Prof. Randal C. Picker, The University of Chicago Law School
December 23, 2011

Possible copyright legislation has been in the news with the House Judiciary Committee’s on-again, off-again mark up of H.R. 3261, the “Stop Online Piracy Act.”  SOPA, and its Senate cousin, the PROTECT IP Act, are designed to limit intellectual property infringement at the price of, depending on whom you believe, the Internet’s very soul.

I should say up front that I don’t have a current position on the merits of SOPA and PROTECT IP.  I am not in a position, in Siskel-and-Ebert-like fashion, to offer a thumbs up or down on these bills.  There is a great deal going on here and I am sure that my understanding is only partial.  And the bills are a moving target, changing as more commentary comes in.  My job is to try to advance my understanding and sometimes to do that in public in an effort to push forward more general understanding.  Think of this as a progress report on those efforts.

I have been trying to understand the arguments through the eyes of the Internet engineers and have focused on two documents.  The most recent is a Dec. 15, 2011 letter signed by 83 Internet inventors and engineers and the second is a May 2011 white paper on “Security and Other Technical Concerns Raised by the DNS Filtering Requirements in the PROTECT IP Bill.”  And in looking at those documents, I have focused, at least for this post, on trying to separate technical claims from behavioral and political claims on just one issue, namely, how foreign governments that want to engage in censorship will respond to U.S. efforts to implement DNS filtering to prevent copyright infringement.

It is always interesting to be a tourist in another country.  Take two statements from the December letter: “When we designed the Internet the first time, our priorities were reliability, robustness and minimizing central points of failure or control.”  And: “We cannot have a free and open Internet unless its naming and routing systems sit above the political concerns and objectives of any one government or industry.”

Both of these are fundamentally lawless, the second even more so than the first.  I don’t mean lawless in the sense of bad actors, but much more literally: without law.  Taken as a general principle, the first statement suggests that firms offering new products or services should do so without regard to whether applicable law imposes requirements or not.  Law just isn’t – shouldn’t be? – a focus in building new products.  The second statement really does seem to contemplate that governments – democratically constituted or otherwise – will simply cede the Internet field to a self-appointed technocracy.

This is a statement of Internet exceptionalism, a claim that this is an area that should exist outside of law.  A law-free zone.  You can imagine my surprise, as a law professor, at that idea.  I am a little unsure if my reaction is merely parochial – “What?  Something that we don’t control?” – or genuinely reflects the notion that we are a society organized by laws and hence it really is in the most basic sense illegitimate to contend that a particular activity should simply exist outside of law.  And of course, the Internet doesn’t exist outside of law today as Internet firms receive the benefits of a number of safe harbors, such as 47 U.S.C. 230 and 17 U.S.C. 512.

Much of the discussion of SOPA focuses on the way in which U.S. efforts to protect intellectual property on the Internet will enable other governments to engage in broad censorship.  Call this censorship spillover or a censorship externality.  The claim is that the United States will forfeit its moral authority to oppose the censorship of free speech around the world if the United States uses a similar capability in the name of preventing IP infringement.

Consistent with the December letter, the May white paper suggests that DNS filtering may already be taking place in the Middle East and China.  That takes place on the existing infrastructure of the Internet and doesn’t seem to be something that the United States can prevent in the current design of the Internet through some sort of anti-censorship hardwiring.  Instead, the fear is again one of censorship spillover: “In addition, if the U.S. mandates and thereby legitimizes DNS filtering, more countries may impose their own flavor of DNS filtering.”

The U.S. State Department rejects that position.  In an Oct. 25, 2011 letter to Rep. Howard Berman, Secretary Clinton stated that the “State Department is strongly committed to advancing both Internet freedom and the protection and enforcement of intellectual property rights on the Internet.”  She saw “no contradiction between intellectual property rights protection and enforcement and ensuring freedom of expression on the Internet.”

I think the core idea at work there is straightforward.  Consider prisons.  The United States puts people in prison who commit serious crimes such as armed robbery, burglary, and murder.  China may put political dissidents in prison.  No one would contend, I assume, that the United States should stop putting serious criminals in prison even though the Chinese are jailing political dissidents.  One tool, used legitimately here and, in our view, the same tool used illegitimately in China.

At one level, this is the structure of the censorship-spillover argument.  If the United States blocks websites that engage in broad copyright infringement, then China will squelch political speech and no one will be able to separate the activities of the United States and China.  I would hope that argument would fail if we applied the logic to the prison example and would fail here as well.

We need to separate out the censorship-spillover claim from a possible claim regarding anti-censorship hardwiring.  We can’t prevent China from building prisons by not building prisons in the United States.  Prison-building technology doesn’t work that way.  This isn’t a world in which we can only have prisons in both the United States and China simultaneously or prisons in neither.  Even if the United States stopped putting felons in prison, China could still imprison dissidents.  In contrast, if we could build an Internet infrastructure that prevented censorship of political speech in China but the only way to do that was to accept broad copyright infringement in the United States, then we would face a real trade off.

As I parse the December letter and the May white paper, I don’t think that the claim being made by the technologists is that SOPA breaks a current scheme of hardwiring that prevents censorship around the globe.  I also don’t understand us to have on the table my hypothetical Internet in which there is hardwired protection for free speech in all countries.  Instead, censorship is being done now on the existing infrastructure and, according to the white paper, at the DNS level.

As I read the two documents, on censorship spillover I see a behavioral and political claim about how states act and how they attempt to legitimate their acts – not a technological claim.  I am confident that there would be a range of views on that question, but I wouldn’t expect evaluation of that claim to be the exclusive province of an IETF working group.  As a law professor, unsurprisingly, I would instead expect us to turn to Congress and the State Department.

So is the censorship-spillover claim a technological claim?  Is it instead a behavioral and political claim?  If the latter, who should assess that claim?

[Thanks to James Grimmelmann for thoughtful and quick comments.]