Section 512(f) of the DMCA says pretty clearly that anyone who "knowingly materially misrepresents under this section... that material or activity is infringing... shall be liable for any damages, including costs and attorneys' fees", which would seem to apply here; the EFF argues that Universal should have reasonably known that the video obviously constituted fair use. In a Law.com article about the case, attorney Kelly Klaus, representing Universal, countered that "Congress also said that there was another remedy, which is the counter-notice procedure, which is what happened here." But this seems to miss the point -- the DMCA says that the remedies are the counter-notice procedure and an award for attorney's fees. (Klaus's firm did not respond to requests for comment for this article.) Anyway, as EFF staff attorney Corynne McSherry points out, if there were no possible award for attorney's fees against copyright holders who make false accusations, then there would be no disincentive for copyright holders not to file frivolous accusations in the first place.
I'm an EFF member and support their request for attorney's fees, but let's play devil's advocate. Suppose you were an indie musician who sold your songs online, and you found a number of YouTube videos that used your song without permission, so you sent a long list of DMCA takedown notices to YouTube. Included in that list was one video that used only a brief portion of your song, short enough to count as fair use. Is $400,000 a fair punishment for accidentally including one video in your list that wasn't a bona fide copyright infringement?
On the other hand, if the EFF doesn't get their attorneys fees, then they have to eat the cost of the work they did, and that doesn't seem fair either.
The problem is that once you have a $400,000 bill on the table, someone has to pay it, which punishes one or both parties usually vastly out of proportion to any wrongdoing. ($400,000 is almost half of what Reebok had to pay when one of their lead-tainted bracelets killed a child.) Huge attorney's fees awards also limit access to the court system for plaintiffs who might have a reasonable case, but can't afford the risk of having to pay attorney's fees if they lose, and for defendants who might also have a reasonable case, but are under pressure to settle quickly to avoid the risk of a huge attorney's fees award against them.
This suggests an economics / game theory problem: Could you come up with a system that takes into account the incentives of parties on both sides, and that prevents huge legal bills from being generated?
Now, any argument about the legal system usually raises two kinds of objections. The first is that the existing system "works". Well, in many ways it does, but everybody also knows that wealthy corporations and individuals enjoy a huge advantage in the court system, even though courts are supposed to treat all parties equally. So at least in that respect it doesn't "work" the way it's supposed to. The second objection is that it's too hard to change the rules and traditions that are built into legal proceedings, so it's better just to work within the system. True, but that's not the question I'm asking. I'm posing it as a logical brainteaser: If you had carte blance to modify the way that legal disputes were held, could you do it in a way that respects the rights and interests of all parties and still minimizes the legal fees incurred? (Whether I'm right or wrong, my goal is to make this argument more interesting to mathematicians and game theorists, than to lawyers; otherwise, I've failed.)
From a game-theoretic point of view, you might argue that large attorney's fees serve a useful purpose by discouraging frivolous lawsuits. The problem is that the fees don't just discourage frivolous lawsuits but also non-frivolous lawsuits where there's a reasonable chance of losing. On the other hand, a person who is already broke would have little disincentive to file a frivolous lawsuit, since the worst that can happen is that they'd get hit with a huge award for attorney's fees and have to declare bankruptcy, which they might consider worth the risk for a small shot at a million-dollar payout. So assume that attorney's fees are not themselves the best way to deter frivolous lawsuits, and that avoiding large fees in general is still a desirable thing. How do you design rules to achieve that?
I think you could save a lot of money by enforcing a rule that a lawyer is not allowed to seek attorney's fees from the other side for arguing any points that the other side offered to concede anyway. So the incentive would be that if party A's lawyer concedes some point of fact or point of law, and party B ultimately wins the case and an award for attorney's fees, then party B is not allowed to seek attorney's fees for arguing the point conceded by party A's lawyer.
In all of my legal cases where the other side was represented by a lawyer who was getting paid by their client up front, it was clear from reading the other side's briefs (and my own lawyers agreed with me) that opposing counsel had spent a lot of time spinning their wheels and arguing obvious or irrelevant points before getting to the crux of the dispute. If their client wants to pay them for that busy-work, that's between them and their client, but if they had won the case and an award for attorney's fees, I would have objected that they shouldn't be allowed to charge us for time they spent arguing points that we would have given to them anyway. The hypothetical savings from implementing and enforcing this rule, are not trivial.
So how does game theory predict that the two sides would behave under this rule? Suppose MegaCorp is suing or being sued by IndieActivist. MegaCorp's first priority is to win, and if possible to hit IndieActivist with a huge award for attorney's fees to discourage other would-be IndieActivists. MegaCorp doesn't want to lose, but if they do lose, they don't much care about the attorney's fees award they would have to pay to IndieActivist's lawyers. In this scenario, they would be expected to concede very little, disputing trivial points in order to drag out the case as long as possible, hoping that IndieActivist's lawyers would run out of time or money and pressure their client to settle. In other words, MegaCorp would behave about the same as they would under the existing rules.
For IndieActivist, on the other hand, their first priority is to win, but they also care very much about not having to pay a staggering award for attorney's fees if they lose. So they would be expected to concede any points of fact or law, even if favorable to MegaCorp, if those points are so obvious that they don't think the judge would be likely to rule in their favor on those questions anyway. This way, even if IndieActivist loses and has to pay attorney's fees to MegaCorp, those fees would be limited to the time spent arguing the actual point of disagreement that formed the crux of the lawsuit.
Suppose, for example, that Universal had actually sued Lenz for violating Prince's copyright by using a 30-second excerpt of his song in her video. Lenz or her lawyers could have filed a brief conceding all the obvious points that they would expect Universal's lawyers to make: Prince was the holder of the copyright, the copyright had been filed with the Copyright Office, Lenz never sought permission for using the recording, etc. Very quickly, the whole case could be distilled down to: "Show this video to the judge and let them decide if it qualifies as 'fair use'." Any effort spent arguing any points beside that, is wasteful. And if the legal system encourages lawyers to rack up billable hours arguing other points, then the system is wasteful. Concede the obvious, and everybody's costs are kept under control.
This only partially addresses the problem of large attorney's fees, because it still leaves the fees that are generated in the process of arguing points that the other side wouldn't concede. Solving this problem is much harder, because while you can simply eliminate the work that's spent on arguing points that the other side would give to you anyway, you can't eliminate the work spent on points that are genuinely in dispute, you can only try to make that work shorter and cheaper. I've argued for my own fairly complicated remedy in a separate article, but my main point was that legal costs aren't driven up so much by the complexity of the law as by the ambiguity in it. The Windows programming interface, after all, is also very complex, but if you can write a clear description of what you want a simple program to do, you can often get a programmer to write the program for you for dirt cheap. In arguing a legal case, on the other hand, the number of possible outcomes grows exponentially with each point of ambiguity in the law where there's no way to predict how the judge will interpret a particular rule.
But still, even if you can't reduce the ambiguity in how a legal question will be interpreted, you can avoid a lot of unnecessary attorney's fees by distilling the case just down to that particular question. Is it fair use to use a 30-second clip of Prince's song in a video of a dancing toddler? Let the judge decide. But if that's the one and only point that both sides can't agree on, then neither side should be able to bill for time spent arguing about anything else.
Perhaps someone mathematically or logically inclined can come up with a better algorithm for avoiding the billing hours generated by arguing the obvious. I'm not entirely happy with my own solution, because it still allows MegaCorp to concede absolutely nothing, and to try and bleed IndieActivist dry by forcing them to argue even the most trivial points. IndieActivist's lawyer could be reimbursed for that time if they win and get an award for attorney's fees, but they might run out of money or patience before then. To counter this tactic, you could allow either side to seek penalties for Frivolously Arguing The Super-Obvious. If IndieActivist's lawyer wants MegaCorp to concede an obvious point and MegaCorp won't do it, IndieActivist could seek a FATSO penalty, and the judge could decide whether to award them that penalty if the point is really and truly obvious, without deciding on the merits of the case as a whole. The penalty doesn't have to be large enough to hurt MegaCorp, it just has to be large enough to compensate IndieActivist's lawyer for their time, so that MegaCorp can't run them into the ground by forcing them to argue every point unnecessarily. However, economic game theorists might think of some unintended consequence of the FATSO rule. Could MegaCorp flood IndieActivist's lawyer with a gigantic list of requested concessions, so that if IndieActivist's lawyer screws up and forgets to concede one of the points that the judge turns out to consider "obvious", MegaCorp could hammer them with a FATSO award too? It's hard to anticipate all the ways that either party might abuse a new rule of the game.
Meanwhile, under the existing system, while it may be unfair to Universal in some cosmic sense that they have to pay out $400,000 for sending one mistaken DMCA takedown notice, it would be more unfair to force the EFF to eat those costs, and in any case the DMCA does clearly allow for an award of attorney's fees. But it would be better for everyone in the long run -- especially for the EFF and the kind of relatively powerless clients that they usually represent -- if there were more ways to keep legal costs from spiraling out of control in the first place.