William H. Neukom Professor of Law
Stanford university Law School and Director of the Stanford Program in Law, Science, and Technology
September 28, 2005
Protecting Copyright and Innovation in a Post-Grokster World
Testimony Before the Senate Committee on the Judiciary
September 28, 2005
Professor Mark A. Lemley, Stanford Law School
The Need for Balance
Copyright owners fund the creation and distribution of new works of art, music and literature that make the world a better place. Copyright owners are rightly concerned with infringement in the digital environment. In the past ten years, Congress has created a wide variety of regulations designed to help solve this problem. Those laws have helped, and yet the problem remains. Copyright owners accordingly want Congress to do still more to help them reduce copyright infringement.
Technology companies create innovative products that make the world a better place and drive a significant fraction of the U.S. economy. Unlike copyright owners, they have not regularly sought help from Congress. Instead, they for the most part want to be left alone to innovate and sell products.
Unfortunately, those interests increasingly come into conflict. Copyright owners increasingly sue not the people who actually engage in infringement, but who provide goods or services that can be used by others to infringe. They have cast their net ever wider in their effort to stop infringement, suing not only those who provide services directly connected to infringement, like Napster, but makers of consumer electronics devices that play music, sellers of software that can be used to infringe, Internet service providers, search engines, telephone companies who own the wires, venture capitalists who fund companies that do these things, and even law firms that advise them.
Copyright owners are attracted to suing these intermediaries because it is cheaper and easier than suing the millions of American consumers who download music without paying. But doing so comes at a cost. If writing software, or making a consumer electronics device, or running a search engine, can subject someone to statutory damages that can run into the billions of dollars, or perhaps even criminal liability, innovators and entrepreneurs will think twice before developing or marketing a new product that might subject them to a lawsuit. And since the courts have interpreted copyright law not to have any corporate veil, someone who runs or simply works for such a company could lose their house and their family's retirement fund. The threat of a lawsuit will deter not just innovators developing technologies with illegal uses, but those who develop technologies with both legal and illegal uses and those who don't yet know how the market will use their technology. The list of such dual-use technologies is long and distinguished: broadband Internet service, the iPod, TiVo, CD burners, and computers themselves, to name just a few.
While reducing copyright infringement is an important goal, it cannot and should not be the only goal of public policy. Congress should also be concerned that overzealous enforcement of copyright will create a hostile environment for technological innovation and entrepreneurial business models. It should strive to balance these important interests, providing effective copyright protection but also preserving an environment in which innovation can thrive.
Nor can Congress simply rely on assurances from the copyright industry that they will foster innovation themselves, or target only "bad" and not "good" innovations. The content industry has proven short-sighted, time and again trying to stifle technologies that ultimately proved beneficial not only to society but even to copyright owners. They tried - and fortunately failed - to shut down jukeboxes, radio, cable television, the VCR, and the mp3 player. Perhaps it should not surprise us that publicly traded companies should have a short-run focus, looking at this quarter's bottom line and not what will benefit society in the long run.
Further, the content industry is not monolithic. Innovators need to worry not just about what the major music and movie studios will do, but also what a variety of fringe content owners and tens of thousands of writers will decide to do. Many of the cases that demonstrate the worst overreaching - suits against search engines, for example - are brought not by the major studios but by pornographers or idiosyncratic individuals. If copyright law gives those individuals the power to stop innovation or to control its direction, new technologies will exist only if all of those thousands of content owners sign on. The difficulties legal music services such as iTunes have faced in trying to clear all the necessary licenses for content are but one example of the problems that broader copyright protections can create.
The Uncertain Impact of Grokster
The legislative landscape changed significantly in June, when the Supreme Court decided MGM v. Grokster. In that case, the Court created a new cause of action for inducement of copyright infringement previously unknown in copyright law. In doing so, the Court sought to hold liable particular peer-to-peer (p2p) software providers that it viewed as bad actors without exposing the entire technology industry to the debilitating risk of liability that would come from abandoning or limiting the Sony safe harbor for devices capable of a "substantial noninfringing use." In short, it sought to achieve the balance I have just described.
Whether the Court succeeded in creating a middle ground remains to be seen. Much will depend on how the Court's open-ended, multi-factor test for improper purpose is interpreted in the lower courts. Much will also depend on how far copyright owners seek to take the new doctrine, and whether they overreach. For this reason, it is premature to propose legislation to correct deficiencies in the new inducement test.
Nonetheless, three questions left open by the Grokster opinion bear close scrutiny, and may ultimately require correction or clarification by Congress.
1) What is required to prove improper purpose? The Grokster opinion found evidence of bad intent on the basis of allegations that the p2p defendants were founded in order to reach former Napster users, most of whom were engaged in copyright infringement; that the defendants did not redesign their software to minimize infringement; and that the defendants, like most companies, made more money the more users they had, and therefore had a motivation to permit or encourage infringement. A narrow reading of the Court's opinion would require the confluence of all or almost all the factors at issue in Grokster. On this view, only a company whose business model was designed to promote infringement need worry about inducement liability.
By contrast, a broad reading might find liability based on some subset of the three, such as making money from software knowing it was used by infringers. If courts do read Grokster so broadly, it would create real problems for legitimate technology companies, many of whom could be threatened with suit on the basis of conduct that is as consistent with legitimate as with wrongful intent. Should that happen, Congress may need to intervene in order to protect innovation.
2) What conduct is required? The Grokster opinion focused its primary attention on the question of the p2p defendants' purpose. It finessed the question of what conduct must be coupled with an improper purpose in order to find infringement. At various points, the Court made it clear that to be liable for inducement a defendant must do more than simply sell a product capable of both legal and illegal uses with bad intent. It recited conduct it believed the defendants had engaged in, including advertisements allegedly targeting infringers and the provision of tech support to people who were infringing. The requirement for some affirmative conduct that induces (ie. causes) infringement by others follows naturally from the patent standard, which clearly requires affirmative conduct beyond simply the sale of a product. [See my draft article, "Inducing Patent Infringement," attached]. And since the Court unanimously reaffirmed the Sony "safe harbor" in footnote 12, it would make little sense to interpret the new inducement test in a way that would effectively overrule Sony.
But at other points in the Court's opinion, the Court suggested that the sale of a lawful product standing alone might suffice to prove inducement if were coupled with bad intent. If courts read Grokster to cover sales alone, they will doom any innovator to the risk of liability whenever a plaintiff can find a memorandum from the marketing department or conduct by tech support staff that could support a finding of bad purpose. No matter how valuable the product, the technology company would risk a lawsuit if any copyright owner decided it wanted to sue. Should courts take this route, Congress may need to intervene in order to protect legitimate technology companies.
3) What state of mind must exist regarding infringement? The final question is what a defendant must intend - merely the acts they encourage, or actually to promote infringement? The question didn't arise for the Court in Grokster because it seemed clear that p2p file sharers were in fact infringing. But in other circumstances, companies will intend to encourage copying but believe that that copying is legal, for example because it is a fair use or has been authorized by the copyright owner.
In such circumstances, it would not make sense to hold the company liable. As Jonathan Band writes in the Computer and Information Lawyer:
Thus, it is unlikely that a person can be liable for inducement if he advertises a use that he reasonably believes to be a fair use, but which turns out to be an infringement. The language of the opinion suggests that liability should attach only if the defendant had the specific intent to cause infringement: "the object of promoting its use to infringe," id. at 2770, "their principal object was use of their software to download copyrighted works," id. at 2774, "an actual purpose to cause infringing use," id. at 2778, "statements or actions directed to promoting infringement," id. at 2779, "purposeful, culpable expression and conduct," id. at 2780, "a message designed to stimulate others to commit violations," id., "acted with a purpose to cause copyright violations," id. at 2781, "a principal, if not exclusive, intent on the part of each to bring about infringement," id., "intentional facilitation of their users' infringement," id., "unlawful objective," id., "the distributor intended and encouraged the product to be used to infringe," id. at 2782, n.13, "a purpose to cause and profit from third-party acts of copyright infringement," id. at 2782, and "patently illegal objective." Id.
Further, as I explain in my attached article on inducing patent infringement, the majority view of the courts in patent law requires intention to cause another to infringe, not simply intention to cause copying that might or might not be infringement. It seems logical to understand copyright law the same way, since Grokster created copyright inducement from the patent analogy.
If the courts get this wrong - if they interpret Grokster to make illegal any intent to encourage copying, even if the defendant believes that copying to be legal - the consequences for companies like search engines could be significant. In that case, Congress may need to intervene in order to protect innovation.
The Problem Won't Go Away Because of Grokster
The above comments make it clear that Grokster is not a perfect solution for technology companies. Interpreted narrowly, to cover only businesses purposefully built on supporting infringement, it is a rule that most technology companies can live with. But if copyright owners try to expand the reach of the opinion they will create an environment in which innovation is difficult if not impossible.
Nor is Grokster a panacea for the copyright owners. While it provides a new legal tool they can use against some p2p companies, it will not eliminate copyright infringement in the digital environment. It won't eliminate p2p file sharing, and indeed the Supreme Court was careful to note that p2p technology itself was a good thing. Nor can inducement liability be used against infringement through decentralized software systems not run by a company that can be found liable for inducement.
As a result, neither the problem of copyright infringement nor the threat to innovation posed by overzealous copyright enforcement will go away. Congress and the courts will have to continue to try to solve one problem without making the other worse.
While in my view it is too soon to propose legislation dealing with Grokster's new inducement standard, there are a number of things Congress might do to help ease the burdens the current situation places both upon copyright owners and innovators.
1) Make it easier for copyright owners to target direct infringers. Part of the reason copyright owners target intermediaries and innovators is that they find it difficult to find and target the people who are actually doing the infringing. Anything Congress can do to help stop the direct infringement will relieve the pressure on innovators. In an article published earlier this year entitled "A Quick and Inexpensive System for Resolving P2P Copyright Disputes," Tony Reese and I propose to use the Copyright Royalty Judges Congress created last year to administer a quick, simple and cheap system for identifying and punishing high-volume illegal file traders. [A copy of the article, including proposed legislation, is attached]. If copyright owners have an effective way of targeting direct infringers, they will have less desire or inclination to go after intermediaries in a way that threatens innovation.
2) Make it easier to clear rights in the digital environment. Consumers want online music and other content. The unwillingness of copyright owners to fill that need in the late 1990s and the early part of this decade contributed to the problem of rampant copyright infringement. Conversely, copyright enforcement will not be effective unless it is coupled with a cheap, easy, legal alternative. While third parties like Apple have made significant strides in the last two years in putting legal music online, they are hampered by divided ownership of rights and a labrynthine set of rules that make it difficult even for people who want to make content available online to do so. Congress has begun considering how to simplify these rules. I urge you to continue this effort, and to resist efforts by special interests to complicate the ownership or licensing rules in order to protect their historic piece of the pie.
3) Insulate technology companies from unreasonable liability. The remedies currently provided in copyright law are written with direct infringers in mind. They provide a windfall to copyright owners in order to deter infringement by providing statutory damages and even criminal liability. Those remedies make sense when applied to direct infringers. They make considerably less sense when applied to companies that are not themselves infringing copyright, but merely providing a product or service that others can misuse. This is particularly true in the digital environment, since because of an accident in the way statutory damages are calculated anyone who is found liable for indirect infringement on the Internet faces liability of billions of dollars. If you are at risk of losing your whole company (and your house and your kids' education), even a very small chance of liability will be enough to deter valuable innovation.
Congress could relieve much of the pressure copyright law puts on innovation by limiting liability for indirect infringement to the actual damages caused by any such infringement. This would compensate copyright owners for their losses, and force indirect infringers to bear the cost of any harm their conduct causes, but would not over-deter innovation. And since copyright owners could still sue direct infringers for statutory damages, deterrence of infringement would remain intact.