We are pleased to publish this guest post on the lessons of the SOPA/PIPA/Megaupload fight by Yochai Benkler, Berkman Professor of Entrepreneurial Legal Studies at Harvard, faculty co-director of the Berkman Center for Internet and Society, and author of The Wealth of Networks and The Penguin and the Leviathan.

Lesson 1: The Networked Public Sphere comes to Washington.

On Wednesday, January 18, 2012, a new model of politics succeeded in bringing to a halt legislation that had been pushed by some of the most powerful industry lobbies in Washington, which began its life with broad bi-partisan support in both chambers of Congress. The Stop Online Piracy Act (SOPA) and the Protect IP Act (PIPA) were to be the most significant changes in intellectual property law since 1998, when a slew of laws, most importantly the Digital Millennium Copyright Act (DMCA) and the No Electronic Theft (NET) Act, first set the terms of engagement between the twentieth century incumbent cultural industries—mostly Hollywood and the recording industry—and the new forms of culture and creativity in the networked environment. By Friday January 20th, as Harry Reid announced that PIPA was not going to the floor of the Senate, Chris Dodd was raising a flag of truce on behalf of the Motion Picture Association of America (MPAA); suddenly, in the face of overwhelming political power the likes of which he told the New York Times he had not seen in forty years, the MPAA was ready to talk. But he wanted to talk to the tech industry; ignoring the fact that while the industry played a critical role in the opposition, it was the coalition combining for-profit and nonprofit; organized and decentralized; social-networked and blog/twitter-based action that successfully blocked the power of concentrated money on K Street.

It's not up to the tech industry to negotiate on behalf of the millions of people who came to care.

Sure, Silicon Valley upped its lobbying game, and that was an important part of the story. But the turning point came when people who cared harnessed the network to concentrate human action. Whether it was the consumer boycott that put pressure on GoDaddy to abandon its support for the legislation, which caused other technology firms to shy away from supporting it; the constituents writing letters to their senators and representatives; or the remarkable democratic debate among hundreds of committed editors that preceded Wikipedia's decision to shut down for a day, what we saw was that even in this day, when money is so powerful in American politics, people acting in concert can have a real impact.

For almost twenty years, the copyright industries have pushed hard against weak opposition, and extended the scope, reach, and aggressive enforcement of copyright to contain networked technology and resist networked culture. The political calculus seems to have changed drastically this week, and we need to understand how to exploit and harness the changing winds to expand and lock in this initial victory.

Lesson 2: Hollywood and the recording industry don't like traditional copyright law, balanced by courts under due process constraints.

Copyright seems to be too balanced for the industry's taste. Traditional copyright law has too many balances; too many reasons judges might prevent Hollywood from just shutting the whole thing down so people can be made to sit quietly on their couches and pay up. The bills were designed to try to create new pressure points that would allow either copyright owners or their associated functionaries at the Justice Department to kill threatening sites, without having to go to the trouble of identifying specific infringements or proving anything to a court. From the very beginning, in September of 2010, the first Senate bill, the Combating Online Infringements and Counterfeits Act (COICA), each successive version of the Act tried to avoid the inconvenience of actually having to prove that the site being targeted violated copyright law before inflicting mortal damage on that site. In that earliest version, the Justice Department was supposed to create a blacklist of “bad actors” by mere allegation; no proof necessary before these sites would start to be blocked. By the time SOPA was introduced, that power was directly granted to copyright owners for blocking payment systems and advertising, and again to the Justice Department, with slightly higher constraints, for blocking DNS service to the sites. Fundamentally, the aim of these laws was to replace the balance of copyright with a unilateral power to hobble or shut down whole sites suspected of helping piracy before a final determination of actual copyright liability, subject to all the balances that copyright has traditionally required between the rights of copyright owners and the rights of the public and later creators to use and build on the culture in which they are immersed.

Lesson 3: As the networked environment resists control, more of the flow of networked economy has to be sucked in to the enforcement vortex.

The Net is proving much harder to control than the industries anticipated when they got the Digital Millennium Copyright Act DMCA passed in 1998. In order to actually control materials on the Net, SOPA and PIPA tried to harness a range of technical, economic, and bureaucratic platforms, aimed to impede the functions of an ever-more-vaguely defined set of targets. Technical platforms included most prominently the DNS service and registrars and the search engines. Business platforms included payment systems and advertising systems. In order to achieve effective enforcement in a global digitally networked environment, Hollywood seems destined to try to draw an ever-larger set of platforms and actors into the risk of potential copyright and near-copyright liability.

Lesson 4: Freedom of speech is bound with freedom in the network. It is impossible to achieve the kind of control Hollywood wants without severely undermining the freedom the Internet makes possible.

Because control is hard and IP enforcement pulls in more of the platform providers and technologies into its enforcement effort, it necessarily makes the Net more controllable. We saw this most clearly in the DNS filtering provisions that were perhaps the most frightening of the provisions, and those that raised the most aggressive opposition. Freedom to develop and use new, control-resistant, decentralized technologies is absolutely central to what made the Internet the platform of freedom that it has become. As long as the copyright industries insist on trying to reshape the Internet to make it controllable in order to serve their interests, there effort will be fundamentally in conflict with Internet Freedom. To understand this, we need but quote Larry Lessig's decade-old “code is law,” or Eben Moglen's memorable “Freedom of the press, freedom of information, freedom of thought itself are now 'implemented' rather than 'declared', 'protected' or 'guaranteed'.”

Lesson 5: Criminalization has become a Golem/Frankenstein, hunting technological innovation through its own internal bureaucratic logic.

One of the most brilliant moves the industry made over the course of the past decade has been to lobby for, and obtain, the creation and funding of various positions within the Government whose career success depends on harnessing the resources of the federal government to carry out critical parts of the industry's assault on threatening technologies and the more aggressive users. This began in 1998 with the DMCA and the NET Act, which fundamentally shifted criminal copyright from being law aimed at commercial enterprises that actually stamp out thousands of illegal copies, to something that could sweep any teenager who swapped files too enthusiastically. We first saw the effects of this strategy when the Justice Department dramatically increased the number of criminal copyright infringement actions it brought in tandem with the recording industries surge of civil suits against individual file sharers, from 2004 to 2008.

Criminalization got a major push from the PRO-IP Act in 2008, with the expansion of funding for more criminal enforcement and coordination, the creation of an IP czar in the White House, and the importation of forfeiture law from drug enforcement to IP enforcement. We saw the result of this in the crackdown in Megaupload on the day after the major Internet anti-SOPA protests. There is a large number of functionaries throughout the federal government, most importantly the Justice Department's Computer Crime and Intellectual Property Section, a White House IP czar, and an IPR Center housed in the Department of Homeland Security, whose professional success, irrespective of the policy position of any given administration, are measured by (a) how threatening we think Piracy is, and (b) how many large prosecutions they are able to bring.

The creation of these bureaucratic incentives means that even if Congress now passes no more expansive laws; even if the industry shifts away from business models that are actually sensitive to copying, as the recording industry already has and movies are well on their way to doing, prosecutions against companies too small and too new to be “too big to fail” will continue apace. How else will the bureaucrats justify their existence? As the GAO found in its 2010 study of the costs of piracy, these agencies do not really create credible numbers of their own on the size of the real threat; they rely unquestioningly on industry studies. As for these studies, the GAO politely said that they were bunk. The truth is we simply do not know whether “piracy” is a significant problem at all, and the growth and success of the industries it supposedly harms belays that claim more effectively than any theoretical or methodological critique of the argument that piracy is important.

In between January 18’s protest and January 20th’s surrender by Dodd, on Thursday, the Justice Department showed that it didn't need any new powers to shut down and seize the assets of Megaupload in Hong Kong and New Zealand. Putting together an indictment made of what lawyers called a parade of horribles, including extensive insinuations of illegality attached to quite plausible and legitimate design decisions of cloud-based backup and file-transfer sites alongside descriptions of the less-than-savory characters actually behind Megaupload, the Justice Department seized assets and obtained detention of the company and its principals.

Lesson 6: The Supreme Court's alteration of the Sony standard in Grokster, coupled with the new seizure powers passed after it, create genuine risks to technological innovation.

While Megaupload may be run by unsavory characters, the prosecution of the site, and even more so, the seizure of its assets before any of the prosecutors' allegations have been proven, is a warning call that the logic of the Criminalization Golem is already on the move. Let's be clear. It is almost certain that under the traditional law of contributory copyright infringement as the Supreme Court laid it out in the Sony Betamax case, Megaupload would not be liable for even civil copyright infringement, much less criminal. There, the Supreme Court refused to succumb to Hollywood's panic of this new technological threat, the VCR, and held that as long as the technology has substantial non-infringing uses, its creator would not be held liable for the fact that users were also using it for infringement. Seized in the post-Napster panic, the Supreme Court revised that holding in its Grokster case in 2005, adding a subjective “inducement” arm to the test: if there was evidence that the site was inducing infringement, it could be held liable. It was apparently in an effort to fit this arm that the prosecutors in the Megaupload indictment used various technical features of the site, which have perfectly legitimate uses—such as allowing publication of a public link to a file; avoiding duplication of storage by generating separate links, for separate users, to files; the absence of a search function; or the development of an anti-child pornography filter but not one for (the less deterministic fact of) copyright infringement as evidence of nefarious intent to induce copyright infringement.

The result has been that at least some legitimate online storage sites have begun to look at their legitimate features to see what could be interpreted by a court reading a one-sided indictment as it issues a temporary seizure order as evidence of “inducement.” Given that the Supreme Court's own opinion in Grosker suggested that simply using the suffix “-ster” was useful evidence of intent to induce infringement, and that, following suit, several the elements included in the Megaupload indictment have innocent uses, the chilling effect on new, small sites of the risk of having everything taken away on the basis of an indictment (and remember, Grand Juries provide no meaningful protection; they merely offer a veneer of legitimacy to what are, at root, unproven accusations). By the time they have had their day in court, their business is destroyed; their user-base moved on to the next site.

Where do we go from here?

SOPA/PIPA/OPEN are not the proper baseline for the new settlement.

We are already hearing strong appeals to develop a compromise that is somewhere between SOPA/PIPA and the Online Protection and Enforcement of Digital Trade Act (OPEN) introduced by Representative Darrel Issa and Senator Ron Wyden. While Sen. Wyden and Rep. Issa did a fantastic service to the Internet in holding the line and presenting an alternative when Hollywood's Juggernaut was still making the sound of inevitability, the political sands have shifted. In the veto-rich system of the American legislative arena, SOPA and PIPA are dead and can be kept dead indefinitely after the lesson the people taught Congress and the lobbyists last Wednesday. Hollywood needs no more power to control the network, and the damage that will be done to freedom from any further extensions is too great to justify, given the uncertain and unproven scope of the “piracy” problem.

The OPEN Act avoids the very worst of SOPA/PIPA: it creates a process whereby an administrative inquiry and adjudication would precede any shutdown; it tightens the definition of the targets of actions, and it limits the range of actors drawn into enforcement proceedings to payment providers and advertising providers, critically leaving the DNS service alone.

But OPEN still buys in to the basic assumption that the interests of Hollywood justify drawing an ever-growing universe of complementary systems, like payment and advertising, into their enforcement dance against the piracy menace, without evidence that that menace is significant enough to justify the expansion. It cannot prevent the International Trade Commission from becoming just another captured agency doing the bidding of the people they perceive as their clients; and it retains the model of creating immunity for the critical services, in this case payment and advertising, for “voluntarily” cutting off services on suspicion of infringing activities, setting up the conditions for the creation of a private blacklist.

Given our long experience under the DMCA and NET Act with how aggressive the copyright industries have been in their enforcement efforts, and how uncertain interpretation of legislation can be after it is passed; given the strong signals SOPA/PIPA sent that the industry is continuously seeking new ways of short-circuiting the balances of traditional copyright law to gain ever-more control over networked creativity so as to prevent any possible spillage, there is no reason to give Hollywood yet a new platform with which to challenge innovation and networked creativity. At a minimum, if international piracy and these new means of enforcing against it are indeed so important to the industry, then adding these kinds of provisions should only come as part of a renegotiation of the whole of the system, to roll back some of its worst excesses.

The starting point for negotiation cannot be that everything the industry got while networked citizenry was weak and dispersed is sacrosanct, and the only things on the negotiating table are Hollywood's shiny new regulatory toys. The politics have changed. Everything should be up for renegotiation, or we should use the blocking power of the network in conjunction with the veto-rich environment that is the American legislative system to prevent any additional creep from today's baseline.

There is an opportunity to harness the new political energy to reverse core institutional elements of Hollywood's decade-and-a-half-long land grab.

The newfound political power needs to be directed not only at these most recent excesses of the persistent efforts by the twentieth century incumbents to set the terms of cultural exchange in the twenty-first century; it needs to be directed more fundamentally at preserving the freedom of expression and a freedom to operate in the networked information economy and society.

As a practical matter, as Hollywood beats a disorderly retreat from law week's battlefield, we need to be thinking not about what compromises to make around SOPA/PIPA and the OPEN Act, but about what the architecture of freedom in the networked environment requires of copyright law more generally. We need to develop an actionable legislative agenda.

Here are four core targets that I think should be part of such an agenda. Others obviously will have different ideas; but at least I hope we can focus the conversation of where to go from here on the basic strategic interventions we need to make to de-regulate innovation on the Net and reverse the dynamics whereby securing the business model of twentieth century content industries threatens freedom, creativity, and innovation in the twenty-first.

• Legislatively re-instate the Sony doctrine and reverse Grokster. Technology developers should only be liable for copyright infringements by users if there are no substantial non-infringing uses of the technology.
• Decriminalize copyright to per-1998 levels: put the Golem to sleep. Return the definition of criminal copyright to require large scale copying for commercial gain; reduce the funding to criminal enforcement and reduce the presence of federal functionaries whose role is to hype and then combating the piracy threat. In particular, as calls to shrink the federal government abound, it is critical to include in every legislation downsizing the federal budget provisions that would defund and eliminate most of the burgeoning apparatus of multi-agency criminal enforcement of copyright. The most direct pathway to this will be in appropriation bills, to defund implementation of PRO-IP until a more balanced substantive approach can be worked out.
• Create a fair use defense to the anticircumvention and antidevice provisions of the DMCA. Users should be exempt from DMCA liability if they propose, in good faith, to make a fair use of the encrypted materials. Decryption and circumvention providers should be exempt from liability on the model of the Sony doctrine, if there are “substantial non-infringing uses” for the circumvention technology or device they offer. This would fix a much older overreach by the industry, from 1998, that has been very slowly and imperfectly loosened by the Librarian of Congress under powers to exempt certain uses from liability.
• Rein in the international trade pathway for copyright extension. Another pathway, similar to criminalization in the sense that it harnesses federal functionaries to help the industry, distinct in the set of functionaries it harnesses, has been international trade. Through a set of trade agreements, both bilateral and multilateral, the U.S. government has pursued the passage of requirements more stringent than it could itself pass in the U.S. The recent adoption of SOPA-like laws in Spain is one example, as is the notorious Anti Counterfeiting Trade Agreement (ACTA). We need a law that would prohibit secret negotiation of IP-related provisions in international agreements, and a law that prohibits the U.S. from entering agreements that require of ourselves or our trading partners more restrictions on the public domain than then-current U.S. law permits.

The American political system has strong ratchet effects. It takes a lot less power to resist legislation than to pass it. This time, the mobilization was powerful enough to stop the ever-expanding IP ratchet from moving one more notch up. Future battles, whether precisely along the lines I propose here or along other lines, need to take advantage of the ratchet effect for positive, as opposed to merely blocking, purposes. We need to develop a distinct agenda, and focus mobilization efforts around ratchet points—institutional, organizational, or substantive law—so that the mobilization can lead to ratcheting back controls, no less than it was used over the past two decades for ratcheting up these control in the name of securing the copyright industries past business models.

Lesson 7: The moral authority of the networked citizenry vs. the power of money.

The victory of last week was an exhilarating example of citizens coming together and pushing back on the interests of a well-funded industry lobbying campaign. It's too soon to tell yet whether what happened was that elites in the tech industry used their platform power to harness the population to their sectoral interest, or whether the citizens of the Net harnessed major players in the technology industry to gain greater freedom. Probably what happened was a little bit of both. But coalitions are always a bit messy, and it is often hard to tell whose agenda is really gaining the upper hand. For now, at least, it seems that the interests of the core of the technology industry and the interests of individuals and communities that have come to rely on the Internet as their primary platform for freedom and creativity are sufficiently aligned that the coalition can hold for a while, on a range of issues, including all those I raise as targets for common action. But networked citizens cannot and should not be sanguine that the alliance will always be so smooth. Nor should the industry take support for granted, or assume that it can rely on its own power and lobbying.

The power we saw in the hands of networked people is a fundamentally more legitimate source of power than corporate money. Democracies are by and for the people. We believe in one-person, one-vote; and while corporate organizations are enormously useful, and can make us more effective in the pursuit of our life plans and dreams, at root it is us, human beings, flesh and blood, who are the foundational constituents of a democracy. That is why Wikipedia played such a critical role: unlike all the other major sites that shut down. Wikipedia is not a company; Wikipedia, for this purpose, functioned as a mini-democracy within a democracy, where people who continuously volunteer for the public good came together to do something new for the public. Wikipedia represents a moral force that no commercial site can ever hope to replicate. Some sites, like Reddit, are sufficiently based on users that they can structure their future protest actions as democratic debates, letting users decide. Extending the debate and collective decision-making feature of the Wikipedia decision to other platforms should play an important role in the future, and will also help to solidify the alliance between networked citizenry and the companies that provide the infrastructure of networked discourse. If the technology industry wants to continue its battle with Hollywood as a battle among paid lobbyists, it may do so, likely at its own peril. But if the industry wants to be able to speak with the moral authority of the networked public sphere, it will have to listen to what the networked public is saying and understand the political alliance as a coalition.

The greatest hope from the events of the past two weeks is that we are beginning to see a re-emergence of the possibility of a truly engaged citizenry after decades of the rise of lobbying and money. I suspect that it is too soon to go after legislative changes that target that ambitious goal directly, as Micah discussed yesterday ("After SOPA/PIPA Victory, Tech is Thinking of Tackling Political Reform.") But if we can use the enthusiasm and focus to make significant inroads in this narrow and specifically actionable item, perhaps we will also begin to hone a more general a new model of democratic participation for a new generation. A model of citizen participation that is as far from the couch potato's passivity as the Internet is from broadcast.