Digital Copyright

CHAPTER NINE: The Bargaining Table

When the government released the White Paper in September 1995, it looked as if the White Paper’s recommendations were all but a done deal. Immediately after the report’s release, implementing legislation was introduced in both houses of Congress with bipartisan support, and Commissioner Lehman confidently predicted easy enactment before spring. It didn’t turn out that way. Library groups, online service providers, consumer organizations, writers’ organizations, computer hardware manufacturers, Internet civil liberties groups, telephone companies, educators, consumer electronics manufacturers, and law professors registered early and fervent objections. They collaborated with one another in opposing the legislation, and used a tool that the supporters of the White Paper were not yet in a position to exploit: they organized, planned, and lobbied using the Internet.

The digital copyright enhancement legislation introduced at the administration’s request in the wake of the White Paper was minimalist in form if not effect. It would have added about one thousand words to the existing statute. The new words would have expanded copyright owners’ exclusive rights to distribute their works to encompass transmission, and would have protected copyright owners’ ability to use mechanisms to prevent or discourage infringement. The bill would have made it illegal for anyone, for any reason, to make, sell, or give away a product designed to enable people to get around such a mechanism, even when that product had legitimate uses. Precisely what protective mechanisms the bill contemplated was unclear. The White Paper mentioned the use of passwords to restrict access to works, the distribution of copies of digital works with encoded information limiting the uses that might be made of them, and the introduction of systems for tracking and monitoring all uses of copyrighted works.

The effect of the bill, had it been enacted, would have been more far reaching than initially appeared. Historically, as I’ve explained earlier, statutory copyright rights have been phrased broadly, but made subject to a variety of broad and narrow exceptions, limitations, and privileges. The new statutory rights, however, were subjected to no exceptions, limitations, or privileges.[1] Defeating a monitoring and tracking system by using a false name, for example, to conceal your interest in reading information about hemorrhoids, or herpes, or HIV would have violated the bill’s anticircumvention provisions.

Because computer-mediated uses of works in digital form can be subjected to extensive restrictions using software tools, the upshot would have been to give copyright owners far more control over use of any works in digital form than they had ever had over analog uses. But the language of the bill was not limited to digital works. Libraries objected that the bill was phrased broadly enough to cover any system purporting to limit the uses consumers might make of conventional books or music.

In the weeks following the release of the White Paper and the introduction of its implementing legislation, Peter Jaszi, a law professor at American University in Washington, held informal consultations with like-thinking law professors and representatives of library organizations to see whether there was any possibility of mounting an effective opposition to the White Paper’s proposals. (I confess to being one of the law professors.[2]) The copyright owner lobbies had responded to the White Paper with a flurry of supportive press events. Naysayers were few: a large number of copyright law professors, most of the library organizations, cyber-libertarians, and some computer scientists had registered quiet opposition without getting much attention from politicians or the press. Adam Eisgrau, the newly hired lobbyist for the American Library Association, suggested to Jaszi that Congress regularly ignored the efforts of nonprofit and educational organizations. What it would take to get serious congressional attention, Eisgrau believed, was the opposition of commercial and business interests.

Jaszi asked likely White Paper opponents to come to a meeting at American University billed as a roundtable discussion of the White Paper’s recommendations. The purpose of the meeting was to organize a coalition among business and nonprofit entities unhappy with the White Paper, to combine their efforts into an effective opposition. Jaszi had invited representatives of library organizations, online service providers, telephone companies, computer hardware and software manufacturers, consumer electronics companies, and civil rights and consumer protection organizations. All of them, he believed, ought to find something objectionable in the White Paper’s proposals. Copyright-holder lobbies had perfected a strategy of working out their differences privately and then presenting Congress with a united front. The interests of the White Paper’s likely opponents were probably too dissimilar to enable them to stand together in the long term, Jaszi thought, but, so long as they shared the short-term goal of preventing the enactment of the White Paper’s recommendations in their current form, they might be able to accomplish more working together than in their individual capacities.

By the end of the afternoon, several of the invitees had agreed to a temporary, informal alliance, and had settled on a name: the Digital Future Coalition, or DFC for short. At Eisgrau’s suggestion, the group’s early efforts emphasized the commercial and business interests among its members. Nonetheless, onlookers perceived DFC to be essentially a library and law professor effort, and with reason. In its early months, DFC was a two-man show. Eisgrau and Jaszi coordinated its activities out of the American Library Association’s Washington office, and, with the help of members of the library and law professor communities, wrote or assembled all of its materials. In fact, however, while the DFC’s early funding relied largely on a private foundation grant, a large chunk of its eventual operating budget and the majority of its legislative strategy derived from the Home Recording Rights Coalition (HRRC).

The Home Recording Rights Coalition styles itself a grass-roots lobby organized to protect consumers’ right to private home audio- and videotaping. It represents consumer electronics manufacturers, wholesalers, and retailers: makers and sellers, in other words, of recording devices, whose business depends on consumers’ legal ability to record. Operating through the HRRC, the consumer electronics industry had used the public’s desire to make free copies of music and movies as a tool to block copyright owner’s efforts to prevent unauthorized copying, while making bundles of money selling devices that facilitated it. Organized in the early 1980s in response to the Sony Betamax litigation, in which movie studios unsuccessfully sued the manufacturer of the Betamax VCR for copyright infringement, the HRRC established itself as a player entitled to sit at the copyright bargaining table. It lobbied to prevent the enactment of bills prohibiting videotape rental, or the manufacture and sale of dual-deck recording devices. It blocked enactment of the Audio Home Recording Act until copyright owners made a variety of concessions to equipment manufacturers.

The professors knew the law, knew the White Paper, and could criticize it to their peers[3] and write mainstream critiques for the popular press.[4] The library groups had a geographically dispersed membership who took copyright law extremely seriously, and could easily be brought up to speed on the threat posed by the legislation. The HRRC had a sophisticated lobbying and public relations machine already in place, and a significant commercial interest in the outcome.

The DFC gained significant initial credibility when it recruited representatives of the technical community. A significant portion of the opposition to the White Paper came from people and groups who were Internet-savvy. The supporters of the legislation, by and large, were not. Much of the early opposition to the bill was effective precisely because opponents knew the Internet and could use it to get the word out—to scientists, to journalists, to writers and students, and lawyers and cyber-libertarians. Initial press coverage of the White Paper Report had ranged from deferential to reverential. As more criticism hit the Internet, the press treatment got nastier. Popular press accounts accused copyright-holder interests of grabbiness,[5] and the atmosphere became decidedly less pleasant. At one point in the debate, Commissioner Lehman was widely reported to have privately threatened one of his high-profile detractors with grievous bodily injury.[6]

Soon, supporters of the implementing legislation began to back away from the White Paper’s analysis, and suggest that everyone ignore the White Paper and focus exclusively on the language of the bills. The Patent Office, the Copyright Office, and key senators and members of Congress encouraged private negotiations among opposing interests to reach compromise solutions, but compromises were hard to come by. Copyright owner groups, having been promised the moon by Commissioner Lehman, were unwilling to settle for a smaller chunk of some asteroid; opponents of the bills, for their part, had little to lose by delaying the legislation, and therefore had no incentive to compromise on unattractive terms. An additional complication was that the White Paper’s strategy of claiming that existing copyright law already gave copyright owners all the additional rights they were seeking had generated a heated argument over the deceptiveness of that description, and that fight had quickly gotten ugly. The resulting atmosphere of mistrust made good faith negotiations difficult.

If the world of legislative politics were a more sensible place, perhaps the supporters of the digital copyright enhancement legislation could have added privileges, limitations, and exceptions to the new provisions that were analogous to the ones in the extant law, and everyone could have gone home early. What happened instead was what always seems to happen: intensive negotiations among supporters and detractors of the bill led to a proliferation of narrow, stingy, conditional privileges and exceptions that apply to the folks who insisted on them, but not to you and me.

As soon as it became clear that the White Paper implementing bills would not merely sail through Congress unopposed, supporters of the legislation began negotiating with opponents in a variety of fora. The “serious” negotiations—the ones perceived to be necessary to ensure the enactment of legislation—involved the motion picture industry, the music recording industry, the book publishers and the software publishing industry on behalf of the “content owners,” and the online and Internet service provider industry, the telephone companies, the television and radio broadcasters, computer and consumer electronics manufacturers, and libraries representing the “user interests.”

The consumer electronics and library groups were perhaps less equal opponents than the others. Long negotiating experience had led the content owners to view consumer electronics groups as manufacturers with the souls of pirates. The Home Recording Rights Coalition had proved unwilling to settle on terms acceptable to the content owners in the past, and could mobilize significant grass-roots opposition working through local distributors of tape and video recorders. From the content owners’ vantage point, it may have seemed more profitable to try to discredit the consumer electronics groups than to negotiate with them. Playing on xenophobic themes that had served them well in the past, representatives of content owners dismissed the Home Recording Rights Coalition as a front for Japanese manufacturers eager to make a buck off of American material.

A long copyright history of negotiations with libraries may also have persuaded content owners that library groups were easily marginalized and not a significant threat. Library groups had a history of settling for very little. Commissioner Lehman had no compunction about criticizing as out of line and out of touch libraries’ demand that in a digital age their patrons should continue to have essentially free use of valuable material. The effort made to accommodate library interests was accordingly modest, and seemed to focus primarily on first dividing libraries from the commercial opposition and then buying them off cheaply. Thus, Commissioner Lehman initially suggested that libraries’ fair use concerns should be addressed by encouraging them to meet with publishers to negotiate a mutually satisfactory solution, and that Congress therefore need not take up any of the libraries’ objections. When those negotiations failed to produce any agreement, Lehman offered a modest amendment expanding a library’s established privilege to make a single facsimile copy in order to preserve or replace an out-of-print work. Lehman’s proposal would have permitted the library to make up to three copies of a single work—a necessary expansion if the Working Group’s position that RAM copies were actionable should prevail.

The folks who make tape recorders and run libraries have many altruistic motives, but they also have their own agendas. Their interests may often accord with the public’s, but where they diverge, the electronics industry and the libraries will look out for themselves. They are not, in other words, effective substitutes for a public advocate. In earlier legislative sagas, public interest groups had not been interested enough in copyright to try to get involved. This time it could have been different. Thanks largely to the early efforts of the DFC, consumer groups, public interest organizations, and Internet civil liberties groups had weighed in in opposition to the White Paper’s proposals. The Consumer Project on Technology suggested that the bill would require Internet service providers to monitor their subscribers’ private electronic mail and to censor content passing through their systems. The Electronic Privacy Information Center objected that the anticircumvention provisions would undermine privacy and impede computer system administration. The National Education Association warned that the bill would gravely impede distance learning. The Electronic Frontier Foundation complained that the White Paper’s recommendations would inhibit browsing on the World Wide Web and make encryption research illegal. Mobilized by DFC and library group action alerts, a variety of organizations with no prior investment in copyright issues advised their members that the legislation would make browsing the Internet illegal, and would subject computer system operators to ruinous liability.[7] It made little difference. Public interest groups weren’t regulars at the copyright negotiating table. Supporters of the legislation essentially ignored them. Many of the groups participated to a greater or lesser degree in the DFC’s efforts, but were never invited to participate in the private negotiations over the legislation. Nor, for that matter, did the DFC ever gain its own seat at the table. Its efforts were limited to behind-the-scenes coordination of its members’ activities.

That left the Internet and online service providers, the telephone companies, the broadcasters, and the manufacturers of computer hardware to deal with. The White Paper had suggested that service providers and phone companies would be liable under current law whenever their facilities were used to transmit or reproduce infringing content. It recommended that no change in the current liability standard should be made. The clear implication was that henceforth, this sort of liability would give content owners a deep pocket to sue; fear of liability would drive service providers to agree to a variety of measures designed to choke off, deter, or avenge infringement by their customers. While the content owner groups insisted that fears of ruinous liability were speculative and any move to relax the traditional strict liability standard was premature, one content owner proved them wrong. The Church of Scientology brought suit against a variety of service providers for transmitting the contents of church documents that their subscribers had posted.[8] (The documents revealed secret church teachings explaining that human beings were in fact the reincarnation of dead space aliens who had been kidnapped to earth many millennia ago and then murdered by an intergallactic tyrant. For some reason, the church was upset that the documents had been revealed to the general public.) The church sought orders compelling the removal or destruction of all unauthorized copies. As those cases worked their way through the courts, companies in the Internet and online service business started paying attention. Finding some compromise that was minimally satisfactory to service providers and telephone companies soon emerged as a necessary precondition to enacting legislation.

Copyright owners remained unwilling to let service providers off the hook, and the providers and telephone companies were determined that the bill not move until their interests were addressed. As the legislation got into trouble, proponents of other unrelated copyright reforms came forward to decorate the administration bill with their own additions, seeking a twenty-year extension to the copyright term, or a broad privilege to play copyrighted music in restaurants and bars, or a narrow privilege to enable computer service businesses to fix broken computers.[9] A variety of different actors tried to broker deals, but nobody was yet willing to settle. By the summer of 1996, the effort to enact NII copyright legislation in the 104th Congress was stalled.

Meanwhile, Commissioner Lehman’s initial confidence that the United States would have the project wrapped up before spring had led him to press for an international diplomatic conference in Geneva hosted by the World Intellectual Property Organization (WIPO),[*] which administers the major international copyright treaties. Lehman hoped to use the conference to persuade the rest of the world to make the United States approach the basis for a new world copyright treaty. He had spent enormous time and energy to assure that any draft treaty presented to the conference for adoption would embody the reforms proposed by the White Paper. The diplomatic conference was scheduled for December; a draft that reflected the White Paper’s recommendations was about to be distributed. The domestic legislation, however, was not moving. The commissioner, therefore, decided to attack the problem the other way around. He focused his attention on getting his agenda adopted by the WIPO member nations, reasoning that when the United States signed the treaty, Congress would be obliged to adopt implementing legislation in accord with the White Paper’s recommendations.

Lehman therefore pushed a draft of the new treaty that would have required signatory nations to implement the controversial elements of the White Paper’s recommendations. One proposed article would have defined actionable reproductions to include all temporary RAM copies. Another would have guaranteed a comprehensive right to communicate works to the public, which would have incorporated Lehman’s proposal for a broad transmission right. A third article would have required countries to prohibit the manufacture or distribution of any devices or services designed to defeat technological copy-protection measures. In addition, the draft included a provision requiring nations to prohibit removal or alteration of “electronic rights management information,” that is, digital code identifying a work, its author, and the owner of rights in the work. All of these proposals echoed similar substantive elements of the original White Paper legislation. Finally, the draft contained a proposed article limiting the scope of fair use or other exceptions to copyright.[10]

Opponents of the White Paper legislation perceived the effort to foist the substance of the White Paper on the United States in the form of a new intellectual property treaty as a sneaky trick. A number of them shifted their lobbying efforts to the international arena in the hope of influencing the treaty process. Meanwhile, developing nations were unenthusiastic about the more expansive proposals. When the treaty conference convened, the majority of nations proved unwilling to sign on to the U.S. delegation’s vision of fortified copyright in cyberspace, and a number of the most controversial parts of the package were voted down in Geneva. Ultimately, the treaty text adopted by the conference incorporated few of Lehman’s ambitious proposals, and even those were substantially diluted. The proposal on temporary copies was eliminated. The new right of communication to the public was limited; delegates adopted language exempting any firm that acted as a mere conduit by providing transmission facilities. The broad technological protection proposal was weakened: while the proposal supported by the United States had prohibited the manufacture, sale, or distribution of devices or services to circumvent technical protections, the ultimate treaty language required only that signatory nations offer effective legal remedies for circumvention in connection with activities that themselves violate the copyright laws. The proposal limiting fair use and other exceptions was transformed into a proposal authorizing the extension of privileges like fair use in order to ensure their effective exercise in the digital environment. The United States signed the treaty adopted by the conference.

Treaties in the United States are not self-executing. They require Congress to enact laws that implement them. Back in the United States, therefore, negotiations shifted to the shape and language of treaty-implementing legislation. Copyright owners, disappointed by the circumscribed provisions included in the final treaty, nonetheless hoped to use the treaty as a platform to achieve more expansive objectives. At the same time, the controversy surrounding the White Paper legislation suggested that the approach most likely to result in expeditious treaty ratification was to introduce a “clean” or minimal bill that effected the legal changes prescribed by the treaty without a bunch of extra stuff. The U.S. Department of Commerce preferred the “clean bill” approach. The Clinton administration had committed itself to a general game plan in connection with all Internet regulation that required it to identify what needed to be done to facilitate electronic commerce, to do that, and to do as little as possible except for that. After the bruising copyright fight in the last Congress, it wanted to satisfy the Hollywood and Silicon Valley communities but did not want to have to expend significant political capital to do so.

If copyright owners wanted major improvements on the treaty’s prescriptions, then, they would need to be clever about how those improvements were framed. The White Paper had predicated its expansive rendition of copyright owner control on a broad interpretation of the classic reproduction right and a proposal for a new transmission right, but both of those elements of the draft treaty had been watered down considerably in the final version. Rather than expending further effort on shoring up the reproduction and transmission rights as a step in implementing the treaty, it seemed more prudent to retreat to the position that U.S. copyright law already provided expansive rights to control all transmissions and temporary reproductions.[11] At the same time, the new anticircumvention provisions adopted as part of the treaty might be used as a basis for greatly enhanced copyright owner control.

The original White Paper legislation would have prohibited the importation, manufacture, or distribution of any device or service whose primary purpose or effect was to circumvent any technological measure that prevented copyright infringement.[12] The actual treaty language was narrower: it required only that the signatory nations “provide adequate legal protection and effective legal remedies against the circumvention of effective technological measures that are used by authors . . [to] restrict acts … which are not authorized by the authors concerned or permitted by law.” Arguably, United States law met that standard. Copyright infringement accomplished through circumvention was already actionable as copyright infringement. In addition, courts imposed liability for knowing facilitation of copyright infringement on producers of devices that had no substantial noninfringing application.[13] Such liability was narrow: In the Sony Betamax case, the United States Supreme Court had refused to hold the manufacturer of a VCR liable merely because the machine could be used to make illegal copies, given that it was also “widely used for legitimate, unobjectionable purposes.” The manufacturer should be liable, the Court explained, only where a device was incapable of substantial noninfringing uses.[14]

The availability of infringement actions against circumventers who succeeded in violating copyright owners’ rights, together with the possibility of suing makers of devices that had no legitimate use, met the standard set by the treaty. Copyright owners maintained, however, that the obligation to provide “effective legal remedies” required the United States to give them the legal ability to prevent circumvention of technological protection measures from occurring at all, by first prohibiting any circumvention of technological protection, without regard to the reason for it, and second, by making any devices or services that facilitate circumvention illegal—regardless of whether the devices or services were used for legitimate purposes.

This would give copyright owners even more extensive control over the use of their works than had been proposed by the White Paper. Once a work was protected by a technological defense, the copyright owner could prohibit and prevent unauthorized use of that work, even where the use would be legal under the copyright law. If circumvention was illegal whatever the reason, it would be illegal to circumvent technological protection to make fair use of a work, or to extract uncopyrightable ideas or facts. If a work were distributed with a password that functioned to limit access, selling or loaning the work along with the necessary password could be deemed illegal circumvention. If the copyright in a work enclosed within a technological protection measure expired, users could still not circumvent the protection to gain access to the unprotected work—even if doing so were legal, the prohibition on devices or services that facilitated circumvention would make circumvention impossible for all but the most expert hackers.

User interests responded with alarm. After all, expansive anticircumvention language had been rejected by the international convention in favor of a provision that promised effective legal remedies only against those acts of circumvention committed in connection with copyright infringement. Indeed, they argued, the treaty expressly invited signatory nations to expand exceptions like fair use to ensure that they continued to be meaningful in a world in which many important copyrighted works were digitally encrypted. Armed with those treaty concessions, the coalition opposing the bill insisted that attacking devices or services was inappropriate and unwise. All the treaty required, and all that made policy sense, was to give copyright owners remedies against people who circumvented technological protection in aid of infringement and redress against others—including device makers and sellers—who deliberately facilitated circumvention for an infringing purpose. Moreover, the DFC suggested, in view of the provisions in the treaty supporting the extension of appropriate copyright limitations, the implementing legislation ought to include language extending fair use to any prohibition on circumvention. If the reason a person circumvented a copy protection measure was to make fair use of the protected work, then that circumvention should itself be legal.

The White Paper had dismissed any call for a fair use exception to anticircumvention legislation with the observation that “the fair use doctrine does not require a copyright owner to allow or to facilitate unauthorized access or use of a work.”[15] Copyright owners expanded on that theme. Fair use, they suggested, might permit some use to be made of an authorized copy of a work in very special cases, but it didn’t permit the theft of a copy. If someone wanted to buy a book, or borrow it from a library that had purchased it, he could then make fair use of it, but fair use didn’t authorize him to break into the author’s house to steal her personal copy. Breaking into technological protection, they argued, was like breaking into a home or stealing a book, and had never been permitted by the fair use doctrine. The housebreaking metaphor proved effective. It was also misleading. The thing about houses is that property laws give homeowners legal control over who gets to come in. A homeowner may therefore say: “My painting may be in the public domain but I don’t have to let you into my locked home to see it.” Backed up by that legal control, she can use protective devices—locks, burglar alarms, electrified fences, vicious attack dogs—to keep outsiders out of her home and away from her painting. The property laws about home ownership are what gives the locks and other devices their legitimacy.

Without those property rights, however, the metaphor collapses. Imagine, for example, that somebody used a lock or other protection measure (a well-trained attack dog, say) to prevent strangers from viewing some painting she didn’t own in some place she didn’t own. If I were to set my vicious attack dog to keep folks away from the Mona Lisa in the Louvre Museum, the guards would simply shoot it. The housebreaking metaphor, therefore, was inapt to support legal recognition of technological protection measures designed to prevent uses that did not invade anyone’s property rights. Using the housebreaking metaphor allowed proponents of unconditional protection against circumvention to skip right past the question whether what was inside that lock was something they were entitled to prevent people from seeing.

The United States Department of Commerce, which was superintending the task of drafting a treaty implementing bill, came up with what appeared to it to be a compromise: It would reformulate circumvention to encompass two different things. First, there was circumvention of technological protection in order to gain unauthorized access to a work. That, the department reasoned, corresponded to stealing a book or breaking into a house. It should be flatly illegal. Not only should the law prohibit trafficking in devices or services designed to facilitate that sort of circumvention, but the law should impose liability on any individual consumer who circumvented technology to gain unauthorized access. Second, there was circumvention of technological protection on a copy of a work that the circumventer was entitled to gain access to, in order to make use of the work in some way that might turn out to be copyright infringement or might turn out to be fair use. In that circumstance, it would be excessive to impose criminal liability for circumvention on individual infringers, because they would already be subject to stiff liability for copyright infringement. It was still necessary, however, to prohibit the making and selling of devices or services to facilitate this kind of circumvention, in order to prevent the widespread marketing of piracy devices under the pretext that they had some noninfringing purposes. The department apparently did not stop to wonder how consumers could engage in circumvention of copy controls for non-infringing purposes, if all devices and services to facilitate that circumvention were illegal.

The Department of Commerce incorporated its two-pronged approach into its draft treaty implementing bill. In its original formulation, circumvention to gain unauthorized access was intended to be limited to initial access: hacking into a work one had no right whatsoever to read, view, or hear. User interests pressed the administration to make that limitation explicit, and copyright owners balked. What about pay-per-view? If a copyright owner chose to distribute a digital version of a movie along with an electronic license to view the work once, and sell licenses for further viewing, someone who had bought a single view should not be permitted to hack the password protection and obtain unlimited viewings free of charge. The more that user interests pressed for some limitation on the copyright owners’ control of access, the more adamant content lobbyists became that any limitation would be unfair and intolerable.

The bills introduced at the administration’s request at the beginning of the 105th Congress were styled treaty implementation bills; they were about twice as long as the bills introduced in 1995, but contained many similar substantive provisions. Commissioner Lehman insisted that the failure of the WIPO conference to adopt his more expansive proposals demonstrated the need for United States leadership on these issues.[16] The commissioner suggested that the United States had a narrow window of opportunity to exercise world leadership by showing our trading partners, through the enactment of potent implementing legislation, that the United States interpreted the treaties to require them to take effective steps to prevent piracy of American property. Lehman argued to Congress that other nations would not act to prevent piracy of United States works until the U.S. Congress demonstrated leadership by enacting tough antipiracy laws, that, for example, made it illegal to defeat copy protection (or to market devices or services that do so) for any purpose whatsoever. Representatives of the motion picture and recording industries backed up the commissioner’s arguments with prophecies of widespread international piracy unless Congress acted quickly. The world’s eyes, they said, were on America.

Telephone companies, commercial Internet service providers, libraries, and schools insisted that an agreement setting up a safe harbor for online service providers was a precondition to the enactment of implementing legislation. Content owners, however, resisted linking the expansion in digital copyright with relief for Internet service providers. Not only were the two subjects legally unrelated, they argued, but content owners’ only meaningful weapon against online pirates was their ability to use the risk of strict liability (liability imposed regardless of fault) for subscribers’ copyright infringement to persuade Internet service providers to assist their enforcement efforts. It soon became clear to content owners, however, that the legislation could not move without a solution to the problem of Internet service provider liability. With House and Senate staffers acting as intermediaries, content-owner groups traded proposals with telephone companies and commercial Internet service providers. Early on in this series of negotiations, a staffer suggested incorporating a privilege for individual consumers who unwittingly viewed material that turned out to be infringing. Neither content owners nor service providers thought that that would be a good idea. Finally, after more than three months of intensive bargaining, the content owners and commercial service providers and telephone companies engaged in a last burst of direct negotiations. They reached a deal.

Content owners agreed that Internet service providers should not be liable for their subscribers’ infringing transmissions so long as the provider had no reason to suspect infringement was taking place, on the condition that the service provider agreed to shut down copyright violators and remove infringing material as soon as a content owner notified it of a violation. The deal did not require complaining copyright owners to substantiate their claims of infringement. Service providers agreed to turn identifying information about accused copyright violators over to complaining copyright holders, and received a complete exemption from liability in suits by subscribers complaining that their material had been improperly removed in response to a copyright holder’s complaint, or that their access had been unreasonably terminated for material wrongly alleged to be infringing. Both sides pronounced themselves satisfied.

The agreement was incorporated into a new WIPO copyright bill to be introduced in the Senate. The new bill was provisionally scheduled for markup by the Senate Judiciary Committee even before the language of its provisions took final form. Interest groups that had opposed the legislation were told by its supporters and by House and Senate staff that if they wanted to reach a bargain they needed to do so immediately, because the bill would be sent to the Senate floor either with them or without them. Most of them dealt. The form the bargains took was the addition of a variety of narrow carve-outs for interests that agreed that, in return for them, they would support the bill.[17]

The Digital Millennium Copyright Act, introduced in the Senate as S. 2037, followed the basic structure of earlier digital copyright bills, prohibiting any “circumvention” of a “technological protection measure that effectively controls access to a work” for any reason, and the manufacture or provision of “any technology, product, service, device, or component” designed to assist in circumventing technological protection. The technology, product, or service prohibition extended to both access-control protection systems and copy-control protection systems.[18] A second provision prohibited unauthorized alteration of “copyright management information”—information identifying a work, its author, its copyright owner, and any terms and conditions of use.[19]

There were narrow exemptions added for law enforcement activities, radio and television broadcasters, and cable systems. Computer software publishers received a narrow exemption that allowed them to circumvent access protection in order to analyze a computer program to enable the creation of a compatible program. Case law held that fair use permitted technical copyright infringements committed in the course of analyzing or reverse-engineering a computer program, even if the reason for the analysis were to permit the creation of a competing product, but the anticircumvention exemption was intentionally drafted to permit reverse engineering only in far more limited circumstances. Libraries received an un-asked-for and unwanted privilege to circumvent access controls for the sole purpose of browsing a protected work to decide whether to purchase it. (The exception thus neatly implied that ordinary citizens had no privilege to browse.) The portion of the bill imposing civil and criminal penalties for trafficking in technology designed to aid in circumvention, however, included no exceptions permitting such technology to be supplied to libraries or law enforcement officers in order to enable them to take advantage of their statutory privileges. Oddly, there was an exception permitting trafficking in technology to assist software designers in the exercise of their limited circumvention privilege, despite the fact that they were the only privileged group likely to have the expertise to do so without outside help. Everyone else entitled to circumvent was apparently expected to develop the facility to do so in-house.[20] The Senate Bill, introduced as S. 2037, was reported favorably by the Senate Judiciary Committee on the same day it was introduced. It went to the Senate floor a week later, where it passed 99–0.

If you’ve been keeping score as you read, you will have noticed that the interests who had not yet made a deal were the consumer electronics companies, libraries, universities and schools, and civil liberties and consumer organizations. Stymied in the Senate, they went back to the House. Representative Rick Boucher (D-Va.), who had introduced an alternative WIPO implementation Bill that opponents of S. 2037 viewed as more responsive to their concerns,[21] had been unsuccessful in persuading the rest of the House Judiciary Committee to amend the administration’s bill to answer their objections. Boucher, however, also sat on the House Commerce Committee. The Commerce Committee, which had long exercised jurisdiction over all telecommunications legislation, had expanded its jurisdiction over the past several years to embrace the Internet and electronic commerce. Since those issues were at the heart of the WIPO bill, the Commerce Committee might be persuaded to interest itself in the legislation. The consumer electronics lobby approached Commerce Committee leadership. Commerce normally left copyright matters to the Judiciary Committee, but the WIPO bill reported out of the House Judiciary Committee went far beyond copyright, containing prohibitions against the manufacture or sale of devices. Shouldn’t such provisions be under the jurisdiction of the Commerce Committee? Committee chair Tom Bliley (R-Va.) and ranking member John Dingell (D-Mich.) agreed that they should, and asked the House leadership to refer the WIPO bill to the Commerce Committee.

The content community didn’t like that idea. Things had gone well for them over in the Senate, and they’d hoped for a quick rubber stamp by the House. At best, a referral to the Commerce Committee would delay the enactment of the WIPO bill; at worst it might result in changes to it. Mitch Glazier, chief counsel for the House Judiciary Committee Subcommittee on Courts and Intellectual Property, didn’t like the idea either. Glazier had been in charge of the House digital copyright legislation since the first bill was introduced in 1995. He had very warm relations with the content community,[22] the confidence of subcommittee chair Howard Coble (R-N.C.), and the ear of Judiciary Chairman Henry Hyde (R-Ill.). After two years of further negotiations, and the help of Senate staff, he had finally reached a compromise that Glazier felt ought to satisfy any legitimate objections. He was concerned that the request for a referral had been made in the interest of delaying action on the legislation, and was impatient for the bill he had worked so hard on to become law.

Glazier persuaded Coble and Hyde that the request for a referral was a power grab by the Commerce Committee, and the congressmen raised objections. The upshot was that the House leadership agreed to make the referral to Commerce for a limited period of a few weeks, but, by the time the Commerce Committee got hold of the bill, it was already annoyed at the effort expended to prevent it from doing so. The Subcommittee on Telecommunications, Trade, and Consumer Protection scheduled an immediate hearing. During the hearing, supporters of the bill suggested gently but repeatedly that the members of the subcommittee should leave the business of making copyright law to their colleagues in the Senate and in the House judiciary committees, who understood it. Even a minor change in the text of the bill passed by the Senate, they warned, would cause the entire edifice to come crashing down, and would destroy America’s best chance to prevent widespread Internet piracy of its valuable intellectual property.

Subcommittee members made it clear that they did not want to do anything that would prevent the bill’s enactment. Nonetheless, they insisted, the bill’s opponents had raised some serious concerns that needed to be addressed. The crux of the problem was the issue of fair use. There seemed to be a difference of opinion as to whether fair use would survive the enactment of the anticircumvention provisions. The legislation passed by the Senate forbade individuals to circumvent access or copy protection systems, regardless of their reasons for doing so. Would the privilege of fair use permit circumvention in order to make fair use of a work? Would fair use allow circumvention in order to gain access to unprotected material bundled with protected expression inside of a single copy protection envelope?

Content owners responded that the anticircumvention provisions would have no effect on fair use. They resisted any suggestion that a fair use privilege be written into the legislation, however, insisting that any privilege to circumvent, even for fair use purposes, would “provide a roadmap to keep the purveyors of ‘black boxes’ and other circumvention devices and services in business” and would “reduce the legal protection for these key enabling technologies to an inadequate and ineffective level.”[23]

In the face of the impasse, Commerce Committee members indicated that the content community was being unreasonable. Perhaps in the week remaining before the Commerce Committee’s jurisdiction expired, they suggested, the content community might sit down with consumer electronics groups, libraries, universities, civil liberties and consumer organizations, and encryption researchers to see whether the problems could be solved.

The bill’s supporters took a calculated risk: they decided not to negotiate any further. Content owners had many powerful friends in the House and Senate leadership. Ultimately, they believed, the Commerce Committee was unlikely to have the stomach to block the legislation. If the content community made a deal, it might have to stick with it. If, on the other hand, it stood firm, then any unwanted amendments attached by the Commerce Committee could be removed either on the way to the House floor or in conference with the Senate.

When the subcommittee met the following week to mark up the bill, therefore, nothing had changed except for the feelings of Commerce Committee leadership. What had been mild irritation at the content community’s disregard of obvious Commerce Committee jurisdiction had grown into exasperation with what seemed to committee leadership to be unmitigated arrogance. Content was sending the Commerce Committee the message that it, and what it chose to do to the bill, didn’t matter. The committee leadership suggested that the content community think again. Having made their exasperation clear, however, committee leaders were unwilling to monkey with anything fundamental; they put their stamp on the bill by tinkering around the edges. The subcommittee approved amendments to permit circumvention for the purpose of protecting personal privacy and to relieve manufacturers of consumer electronics of any obligation to implement any and all technological protection measures a copyright owner might devise. The subcommittee adopted an amendment requiring the secretary of commerce to make annual reports to Congress on whether the enactment of anticircumvention measures was impairing individual users’ access to copyrighted works. More basic amendments to privilege encryption research and to make circumvention suits subject to traditional copyright defenses, including fair use, were introduced but not put to a vote. If the parties could reach a mutually agreeable solution to those problems before the full committee markup the following week, well and good. If not, the amendments would be considered.

Software publishers were initially reluctant to engage in further negotiations, but ultimately they were able to reach a deal on encryption research. Publishers, motion picture producers, and record companies, on the other hand, were unwilling to compromise on fair use, and trusted Senate Judiciary Chairman Orrin Hatch (R-Utah) to get them out of any uncomfortable amendments attached by the House Commerce Committee during markup. That attitude didn’t do anything to assuage the irritation of Commerce Committee leadership. In view of the content community’s demonstration of disdain for the Commerce Committee’s jurisdiction, Chairman Tom Bliley and ranking member John Dingell asked for and got a four-week extension.

But the content community, which felt as if it had been forced to reach too many deals it wasn’t thrilled with, did not want to bargain any further. Allen Adler, the chief lobbyist for the book publishers group, was frankly resentful that the Commerce Committee had dared to insist on exercising jurisdiction in the first place. He found it outrageous that Commerce Committee members, who had far less experience on copyright bills than their colleagues on the House and Senate judiciary committees, would insist that the content community make a deal that would satisfy libraries, universities, or consumer electronics manufacturers. Adler insisted that it was time for the House leadership to put an end to this turf battle between the Commerce and Judiciary committees before it jeopardized the legislation’s chances of enactment in the current session of Congress.[24] Lobbyists for Hollywood requested an urgent meeting with Speaker of the House Newt Gingrich (R-Ga.), and reported that they had received his personal assurance that the WIPO bill would pass Congress that year. Meanwhile, content lobbyists made the rounds, characterizing library, university, and consumer electronics proposals as scandalous, unprecedented, and unabashedly greedy. Under pressure from Commerce Committee staff, content owners agreed to sit down at the table with libraries and universities, but refused to make any substantive concessions. By the day before the full committee markup, no deals had been reached and further talks seemed fruitless. Commerce Committee staff circulated the text of amendments it would recommend for committee adoption in the event no compromise emerged. At that point, talks finally began in earnest.

The markup was scheduled for 10:00 a.m. Thursday, July 17. As negotiations continued, the Commerce Committee postponed its markup until 2:00 p.m., and then until 4:00 p.m. and then, finally, until 10:00 the following morning. Talks persisted throughout the evening, with staffers pressuring parties to reach some sort of deal on fair use. Content owners continued to refuse to consider subjecting the anticircumvention provisions to traditional copyright defenses. After midnight, libraries and content owners reached a compromise that nobody liked, but everyone agreed to live with.

The compromise on fair use nowhere mentioned the phrase “fair use.” Devices and services that facilitated circumvention would still be made illegal, and trafficking in them willfully or for commercial gain would still be made criminal, but the provision prohibiting end users from circumvention would be replaced with one directing the Department of Commerce to promulgate regulations forbidding any person to circumvent technological protection measures. The Commerce Department was to be instructed to conduct biennial studies directed toward identifying classes of copyrighted works that should be exempted from the regulations because of adverse impact on users of those classes of works. Significantly for members of the Commerce Committee, the Commerce Department negotiations would fall under the continuing oversight jurisdiction of the House and Senate commerce committees, although copyright matters would in general remain the business of the judiciary committees.

At 10:00 a.m., when the full Commerce Committee met officially to mark up the Digital Millennium Copyright Act, no further important opposition to the bill remained. The Commerce Committee adopted language incorporating the bargain, and voted unanimously to send the newly amended bill to the House floor. That set the stage for a turf battle between the Judiciary Committee, which had adopted one version of the legislation back in March, and the Commerce Committee, which had just voted out a significantly revised one. At stake were not only the character and shape of digital copyright law, but also the disposition of enormous sums of lobbying and campaign contribution money expended by the major copyright-affected industries.

As part of the Commerce Committee understanding, all parties had agreed to support the Commerce Committee version of the legislation in preference to any earlier versions. The initial reaction to the details on the Commerce Committee deal from those not party to the bargain, however, was negative, and the content owner organizations started backing away from the deal. House and Senate Judiciary Committee members objected to vesting decision-making authority in the controversial Department of Commerce, which Republican leaders had recently tried to abolish.[25] The onerous prescribed rule-making procedure drew criticism from all sides; under the newly drafted compromise, the Commerce Department was to precede each biennial rule making with a lengthy trial-type hearing before an administrative law judge, in which each of the myriad affected parties could call witnesses to testify and could cross-examine the witnesses called by others. The elaborate proceedings could easily have taken years. Crucial ambiguities in the agreed-upon text of the bill inspired attacks laced with worst-case scenarios.

Mitch Glazier, chief intellectual property counsel for the House Judiciary Committee, took charge of the negotiations, determined to restore the legislation to something closer to the version his committee had approved. With the support of House Republican leadership, he spoke with affected parties as well as the staff of both House committees with an eye to getting the bill back into shape and through a vote on the floor of the House before Congress broke for its summer vacation in early August.

Copyright bills never seem to get shorter, clearer, or less complicated when “improved” through the negotiation process. As initially introduced, H.R. 2281 and its companion Senate Bill were both about three thousand words long. The version of H.R. 2281 adopted by the Judiciary Committee in March had grown to more than four thousand words. By the time that the Digital Millennium Copyright Act passed the Senate, the legislation had ballooned to something in the neighborhood of ten thousand words. After the House Commerce Committee had finished its work, the bill comprised more than twelve thousand words. The version of the legislation that Mitch Glazier prepared for vote of the full House contained more than twenty-five thousand words. As the legislation passed through his hands, Glazier succumbed to the temptation to load it up with a variety of unrelated measures pending before the Judiciary Committee, including a provision to, for the first time, give federal intellectual property protection to the facts and data contained in databases and a measure to give copyright-like protection to boat-hull designs. The legislation had been assembled in a hurry, and was riddled with ambiguities, internal inconsistencies, contradictions, and obfuscatory prose. On purely aesthetic grounds, it was an ugly piece of work. It retained the procedure requiring rule making by the Department of Commerce, and thus allowed the Commerce Committee to exercise oversight jurisdiction, but it gutted many of the safeguards that the library and education communities had bargained to make part of that procedure. Glazier had come up with a formulation that satisfied both House Commerce Committee members and his friends in the content community, without giving in to libraries, universities, or consumer groups. It passed the House essentially without debate.

Glazier’s solution did not please the Senate leadership, who objected to any role for the Department of Commerce. The Senate refused to pass the House bill, insisting instead on the version of the bill that it had passed four months earlier. The end of the 105^th Congress was only weeks away, and, somehow, the House and Senate needed to agree to enact the same version of the legislation. The bill’s proponents focused on ensuring that the right Senators and Representatives were appointed to sit on a House-Senate Conference Committee, so that a reconciled and improved version could be rushed through both Houses of Congress in the final days of the session. The bill’s opposition concentrated on strategies that might delay the appointment of a conference committee or ensure that it included members unlikely to accept the revisions that the content community demanded.

By the time the conference committee met, only a few weeks remained before the election recess, and committee members from both parties were in a hurry to get a deal done and passed before they left town. The bill they put together was a hodgepodge, incorporating bits and pieces of both versions. The conference committee jettisoned some of Glazier’s late additions, but added other last-minute bargains. In final form, the Digital Millennium Copyright Act included nearly thirty thousand words and ran to more than fifty pages.[26]

As signed by the president on October 28, 1998, the DMCA incorporated two key provisions and a host of private side deals. The first key provision was the Internet service provider safe harbor, which spelled out the conditions under which service providers could avoid liability when infringing material passed through their systems. The statute identified distinct categories of problematic events that might be able to qualify for a privilege: transitory communications, system caching, hosting of subscribers’ files, and technical infringements committed through the use of search engines and other information location tools.[27] It set different rules and conditions for absolution, depending on which category the offending conduct fit into. There were further special rules and conditions for nonprofit educational institutions. None of the categories, rules, and conditions made much sense on their own terms. Rather, each set gave the wary Internet service provider an opportunity to jump through a long, complicated series of hoops and thereby avoid liability.[28]

The second major part of the law was the anticircumvention provisions. Effective immediately, it became illegal to “manufacture, import, offer to the public, provide, or otherwise traffic in any technology, product, service, device, component, or part thereof” designed to circumvent copy-protection or access-protection technology. Effective October 2000, it would be illegal for any individual to circumvent access-protection technology—that is, measures preventing unauthorized access to a work. Borrowing your brother’s password so that you can read a publication he subscribes to but you don’t is now illegal. So is using a widely distributed software utility that would permit you to view a DVD movie you purchased on a player manufactured and sold in a different region of the world and licensed to play only DVDs from that region. The statute did not prohibit individual consumers’ circumvention of copy-protection technology—that is, measures that prevent infringement of the copyright in a work. Individuals may still, if their purpose is otherwise lawful, devise a trick to defeat Macrovision, which seeks to prevent copying of commercial videotapes. They may try to save material that is posted on the Web in a hard-to-save format. They must, however, come up with ways to do this on their own. Distributing technology designed to help people circumvent copy-protection technology, whatever their reason for needing it, is illegal and in some cases criminal.[29]

The statute leaves it unclear how far the access-anticircumvention prohibition extends. That presents a problem: If “access” is understood to refer only to initial access, the statute’s distinction between circumvention of access-protection technology and circumvention of copy-protection technology (almost) makes sense. If, however, “access” includes all subsequent actions to gain access to a work, the ban on circumvention of access-protection swallows up circumvention of copy-protection as well, since one will normally need to gain access to a work in order to engage in any use of it, fair or not.

The reason for delaying the effective date of the provision prohibiting individual circumvention to gain access was to permit the rule making— now to be conducted by the Librarian of Congress in consultation with the Copyright Office and Commerce Department (thus preserving both Commerce and Judiciary Committee jurisdiction and the associated generous campaign contributions)—to identify classes of works, if any, that should be temporarily exempt from the prohibition. It would remain illegal, however, to distribute devices or perform services designed to permit individuals to engage in circumvention of access-control technology even for works ruled exempt in the rule-making proceeding.[30]

I’ve described the process in mind-numbing detail because it appears to be inexorable. Copyright legislation written by multiparty negotiation is long, detailed, counterintuitive, kind to the status quo, and hostile to potential new competitors. It is also overwhelmingly likely to appropriate value for the benefit of major stakeholders at the expense of the public at large. There is no overarching vision of the public interest animating the Digital Millennium Copyright Act. None. Instead, what we have is what a variety of different private parties were able to extract from each other in the course of an incredibly complicated four-year multiparty negotiation. Unsurprisingly, they paid for that with a lot of rent-seeking at the expense of new upstart industries and the public at large.

Even when interest groups start out with the high-minded intention of not only using the rhetoric of the public interest but actually fighting for the public interest, they end up settling for something that sells the public short. When the groups involved in the DFC agreed to withdraw their opposition to the Digital Millennium Copyright Act in return for modest concessions, most importantly the periodic government rule making that was the act’s substitute for fair use, they believed the deal they had made was the best deal that they could get. By that point in the process, it probably was. Ironically, the resulting law is substantially more pernicious than the bill originally proposed by the Lehman Working Group’s infamous White Paper. The original Lehman bill granted copyright owners sweeping new rights, but its silence on available exceptions invited the courts to apply copyright’s traditional limitations. The DMCA also grants copyright owners sweeping new rights. Its laundry list of narrow exceptions, however, discourages the inference that the classic general exceptions and privileges apply. The original Lehman bill was breathtakingly expansive but it was short. It didn’t improve the copyright law’s general level of incomprehensibility, but it didn’t greatly exacerbate it either. The DMCA is long, internally inconsistent, difficult even for copyright experts to parse and harder still to explain. Most importantly, it seeks for the first time to impose liability on ordinary citizens for violation of provisions that they have no reason to suspect are part of the law, and to make noncommercial and noninfringing behavior illegal on the theory that that will help to prevent piracy.