Category: Law & Policy

Amicus briefs were filed recently in the 4th Circuit Court of Appeals in the case of BMG Rights Management v Cox Communications. In November of 2014, BMG sued Cox (an ISP) for contributory copyright infringement, and a US District Court found for the plaintiff in December of 2015, awarding $25 million in damages. The suit was based on evidence that Cox was willfully ignoring and/or failing to address the use of its service by repeat infringers.

“Digital rights” groups and (let’s be honest) people who support piracy decried the outcome, which Cox has now appealed. A decision may expected by early Spring, and if the court were to find Cox’s arguments persuasive, this would have a very damaging effect for rights holders—further aggravating the weakness in the DMCA as a mechanism of enforcement.

Having failed in the lower court to convince either judge or jury that Cox had sufficiently maintained its liability shield (safe harbor) under the DMCA, the company now seeks to argue on appeal that the DMCA says something other what it says.  And true to form, Public Knowledge and the Electronic Frontier Foundation have chimed in (via joint amicus brief) to propose that if the Cox ruling is upheld, it could lead to disenfranchisement of people from internet access and…y’know…destroy free speech. Again.

Anyway, let’s review.

Although one might get the idea from general discussion that DMCA is either a blanket liability shield for ISPs or a blanket takedown mechanism for rights holders, it is neither of these things. Instead, the DMCA statutes define the conditions and responsibilities of service providers with regard to users uploading or accessing unlicensed, copyrighted material onto their platforms. In simple terms, the law states how an ISP may conditionally retain its safe harbor liability shield, and it is these conditions which tend to get lost in the broader reporting on DMCA-related stories.

A service provider like Cox, which sells internet access to consumers, generally would not be concerned with hosting infringing material the way a platform like YouTube will be, but they can be liable for contributory infringement if the company is aware of subscribers using its service to repeatedly infringe copyright and takes no action to stop the infringing activity. The statute in DMCA §512(i) states that a service provider must have “a policy that provides for the termination in appropriate circumstances of subscribers and account holders of the service provider’s system or network who are repeat infringers.”  Note that the DMCA explicitly anticipates conditions by which the provider must eventually terminate the accounts of repeat infringers—logically, those who refuse to stop after some amount of warning.

So, for all of EFF’s and PK’s dramatics about the cruelty of account termination—their brief compares it to “cutting off a tenant’s water”—one might get the idea that in writing the DMCA, Congress never imagined such a remedy; but termination is precisely what the law says.  (A more reasonable comparison would be made to the prospect of losing a driver’s license, which can be quite damaging to an individual, but which is also a penalty imposed only after some degree of willful abuse.)  What the DMCA does not say is how much infringement makes a “repeat infringer,” or how an ISP must design its policy for addressing repeat infringers where account termination is a possible consequence.

This ambiguity can be exploited by service providers, and in this case, BMG presented substantial evidence to the court that Cox’s policies seemed purposely designed to avoid taking action against repeat infringers within a reasonable interpretation of the DMCA.  The district court opinion written by Judge Liam O’Grady states, “Unfortunately for Cox, the record was replete with evidence that foreclosed any assertion by Cox that it had reasonably implemented a repeat infringer policy.”

Who’s a repeat infringer?

In its appeal, Cox seeks to argue that “repeat infringers” in DMCA §512(i) can only mean “subscribers who have been found liable for infringement by a court or a jury on more than one occasion.” In other words, BMG’s evidence of users consistently accessing unlicensed material does not make those users “repeat infringers” unless they’ve already been found guilty of infringement in a court of law—more than once.

Cox is relying on a specific interpretation of the word “infringer” in the statute, hoping that the appellate court will agree that a “repeat infringer” can only be an individual who has lost at least two copyright infringement cases in court.  That population may be a number barely large enough to fill a small cafe, which is considerably smaller than the billions of users anticipated by the architects of the DMCA. And because it is obvious to any reasonable person that one can be guilty of a violation without being held liable—if you’re let go for speeding with a warning, it doesn’t mean you weren’t speeding—any reasonable person should conclude that Congress’ use of the word “infringer” in this case was meant to describe individuals engaged in unlicensed access to, or use of, copyrighted material, even if the rights holders do not intend to pursue litigation against them.

It is a truly bizarre argument, which overtly pretends the DMCA is something other than what it is.  The background, intent, and language of the law is known, by the parties involved in its writing, to have been designed as a process by which ISPs and rights holders would collaborate to mitigate infringement—no matter where the users reside or who they are—without costly litigation.  For example, a “repeat infringer” under the DMCA can easily be—and often is—a user in a foreign country who has never seen a U.S. court, let alone been a named party in a U.S. copyright case.

The DMCA’s existence is based partly on an understanding that worldwide users would inevitably infringe—either willfully or unintentionally—but that the ISPs and rights holders would have a mechanism for removing infringing files or stopping infringing activity without anyone getting sued—as long as all parties met the conditions in the agreement.

Knowledge of Infringement

If you call someone on their cell 3-4 times and they don’t answer, there may be any number of reasonable explanations.  If you call them 50 times and they don’t answer, they’re ducking your calls—or in legal terms, they’re engaging in “willful blindness” by choosing not to hear what you have to say.  So, what about a few million calls?  At issue in the lower court decision was the fact that Cox simply ignored millions of notices sent by Righstcorp on behalf of BMG containing IP addresses and other corroborating information demonstrating repeat infringement by numerous subscribers.

Cox’s rejection of these notices was deemed “willful blindness,” which is the legal equivalent to taking affirmative action to infringe, hence the charge of contributory infringement.  On appeal, Cox contends that the notices, which they chose to ignore, would not constitute “knowledge of infringement” anyway, thus seeking a standard of “knowledge” so narrow that it would effectively excuse all ISPs from adopting any kind of anti-infringement policy as mandated by the DMCA. 

The Betamax Argument

Cox further argues that as a conduit provider, they can only be held liable if they “actively encourage or induce infringement through affirmative acts.”  In this regard, Cox relies on a very broad reading of Sony Corp v Universal Studios (1984), which held that Sony could not be liable for copyright infringements that may be committed by users of its VCRs.  There are several parts to the Sony ruling, including the Court’s holding that because the VCR could be used for “substantial non-infringing purposes,” Sony could not be liable for any infringing uses unless it actively induced or encouraged that infringement by its customers.

Cox now seeks to argue, by the same principle, that because internet access may be used for “substantial non-infringing” purposes, they should be held to the same standard as Sony because they also did not induce its users to infringe.  Once again, Cox’s argument seeks to bypass the terms of the DMCA with an argument that, if upheld, would unconditionally absolve all ISPs of liability unless they promoted infringement in their marketing.

The fact that the internet, writ large, is used for substantial non-infringing purposes is immaterial.  To stick with automotive analogies, just because cargo trucks are used for substantially legal purposes, this has no bearing on the liability of a trucking company, if it were to turn a blind eye to some of its drivers transporting contraband across state lines.

Nevertheless, Cox—with the hyperbolic assistance of EFF/PK—seeks to argue that if this hypothetical trucking company pays a penalty, loses it’s license, or fires the named truckers, that will lead to the end of trucking itself, and we all starve. If 100 Cox subscribers lose their access due to their infringing activity, it has no more bearing on the internet and its billions of users, than if a different 100 subscribers lost their access due to non-payment for the service.

As argued in the brief filed by the Copyright Alliance, “If Cox’s view was the law, then as long as it was not actively inducing or promoting infringement, Cox could throw each and every infringement notice it received straight into the trash, and the “Abuse Group” charged with addressing online piracy could knowingly permit active infringement without creating any risk of liability to Cox.”

To summarize, the Cox argument boils down to the following:  1) repeat infringers are not repeat infringers; 2) even if they were repeat infringers, we could not know they were repeat infringers from the evidence presented; 3) even if they were repeat infringers and we knew about them, we aren’t liable because we didn’t tell them to infringe.

Implications of this Case

This effort to treat the safe harbor as an unconditional liability shield is generally where large ISPs have tried to move the conversation, both in the courts and in the public dialogue. But the DMCA was never meant to provide a free ride for service providers, although it has inadvertently produced that result to a greater extent than anticipated by its authors.

As Copyright Alliance also observes in its brief, if a rights holder the size of BMG has no remedy in a case in which a service provider has been shown to have circumvented the provisions in the law, then independent rights holders truly have no hope of protection whatsoever in the evolving digital market.  Instead, independent rights holders need the DMCA to be made more effective than it is by revising some of the ambiguity in the statutes, which leads to the kind of bad-faith policies on the part of ISPs that are apparent in this case.