I have never forgotten the dictum of one of my law professors – a professor of tax law, not surprisingly – that fairness breeds complexity. His point was that the complexity of the law is often because it evolves to account for the different circumstances in which real people encounter it. We feel, for example, that parenthood is a social good, but we know it is expensive. So, to encourage families, exemptions for children and other dependents are added to the tax law. The more we strive to recognize how people are differently situated, the more complex our laws become.
In a couple of recent Supreme Court cases dealing with copyright, however, we have seen decisions that seem to forget this basic concept, and thus to oversimplify some copyright principles. The decision handed down in Cox Communications Inc. v. Sony Music a few weeks ago, while it reached the right result, in my opinion, is a recent example. And we have seen this same approach in an earlier copyright decision, which, like the one in Cox, was also written by Justice Thomas – Star Athletica v. Varsity Brands.
The Cox case raised the question of contributory copyright infringement by internet service providers (ISPs) when infringing material is uploaded to their networks by users. Congress attempted to deal with this problem almost twenty years ago, when it added the notice and takedown provisions for ISPs to section 512 of the copyright law. The result is that an ISP can only be liable if it has “actual knowledge” of a specific infringement, which it would have through the notice process.
Almost from the start, these provisions to exempt ISPs from contributory liability have been overwhelmed. The share volume of notices, often generated automatically by robots that search for alleged instances of infringement, is staggering. And courts have struggled to decide when the knowledge is sufficient to force an ISP to act. For example, is an IP address enough, even though we know that IP addresses do not actually identify a specific user? The result has been a lot of lawsuits, which usually target ISPs (that’s where the money is) for not acting, not acting fast enough, or not acting harshly enough against the alleged infringer.
In Cox, the complaint was that Cox, the ISP, failed to act to remove users whose IP addresses they supposedly knew were associated with infringement. As I have said, this is a frequently litigated issue, and the situations involved can be quite complex. Nevertheless, Justice Thomas imposed a simple rule to try and clear the confusion, asserting that the only test for contributory infringement is whether or not the ISP intended their network to be used for infringement, which can be shown by evidence that they induced the infringement or tailored their system for infringement.
Because there was no evidence that Cox did intend infringement, this case, in which the lower court had found Cox liable for contributory infringement, was reversed. I believe that was the right result; liability for ISPs that is premised on the actions of users would be devastating to the internet, and to all the useful things we can now accomplish with it. But I am concerned that the overly simple approach, clearing out all the previous discussions and imposing a single, easy rule, will collapse under the same weight of circumstances that bedeviled previous applications of the notice and takedown regime.
It is legitimate to ask, after Cox, how ISPs should treat takedown notices. Does the “actual knowledge” standard still apply? It is part of statutory law, after all. But the Court has now told us that a higher standard now applies to create ISP liability – intent. Can those ISPs now ignore takedown notices in spite of section 512? Is that part of the law no longer relevant? I am glad to see ISPs protected, but I fear that this test for contributory liability simply cannot account for all the different pressures that will be brought to bear on it.
In the previous case I mentioned, the 2017 ruling in Star Athletica, Justice Thomas took a similar approach, reducing a lot of different tests that had arisen around the rule that “useful articles” are not eligible for copyright protection to a simple formula. In trying to decide how to separate what is truly artistic in an article from what is “merely” useful, a welter of different tests had arisen. As in Cox, there was very little statutory language for the Court to interpret, and Justice Thomas asserts that the problem is easy, and the lower courts are overthinking it; he gives us instead his rule for all situations. The problem, as I see it, is that all those extenuating circumstances that led to complications will simply arise again in subsequent cases. The test Justice Thomas outlined seemed simple, but, even in the case the Court was deciding, it proved difficult; in my opinion, Justice Thomas misapplied his own test to the facts of Star Athletica. So I doubt that the simple rule about useful articles will last for very long before efforts to be fair complicate it again.
Justice Thomas may well have a noble goal in mind – that the law should be simple and easy to understand. We might all wish that was the case. But the history of our law, and the society which creates it, shows us that things always become more complicated. Fairness, indeed, often requires such complexity.
My title suggests an overreaction to go along with this oversimplification. Indeed, I think the Cox case did induce exactly such an overreaction, from no less a person that our Register of Copyright, Shira Perlmutter. Only a month ago, I had the opportunity to hear Register Perlmutter speak at an event celebrating the 50th anniversary of our current copyright law. In that talk, Perlmutter traced how the copyright act has adapted to new technologies over the years since it was adopted in 1976. Her conclusion was that the law was doing well, and that a major revision of copyright was both unlikely and unnecessary. But then came the Cox decision, and an apparent turnaround in the attitude of the Register. This blog post from the “Complete Music Update” news site suggests that, now, Perlmutter believes that the moment has come for Congress to act to preserve ISP liability and protect the content industries. Apparently, she said as much on May 12 to the Senate IP Committee, asserting that “actual knowledge” should continue to be the standard for ISP contributory liability, and that Congress should rewrite the law to make it so.
To me this seems like an overreaction, born of the incestuous relationship between Big Content and the Copyright Office. If the Cox decision oversimplifies the rule for such liability, I think that time and circumstances will push the rules back toward the actual knowledge standard. Indeed, ISPs may not even change their current practices around take-down notices in any case, since they already have procedures in place, and those practices offer a sense of security. So, if Cox represents a well intentioned but likely futile oversimplification, Register Perlmutter’s remarks seem to embody a partisan overreaction.