Draft as of March 1997

Made available to CSTB IP study committee for its work

Please do not cite, copy, disseminate, or disclose to others without permission of the authors

TECHNOLOGICAL PROTECTION FOR COPYRIGHTED WORKS 

by

Pamela Samuelson^* and Benjamin D. Black^**

"Something there is that doesn't love a wall."

INTRODUCTION

Many copyright owners feel threatened by digital technology because works in digital form are vulnerable to uncontrolled replication and dissemination in global networked environments. Digital technology is, however, not just part of the problem; it may also be part of the solution. IBM, as well as numerous start-ups, are actively developing digital technologies that control access to, and monitor the usage of, copyrighted works in order to control digital copyrighted works in a manner that is not currently feasible in the print. The creators of electronic copyright management systems ("ECMS") foresee a time when technology will allow a copyright owner to control, sell and license practically every use of a digital work. More generally, "access control technology," like encryption or password systems, will be widely used by anyone seeking limiting access to digital works, regardless of whether the copyright owner is facilitating a transaction with a consumer. These technologies hold much promise as a way to relieve copyright law of the burden of regulating a wide range of uses of digital information products, most of which, by virtue of the highly decentralized nature of existing digital networks, would be difficult for the law to regulate effectively.

Although prospects for technological protection for copyrighted works are bright, technology is not a panacea. Copyright owner’s principal concern is that what one technology can do, another technology can generally undo. Although domestic court decisions currently permit copyright owners to control the manufacture and sale of technologies having no substantial use except to enable infringement, some publishers regard this rule as inadequate to protect their works against technological pirates. In response, the Clinton Administration has sought to amend the copyright law to prohibit the act of circumventing the technological means that copyright owners choose to protect their works and to outlaw the development and dissemination of products or services that third parties could use to circumvent these technological protection systems. This so-called "anti-circumvention provision" enables copyright owners, like publishers, writers, software companies and movie studios, to bring a civil cause of action under the Copyright Act when these new rights are violated. This Article analyzes the policy issues raised by this proposed legislation.

The anti-circumvention provision is part of a recent spate of legislative activity responding to the perceived threat digital technology poses to the economic interests of copyright owners. These proposals seek to create novel and expanded forms of copyright liability, including legislation intended to impose liability on on-line service providers (OSP) and new database protection, in addition to recently enacted new criminal copyright liability for purely non-profit activity. The unusual level of recent activity is indicative of the passions that digital technology has unleashed in the content communities.

The anti-circumvention provision is the result of this passion and it is novel as well; it uses copyright law to protect access to a copyrighted work rather than the original expression contained in a work. This shifts represents a fundamental change in the nature of copyright protection. It represents copyright liability in the absence of a copy. Some scholars have coined the term "paracopyright" law to describe these new forms of copyright liability. This term, however, arguably obfuscates the true import of the proposed anti-circumvention provision because the prefix "para" implies that the new law merely goes beyond that which already exists. Rather, "pseudo-copyright" law might be a more apt term because the provision creates a new right that does not require a copy to violate: the right to prohibit access to copyrighted works without the authority of the copyright owner and to outlaw certain access-enabling technologies.

Copyright owners want to use ECMS, in conjunction with contract law, to control each and every use of a copyrighted work in digital form and to charge for each use. Access to the work will be controlled on the terms established by the copyright owner through electronic licenses a user must agree to prior to gaining access to a work, which state contract law will enforce. Lawrence Lessig, among others, has recently warned that access control technology has the potential to privatize copyright law, making the exercise of user rights in cyberspace impossible and the threat of copying, negligible. In other words, digital intellectual property may be nearly perfectly protected. In such a world it may be that copyright law will serve little more than as a law that protects access to the copyrighted work, rather than the original expression in the work itself. Contract law and technology will do the rest.

The ability of copyright owners to use technology and contract law to provide perfect protection for copyrighted works would revolutionize intellectual property. Copyright law has never perfectly protected copyrighted works. Copyright has temporal limitations; Article I, Section 8 of the Constitution authorizes Congress only "to promote the Progress of Science by securing for limited times to Authors the exclusive right to their respective Writings" There are policy limitations; the fair use doctrine states that using a work "for purposes such as criticism, comment, news reporting . . . scholarship, or research, is not an infringement of copyright." Copyright does not protect the idea, procedure, process, system, method of operation, or discovery contained in a work, allows libraries and archives to make archival copies, and enables owners of particular copies to sell or otherwise dispose of them. And there are practical limitations; traditionally, a copyright owner could not control the non-infringing personal use of a purchased work—sharing it with a friend, reading a work multiple times, or playing a record until the grooves were worn. In addition, some percentage of infringing activity, primarily private, non-commercial copying, was always beyond the ability of copyright owners to stop. In short, copyright law "reflects a careful, expressly-drawn balance between private (author’s) rights and public rights." The public retained the ability to use copyrighted works in a normal manner, so long as the use did not interfere with the economic expectations of copyright owners.

ECMS, buttressed by an anti-circumvention provision, will enable copyright owners to enjoy all the private rights copyright confers, while denying users the ability to exercise the public rights. We are justified in asking whether this is a good thing. Copyright law is about more than maximizing economic returns for authors. The purpose of copyright is to promote learning and knowledge by encouraging the "widest possible creation and dissemination of literary, musical and artistic works." Some scholars have argued accessibility is a central part of copyright’s quid pro quo; authors receive the economic benefits of copyright protection in exchange for making the copyrighted work accessible to the public. "If access is denied to the public for a work where the author is receiving the economic benefits made available by the copyright system, then the goals of copyright are not being served."

This Article argues that the Clinton Administration’s current approach to regulating circumvention is misguided and may transform copyright’s role in the information society in fundamental and undesirable ways. Part II reviews the recent proposals to regulate circumventions. Part III argues that the empirical premise that underlies legislation prohibiting all circumventions is flawed. Part IV argues that outlawing circumvention enabling technologies is unworkable as a legal regime and will discourage innovation in technology markets. Part V discusses how a broad anti-circumvention provision will transform copyright law from a law that encourages dissemination of information to one that suppresses it.

Copyright law stands very much at a crossroads. The choices we make today will become the status quo of tomorrow. Therefore, our ability to grasp some of the more subtle implications of the choices we make now will pay dividends in the future.

All parties involved in the debate over digital copyright agree that electronic copyright management systems that enable copyright owners to control the commercial distribution of their works are soon to be big business. In order for the Internet to become a large-scale commercial distribution system, content providers want "secure and reliable means for delivering information products and services to consumers."

The current push to regulate circumventions and enabling technologies evolved from the Clinton Administration’s effort to implement the National Information Infrastructure (NII), which is described in the White Paper, and its predecessor, the Green Paper. The White and Green Papers saw the Internet as a prototype for the NII, whose development the Clinton Administration aims to promote. The Clinton Administration originally proposed a broad circumvention provision that prohibited any device or service that had the "primary purpose or effect" of circumventing an access control system that protected a copyrighted work ("White Paper Section 1201"). Notably, the White Paper, and the Green Paper for that matter, in first proposing the provision, were almost completely devoid of policy analysis, relying simply on the potential of mass violations and ensuing costs to justify a blanket prohibition on all circumvention-enabling technologies. The Working Group simply stated that it found, that the existence of enabling technologies would undermine author’s incentive to create, though without specifying the evidence upon which this finding was based. Rather than considering the public policy implications of the act of circumventing in various contexts—such as for news reporting or computer security—the Group concluded that all enabling technologies had to be prohibited. Notably, White Paper did not, in fact, propose a per se prohibition on the act of circumvention, only technology enabling inventions.

When the Administration attempted to implement the proposed legislation to regulate circumventions, it ran into spirited opposition from the academic and technological communities. Some opponents argued that the proposed provision threatened the public domain and fair use; others argued that the proposals posed a grave threat to nascent technology firms that developed products that third parties could use to circumvent copy protection systems. The proposal stalled in committee.

In December 1996, the World Intellectual Property Organization ("WIPO") hosted a conference to craft a protocol to amend international copyright law to reflect developments in digital technology. Undeterred by its failure to win passage in Congress, the delegation from the United States urged the inclusion of a provision nearly identical to the one offered in the White Paper in the major international copyright treaty, the Berne Convention. The Clinton Administration publicly stated that its agenda was to use the international treaty and ensuing ratification debate to get a "second bit of the apple." Once introduced, the proposals met with opposition because of concerns with chilling technology development and undermining the public domain and fair use. After debate, the treaty participants rejected the broad approach advocated by the U.S. delegation. In response to media companies’ that wanted some language in the treaty, the U.S. delegation brokered a compromise provision, Article 11, which was eventually accepted. Article 11 states:

Although some countries used copyright law to regulate the circumvention of copy protection systems, or technologies that enabled infringement and had no substantial noninfringing use, this provision was novel in an international intellectual property treaty. Rather than dictate a particular set of laws or rights signatories had to enact, WIPO delegated the decision on how to regulate circumvention and facilitating technologies to the discretion of the individual countries. Article 11, by its terms, requires a result—adequate legal protection and effective legal remedies—not a particular regulatory scheme.

The debate then moved back to the domestic realm to determine whether and how to comply with Article 11. Within Congress, two competing visions emerged. The Clinton Administration’s WIPO treaty implementation legislation, which Representative Coble introduced in Congress, included a new civil cause of action prohibiting circumvention under the Copyright Act. Like the White Paper, Section 1201 ( "Coble Section 1201") enabled a copyright owner to sue any person who manufactured, imported, or otherwise trafficked in any technology, product, service or device that:

In addition, the Coble Section 1201 went beyond the White Paper by proposing a new cause of action prohibiting the act of circumventing a technological protection system that a copyright owner used to restrict access to his or her work. Further, any person who willfully and for purposes of commercial advantage or private financial gain violates Section 1201 is subject to a five year jail term for their first offense and a ten year jail term for their second.

In the Senate, another voice emerged. Senator Ashcroft proposed a more limited approach than Representative Coble and introduced a circumvention provision ("Ashcroft Section 1201") that prohibited only circumvention that facilitate an act of infringement. In relevant part, the provision states:

These approaches demonstrate the stark divergence of views that exist concerning the proper role copyright law should play in regulating the circumvention of access control systems. The Ashcroft Section 1201 (1) prohibits only certain intentional conduct (e.g., the act of circumventing technological protection with the purpose of infringing copyright); (2) requires an underlying infringement or intent to infringe a particular copyrighted work, and (3) limits protection to a relatively narrow set of copy protection technologies (those that incorporate encryption technology or that are sufficiently incorporated into the work that they cannot be removed without changing the work).

In contrast, Coble Section 1201 (1) uses a flat prohibition on all circumvention of technological protection, without regard to whether the person’s purpose is to infringe copyright; (2) imposes liability based on the existence of a potentially harmful technology; (3) does not require an underlying act of infringement and (4) protects an unspecified, and potentially large, number of technologies that, in their "ordinary course of operation," require the authorization of the copyright owner to gain access to the work. The Coble Section 1201 is significantly broader than the Ashcroft Section 1201 in both the range of conduct it prohibits and the number of technologies to which it applies.

The question remains whether WIPO requires the broad approach embodied in the Coble Bill or the more measured Ashcroft Bill. Despite the rejection of the broad provision in Congress and the concerns raised by the international community, the Administration sought broad treaty implementation legislation on circumvention. The Administration has an international agenda; copyright owners wanted quick passage of their broad implementation legislation to "provide a very powerful model for other countries to emulate in revising their own national laws." Ironically, as the international treaty gave the rightsholders a second bite at the domestic apple, the corporate copyright owners evidently see a broad domestic provision as providing an opportunity to undo the WIPO compromise--a second bite at the international apple.

Article 11 does not require a particular regulatory scheme. Compliance requires only effective legal protection and allows signatories’ in their discretion to determine what that means in light of their traditions and history. The remainder of this Article discusses why it is better public policy to regulate circumventions that facilitate infringement than to create a blank prohibition on all circumventions and access-enabling technologies.

A key difference between the competing paradigms for regulating circumvention and circumvention-enabling technologies—the Coble and Ashcroft Section 1201s—is whether copyright should prohibit all circumventions and enabling technologies, or whether copyright liability should attach only to those circumventions that facilitate acts of infringement. Coble Section 1201 regulates conduct that has traditionally not been reached by copyright law (i.e., conduct that is independent of an act of copying another person’s original expression) and may not involve a copy or misappropriation of intellectual property by any individual in the same course of conduct that gives rise to a claim. Such an act, by itself, might not cause a copyright owner any money damage. By comparison, the Ashcroft Section 1201 generally would impose liability only when the circumvention conduct was at least nominally connected to a violation of one of a copyright owner’s exclusive rights. This section argues that, because the full costs and benefits of circumvention are not yet known, a blanket prohibition on all circumventions and circumvention enabling technologies is premature and potentially bad public policy.

When the circumvention issued is framed in terms of those works the public considers the central beneficiaries of copyright protection—books and movies—the blanket protection of these systems may seem reasonable at first glance. After all, if a person circumvents an ECMS to steal a book or a song and sells the copies, the very heart of copyright law is implicated. That person should be liable to the copyright owner.

Most policymakers would agree, however, that the first step in proper legislative methodology—the reasoned formulation of public policy—is gathering data to fully understand the empirical realities that underlie the need for particular legislation. As the Coble Bill is currently formulated, it contains no limiting provision that separates what might be called "justifiable" circumventions from "unjustifiable" circumventions. The bill makes the act of circumventing access control technology per se illegal. Therefore, the proponents of legislation prohibiting all circumventions, and technology primarily designed to enable circumvention, must believe that the relevant societal costs and benefits of circumvention warrant the blanket prohibition. Logically, such a determination cannot be made without considering whether circumventions can benefit society in some circumstances.

The debate over circumvention has contained little, if any, systematic policy analysis of the role that circumventions, as socially beneficial conduct, might play in the future of the information society. As noted earlier, the White Pape did not consider whether circumvention might be socially beneficial in some contexts. The congressional hearings on Coble Section 1201, while comprised of witnesses who had suffered admittedly unfortunate violations of their rights, involved only anecdotal evidence. For example, one witness described a device called the "Game Doctor" which makes it possible for someone to copy software stored on a game cartridge to a floppy disc and then upload it to the Internet. Similarly, another witness described how his company distributed a working copy of a leading software program on-line with a "timer" program that gave prospective customers the ability to test drive the program for 21 days. The company soon received an E-mail threatening to publish instructions on how to disable the timer unless the company paid $20,000.

For better or worse, Congress has extended copyright protection far beyond books, music and movies--to computer programs and other works that have societal significance that does not necessarily derive from the original expression copyright law protects. If one considers the varied contexts in which access to a copyrighted work may be necessary in the information age, it becomes clear that the central premise of those who desire to prohibit all circumventions and access-enabling technologies is open to question. In this section, we posit a number of circumstances in which society may benefit from being able to circumvent access control technology.

In paper form, once a work is distributed to interested readers, copyright law does not protect the author from criticism by those who obtain access to the work. In the near future, ECMS will increasingly provide content owners the ability to craft carefully how content is used and manipulated. One activity copyright owners might choose to restrict is the ability of readers to criticize the ideas contained in their works. In February 1995, the Church of Scientology sued Netcom for infringement of the Church's copyrights because an ex-minister allegedly uploaded more than 100 pages of secret church material to the Internet. Ten years ago, if the Church of Scientology wanted to distribute its teachings to gain followers, it faced a choice: distribute its religious materials only to small numbers of people under tightly controlled circumstances or widely distribute the materials but risk exposing itself to potential criticism and commentary. Remember, however, that anyone can use an ECMS—not just corporate copyright owners. If the Church of Scientology distributes their work in digital form protected by ECMS, it can eliminate access to its works if a user criticizes them. In addition, the Church will have a user sign a digital license as a condition of access that will expose them to liability if the user criticizes the Church or allows anyone else to criticize them. If a scholar obtained the password from a licensee and accessed the work to criticize it, the Church could sue the licensee for violating the contract and the scholar for violating the anti-circumvention provision. In this situation, copyright law, by prohibiting all circumventions, is put in the awkward position of allowing the Church, or even a racial hate group for that matter, to disseminate their ideas with little or no fear that those ideas will be subject to scrutiny. Copyright’s only function is to restrict access to the work. No result could be more contrary to copyright’s constitutional purpose. In this situation, perhaps circumventing the system for the purpose of criticizing the work may be justified and in the public interest.

In Sega Enterprises, Ltd. v. Accolade, Inc., Sega produced a computer console for the purpose of playing computer games that were compatible with the console. To maintain a monopoly over production of the games, Sega placed a small piece of code in the program that the console had to find before the game would operate. Accolade decompiled Sega’s programs and discovered both the functional components of the program necessary for interoperability with the console and the required initialization code. Sega filed suit claiming copyright infringement because Accolade decompiled or disassembled Sega programs--actions that necessarily involve the making of copies--in order to get access to information necessary to achieve compatibility with Genesis machines. Even though the videogames Accolade produced after decompiling the programs did not contain infringing code from Sega's programs, Sega nonetheless sought and obtained an injunction prohibiting Accolade from marketing the programs developed through decompilation as "fruit of the poisonous tree" of the intermediate copies. The district court held that these copies could not be fair use chiefly because of Accolade's commercial purpose in making the copies and competitive harm likely to result to Sega's market.

The Ninth Circuit Court of Appeals reversed, holding that decompilation or disassembly of a computer program for a legitimate purpose, such as to get access to information necessary to develop a program that will interoperate with another program, constitutes a fair use. Although Accolade had a commercial motive for decompiling Sega's programs, the court decided that Accolade's primary purpose in making copies of Sega’s programs was to study the ideas and information in the program. The court recognized that a ruling in Sega's favor would be tantamount to giving Sega patent-like protection over the functional requirements needed to achieve compatibility, a result contrary to Congress’s intent to protect only expression in programs, not ideas or other functional design elements. Having used the information obtained through reverse engineering to develop its own original, noninfringing games, Accolade deserved the same chance to compete in the marketplace as other authors who had taken ideas, but not expression, from other works.

Consider how Sega might recast its dispute with Accolade if Congress enacts Coble Section 1201. Sega might well argue that it had chosen to distribute its computer programs in machine-readable form as a means of protecting the copyrighted text and the ideas embodied in it from unauthorized use or disclosure. This arguably makes the machine readable format a "technological protection measure [that] effectively controls access to a work." Because the act of decompilation avoids and bypasses this measure, decompilation potentially constitutes a circumvention subject to copyright liability under Coble Section 1201(a). In addition, the decompilation process uses computer programs that, in a sense, reconstruct the source code by examining the behavior of the subject machine readable code. The programs are tools used to aid the decompilation process and thus are potentially subject to Coble Section 1201(a)(2)(A) as well because they were "primarily designed" to circumvent the machine readable format to gain access to the copyrighted work (the code). Taking the plain terms of the provision seriously, for liability to attach it does not matter whether intermediate copies are a fair use; the act of circumvention is enough.

The decompilation debate has been extensive and won’t be repeated here. It suffices to say that, decompilation is a circumvention of an access control system that is in the public interest. The "lock-out" program at issue in Sega served only to bar others from utilizing ideas and functional elements contained in the code. Scientific progress—one goal of copyright—requires competitors to have access to each other’s ideas so that each may build upon the work of the other. The ability of software engineers to achieve interoperability in information technology systems requires them to conduct extensive reverse engineering of software-based products. As one information technology leader explained at the WIPO treaty implementation hearings:

Programmers use reverse engineering to teach students programming skills and to repair malfunctioning software. Outside the information technology industry, reverse engineering competitors products is standard practice.

A circumvention provision that prohibits access to the ideas contained in code might thus stifle innovation in the software industry, protect established companies against competition, and perpetuate the power of firms that already dominate a given niche. Given the current dominance of American information technology companies in world markets, European and Asian software manufacturers should find this possibility especially disturbing.

In contrast, Ashcroft Section 1201 would allow software companies to decompile because the legislation limits liability to acts which facilitate infringement and expressly exempts computer programs. If a programmer decompiled a computer program, there would be no liability because the intermediate copy made in the process of decompilation would be a fair use. In addition, the "tools" used by the programmer are specifically exempted from liability as computer programs. Decompilation thus provides a key reason why the focusing on infringing conduct, rather than outlawing technology, is better public policy.

It is tempting to analogize a copyrighted work protected by an access control system to physical property protected by a fence. Using this metaphor, an unauthorized access (read as circumvention) is akin to a trespass. In the physical world, however, the owner of land does not have an absolute right against trespass; there is an ancient principle in tort law that one can trespass on another’s private property if the trespass is reasonably necessary to avoid harm to himself or others.

The same rationale might apply to a copyrighted work that is protected by an access control system. For example, if a software pirate copies software and then protects the pirated software from discovery by employing access control technology--selling the password only to other pirates--the copyright owner will not have the ability to ascertain whether his or her work has been pirated. If a copyright owner has a reasonable belief that technologically protected material is infringing, the owner should have the right to circumvent the system. Indeed, it would be ironic if Microsoft hacked through a pirate’s ECMS system and sued for infringement only to be subject to a counterclaim under Coble Section 1201.

It is not difficult to posit other examples of access-by-necessity. When someone dies, family members will need to gain access to personal materials that may be technologically protected. A business dependent on a program that has crashed may need immediately to bypass access control technology to avoid harm to its system. Parents may need to bypass access control systems to ensure that their children are not viewing improper material. Workplaces may want to monitor information on worker’s files to protect against economic espionage. In all these circumstances, it is an open question whether society’s interest in access outweighs the copyright owner’s interest in privacy and control. An access control system does not care what content lies within its electronic fence. If Coble Section 1201 is enacted, however, it is by no means clear that this technology will be available to bypass these fence when needed.

In some cases, we as a society have already determined that the public benefit of disseminating information outweighs the costs of wrongfully accessing that information. In the "Pentagon Papers" Case, for example, the New York Times and the Washington Post obtained a stolen classified forty-seven-volume Pentagon study entitled "History of U.S. Decision-Making Process on Vietnam Policy" and began to publish excerpts. When the government attempted to enjoin the publication of the excerpts, the Supreme Court refused because the government had failed to meet its "heavy burden of showing justification" for the imposition of a prior restraint. The fact that the papers were wrongfully accessed meant nothing to the Supreme Court’s analysis—there was no justification for the prior restraint.

Consider how the Pentagon Papers case would play out today. Works produced by the government are generally excluded from copyright protection and are in the public domain. If the Pentagon Papers were on a disk protected by an access control system, current copyright law would in no way stop the New York Times from accessing the disc. Once the disk is in the hands of the media, it seems obvious that the public’s right to know the information on the disk outweighs the interests of the third-party copyright owners who use that access control system to protect their works against potential pirates. In this context, the media must have the means to circumvent the system and the access-enabling technology must not be outlawed because pirates might use the same technology to steal other unspecified copyrighted works. As sensitive information is increasingly hidden behind technological walls, to perform its function in society, the media will need to become more adept at circumvention, not less.

Computer security is vital to ensure secure digital networks and the digital works held on them. Yet, the need for computer security may require the tools to circumvent systems that certain copyright owners may use to protect their works. At the congressional hearings on Coble Section 1201, one witness provided a good example of how digital networks are causing the convergence of copyright law and security policy:

Most people would agree that if a computer program is placed on your hard drive by a third party, you should have the absolute right to disable it. Similarly, copyright owners are planning to use ECMS to gather information on their customers and build complex consumer profiles to sell to companies interested in consumer behavior. A consumer is justified in circumventing any ECMS in order to disable the profiling function if their purpose is to protect their anonymity. And to enable unsophisticated users to protect their anonymity, a technology company should be able to create a product that makes disabling the system easy.

In addition, research on computer security requires the availability of access-enabling technology. There is currently considerable research being done on the development of secure systems. This research requires researchers both to circumvent secure systems and to develop technologies that others could potentially use to circumvent them. Consider whether a computer science professor and the learned society which agreed to publish the professor's article about a major security flaw in a commonly used encryption system might violate Coble Section 1201. If some copyright owners use this technique as technological protection measure, publication of the article could be regarded as "trafficking in [a] technology" or a "service" that enables the circumvention of the copy-protection system.

The point is not that the tools to protect against "cookies," disable data collection systems, or engage in encryption research would necessarily violate Coble Section 1201. These examples show that, in some circumstances, there may be justified and legitimate circumventions of access control technology.

As discussed earlier, the Copyright Act provides only a limited grant of exclusive rights to copyright owners for a limited amount of time. These rights are part of a careful balance between the private rights and public rights embodied in the copyright act. Is it good public policy to allow copyright owners to avoid all public rights by placing their works behind technological walls? Lawrence Lessig stated the problem eloquently:

Digital technology will vest in the copyright owner complete control to define the parameters of use and an anti-circumvention provision threatens to prohibit any user interference in this control. How can we maintain the existence of statutory limitations on copyright owner’s exclusive rights in digital networks?

One way is to enact legislation that prohibits copyright owners from limiting specified user rights in digital works. For example, the EC Directive on the Legal Protection of Computer Programs prohibits using contracts to deny a lawful user the ability to make a back-up copy of a program or the right to observe, study or test the function of the program. These limitations on the freedom of contract resulted from a fear that allowing such contractual provisions threatens individual freedoms and free flow of information. Some behavior, such as criticism and commentary, are simply too central to a free society to allow a copyright owner to control in order to maximize their incentive to create. In the alternative, Congress could create an affirmative defense that a user could raise against a circumvention claim when a specific act of circumvention is in the public interest

If these alternatives are not palatable, Congress could grant users an statutory right to access digital works if they demonstrate that they used work in accordance with rights granted under the Copyright Act. If a person can access a work to use it without violating copyright, it is difficult to understand why copyright law should prohibit this behavior. The court would look at the conduct in specific cases to determine if copyright has been violated, just as it always has.

The White Paper Section 1201 included a provision that may have distinguished circumventions that violated copyright law from those authorized by copyright. The White Paper would have imposed liability on providers of technologies or services if their "primary purpose or effect" was to circumvent a copy-protection system "without authority of the copyright owner or the law." Thus, if the primary purpose and effect of a technology was to enable a user to exercise back-up rights, liability would not attach. Although Article 11 allows this language, Coble Section 1201 does not contain it. In contrast, Ashcroft Section 1201 would allow users to make a back-up copy, for example, because there would be no underlying act of infringement.

The stakes are perhaps greatest in the fair use context. Many scholars have recognized the threat a broad anti-circumvention provision poses to fair use rights. Obviously, one cannot make a fair use of a copyrighted work without access to that work. And some commentators question whether fair use doctrine exists at all in future digital works because of the technological ability to license works easily and efficiently. Rightsholders go a bit further and accept that some forms of technological protection for digital works may make the invocation of fair use impossible. In the White Paper's view, for example, rightsholders have no obligation to make their works available in a form that enables fair uses to be made of them.

A fair use defense cannot defeat a Coble Section 1201 violation. Fair use is an affirmative defense in an action for infringement. One can violate Coble Section 1201 in two ways: an act of circumvention or by trafficking in enabling technology. Section 1201(a) and (b) therefore create entirely new wrongs that do not require an underlying copyright infringement. If a defendant asserts a fair use defense in a Section 1201 lawsuit, the copyright owner’s response will be twofold. First, fair use is irrelevant because Section 107 states only that certain enumerated uses do not constitute an infringement of copyright and proving infringement is not necessary under Coble Section 1201. Second, regardless of whether the defendant’s actual use of a particular work was fair, the access-enabling technology nonetheless threatens the technological protection measure that they had chosen to protect their other works and thus violates Section 1201(b).

The fair use doctrine has existed for a century without undermining authors’ incentives to create new works; there is no reason why the doctrine does not apply to digital works. In the normal situation, a user makes use of a copyrighted work, without the authorization of the copyright owner, and then argues that the particular use ought be considered fair because of the factual circumstances that led to the use. There is no reason why an author should be able to avoid the fair use of his or work by merely by transforming into digital form and placing it behind a digital wall. Without the ability to circumvent, perhaps a user who wants to make a "fair use" can go to court and request a mandatory injunction to force the copyright owner to allow access to a work. But courts are likely to be resistant to such claims because such mandatory injunctions are traditionally disfavored remedies.

By comparison, if we prohibit only circumventions that facilitate infringement, concerns about enforcing fair use rights in the digital context largely disappear. The individual can circumvent the access control system and use the work. The copyright owner can bring an action against the individual for facilitating the infringement and the individual can raise fair use as a defense. Let people circumvent and infringe and then take the serious risk that a fair use defense won’t work. When someone facilitates the use of a work in a manner not allowed by the copyright act, they will be liable for infringement. This way, the fair use defense will continue to operate as an equitable rule of reason, policing the outer boundaries of copyright law in the digital world, just as it does in the rest of the world.

This section has demonstrated that there are a number of instances where the ability to circumvent, and the concomitant need for technology that enables circumvention, is in the public interest. Certain circumventions in no way interfere with the economic interests of the copyright owner and thus should not be subject to liability under the Copyright Act. Framing the circumvention issue in terms of only digital commerce does not capture the full extent of the issue. As computer programs increasingly play functional role in the daily lives of people and businesses, and more sensitive and important information is placed behind digital locks, the need to bypass access control systems may well evolve into a complex public policy problem of its own. Such a problem may require a varied, nuanced approach to balance all the interests involved.

One response that a critic may make to the idea that some circumventions and access-enabling technologies are in the public interest is that Coble Section 1201 will not prohibit the "good" circumventions. This response requires an analysis of how Coble Section 1201 will be applied in litigation.

This argues that the Coble Section 1201 creates an unworkable regulatory system and, ultimately, is bad public policy. First the section discusses the rationale behind the current contributory infringement standard. Second, the section argues Section 1201 litigation will be based primarily on speculation regarding the future effects of a given product. Third, the section discusses the serious standing problems raised by the direct regulation of access-enabling technologies. Fourth, the section argues provision creates the danger of "strike-suit" litigation that will hamper competition and innovation by threatening small technology firms.

In Sony Corp. v. Universal City Studios, Inc. the Supreme Court found that the manufacturer of Betamax machines was not liable as a contributory infringer for private, non-commercial copying of television programs because the product was capable of substantial noninfringing uses. Copyright scholars largely agree that the proposed Coble Section 1201 will overrule the contributory infringement standard set forth in the Sony decision. While Congress obviously has the power to override Supreme Court rulings, prudence would suggest that it should first consider the policy reasons that prompted the adoption of a contributory infringement standard for copyright law that allows copyright owners to control the manufacture and sale of technologies only when these technologies lack substantial noninfringing uses.

Because the copyright statute did not have a contributory infringement provision, the Sony Court had to decide whether to "borrow" the contributory infringement standard of trademark or of patent law. Trademark law imposes liability upon vendors who sell products with actual or constructive knowledge that some purchasers will use the products for infringing purposes, even if the products can also be used for noninfringing purposes. Under this standard, Sony probably would have been liable for contributory infringement because it sold video tape recording (VTR) machines with knowledge that some consumers would use them to make infringing copies of copyrighted works. Patent law, by contrast, gives patent owners the right to control sales of products only if infringement is virtually all for which the products are useful. If products are capable of substantial noninfringing uses, patent owners cannot control them under the "staple item of commerce" doctrine, even if the primary use of and market for the product is to enable consumers to infringe the patent.

Citing "the historic kinship between patent law and copyright law" and the public interest in access to technologies that have substantial noninfringing uses, the Sony Court decided to import the patent contributory infringement standard into copyright law. The Court observed that a balance must be struck "between [the rightsholder's] legitimate demand for effective--not merely symbolic--protection of the statutory monopoly, and the rights of others freely to engage in substantially unrelated areas of commerce." As long as a technology was capable of substantial noninfringing uses, the Court decided that copyright owners should not be able to regulate sales or uses of the technology.

Patent law does not shield from liability manufacturers of technologies with both infringing and noninfringing uses out of a desire to molly-coddle would-be infringers, but because the public has a legitimate interest in having access to aticles of commerce that can be put to noninfringing uses. The "substantial noninfringing use" standard evolved from a legal fiction. Early courts used the sale of a nonstaple as an evidentiary device; courts inferred from the fact that a person sold an article of commerce with no substantial noninfringing use that the defendant/seller intended to infringe the plaintiff’s patent. Over time, however, the rationale shifted from an evidentiary presumption to public policy based on free access to staple articles. As A. Samuel Oddi explained:

At the same time, Congress limited the patent monopoly by requiring direct infringement in order for a contributory infringement claim to arise. Congress thus wanted to ensure that patentees would not be able to misuse their patent monopolies by using contributory infringement lawsuits to control the sale of unpatented items. The line was chosen to limit the effects of the statutory monopoly to the monopoly itself. As the Supreme Court has explained, "The policy of free competition runs deep in our law." If the doctrine of contributory infringement was extended any farther, competition in staple products would be hindered and, logically, the public denied access to products that could be put to non-infringing uses. Extending the patent monopoly is also against the public interest because it allows the patentee to gain economic returns in excess of the public benefit created by the invention itself.

The Sony Court felt that these same policies were relevant copyright. The Court noted that noninfringing uses of VTRs included making copies of uncopyrighted television programs, programs whose copyright owners had no objection to time-shift copying, and private, noncommercial copies of copyrighted programs for time-shifting purposes. The Court regarded such private, noncommercial copying as presumptively fair, a presumption that should only be overcome if plaintiffs produced proof of some meaningful likelihood of harm arising from the user activity. As the Court explained, "a use that has no demonstrable effect upon the potential market for, or the value of, the copyrighted work need not be prohibited in order to protect the author's incentive to create. The prohibition of such noncommercial uses would merely inhibit access to ideas without any countervailing benefit."

Coble Section 1201 extends the copyright monopoly to staple articles of commerce. If an article is "primarily designed" to circumvent, it violates Coble Section 1201 even if it has substantial noninfringing uses. And a product that has only a limited "commercial" purpose other than circumvention, may have substantial noncommercial uses like the private copying at issue in Sony. Because the copyright owner need not suffer any harm to outlaw the offending item, a copyright owner can outlaw a technology, and deny others technology that may have legitimate uses without any corresponding benefit to society.

Litigation under Coble Section 1201 also raises serious problems of proof for litigants attempting to protect their technologies from being outlawed by copyright owners. For example, because a violation of Section 1201 may occur at the product design stage, a copyright owner can bring suit without having suffered any damage from a given technology. Coble Section 1201 imposes liability based on a prediction that costs of the infringement will, if the technology becomes available, outweigh the non-infringing benefits of that technology to the public. There are reasons to doubt that litigation is an effective mechanism to make this determination.

First, litigation is a poor mechanism for considering whether the impact of a given technology on copyright owners outweighs the public’s interest in access to that technology. The Sony case, which presents the basic fact pattern that is likely to occur in Coble Section 1201 litigation, demonstrates the problems inherent in this form of litigation. Sony shows that corporate rightsholders are willing to attack a technology they view as threatening before it is clear how that technology will be used in the marketplace. Plaintiffs may argue that the technology at issue, at the time of trial, has only a limited commercial purpose other than circumvention. But future uses should be considered in this determination. Liability under Coble Section 1201 may thus turn on speculative predictions about the likely uses of a given technology in the future.

The litigation process is an extremely poor mechanism for predicting the future uses of a given technology. When Universal brought the Sony litigation in 1979, it planned to distribute movies on discs. Both plaintiffs admitted that they had not suffered any monetary damage as a result of VTRs prior to instituting litigation to outlaw that technology and their experts were unable to prove that they would suffer harm in the future. Today, videocassettes provide movie producers more revenue than ticket sales. A court has inherent limitations in the type of evidence it can consider. Even with the fact-gathering capabilities of agencies and Congress, the government is unable to accurately gauge future technological developments. The idea that we can make similar predictions through litigation is absurd.

Second, Coble Section 1201 presents difficult problems of proof once an action is initiated. To prove that a given technology was "primarily designed" to circumvent or has only a "limited commercial purpose" other than to circumvent, corporate copyright owners are likely to gather data concerning how third parties use the product. In Sony, Universal and Disney used private investigators to gather information on how VTRs were being used by third parties, and conducted extensive surveys. Litigation under Coble Section 1201 will likely devolve into competing surveys and expert testimony.

Third, it is not clear what evidence should be relevant in making the determination of a technology’s likely future purpose and uses. Should a court should consider whether a given technology is capable of noninfringing uses, or whether to users subjectively intend to use, or actually do use, a technology for noninfringing uses. Is evidence of actual uses relevant to prove what the product was primarily designed to do?

Along these lines, the provision creates a new distinction between commercial and noncommercial uses. A technology that has a significant commercial use will not violate the Coble Section 1201, but a technology that has only non-commercial uses will violate the Coble Section 1201. This distinction is aimed directly at the private, noncommercial conduct the Court found relevant in Sony. This distinction is unworkable. What constitutes a "limited commercially significant" purpose? For example, is the exercise of statutory rights (i.e., § 117 backup rights) "commercial"? In the fair use context, courts and scholars have already criticized the commercial/noncommercial distinction as "simplistic and inherently ambiguous."

The Coble Section 1201 creates expansive standing. The concept of standing has been expressed in myriad ways. Article III requires a plaintiff to "’show that he personally has suffered some actual or threatened injury as a result of the putatively illegal conduct of the defendant’. . . the injury ‘fairly can be traced to the challenged action’ and ‘is likely to be redressed by a favorable decision.’" One function of standing is to ensure that the parties to litigation have a sufficient stake in the outcome of the controversy at hand. An analysis of standing under Coble 1201 demonstrates that quite literally thousands of copyright owners may have standing to bring suit against companies who create enabling technologies.

Coble Section 1203(a) authorizes "Any person injured by a violation of section 1201 [to] bring a civil action." Coble Section 1201 creates a strange type of statutory injury: manufacturing or trafficking in any "technology, product, service" primarily designed to circumvent, or with a limited commercial purpose other than to circumvent a technological protection measure that effectively controls access to a copyrighted work. Two parties probably have standing to bring a civil action under Coble Section 1201: individual copyright ownrers and associations like BMI and ASCAP. The allegation that a potentially threatening technology is in violation of the statute is probably sufficient to confer standing to upon any copyright owner who uses that access control system,which could number in the thousands. The Green Paper expressly supported this, stating that any copyright owner whose copyrights could be infringed by a particular circumvention technology or service would have standing to sue, regardless of whether the challenged technology or service had ever actually been used to make infringing copies of any copyrighted work, let alone used to make infringing copies of that copyright owner's works.

Creating such expansive standing is troubling. Consider the facts in Vault v. Quaid. Vault had developed a software-based copy-protection system called Prolok intended for sale to other software companies. Vault expected its customers to use Prolok as a form of technological protection so that consumers could not make unauthorized copies of the application software Vault's customers were selling. After Quaid developed Ramkey, a software product that included a component that could "spoof" Vault's copy-protection system, Vault sued Quaid for copyright infringement.

Who could sue Quaid under Coble Section 1201? Notably, Vault might not have standing. The Green Paper took the position that manufacturers of technologies that could be circumvented, such as Vault, would not have standing. This leaves the numerous copyright owners to initiate the suit. Would they have to organize a class action? One would presume so. Quaid would thus be faced, not just with one lawsuit, but potentially thousands which it would have to defend.

The Sony Court faced the same standing problems raised by Coble Section 1201. The Court pointed out that "this is not a class action on behalf of all copyright owners who license their works for television broadcast, and respondents have no right to invoke whatever rights other copyright holders may have to bring infringement actions based on Betamax copying of their works." The Court characterized as "extraordinary" the claim that "the Copyright Act confers upon all copyright owners collectively, much less the two respondents in this case, the exclusive right to distribute VTR's simply because they may be used to infringe copyrights" and to declare them "contraband." The Court also observed that, if Universal and Disney prevailed in the lawsuit, it would frustrate not only consumer interests in using VTR's, but also the interests of the many copyright owners who wanted consumers to use VTRs for time-shifting purposes as a way to enlarge audiences for their programs and enhance advertising revenues for television audiences.

As discussed earlier, the provision leaves firms vulnerable to strike-suits based purely on speculation. Notably, anyone who wants to challenge a technology can create standing merely by purchasing a threatened access control technology and using it to protect a copyrighted work. It is easy to allege that a particular technology was designed to bypass a protection system. More importantly, when a technology is new and its future uses unclear, it is vulnerable to the claim that it has a only limited commercial purpose other than circumvention. If thousands of copyright owners can bring an action against a technology company, its likely that one or more than one may do so, even in circumstances where they are unlikely to win in the end.

Once Section 1201 creates new, prospective liability based on the predicted and vague future effects of a given technology, the danger of "strike suits" becomes evident. Again, the Sony case is instructive. Disney and Universal did not necessarily want to outlaw VTRs; rather, they requested the grant of "a continuing royalty pursuant to a judicially created compulsory license" Under Coble Section 1201, perhaps rightsholders will want similar relief.

Although copyright owners might admit that private, noncommercial copying is, for now, beyond their practical power to control, they claim that they have the legal authority to stop such use. Corporate rightsholders have capitalized on this legal authority. In 1992, corporate rightsholders successfully lobbied Congress to force manufacturers of digital audio recording devices to pay a royalty for each sale of devices covered by the statute to compensate them for private, noncommercial copying. In addition, they forced design changes to the DATs that degraded the quality of the sound the machine could produce.

Section 1201 creates the opportunity for well-funded copyright holders to repeat what happened with DATs. Coble Section 1201 may be difficult to decide on a motion for summary judgment. Given the generosity of the remedy provisions--allowing recovery of actual damages, profits, statutory damages, and up to treble damages for repeated violations, as well as the grant of injunctive relief and recovery of costs and attorneys' fees--electronics equipment manufacturers, software developers and information service providers, are understandably anxious that the broad anti-circumvention provision may make them vulnerable to lawsuits by nervous or exploitative copyright owners.

Chris Byrne, the Director for Intellectual Property for Silicon Graphics, testified at the congressional hearings:

For many small software companies, from whom some of the most innovative products on the market come, the potential for litigation under Coble Section 1201 could deter development of information technology products and services having substantial noninfringing uses.

Some years ago, when the motion picture industry tried to persuade Congress to pass an anti-circumvention provision very similar to that now pending in Congress, some supporters of appropriate regulation of circumvention technologies were critical of that proposal because of the potential impact of a very broad anti-circumvention provision on investments in information technologies:

A copyright infringement standard that does not prohibit technologies with substantial noninfringing uses appropriately limits the copyright monopoly and balances the competing interests. As a result, Congress should consider adding this limitation to any anti-circumvention provision it chooses to enact.

If a circumvention of a access control system constitutes a infringement of copyright, one problem that arises is the possibility of providing copyright protection to works, and the information contain in copyrighted works (i.e., the public domain) that are not otherwise protected by the Copyright Act.

Consider what happens if a particular access control system is used to protect both copyrighted and uncopyrightable works. A copyright owner chooses to use the same encryption algorithm for six works, five copyrighted and one in the public domain. That algorithm constitutes a technological protection measure that controls access to a copyrighted work—in fact, five copyrighted works. At the same time, the algorithm also protects the public domain work. A person then develops a technology that a third party uses to access the public domain work. Under the Coble Section 1201(b), whether or not liability is imposed on the manufacturer of the access-enabling technology will depend on two difficult fact questions: (1) was the work "primarily designed" to circumvent copyrighted works or public domain works or (2) is the access of the public domain works only a limited commercial purpose. The answer is quite unclear.

An incidental byproduct of using the same algorithm for both copyrighted and public domain works would seem to be the provision of copyright-like protection for public domain works in digital form. If this logic is correct, anyone wanting to market public domain works would understandably find it attractive to use standard copyright industry encryption techniques when distributing public domain materials. Although one might object on equitable or economic grounds that the law should prohibit the access to the work once its encased in technology, using copyright law to protect public domain works seems to run directly counter to notion that copyright grants authors a monopoly for a specifically limited duration. Access control technology creates the means for perpetual copyright terms.

The Green Paper acknowledged the legitimacy of "some concerns regarding [the anti-circumvention] proposal, particularly with regard to works whose term of copyright protection expires but are still protected by anti-copying systems, and works in the public domain." The Clinton Administration has suggested that, even if some firms used technological means to protect public domain works, this protection would attach "only to those particular copies--not to the underlying work itself." Anyone who wished to copy the public domain work could always do so from a source unencumbered by technological protection. This response, of course, presumes that a copy exists that is unhampered by technological protection, and also begs the question of why copyright law, which protects only original expression, should protect expression that is not protected by copyright if it is accessible and technological protection for it can be bypassed without infringing copyright. In addition, Coble Section 1201 may incent firms to acquire and hide away or destroy extant unencumbered public domain copies of rare works in order prevent members of the public from gaining alternative access to them.

The underlying, unspoken theme implicit in this particular argument is that the public domain should exist only in paper form. If technological walls are protected by copyright law, the public domain is an artifact of the print era that will not develop in the digital environment. . Historically, the public domain enabled competitive publication of works after its author and first publisher had recouped their investments and made sufficient profits to justify further investments in the production and distribution of creative works. People remain free to place public domain works on the Internet and try to use technology to sell it. Someone who hacks through the ECMS and creates a competing digital product would almost certainly be liable under state unfair competition and business practice statutes. But using copyright law to protect the envelope in which they wish to deliver a public domain work distorts the traditional purpose of copyright law.

Granted, there are practical reasons to doubt whether works developed and distributed in electronic form will ever enter the public domain. By the time works first published in electronic form reach copyright expiration dates, it is far from clear that the media and equipment now used to read those electronic copies will still be available. In addition, owners of copyrights in works published in electronic form may well decide to program copies to self-destruct prior to the date of expiration. Following the expiration date, holders of these now-expired legal rights might choose to make the works available only by online access under restrictive contract terms, such as those West Publishing now employs with public domain statutes and judicial decisions.

It is likely that contract law, in combination with access systems, will increasingly provide content owners the ability to carefully craft how content is used and manipulated and to charge for each use. A number of commentators are concerned that, because of great inequities in bargaining power on the Internet, contract formation will be highly one-sided in favor of large corporate rightsholders. Outlawing all circumventions and circumvention technologies encourages the development of contractual liability. That is, of course, the point. Is this the type of information society in which we want to live?

The fair use defense is threatening to disappear in the digital world. As discussed earlier, access control systems will create manageable licensing schemes, which some see as rendering fair use unnecessary. It is fair to question whether we want to live in an information society where fair use does not limit the reach of the copyright law. As the Supreme Court has stated, "This ‘equitable rule of reason’ permits courts to avoid rigid application of the copyright statute when, on occasion, it would stifle the very creativity which that law is designed to foster." Thus, courts attempt to tailor copyright law in the public interest. For example, the fair use doctrine provides stronger copyright protection for creative works than informational works, protects parodies, is used allows de minimis uses of copyrighted works, and allows newspapers to use copyrighted works, even in their entirety at times, for the purpose of reporting the news. Prohibiting all circumvention renders these limitations nugatory. If you eliminate fair use, you have eliminated the role the public interest plays in copyright law.

Once copyright owners use access control technology to force users through licensing schemes, there exists the possibility that the only role left for copyright will be to prohibit circumvention of access control technology. Rather than a regime designed to encourage the dissemination of information, copyright may become only a regime that prohibits access to copyrighted works. That would be an ironic twist, given the statute’s history. In this potential future, copyright law will regulate the activities of computer hackers, protect trade secrets, and for guard citizens against unwelcome intrusions into their private information spaces. It has the potential, in other words, to transform copyright law into a general purpose misappropriation statute.

Consider the famous case of Robert Morris and his "Internet worm" under the terms of the proposed anti-circumvention provision. Password software is a popular mechanism to prevent unauthorized access to and use of electronic information stored on computers; thus, any anti-circumvention would probably apply. Morris' program was "primarily designed" to circumvent the password system. Morris used his worm to bypass the Unix password system and access approximately 6000 computers nationwide. So, he could be sued under Coble Section 1201.

Certainly, Morris should be subject to liability for his in actions; after all, this conduct was punishable under existing law. However, the owners of the computers, making the reasonable assumption that these computers contained protected works, could all have sued Morris in federal court. Morris would have been liable for statutory damages, at least, to each plaintiff. A blanket prohibition on circumvention thus converts copyright law into a powerful anti-computer hacking statute. What does this have to do with copyright law? In addition, computer hacking can currently "be charged under at least forty different federal statutes," including the federal criminal code, the National Stolen Property Act, mail and wire fraud, the Electronic Communications Privacy Act, and the Telecommunications Act of 1996. Properly punishing illicit computer hacking is certainly important, but using copyright law to regulate computer hacking seems far from copyright’s constitutional purpose: the creation and dissemination of copyrighted works.

In the past, copyright and trade secrecy law regulated very different kinds of works in very different ways. Trade secrecy law protected many documents containing secrets against the use of improper means to obtain the secrets, as well as against disclosures in breach of confidential relationships or contracts. Federal copyright law has largely protected published works against unauthorized copying. Although authors are often said to have had common law protection against unauthorized publications of their manuscripts, it was not until 1978 that federal copyright law undertook to protect every original work of authorship from the moment of its first fixation in a tangible medium. Copyright protection now lasts from that first fixation to fifty years after the death of the author.

Because copyright law now spreads its protective mantle over unpublished works as well as published ones, new opportunities exist for firms to use both copyright and trade secrecy law, or whichever of the two would best serve its interests in a particular situation. Owners of electronic documents containing secret information may find some advantage in re-conceiving their legal protection strategies away from reliance on trade secrecy alone. Technological means to protect digital secrets will also be an attractive strategy, whether a firm conceives of its content in copyright or trade secret terms. If owners of trade secrets protect digital copies of their secrets through an encryption algorithm or other technology commonly used by the copyright industries and conceive of their content as protected by copyright law as well as trade secrecy law, then technologies or services that might be used to circumvent or bypass this technological protection might be challengeable under the anti-circumvention provision. Notably, federal law already provides protection for trade secrets that overlaps with the circumvention provision, demonstrating the convergence of these areas of law. In other words, copyright law may become no more than a form of federal privacy law, despite the fact that state privacy law currently exists to vindicate this interest.

It would be unfortunate if enacting an anti-circumvention provision deflected the focus of copyright analysis away from public policy balancing that characterizes so many important copyright precedents. Sony, Vault and Sega demonstrate that courts have generally been less receptive than publishers had hoped they would be, refusing to allow copyright law to be used to regulate technologies except as explicitly authorized by Congress. These cases illuminate the need to strike a balance in copyright law among competing interests. These cases reflect a tradition that views copyright as a limited grant to authors for the purpose of achieving a larger societal goal, one embedded in the U.S. Constitution, of promoting knowledge and public access to it. When new technologies have raised questions that are not easily answered by existing law, courts have tended to construe the law in light of these larger societal purposes. These larger purposes of copyright law and the notion of copyright as a law that strikes balances are given short shrift when Congress is faced with the claims of rightsholders that the digital environment threatens their economic viability.

The Clinton Administration wants to make the power of rightsholders in the electronic environment truly awesome. The White Paper treats every electronic access to or use of a copy of a copyrighted work as a prima facie infringement, regards the first sale doctrine as inapplicable in the electronic environment, and fair use unavailable if the use can be licensed. Scholars have documented that, in many areas, copyright owners have already successfully changed copyright law to restrict access to ideas and information:

Technological protection is an important component of this larger maximalist strategy to shift power away from consumers and toward publishers by placing information under private control, not just expression.

There is, however, an alternative perspective on copyright law, one that conceives of fair use as a "right" of users, and public access to public domain works and unprotectable elements embodied in copyrighted works as a critical part of the "quid pro quo" that has long been part of copyright tradition. A different way of approaching circumvention legislation is to ask whether these "rights" can be protected in the digital environment without undermining the incentive of rightsholders to create and disseminate copyrighted works.

The effects of technological protection depends to a large extent on the development of markets for digital information products. Perhaps the markets and competition will mitigate how copyright owners will use technological protection. What killed the market for Vault's software and similar products was not its loss in litigation, but consumer preferences for copies of programs unhampered by copy-protection systems. Perhaps the technology will be so expensive, or difficult to implement, that copyright owners will not be able to use them. Some commentators argue that rights holders should focus more on protecting revenues rather than protecting bits. Others argue that protected content will be so cheap that technological protection will not raise serious public policy concerns. Notwithstanding the vulnerable nature of software products, and the absence of commercially acceptable forms of technological protection for these products, the software industry as a whole has prospered. The same may very well be true for information products generally. Until policymakers can document the nature of specific market failures caused by digital technology, perhaps creating new causes of action under the copyright law is premature.

IV. CONCLUSION

What I was walling in or walling out.

We should realize the path that the White Paper and the Coble Bill have charted are not inevitable, but rather are choices that will influence a future that is still ours to control. John Seely Brown, who is the Chief Scientist of Xerox Corp. and the head of its Palo Alto Research Center recently observed:

Neither the threat posed by circumvention nor public benefits of circumvention is at this moment at time clear. Prudent public policy does not spring from speculation and fear. Professor Ejan Mackaay, an economist and copyright scholar with a deep interest and appreciation of the challenges and opportunities of emerging digital networks, as well as other commentators, suggests that it is simply too early to tell whether legislation is needed. Let copyright owners use technology to build "fences" around their works and explore new markets. If the fences they use are inadequate to protect against market failure, the legislative process can move forward based on empirical realities, rather than speculation. For example, encryption technology, which holds such promise in the eyes of rightsholders to control distribution of digital work, at the same time threatens them by making it possible for pirates to disseminate digital works without detection. The nature of the problems just aren’t clear.

Digital technology is developing at such a rapid pace and is so complex that the Coble anti-circumvention solution may be inappropriate. Rather than rush to enact blanket prohibitions, which may do a considerable amount of unintended mischief and propel copyright policy in a direction for which it is currently ill-equipped, Congress should take more the measured approach of the Ashcroft Section 1201. If this protection is not enough, there will always be time to fix it in the future.