to make full copies of important music more widely available on college campuses and recognized that doing so required licenses rather than reliance on fair use. They encountered rights clearance nightmares (which, with tenacity and persistence, they resolved, but others with fewer resources and less skill might have more trouble overcoming such problems). [...]
[...] In the spirit of living up to my own principles, I made a couple of very minor contributions to the world of open access primary legal source materials in the fall. The process of preparing the handout for my fall IP survey course left me with electronic versions of several recent cases that I wanted to cover in class, so I picked a couple of them and put them online at Wikisource. Here they are: BMG Music v. Gonzalez, 430 F.3d 888 (7th Cir. 2005) (a case I’ve briefly mentioned on this blog a couple of times); and Lexmark Int’l, Inc. v. Static Control Components, Inc., 387 F.3d 522 (6th Cir. 2004) (which I’ve blogged about here and here). I wrote up a short summary of the Gonzalez case for Wikipedia, and minimally edited this already very thorough entry on the Lexmark decision. [...]