Over at Techdirt there’s a write-up of the latest comment I submitted on behalf of the Copia Institute as part of the Copyright Office’s study on the operation of Section 512 of the Digital Millennium Copyright Act. As as we’ve told the Copyright Office before, that operation has had a huge impact on online free speech. (Those comments have also been cross-posted here.)
In some ways this impact is good: providing platforms with protection from liability in their users’ content means that they can be available to facilitate that content and speech. But all too often and in all too many ways the practical impact on free speech has been a negative one, with speech being much more vulnerable to censorship via takedown notice than it ever would have been if the person objecting to it (even for copyright-related reasons) had to go to court to get an injunction to take it down. Not only is the speech itself more vulnerable than it should be, but the protection the platforms depend on ends up being more vulnerable as well because platforms must risk it every time they refuse to act on a takedown notice, no matter how invalid that notice may be.
Our earlier comment pointed out in some detail how the current operation of the DMCA has been running afoul of the protections the First Amendment is supposed to afford speech, and in this second round of comments we’ve highlighted some further deficiencies. In particular, we reminded the Copyright Office of the problems with “prior restraint,” which the First Amendment also prohibits. Prior restraint is what happens when speech is punished before there has been any adjudication to prove that it deserves to be punished. The reason the First Amendment prohibits prior restraint is that it does no good to punish speech, such as by removing it, if the First Amendment would otherwise protect it – once it has been removed the damage will have already been done.
Making sure that legitimate speech cannot be removed is why we normally require the courts to carefully adjudicate whether its removal can be ordered before its removal will be allowed. But with the DMCA there is no such judicial check: people can send demands for all sorts of content to be removed, even if it weren’t actually infringing, because there is little to deter them so long as Section 512(f) continues to have no teeth. Instead platforms are forced to treat every takedown notice as a legitimate demand, regardless of whether it is or not. Not only does this mean they need to delete the content but, in the wake of some recent cases, it seems they also must potentially hold each allegation against their user, regardless of whether it was valid or not, and then cut that user off from their services when they’ve accrued too many such accusations, again regardless of they were valid or not.
As we did before, we counseled the Copyright Office to return to first principles: the DMCA was supposed to enhance online free speech, and it’s important to make sure that all of its provisions work together to do just that. To the extent that it may be appropriate for the Copyright Office to make recommendations on this front, one is to remind all concerned that the penalty articulated in Section 512(f) to sanction bad takedown notices can and should be applied according to a flexible standard, rather than the rigid one courts have lately adopted. In any case, however, the Copyright Office certainly should not be advocating to changes in any provisions or their interpretations that make the DMCA any less compatible with the First Amendment than it has already tended to be.