I The Internet Trinity II Foucault & Digital Libertarianism III Safe Harbours and Unintended Consequences IV Privatised Panopticons and Legalised Enclosures V A Communications Sampler Conclusion

IV

Privatised Panopticons and Legalised Enclosures

I have argued elsewhere that the current government proposals for the "reform" of copyright on the Internet weigh only the costs of cheaper copying rather than its benefits, underestimate the importance of fair use to competition policy and free speech, fail to recognise the unique features of both intellectual property and networked environments, and apply bad economic analysis to an even worse depiction of current law.(50) Leaving aside the virtues or vices of these proposals aside for the moment, I will focus here on the methods by which they were to be implemented.

One of the key problems for any Internet copyright regime is enforcement. The Internet trinity I discussed earlier would seem to apply with particular strength to the problem of policing copyright on a global distributed network. The technology is resistant to control, the subject matter of the regime is intangible and trivially easy to circulate, and both the content and the people regulated by the regime are frequently beyond the jurisdiction of the sovereign in question. The combination of these circumstances has produced a series of warnings that intellectual property law was doomed because neither its conceptual structure nor its enforcement mechanism could survive 'being digital.'(51) The best known of these warnings is also the best written.

The riddle is this: if our property can be infinitely reproduced and instantaneously distributed all over the planet without cost, without our knowledge, without its even leaving our possession, how can we protect it? How are we going to get paid for the work we do with our minds? And, if we can't get paid, what will assure the continued creation and distribution of such work? Since we don't have a solution to what is a profoundly new kind of challenge, and are apparently unable to delay the galloping digitization of everything not obstinately physical, we are sailing into the future on a sinking ship. This vessel, the accumulated canon of copyright and patent law, was developed to convey forms and methods of expression entirely different from the vaporous cargo it is now being asked to carry. It is leaking as much from within as without. Legal efforts to keep the old boat floating are taking three forms: a frenzy of deck chair rearrangement, stern warnings to the passengers that if she goes down, they will face harsh criminal penalties, and serene, glassy-eyed denial.(52)

If one saw these technological transformations as mainly a threat to both the copyright owner and the enforcement power of the state, how would one respond, particularly if one took seriously the difficulties in policing that the Internet trinity points out? One would try to focus on building the regime into the architecture of transactions in the first place -- both technically and economically -- rather than policing the transactions after the fact. More concretely, one would want to escape from the practical and legal limitations of a sovereign-citizen relationship. Thus one might seek out private actors involved in providing Net services who are not quite as mobile as the flitting and frequently anonymous inhabitants of cyberspace. In this case, the parties chosen were the Internet Service Providers. One would pin liability on them and leave it up to them to prevent copyright infringement through technical surveillance, tagging and so on, and to spread the cost of the remaining copyright infringement over all the users of their service, rather than all the purchasers of the product in question.

By enlisting these nimbler, technologically savvy players as one's private police, one would also gain another advantage; freedom from some of the constitutional and other restraints that would burden the state were it to act directly. Intrusions into privacy, automatic scrutiny of e-mail, curtailing of fair use rights so as to make sure that no illicit content was being carried; all of these would occur in the private realm, far from the scrutiny of public law. There are advantages to privatising the Panopticon, it turns out.

Given all these "advantages" it is unsurprising to find that strict liability for on-line service providers became a central feature in the Clinton administration White Paper,(53) the Bills implementing its ideas(54) and the US's proposals for the WIPO treaties in Geneva.(55) The specifics of this proposal were relatively simple. On-line service providers were to be made strictly liable for copyright violations committed by their subscribers -- in part this was done by an expansive definition of fixation so that even holding a document in RAM memory as it was browsed, would constitute the creation of a copy.(56) Clearly then, the relatively more stable versions held in a server's disk cache or stored temporarily in its computers would count as copies. The theory also depended on the notion that we should analogize the on-line service provider to an innocent but infringing photoshop and thus impose strict liability as a direct infringer, rather than analogizing the service provider to a business that rented Xerox machines by which material could be copied illegally, which would be liable only if it was guilty of contributory infringement.(57) Notably this theory was rejected by the only court to have faced it squarely.(58)



In one sense this strategy is very similar to the use of strict liability elsewhere in the legal system -- and of course it can be understood entirely without reference to the Foucauldian gloss. (Although one must note that the the conventional reasons for imposing strict liability are strikingly absent.(59)

With or without Foucault, however, thinking about the use of strict liability as an enforcement mechanism does illustrate the limitations of the Austinian view of the state's exercise of power. (Unsurprisingly perhaps, Austin argued againt strict liability and judges under the influence of Austinian reasoning actually declared that strict liability was not true law.)(60) My central point here is not the undesirability of strict liability for on-line service providers, though the rationale, legal basis and constitutionality of sucha system seem doubtful to me. Rather, I think that the possible impact of a strict liability system on actual privacy, speech and discourse indicates another limitation of the jurisprudence of digital libertarianism. Onee again, the focus on public, criminal and sanction-backed acts by states exercising their power directly, tends to obscure and thus to undervalue the efficacy of efforts that rely on privatised enforcement and surveillance, cost spreading and the use of "material coercion rather than the physical existence of a sovereign."

It is to the latter point that I now turn. One prong of the Administration's plan for copyright on the Net depended on enrolling private actors to act as enforcement agents in a way that sidestepped the rights, duties and privileges between citizen and sovereign. The other prong depended on coating technological anti-copying devices with the authority of the law in such a way as to change the relative powers of current copyright holders on the one hand and their customers and future competitors on the other. The two most important provisions are the "circumvention of copyright protection systems" section and the "integrity of copyright management information" section of the NII Copyright Protection Act of 1995.(61) Similar provisions were proposed by the United States during the WIPO conference.(62)

These two provisions seem on first sight to be entirely unobjectionable. The circumvention section imposes civil liability on importers, manufacturers and distributors of devices the primary purpose or effect of which is to circumvent a copyright protection system.(63) The management section imposes civil and criminal liability on someone who removes or tampers with copyright management information.(64) Obviously technological protections are going to be an important way by which digital intellectual property is safeguarded and these technological protections will include, among other things the kind of deeply embedded information that the management information section protects. Documents will keep track of how many times they are read and may complain if they are read too much or by the wrong person. Pamela Samuelson calls these "texts that rat on you." Digital books sold to one person may be encoded so that they can't be read by someone else on another computer. Given the possibility of documents that have the copyright details bound into in every packet of data, and which also check themselves to be sure that no alterations have been made, quotation may be perceived as alteration. (Presumably internet service providers would also be encouraged to introduce some system of scanning which looked for altered or unauthorised packets of data.)

The point about all of this, is that there will be a continuing technological struggle between content providers, their customers, their competitors and future creators. Obviously it will sometimes be in the interest of content providers to make it as hard as possible for citizens to exercise their fair use rights. They will try to build technological and contractual fences around the material that they provide, not just to prevent it being stolen, but to prevent it from being used in ways that have not been paid for, even if those uses are privileged under current intellectual property law. They may want to stop their competitors from achieving "interoperability" or prevent their customers from selling second hand versions of their producs. The technical means to do this can be thought of digital fences. Sometimes those fences will be used to stop clear violations of existing rights. Sometimes they will be used to enclose the commons or the public domain. Thus by making it illegal or impractical for me to go around through or over the fence, the state adds its imprimatur to an act of digital enclosure. The Internet trinity tells us that information wants to be free and that the thick fingers of Leviathan are too clumsy to hold it back. The position is less clear if that information is guarded by digital fences which themselves are backed by a state power maintained through private systems of surveillance and control.

JUMP TO NEXT SECTION


I The Internet Trinity II Foucault & Digital Libertarianism III Safe Harbours and Unintended Consequences IV Privatised Panopticons and Legalised Enclosures V A Communications Sampler Conclusion


ENDNOTES TO SECTION IV

50. See James Boyle, Intellectual Property Online: A Young Person's Guide, 10 Harv. J. L. & Tech. 47 (1996); James Boyle, Shamans, Software, and Spleens 18-20, 51-61, 162-63 (1996); James Boyle, Q: Is Congress turning the Internet into an information toll road? Yes: The Senate would whack away at 'fair use' of electronic documents needed for news and education, Insight, Jan. 15, 1996; James Boyle, Sold Out N.Y. Times, March 31, 1996, § 2, at 2.

51. See Nicholas Negropronte, Being Digital (1995).

52. See John Perry Barlow, Selling Wine Without Bottles: The Economy of Mind on the Global Net,

<http://www.eff.org/pub/Publications/John_Perry_Barlow/HTML/idea_economy_ article.html>

53. United States Department of Commerce, Information Infrastructure Task Force, Intellectual Property and the National Information Infrastructure: the Report of the Working Group on Intellectual Property Rights at 114-24 (Sept 1995) ("White Paper"); James Boyle, Intellectual Property Online: A Young Person's Guide, 10 Harv. J. L. & Tech. 47, 58-111 (1996); Niva Elkin-Koren, Copyright Law and Social Dialogue on the Information Superhighway: The Case Against Copyright Liability of Bulletin Board Operators, 13 Cardozo Arts & Ent. L. J. 345 (1995); Cf., Religious Technology Center v. Netcom, 907 F. Supp. 1361, 1377 (N.D. Calif. 1995) (stating that strict liability for ISPs "would chill the use of the Internet because every access provider or user would be subject to liability when a user posts an infringing work to a Usenet newsgroup." Id. at 1377).

54. See NII Copyright Protection Act of 1995, S. 1284, 104th Cong. (1995), H.R. 2441, 104th Cong. (1995).

55. See WIPO Copyright Treaty, Dec. 23, 1996, CRNR/DC/94 (visited June 26, 1997) <http://www.wipo.org/eng/diplconf/distrib/94dc.htm>; See also, News from WIPO (visited June 26, 1997) <http://www.hrrc.org/wiponews.html> (detailing course of deliberations during the Diplomatic Conference).

56. See James Boyle, Intellectual Property Online: A Young Person's Guide, 10 Harv. J. L. & Tech. 47, 830194 (1996) (discussing MAI Systems Corp. v. Peak Computer, Inc., 991 F.2d 511 (9th Cir.1993)).

57. Id. at 103-04.

58. See Religious Technology Center v. Netcom, 907 F. Supp. 1361 (N.D. Calif. 1995); See also, Playboy Enterprises, Inc. v. Chuckleberry Publications, Inc., 939 F. Supp. 1032 (1996); Sega Enterprises, Ltd. v. Maphia, 948 F. Supp. 923 (1996);

59. We impose strict liability on manufacturers on products for a number of reasons -- one of which is that we believe the state could not possibly inspect every product and every design in the market-place. Simply by forcing manufacturers to internalise the costs of injuries caused by their products, we produce a strong, private set of incentives that in turn encourage internal mechanisms of review and product redesign. See Guido Calabresi, The Cost of Accidents: A Legal and Economic Analysis (1970); See also, Guido Calabresi, First Party, Third Party, and Product Liability Systems: Can Economic Analysis of Law Tell Us Anything About Them?, 69 Iowa L. Rev. 833 (1984); A. Mitchell Polinsky, An Introduction to Law and Economics at 97-106 (2d ed. 1989). Plaintiffs become private attorneys-general. There are however, also some striking differences between the familiar example of the use of strict liability in the tort setting and the imposition of strict liability on internet service providers. In product liability, the conventional range of reasons for imposing strict liability on the manufacturers includes the claims that:

They are generally the cheapest cost-avoiders -- in other words, they are best able to respond to liability for damage by making changes that could prevent the damage

They are generally the best loss spreaders -- in other words, they are best able to pass the cost of unavoidable or cost-justified damage on to the appropriate group, consumers of the good in question.

They are generally in an advantageous position in terms of knowledge and effective power -- at least as compared to the relatively powerless individual consumer. See Escola v. Coca Cola Bottling Co., 150 P.2d 436, 440-43 (1944) (Traynor, J., concurring).

In the online setting, none of these claims is obviously correct. In some cases service providers may be able to prevent illicit copying relatively cheaply without imposing large social costs. On many other occasions however, it seems that the costs of their enforcement may outweigh the benefits -- in the form of transaction costs required to ensure compliance, for example, or draconian restrictions of the fair use privileges of their subscribers so as to be sure that illicit copying is not being carried on. (Since ISP's would pay for all detected copyright infringements, but would not be forced to internalise the cost to their customers of restricting fair use, the incentives would be asymmetrically anti-consumer.) Leaving aside the efficiency costs of enforcement by service providers, there is also the question of whether they are the cheapest cost-avoider. In many cases, the party best situated to avoid the cost of copyright infringement will be the owner of the copyright. Whether by developing technical solutions or by fine-tuning their business plan so as to minimise the incentives to violate copyright in the first place, copyright owners might well be the cheapest cost-avoiders. If that is true, it would actually be inefficient to allow them to rely on another party for enforcement of their rights.

Beyond the question of the cheapest cost-avoider is the question of best loss spreader and here too it is hard to be confident that the ISP's are the appropriate parties. The economic analysts' mantra is "activities should internalise their full costs." If the costs of a good or activity are not passed on to those who use the good or engage in the activity, then those individuals will make inefficient choices. Thus, for example, if the price of gasoline does not reflect the environmental damage done by gasoline, that damage becomes a negative externality, and gasoline is inefficiently priced relative to its "true" costs. Over what group then, should the costs -- i.e. the copyright owner's forgone profit -- of illicit copying be imposed? The inquiry is a fascinating one, with more layers than I can fully explore here. It is complicated by the fact that the "costs" imposed by the illicit copying of an information good are economically different in some ways from the costs imposed by theft of material goods. As a content provider, I can make a rational economic decision to sell my good across some cheap but "leaky" medium, which lowers my costs of advertising and distribution and increases the number of unauthorised copies circulating. I may even believe that some of the unauthorised copies provide a benefit to me -- making my word processing program a de facto standard in the industry or establishing my band as the best known, thus increasing the market for future products. But let us leave aside the joys of pointing out that economic analysis depends on questions of interpretation that cannot themselves be decided according to economic criteria. There is at the very least, strong reason to doubt that users of on-line services, rather than purchasers of the good in question, are the appropriate group over whom the costs of illicit copying should be spread. This would, in fact, actively undermine the competitive incentives to companies to develop their own anti-copying methods.

Finally, the asymmetry of power and knowledge that occurs when Mrs. McPherson confronts the Buick Motor Company, is by no means as clear when Microsoft wants Netcom to do its enforcement work. For all of these reasons, the imposition of strict liability on ISP's does look rather different than its imposition on manufacturers of defective products. If there is an advantage to this scheme, that advantage redounds mainly to the content providers; such a plan would shift enforcement costs from owners and allow them to reap the benefits of the Net without fully bearing its costs.

60. See 2 John Austin, Lectures on Jurisprudence 136 (5th ed. 1885)

61. See infra note ___.

62. See WIPO Copyright Treaty, Dec. 23, 1996, CRNR/DC/94 (visited June 26, 1997) <http://www.wipo.org/eng/diplconf/distrib/94dc.htm>; See also, News from WIPO (visited June 26, 1997) <http://www.hrrc.org/wiponews.html> (detailing course of deliberations during the Diplomatic Conference).

63. See supra note 52 at § 1201.

64. See supra note 52 at § 1204.

JUMP TO NEXT SECTION