Prof. James Gibson, University of Richmond School of Law
September 15, 2011

The Digital Millennium Copyright Act has been around for more than a dozen years now.  Some of its provisions were just weird, such as the one that established sui generis protection for boat hull designs.  Others have had a skeptical reception in the courts, like the anti-circumvention provisions that forbid certain forms of hacking through technological protections for copyrighted works.

But one DMCA provision that has proved popular in both the copyright community and the courts is the notice-and-takedown procedure codified at 17 U.S.C. § 512(c).  When a copyright owner finds that some Internet user has illegally posted its copyrighted content online, the notice-and-takedown procedure allows the owner to send a notice to the Internet service provider (the ISP) that hosts the material and ask that it be taken down.  If the notice contains the right information, the ISP’s compliance will shield it from liability for damages it might otherwise have had to pay.  (Google recently used this DMCA shield to escape liability for its YouTube subsidiary’s use of copyrighted content, as some past IP Viewpoints entries have discussed.)

In fact, this provision has proved so popular that the United States has been exporting it (or a version of it) to other countries by including its provisions in free-trade agreements with such countries as Australia, Chile, Colombia, Dominican Republic, and Singapore.  And in one respect, the notice-and-takedown model might work better abroad than it does in the United States.

To understand why notice-and-takedown might work better outside the United States, one must first understand how it works inside the United States.  At first glance, Section 512(c) seems to call for the input of both copyright owner and Internet user.  A copyright owner spots what it thinks is an infringing use of its work online and then notifies the ISP, asking for the material to be taken down.  To preserve its immunity from liability, the ISP will comply with the request.

At the same time, the ISP will (in theory) tell the Internet user who originally posted the material about the takedown.  That gives the user a chance to object – perhaps because the material is not actually copyrighted or the use was not infringing (e.g., a fair use).  So the Internet user can counter-notify; it can send the ISP a request to put the material back online.  The ISP would have two weeks to restore the material, unless the copyright owner decides to take the Internet user to court.

Both the notice and the counternotice are subject to penalties for false statements, including the possibility of a perjury prosecution and a civil lawsuit for knowing and material misrepresentations.  So again, at first glance, this seems like an evenhanded system.

The problem is that, under U.S. law, the ISP has a lot of incentive to comply with the initial takedown notice, and very little incentive to comply with any counternotice – or even to tell the Internet user about the takedown in the first place.  You see, compliance with the initial notice from the copyright owner gives the ISP immunity against a copyright lawsuit for damages.  This immunity is valuable, because ISPs that knowingly facilitate the copyright infringement of Internet users would ordinarily share the users’ legal liability.  Complying with a copyright owner’s notice therefore helps an ISP avoid a real threat of liability.

In contrast, complying with the counternotice from the Internet user does very little for the ISP, because, under the DMCA, such compliance merely immunizes the ISP from a lawsuit brought by the user for an unjustified takedown.  And there is no need for such immunity, because Internet users don’t really have any legal basis to sue their ISPs for an unjustified takedown in the first place.  The relationship between ISPs and Internet users is governed almost completely by private contracts.  Those contracts are invariably written by the ISP, and they almost never give the subscriber any right to protest an unjustified takedown of material he or she posts online.  There is no real threat of liability, and so the DMCA gives ISPs no incentive to comply with counternotices, or even to tell the user that the takedown has occurred.

In practice, this means that ISPs routinely comply with takedown notices, but the process usually stops there.  Internet users might never learn that the material they posted has been removed – and if they do learn about it, their counternotices can be ignored without legal consequence.

Or so it goes in the United States.  Why might it work differently elsewhere?  Well, the imbalance in the U.S. system results from the absence of any real threat of a lawsuit brought by the Internet user against the ISP.  In a different legal system, however, that threat might actually exist.

In particular, in many countries the right to free speech is enforceable against private parties.  (In contrast, in the United States it is enforceable only against the government.)  If an Internet user’s posting of non-infringing material qualifies as speech – and it easily could – then its unjustified removal would subject the ISP to liability for violating the user’s rights.  Another possibility, particularly in civil law jurisdictions, would be a statute that specifically establishes ISP liability for unjustified takedowns.  Either of these causes of action could provide the real threat of liability necessary to give the counternotification process some teeth (as long as the damages to be awarded were significant).

In short, as the United States rolls out its DMCA model to its trade partners, those partners would be well advised to examine whether their own legal systems can lend even more integrity to the notice-and-takedown system than we have here in the United States.