Note: This HTML text faithfully reproduces the text of the previously published, print version. Notation such as "[p. 557/p. 558]" indicates the print version's pagination. Print version (C) 1998 North Carolina Law Review; reproduced here with permission. HTML derivative work (C) 1998-99 Tom W. Bell.
In this Article, Professor Bell examines the impact of new technologies on copyright's fair use doctrine. The Article examines the prospective capabilities of automated rights management technologies to monitor and track the exchange of information in digital intermedia, such as the Internet, that would enable copyright holders to bill consumers for use of their works. Professor Bell argues that these billing capabilities will cause a transformation in copyright law: a system of "fared use" will radically reduce the scope of the "fair use" defense. Upon examination of the effects of such a transformation, Professor Bell posits that a system of fared use actually may offer freer access to expressive works. Professor Bell argues that allowing copyright owners and consumers to exit copyright law and freely contract under a fared use system in time may reveal a system more beneficial than one preempted by federal copyright law. Professor Bell concludes by urging lawmakers and academics to await the emergence of new automated rights technologies and allow experimentation in the market to dictate copyright law's adaptation to such new technologies, rather than requiring new technologies to adapt to the traditional fair use doctrine.
I. INTRODUCTION 558
II. THE FUTURE OF FAIR USE? 563
[p. 557/p. 558]
A. Automated Rights Management 564
B. Fair Use Shrinks as Licensing Grows 567
1. American Geophysical and Princeton University Press 567
2. The NII White Paper and the NIICPA 571
C. The Dystopian View 573
1. The End of Unbilled Use in Digital Intermedia 574
2. Private Censorship by Copyright Owners 577
III. THE FUTURE REVISITED: FROM "FREE USE" TO FARED USE 579
A. Fair Use Versus "Free Use" 579
1. Fair Use Is Not Free Use 580
2. Fixing Market Failure 581
3. Maintaining Copyright's Quid Pro Quo 585
4. Calculation and Experimentation 590
B. Objectionable Use Remains Fair Use 592
C. The Advent of Fared Use 596
1. The Reciprocal Quasi-Compulsory License 596
2. The Net Benefits of Fared Use 600
IV. CONTRACTUAL MODIFICATIONS OF FARED USE 600
A. The Utility of Fared Use Contracts 601
1. Anti-Criticism Contracts 601
2. Anti-Censorship Contracts 603
B. The Enforceability of Fared Use Contracts 604
1. Fared Use Contracts as Shrinkwrap Licenses 605
2. Fared Use Contracts as Adhesion Contracts 607
3. Preemption by § 301 608
4. Preemption by the Supremacy Clause 610
C. Exit from Copyright 614
V. CONCLUSION 618
"Information wants to be free," claim those who decry the over-zealous enforcement of copyrights.[1] But they [p. 558/p. 559] cannot mean what they say. Information wants nothing at all.[2] This contemporary epigram instead reveals, indirectly, what people want. Perhaps it originated as a mere prediction about the difficulty of enforcing intellectual property rights on the Internet and in other digital intermedia.[3] Facing newly effective means of enforcing such rights, however, those who claim "information wants to be free" increasingly give the slogan a normative spin. Its meaning then boils down to this: people want information for free.[4]
So restated, the catch-phrase still rings true. All else being equal, who would not prefer to get information--that increasingly vital good--at no cost? But, alas, information never comes for free. We can only account for its costs as fully as possible, try our best to minimize them, and allocate them fairly.
The information economy balances fixed and variable costs from several sources. On the one hand, consumers necessarily bear costs when they search for, interpret, and collect information. This holds true even when--perhaps especially when--the fair use defense to copyright infringement allows a consumer to avoid paying cash for the right to use an expressive work.[5] On the other hand, information providers necessarily bear costs when they create, package, and distribute information. They thus seek remuneration and profit through licensing fees, at least so far as the countervailing fair use defense and various practical hurdles will allow.[6]
One method in particular offers information providers a promising way to increase their licensing opportunities: automated [p. 559/p. 560] rights management ("ARM").[7] ARM enables information providers to enforce standard copyright claims mechanically, without resort to the threat of litigation. It also allows copyright owners and others to create and enforce contracts that specify other sets of rights.[8] Although ARM may give information providers newfound power to control the use of their wares, it does not necessarily justify that control. The proper legal response to ARM thus remains an open--and vital--question.
ARM portends far-reaching and unprecedented effects on rights to information in the new digital intermedia. Specifically, ARM threatens to reduce radically the scope of the fair use defense to copyright infringement.[9] ARM will interact with existing legal [p. 560/p. 561] doctrines to supplant fair use[10] with an analogous but distinctly different doctrine: fared use.[11] Part II of this Article details the causes of this transformation and closes with a worst-case scenario.
Part III, in contrast, catalogs the benefits of fared use. It describes the legal and practical scope of fared use and defends it as not only efficient, but equitable.[12] Fared use would make copyrighted works in the digital intermedia available under reciprocal quasi-compulsory licenses. Although consumers might have to pay fees that the fair use defense would excuse in other media, they would in return gain better access to better information. Fared use would not necessarily cost consumers more, however, because the fair use doctrine now imposes considerable hidden costs.[13] Though they might win increased licensing fees under fared use, copyright owners would, thanks to a diminished but still potent fair use defense, have to endure greater exposure to the objectionable reuse of their works.
A default rule of fared use would create a new public bargain between consumers and copyright owners that would largely flow from the interaction of ARM with current law. Part IV argues, however, that consumers and providers of information should have the right to shape alternative licensing systems through private agreements. Although we cannot yet discern exactly what sort of rights to information would evolve under contracts backed by ARM--an excellent reason for lawmakers to avoid premature meddling in the field[14]--a plausible accounting shows that the public, [p. 561/p. 560] as well as the contracting parties, can benefit. Entrepreneurs can create a world where information costs less than it does under fair use, and perhaps even one where the public gets paid to consume information. To encourage experimentation, therefore, lawmakers should not flatly invalidate fared use agreements that skirt preemption. Rather, they should allow information consumers and providers to exit freely from copyright law into contract law.
The impact of technological advances on copyright has generated a large and interrelated body of commentary. It thus bears noting what the present Article does not concern. It does not address how automated rights management might affect consumers' privacy.[15] Nor does it spend much time on the comparatively easier question of how ARM might interact with the first sale doctrine.[16] Despite this Article's occasional cites to the recent NII White Paper,[17] it leaves detailed criticism of that controversial document to others.[18] Broad questions about the justifiability and continued viability of copyright law exceed the bounds of the present inquiry.[19] The feasibility and [p. 562/p. 563] desirability of relying solely on contracts and trade secrets to protect expressive works deserves more attention, but the scope of that particular issue demands a separate article.
Both friends and foes of copyright can find something of use in this Article. It provides the former with a defense of widespread licensing and the latter with a graceful exit to a copyright-free world. Readers who hold the middle ground, preferring fuzzy copyrights and broad fair use privileges, may find the present account a bit bracing. Ultimately, though, they will find that the fared use system described here achieves many of their most cherished policy goals.
However esoteric a topic to the rest of the world, the proper scope of the fair use doctrine[20] evokes heated debate among people who peddle information for a living.[21] A confluence of recent events has especially alarmed scholars, librarians, and other parties who rely on the fair use defense in conventional media.[22] From technical advances in automated rights management, described by Subpart A, and recent developments in copyright law, described by Subpart B, these parties forecast a dystopia, described by Subpart C, where fair use will have virtually disappeared from the digital intermedia.
Owners of conventional sorts of property do not rely [p. 563/p. 564] on the law alone to protect their assets. They also deploy fences, locks, and guards. Automated rights management provides the owners of intangible assets with similar defensive mechanisms, albeit ones built into computer hardware and software and implemented via firewalls, encryption, and passwords. Although ARM researchers continue to develop and experiment with a variety of approaches, the huge market for intellectual property protection virtually ensures that ARM technology will see increasingly widespread use.[23]
ARM appeals to information providers because it stands to give them the power to accomplish two things that hitherto seemed impossible in digital intermedia. First, ARM will make it possible and cost-effective for information providers to enforce standard copyright or trade secret claims. Second, ARM will empower them to enforce contracts that define different or additional rights.[24] Additionally, ARM promises to perform these functions cleanly and effectively, without resort to uncertain and wasteful litigation.
Until very recently, the promise of automated rights management remained just that: a mere promise. With so much to [p. 564/p. 565] gain at stake, however, researchers worldwide have been striving to perfect ARM technology.[25] A wide variety of ARM systems already have left laboratories and entered the marketplace.[26] IBM has convinced thirty companies, including America Online, Yahoo!, and Xerox, to employ its ARM technology.[27] Competitor Electronic Publishing Resources, Inc. ("EPR") has signed agreements with National Semiconductor and the Copyright Clearance Center.[28] As typically occurs with a new networking technology, these and other ARM providers have struck up a variety of shifting consortia.[29] Such activity indicates that copyright owners and information providers soon will have access to robust ARM technology from one or more sources.[30]
Subscription services, which charge for access to proprietary databases by the hour or month, currently account for most of the intellectual property sold through digital networks.[31] Although even [p. 565/p. 566] this sort of electronic tollbooth qualifies as "automated rights management" in a broad sense,[32] most ARM technicians aspire to something more sophisticated. At a minimum, they aim at permitting information providers, and perhaps even individual owners of proprietary data, to sell access on a document-by-document basis.
The simplest such pay-per-use systems offer encrypted documents for sale, or rather the keys to those documents, one at a time. The purchaser of a key gets access to a single locked document.[33] This approach still leaves the purchased information subject to subsequent copying in its original medium, however.[34] More sophisticated ARM systems thus employ methods such as steganography,[35] micropayments,[36] and imbedded applications[37] to give information providers exact and continuous control over proprietary information.[38]
At its most powerful, ARM supports the "superdistribution" of [p. 566/p. 567] proprietary information. In other words, it allows information providers to market documents that disallow certain types of uses (e.g., copying) and provide continuing revenue (e.g., charging 2¢ per access) regardless of who holds the document (e.g., including someone who obtained it post-first sale).[39] Superdistribution thus offers information providers a rather daunting compendium of powers. In practice, of course, no ARM system can guarantee absolute control over information, especially after it escapes digital media.[40] By accident or design, documents inevitably will fall outside the reach of ARM. Even if only partially effective, however, ARM will radically improve the efficiency of licensing practices in the digital intermedia. Consequently, it will have a radical impact on the fair use doctrine.
Current case law makes it harder for defendants to benefit from the fair use defense to the extent that plaintiffs make it easy to pay licensing fees. The Second Circuit in American Geophysical Union v. Texaco, Inc.,[41] expressly supported this view.[42] The Sixth Circuit panel in Princeton University Press v. Michigan Document Services, Inc.,[43] rendered a holding to the contrary, until vacated and subjected to an en banc rehearing. The Sixth Circuit's final, binding opinion strongly endorsed the reasoning of the American Geophysical court.[44]
In American Geophysical, eighty-three publishers of scientific and technical journals joined in a class action copyright infringement suit against Texaco, alleging that the defendant had made unauthorized photocopies of individual articles from their [p. 567/p. 568] publications.[45] Among other defenses, Texaco cited fair use.[46] It lost in the district court,[47] appealed to the Second Circuit, and lost yet again.[48]
The outcome of the case turned on the crucial fourth factor of the fair use defense,[49] which requires a court to consider "the effect of the use upon the potential market for or value of the copyrighted work."[50] Citing the district court's findings of fact, the court of appeals held that the publishers had, with the help of the Copyright Clearance Center, created "a workable market for institutional users to obtain licenses for the right to produce their own copies of individual articles via photocopying."[51] This made it difficult for Texaco to claim that its unauthorized copying had no impact on revenue the plaintiffs might have earned from their copyrights. As the court of appeals explained, "the right to seek payment for a particular use tends to become legally cognizable under the fourth fair use factor when the means for paying for such a use is made easier."[52]
Because he disputed whether the plaintiffs had created a workable market for licensing copies of individual articles, Judge Jacobs dissented. Absent such a market, he opined, the majority's reasoning ran in circles: "[T]he market will not crystallize unless courts reject the fair use argument that Texaco presents; but, under [p. 568/p. 569] the statutory test, we cannot declare a use to be an infringement unless (assuming other factors also weigh in favor of the secondary user) there is a market to be harmed."[53]
The majority opinion rebutted this accusation, explaining that "[t]he vice of circular reasoning arises only if the availability of payment is conclusive against fair use."[54] The majority thus tested Texaco's copying against all of the fair use factors set forth in § 107 of the Copyright Act, and limited consideration of the fourth factor to the impact on "traditional, reasonable, or likely to be developed markets."[55] In this case, the court found, the market in question already existed. Therefore, "[w]hatever the situation may have been previously, before the development of a market for institutional users to obtain licenses," the American Geophysical court reasoned, the development of such a market made it "now appropriate to consider the loss of licensing revenues."[56] The court determined that congressional intent supported its decision.[57]
The majority might have added that, so far as the dissent's accusation of circularity applies, it applies equally well to the presumed alternative of freezing fair use in the face of technological advances. If the law refuses to allow copyright owners to benefit from reduced transaction costs, they will not pursue the technological innovations that would make such reductions possible.[58] Failure to [p. 569/p. 570] upgrade the fair use doctrine would thus ensure that neither producers nor, notably, consumers enjoy all the benefits that the digital intermedia might otherwise offer.[59]
Princeton University Press v. Michigan Document Services, Inc.[60] presented the Sixth Circuit with a copyright infringement case strongly reminiscent of American Geophysical. Princeton University Press and other publishers jointly sued the defendant copy shop for reproducing excerpts from their copyrighted works in course packs it sold to university students.[61] Here, as in American Geophysical, the defendant bypassed a mechanism for paying licensing fees and instead pleaded the fair use defense.[62] In the Sixth Circuit's first attempt at resolving Princeton University Press, the panel reversed the district court's ruling in favor of the defendant.[63] Like the dissent in American Geophysical, the judges found the plaintiff's argument circular.[64] The panel failed to convince its colleagues on the Sixth Circuit, however; sitting en banc, the Sixth Circuit vacated the panel's decision.[65]
In reconsidering the issue, the Sixth Circuit in Princeton University Press closely followed American Geophysical and held that "[w]here . . . the copyright holder clearly does have an interest in exploiting a licensing market--and especially where the copyright holder has actually succeeded in doing so--'it is appropriate that potential licensing revenues for photocopying be considered in a fair use analysis.' "[66] The court similarly agreed with American Geophysical that "[o]nly 'traditional, reasonable, or likely to be developed markets' are to be considered in this connection."[67]
Indeed, the court in Princeton University Press arguably did even more than the American Geophysical court did to ensure that the fair use defense will shrink as licensing opportunities expand. In American Geophysical, the court downplayed the significance of [p. 570/p. 571] the one fair use factor that most strongly supports the rise of fared use--the factor that requires a court to consider "the effect of the use upon the potential market for or value of the copyrighted work"[68]--by interpreting the Supreme Court's decision in Campbell v. Acuff-Rose Music, Inc., as an indication that the Court was "abandoning the idea that any factor enjoys primacy."[69] The Princeton University Press court, however, read the precedents differently, stating that "[w]e take it that this factor, 'the effect of the use upon the potential market for or value of the copyrighted work,' is at least primus inter pares."[70] By emphasizing that fair use should not cut into licensing revenues, the Princeton University Press court helped to ensure that fair use will wither as the use of ARM grows.
Taken together, American Geophysical and Princeton University Press demonstrate that courts considering the question have concluded that the fair use defense should give way when copyright owners can conveniently collect licensing fees. The executive branch addressed this same question in the NII White Paper[71] and reached the same conclusion.[72] The NII White Paper's statutory proposals have, moreover, drawn support in the legislative branch as the National Information Infrastructure Copyright Protection Act ("NIICPA").[73]
[p. 571/p. 572]
The NII White Paper predicts not only the advent of comprehensive ARM systems,[74] but also that such systems may "lead to reduced application and scope of the fair use doctrine."[75] Moreover, it encourages this trend. Following the lead of American Geophysical, the NII White Paper would empower copyright owners to use any "process, treatment, mechanism, or system" to deny access to their works.[76] In response to claims that this would violate the public bargain implicit in fair use, the NII White Paper responds that the doctrine "does not require a copyright owner to allow or to facilitate unauthorized access or use of a work."[77]
To give effect to its arguments on how best to regulate digital communications, the NII White Paper suggests particular amendments to the Copyright Act of 1976.[78] These amendments already have surfaced in Congress, taken word-for-word from the NII White Paper, and proposed in the NIICPA.[79] Rather than expressly requiring that courts limit the fair use defense insofar as opportunities to pay licensing fees expand, the NIICPA merely aims to protect--and thus encourage--ARM systems. It would do so by adding to the Copyright Act a new Chapter 12, entitled "Copyright Protection and Management Systems," including proposed §§ 1201-04.
The NIICPA's proposed § 1201 prohibits devices or services "the primary purpose or effect of which is to avoid, bypass, remove, deactivate, or otherwise circumvent . . . any process, treatment, mechanism or system which prevents or inhibits violation of any of the exclusive rights of the copyright owner under section 106" of the Copyright Act.[80] In effect, § 1201 almost certainly would prohibit anti-ARM systems. The proposed § 1202 bars tampering with [p. 572/p. 573] "copyright management information,"[81] including "terms and conditions for uses of the work,"[82] that copyright owners attach to their digital works. Section 1202 thus also aims to make the world safer for ARM systems. The NIICPA's proposed §§ 1203-04 give teeth to these protections by imposing civil and criminal penalties, including up to $500,000 in fines and five years in jail, for their violation.[83]
The NII White Paper and the NIICPA so far represent mere aspirations--not law. But they powerfully indicate that the current administration intends to give the American Geophysical holding broad effect. Though the intent of Congress remains uncertain, it has not of late shown very much independence in matters relating to digital communications.[84] At any rate, as the holdings in American Geophysical and Princeton University Press indicate, the judicial branch will not need the NIICPA to ensure that the fair use doctrine shrinks in accord with the growth of ARM licensing schemes. Indeed, the NII White Paper and the NIICPA have justly drawn criticism for doing far more to promote ARM than is necessary--or wise.[85]
Parties who directly benefit from appeal to the fair use doctrine in conventional media tend to fear the loss of those benefits in digital intermedia. They thus regard with alarm the technical and legal developments described above,[86] and argue that automated copyright [p. 573/p. 574] management, combined with the doctrine that licensing can supplant fair use, creates the prospect of a copyright dystopia.[87] This section describes, in two related portraits, what friends of conventional fair use fear. Section 1 describes a world in which the fair use doctrine no longer forgives the refusal to pay licensing fees in digital intermedia. Section 2 takes the same scenario a step further, describing a world in which the fair use doctrine does not protect such uses as the copyright holder finds objectionable. Critical analysis of these scenarios follows in Part III.
Parties who make frequent appeal to the fair use doctrine fear that they will not enjoy in digital intermedia the sort of immunity from licensing fees that fair use provides in traditional media.[88] [p. 574/p. 575] Because a copyright does not give its owner the right to require licensing fees for uses falling within the scope of § 107,[89] parties availing themselves of fair use typically regard it as free use.[90] The prospect of paying in digital intermedia for what apparently comes free elsewhere thus evokes contempt, if not horror, from those who re-work information for a living.[91] It seems certain, however, that the [p. 575/p. 576] combination of automated rights management and the legal doctrines set forth in American Geophysical,[92] Princeton University Press,[93] the NII White Paper,[94] and the NIICPA[95] will have just such an effect.
The sort of copyright nightmare that friends of fair use fear unfolds something like this:[96] Imagine that you download some quotes from text on the Church of Technolism's web page, incorporate them in an essay critical of the Church, and publish it on your own web page without seeking the Church's permission. The Church of Technolism objects to your comments and demands that you cease republishing quotes from its online document. You refuse. Although the Church sues for copyright infringement, you rest easy in your faith that the fair use defense will excuse your copying. But it does not.
The court finds not that you have quoted too much, but rather that you have paid too little. Although you quoted no more than what your critique required, you neglected to pay the Church of Technolism for your republication of its copyrighted materials. Citing American Geophysical, Princeton University Press, the NII White Paper, and the recently enacted NIICPA, the court holds that the advent of convenient and cheap means of making licensing payments forecloses your claim that paying for your use would have proven too burdensome. In short, as automated rights management has grown in scope, your fair use defense has shrunk.
As this scenario illustrates, the near future may hold some changes for the fair use doctrine. Whether these changes qualify as beneficial or detrimental will come under consideration in the next Part. The next section will first show, however, that fair use faces something more worrisome than ubiquitous licensing fees. The same events that conspire to end the unbilled use of copyrighted works in digital contexts, together with related factors, threaten to constrain [p. 576/p. 577] severely the sorts of critical commentary and parodies that copyright owners so often find offensive.
Increasingly, consumers in all probability will find that access to information in digital intermedia comes subject to contractual provisions that aim to secure rights more broad than those provided by the Copyright Act.[97] Consumers will find, in particular, that such contracts inhibit or forbid uses that would otherwise arguably fall within the scope of the fair use defense. IBM's entry into the market for ARM services, for example, specifies that consumers signing onto its InfoMarket Service "may not copy, modify, adapt, reproduce, [or] translate . . . any information delivered or accessed via the Service."[98] This proposed agreement clearly aims to foreclose appeals to the fair use defense.
Despite their vocal concerns about losing unbilled use in digital contexts, champions of fair use have made comparatively little noise about the potential impact of such contracts.[99] To bring that impact into focus, let us briefly return to the hypothetical nightmare in progress:[100] The court holds not only that you have no right to use the Church of Technolism's copyrighted materials without paying, but that you have no right to use them without permission. The Church [p. 577/p. 578] has cited not only its rights under copyright law--a claim weakened by the policies underlying the fair use defense--but a contract arising out of your willing use of the particular service that controls access to the Church's web page. That contract gives the Church of Technolism the right to prevent any and all quotations of the text from its web page. Thanks to that contract, you find that you cannot buy your way out of copyright infringement. The Church of Technolism thus wins as its remedy not payment of the lost licensing fees, but rather an injunction on your unauthorized use.
Chilling though this scenario may sound, it remains technically and legally plausible. Copyright owners often evince a desire to prohibit the critical, blasphemous, or satirical reuse of their works.[101] In Campbell v. Acuff-Rose Music, Inc.,[102] the Supreme Court forbade employing copyright law to achieve such censorship, of course. It did not, however, expressly rule out using contract law to similar effect.
Though thin-skinned copyright owners might have it otherwise, mass distribution in conventional media does not lend itself to the imposition and enforcement of such anti-criticism contracts. The digital intermedia, in contrast, deliver information through channels ready-made to impose contracts limiting or forbidding fair use. Moreover, automated rights management makes enforcing those contracts wholly viable from a technical point of view.
As discussed more fully below, however, courts will probably hold that the fair use defense gives consumers the right to engage in limited critical reuses, ARM or not.[103] Fair use will thus continue to play a vital role, albeit a diminished one, in a world of otherwise pervasive fared use. Contracts that interfere with the fair use defense [p. 578/p. 579] might risk federal preemption.[104] Before addressing that particular issue, however, the next Part examines how the fair use doctrine will interact with the pre-contractual self-help measures that ARM will make available to copyright owners in digital intermedia.
This Part critically evaluates the dystopian future of fair use described above and describes fared use as a better and more likely alternative. Subpart A finds that those who decry the end of unbilled use in digital intermedia get the facts largely right but the valuation wrong. The combination of ARM and existing law will indeed make it easier for owners of copyrights in digital works to demand and enforce licensing fees. Consumers have no sound reason to object to paying such fees, however, and they have good reasons to welcome the systemic effects of the resulting market for expressive works.
Subpart B argues that, absent contractual provisions to the contrary, the fair use doctrine will continue to provide some protection to critics, satirists, and others who offend copyright owners. Subpart C explains how this, combined with the ubiquitous license of such uses, will put the digital intermedia under a reciprocal quasi-compulsory licensing scheme. In this spontaneously generated fared use system, information providers license online expressive works comprehensively but endure greater exposure to the offensive reuse of their works.
Because ARM will enable information providers and consumers to contract around this initial version of fared use, it may well end up as no more than the default rule for defining rights to information in the digital intermedia. Considering the likelihood and desirability of that outcome must wait for Part IV, however. Re-evaluating traditional fair use doctrine, and projecting how it will evolve into fared use, merits attention first.
Predictions that consumers will not generally enjoy unbilled access to copyrighted works in digital intermedia appear to assess the future accurately. For reasons set forth above, the advent of sophisticated automated rights management will sharply lower transaction costs for regulating the use of copyrighted materials. The legal doctrine set forth in American Geophysical and Princeton [p. 579/p. 580] University Press, with or without the NII White Paper's broad interpretation of these cases, will limit the applicability of the fair use doctrine online. New technology and current law will thus end up partially supplanting the fair use defense in the digital intermedia. For reasons set forth below, however, even parties who currently benefit from applying the fair use defense in conventional media should welcome this change.
Despite gross misconceptions to the contrary, fair use never comes for free.[105] One way or another, consumers using conventional media must pay to browse magazines at newsstands, to photocopy and distribute newspaper stories for spontaneous classroom use, to search for quotes and type them into articles, and to otherwise avail themselves of the fair use doctrine. Although such acts do not entail paying licensing fees, they inevitably impose a variety of transaction costs--for personal transport, manipulating paper and ink, searching card catalogs, and so on--that follow from the very nature of conventional media.[106] It makes no difference that consumers pay licensing fees in cash whereas they pay fair use's transaction costs in lost opportunities. Economically speaking, a cost is a cost.[107]
The digital intermedia allow consumers to avoid or reduce such transaction costs. Bits flow directly to homes and offices, copy easily into RAM or magnetic storage, forward instantly to destinations worldwide, and submit easily to electronic searches. Transaction costs remain even here, of course. The burgeoning growth of the Internet and other digital intermedia indicates, however, that consuming bits very often costs less than consuming atoms. The increasing reliance of legal academics on commercial online services, CD-ROMs, and the Internet confirms this observation.[108] Those who [p. 580/p. 581] decry the advent of fared use[109] thus err when they imply that it must impose a net cost on consumers.[110] To the contrary, fared use offers a considerable likelihood of providing more and better verified, organized, and interlinked information, at less cost, than fair use does now.[111]
Scholars have explained fair use in at least three ways: (1) as a proxy for a copyright owner's implied consent;[112] (2) as part of a bargain between authors and the public, struck on their behalf first by courts and then by Congress;[113] and (3) as a response to a market failure in private attempts to protect authors' expressions from undue copying.[114] The first of these three explanations has fallen into [p. 581/p. 582] disfavor because it does not explain why fair use protects parody and other uses of copyrighted material that owners find disagreeable.[115] The second explanation receives due consideration below.[116] The present subsection addresses the third explanation of fair use and argues that, as a response to market failure, the fair use doctrine can and should give way in the face of the effective enforcement of authors' rights through automated rights management.
Lawmakers enacted the Copyright Act to cure an alleged case of market failure: creating a work can cost authors a good deal, whereas copying a work costs free riders very little.[117] Absent special protection from such copying, the argument goes, authors will underproduce and the public will suffer. Copyright, as Justice Holmes explained, therefore "restrains the spontaneity of men where but for it there would be nothing of any kind to hinder their doing as they saw fit;"[118] namely, copying others' expressions at will. Perhaps [p. 582/p. 583] in the digital intermedia automated rights management will cure this market failure by protecting authors' works through technological and contractual means.[119] ARM's other curative effects interest us here, though.
Markets, like squeezed balloons, bulge outward where unconstrained. In its attempt to protect authors from the discouraging effects of unfettered copying, copyright law has thus created market failure elsewhere. The costs of avoiding infringement by obtaining permission to use a copyrighted work, and thus avoiding infringement claims, often exceed the benefits of the desired use. Such transaction costs threaten to prevent many socially beneficial uses of copyrighted works from taking place. The doctrine of fair use attempts to cure this particular market failure by excusing as non-infringing a limited (though poorly defined) class of uses of copyrighted works.[120] As Professor Gordon describes it, "courts and Congress have employed fair use to permit uncompensated transfers that are socially desirable but not capable of effectuation through the market."[121]
Understanding fair use as a response to market failure does much to explain the vagaries of its development in the case law.[122] More to the point, it lends support to the holding in American Geophysical. Consistent with the market failure theory of fair use, the court reasoned that "a particular unauthorized use should be considered 'more fair' when there is no ready market or means to pay for the use, while such an unauthorized use should be considered 'less fair' when there is a ready market or means to pay for the use."[123] In other words, the scope of the fair use defense rises and falls with the transaction costs of licensing access to copyrighted works.
Automated rights management radically reduces the transaction costs of licensing access to copyrighted works in digital intermedia. Indeed, as its name suggests, it makes licensing automatic. Insofar as [p. 583/p. 584] it responds to market failure, therefore, fair use should have a much reduced scope when ARM takes effect. Should fair use disappear entirely? Were ARM to abolish transaction costs, Coase's theorem would suggest that the market would internalize all costs and permit only value-maximizing transfers.[124] ARM does not wholly negate search and exchange costs, however, thus perhaps leaving contractual gaps for fair use to fill. Furthermore, Congress employs fair use to redistribute rights in accord with its particular notions of equity. The defense thus excuses certain "socially beneficial" uses, such as criticism and parody,[125] that copyright owners might prefer not to license.[126] Fair use thus remains relevant, even given ARM. Exactly how the two will interact to create a fared use system, and how well that system accommodates the concerns that drive fair use, receives consideration below.[127]
First, though, a cautionary reminder: no forecast of the probable impact of ARM can overlook the fact that, just as ARM technology may shape the law, the law will shape the development of ARM. Although robust and sophisticated ARM technology would likely curtail application of the fair use doctrine, unduly aggressive application of the fair use doctrine might thwart creation of the necessary technology.[128] This risk should concern not only computer scientists and information providers, but artists and audiences, too. Stifling ARM would stifle the wide variety of new expressions and experiences that the digital intermedia promise to inspire, promote, and provide.[129]
[p. 584/p. 585]
As courts and commentators often have noted, the Constitution demands a public benefit as the price for the limited statutory privileges that copyright creates.[130] In contrast to the view that the fair use doctrine represents a second-best response to pervasive market failure, therefore, some commentators regard the doctrine as an integral part of this constitutional quid pro quo.[131] On this view, [p. 585/p. 586] fair use provides a public benefit--unbilled access to copyrighted works--to balance the State's grant of a limited monopoly.
Automated rights management at first appears to threaten this bargain. It seems as if ARM restricts the public's access to copyrighted works in digital intermedia without offering a benefit in return. As this subsection's consideration of the issue shows, however, friends of fair use should not assume that ARM will leave the public worse off. To the contrary, it appears likely to provide a net benefit to the public.
By reducing transaction costs throughout the market for copyrighted expressions, ARM benefits the public both directly and indirectly. Having emanated from an intentionally vague statute[132] and developed in various, occasionally contradictory cases,[133] the fair [p. 586/p. 587] use doctrine necessarily blurs the boundary between valid and invalid copyright claims. High risks of "theft"--here, infringement--increase the insecurity of copyright's protection. Though the resultant uncertainty obviously harms producers and sellers of copyrighted works, it also harms consumers. Academics, artists, commentators, and others desirous of reusing copyrighted works without authorization must borrow at their peril, consult experts on fair use,[134] or, sadly, forego such reuse altogether. ARM's clarifying power directly benefits those who would reuse copyrighted works--and through them their public audiences--by creating harbors safe from the threat of copyright litigation.[135]
Moreover, ARM benefits the public indirectly by increasing the transactional efficiency of the market for expressive works.[136] Like other markets, the market for expressive works does not constitute a zero sum game. And, as Coase observed of markets in general,
[i]t is obviously desirable that rights should be assigned to those who can use them most productively and with incentives that lead them to do so. It is also desirable that, to discover (and maintain) such a distribution of rights, the costs of their transference should be low, through clarity in the law and by making the legal requirements for such transfers less onerous.[137]
[p. 587/p. 588]
ARM, by its systemic improvement of copyright's transactional efficiency, helps us discover and maintain a distribution of rights to expressive works that will increase net social wealth.[138] ARM thus stands to benefit both producers and consumers.
In particular, because it increases the value of expressive works, ARM will put deflationary pressure on the price of accessing them. In general, an asset's current price internalizes the value of its future income stream.[139] Copyrights therefore commonly lose present value because, with the passage of time and their wider distribution, they prove increasingly vulnerable to uncompensated uses.[140] Because it reduces such risks, ARM tends to increase the value of copyrights. But although this windfall might initially accrue to copyright owners, competition among information providers would force access prices downward, toward the marginal costs of obtaining and distributing expressive works.[141] Directly or indirectly, such price pressure would similarly affect the prices that copyright owners can demand.[142] Gains [p. 588/p. 589] that ARM provides to copyright owners would thus pass on to consumers in the form of reduced access fees.[143]
Because ARM will increase the value of copyrighted works, moreover, it will encourage their greater production and improved distribution.[144] Consumers will thus benefit from better access to information.[145] Access providers will improve the information itself, too, increasing its quantity and making it better organized, verified, interlinked, diverse, up-to-date, and relevant.[146] Although this [p. 589/p. 590] cornucopia of information may at first come only for a fee, some of it eventually will fall into the public domain.[147] To judge from current implementations of ARM, copyright owners might very well offer limited free access to their wares in an attempt to draw more extensive (and expensive) uses.[148] Entrepreneurs will undoubtedly create other services, at present utterly and inevitably unforeseen, to attract and satisfy consumers of information.
As such considerations demonstrate, "strict copyright enforcement doesn't necessarily mean people would pay more for viewing copyrighted material."[149] Because automated rights management creates well-defined and readily transferable property rights to information, it puts the power of the market in the service of consumer demand. As Professor Goldstein explains, "there is no better way for the public to indicate what they want than through the price they are willing to pay in the marketplace. Uncompensated use inevitably dilutes these signals."[150] Fared use therefore probably will provide better public access to copyrighted works in digital intermedia than fair use does or could. At any rate, no one can plausibly claim that fared use necessarily would serve the public interest any less well than the existing quid pro quo.
For lawmakers to calculate which intellectual property rights best suit the new digital intermedia would require them to accurately [p. 590/p. 591] measure several variables--many of them already intangible--that rapid technological, social, and economic changes have sent spinning.[151] At a minimum, choosing the proper quid pro quo for copyright will prove difficult.[152] It arguably defies solution entirely.[153]
This difficulty by no means implies that we should forego deliberating about how copyright law affects public policy. It does imply, however, that we should not take it on faith that traditional fair use doctrine represents the one best rule for regulating access to expressive works in the digital intermedia. Regardless of our public input and the sincerity of their deliberations, legislators simply do not and cannot know that much.[154]
Automated rights management allows experimentation with a wide variety of approaches to regulating access to expressive works in digital intermedia. To the extent that ARM merely enforces existing tenets of copyright law, on the one hand, it hardly can give rise to any special objections. On the other hand, to the extent that ARM supports particular agreements between copyright owners, information providers, and consumers, its use merits a presumption of enforceability. Absent proof of a very narrow category of circumstances, such as duress or misrepresentation, we can assume that contracts under fared use reflect the interests of those who choose to enter into them.[155] It seems highly doubtful that such contracts would generate sufficient negative externalities to justify their invalidation.[156] Whether a full-blown fared use system would [p. 591/p. 592] interfere with federal regulation of expressive information remains a separate question, and receives consideration below.[157] If not killed at birth by myopic legal conservatism, however, ARM will remove decisions about which rights best suit the new digital intermedia from lawmakers' hands and put them at the local level--the only place where the requisite, relevant information exists.[158]
If ARM can give information providers the power to monitor various uses and reuses of their copyrighted materials for billing purposes,[159] it can also give them the power to bar such uses as they find objectionable. Technical prowess alone does not justify such censorship, however. Public policy and copyright law pose additional hurdles, and, as this Subpart shows, quite high ones. Whether an information provider could overcome those hurdles by supplementing copyright law with contract law remains a distinct possibility, and one that Part IV considers separately.
The prospect of information providers using ARM to track and prohibit objectionable uses of copyrighted works might well alarm free speech advocates. It seems unlikely, however, that the First Amendment's influence over state action would reach such private action.[160] Thanks to the idea-expression distinction, moreover, courts have found that they do not unconstitutionally hinder free speech in using state power to enforce copyright claims against commentators. As the Supreme Court has observed, the idea-expression distinction "strike[s] a definitional balance between the First Amendment and the Copyright Act by permitting free communication of facts while still protecting an author's expression."[161]
[p. 591/p. 592]
Where the First Amendment proves impotent against ARM-based censorship, however, civil disobedience might have a very real impact. Because ARM's powers would at most reach only cut-and-paste quotations, it would not limit criticism or reportage that merely paraphrases a protected work.[162] Nor, of course, could a copyright owner count on ARM to catch reuses that, innocently or not, include cut-and-paste quotes of works stolen ("liberated," some would say) from ARM protection.[163]
It bears noting that the NII White Paper and NIICPA would, if written into law, considerably raise the stakes of "laundering" ARM-protected documents.[164] That escalation seems unwise, not to mention at cross-purposes to existing law. As currently interpreted, the fair use doctrine excuses some amount of unauthorized cut-and-paste quoting from expressive works protected only by ARM (and the owner's outcry). The NII White Paper and NIICPA propose banning devices that have "the purpose or effect" of circumventing ARM technology.[165] Because this provision would outlaw methods of squeezing fair use out of ARM-protected works, it would penalize the exercise of rights guaranteed by the Copyright Act.[166]
For now, at least, fair use can trump a copyright owner's objections. The American Geophysical court allowed licensing to [p. 593/p. 594] limit the fair use defense only "when the means for paying for such a use is made easier."[167] The court thus conditioned its holding on the observation that "a particular unauthorized use should be considered 'more fair' when there is no ready market or means to pay for the use."[168] In this, it followed the Supreme Court's reasoning in Campbell v. Acuff-Rose Music, Inc.:
The market for potential derivative uses includes only those that creators of original works would in general develop or license others to develop. Yet the unlikelihood that creators of imaginative works will license critical reviews or lampoons of their own productions removes such uses from the very notion of a potential licensing market.[169]
Therefore, although information providers might like to stop wags who evade ARM protection and reuse "liberated" works in objectionable ways, the fair use defense would continue to shield such defendants from copyright infringement claims.
One might well question the economic wisdom of such a policy. Just because an information provider refuses to license a particular use does not mean that no market exists. Rather, it demonstrates that the information provider demands more for that use than anyone wants to pay. Even a flat refusal to deal does not demonstrate market failure. As Professor Gordon has observed, "[a] refusal to license must not automatically justify a right to fair use; markets can function only if owners have a right to say 'no' as well as 'yes.' "[170] Strictly speaking, courts should thus find market failures only when technical barriers--not mere refusals to deal--prevent licensing agreements from taking place.
As the reasoning in the Campbell and American Geophysical decisions on this point demonstrates, however, the fair use doctrine sometimes favors public access over sound economics. Information providers who would prefer to use ARM and copyright law to completely bar objectionable reuses of their work will thus probably find their efforts thwarted by anti-ARM techniques and the fair use doctrine. Here, at least, fair use strikes back at ARM.
Suppose that you defused the ARM protecting a copyrighted [p. 594/p. 595] expression and reused the work without the owner's authorization. Often such an owner, offended, will disdain to charge you a licensing fee.[171] But some owners will, if able, take your coin (and your apology).[172] Can you count on the fair use defense to excuse not only your objectionable reuse of a copyrighted work but also your refusal to pay for that use? Probably so. Reuses that qualify for the fair use defense do not, under § 107 of the Copyright Act, constitute infringement.[173] The Act would thus not obligate you to pay.[174]
Requiring payment in such circumstances arguably makes more sense, for the same reasons that support the spread of licensing generally.[175] Furthermore, excusing non-payment might encourage over-production of reuses that aim, for purely economic reasons, to offend copyright owners.[176] True, parody and other criticism "can provide social benefit, by shedding light on an earlier work, and, in the process, creating a new one."[177] But too much of a good thing is no good at all.
Such theoretical considerations have yet to change copyright law, however. Unless and until they do, it looks as if information providers will have to suffer objectionable uses without remuneration,[178] license them grudgingly,[179] or try to prevent them by [p. 595/p. 596] means of supplementary contracts.[180]
Considered together, the ARM technology described in Subpart A likely will combine with the legal doctrines described in Subpart B to place digital intermedia under a system of reciprocal quasi-compulsory licenses. This spontaneously generated outcome would represent the default rule under fared use. For reasons that this section explains, fared use offers a welcome alternative to the dystopia portrayed by those who would have fair use function in digital intermedia exactly as it has in conventional media. Section 1 explores the scope of fared use's reciprocal quasi-compulsory license. Section 2 describes the benefits of a fared use system and defends it against possible objections.
The technical and legal factors set forth above likely will combine in digital intermedia to supplant fair use with an analogous but distinctly different rule: fared use. In a nutshell, fared use would subject copyrighted material in digital intermedia to a reciprocal quasi-compulsory license.[181]
The impact of ARM on licensing transaction costs, together with the theory of fair use embraced by American Geophysical, Princeton University Press, and the NII White Paper, ensures that consumers in most instances will have to pay for using copyrighted material in digital intermedia. Consumers thus face increased licensing. At the same time, the reasoning of Campbell and its progeny, American [p. 596/p. 597] Geophysical and Princeton University Press, ensures that copyright owners cannot selectively bar what they regard as objectionable uses of their material. Copyright owners must therefore endure increased licensing, too. In sum, fared use establishes, at least in its default mode, a reciprocal quasi-compulsory license.
This license has limits, however. For one thing, it does not require a copyright owner to put up with as much objectionable use as consumers choose to purchase or take. Rather, consumers have a right to only as much access as fair use would traditionally provide. Beyond this, a copyright owner presumably could shut off access by declining to license further. To borrow the example from Campbell, a parodist would have the right to a guitar riff, but not a whole song.[182] Fair use therefore would continue to guarantee access, albeit limited, to expressive works in digital intermedia.
Another example from Campbell demonstrates that copyright owners will face powerful incentives to offer licenses even for uses that qualify as fair. To settle the Campbell litigation, the copyright owner, Acuff-Rose Music, agreed to accept license fees from the defendant parodist, 2 Live Crew.[183] Acuff-Rose did so, moreover, despite the fact that it earlier had disdained 2 Live Crew's offer to pay licensing fees.[184] A representative for Acuff-Rose succinctly explained the motivation for this change of heart: " '[W]e will be getting paid for the song.' "[185] The settlement benefited 2 Live Crew as well, giving it a safe harbor more clearly defined than the Supreme Court's necessarily theoretical opinion could provide.
This outcome suggests that fared use will introduce licensing not only where, thanks to ARM, "the means for paying for such a use is made easier,"[186] but even where "the unlikelihood that creators of imaginative works will license critical reviews or lampoons of their own productions"[187] might otherwise give rise to the fair use defense. Consumers would, of course, retain the right to refuse to pay for uses [p. 597/p. 598] that qualify as fair.[188] But they will no doubt find the prospect of clarifying their rights appealing, and payment via ARM not at all burdensome. In similar fashion, the prospect of increased revenue and of escaping the uncertainty of fair use may very well encourage copyright owners simply to license all uses of their works.[189]
Given the continuing high transaction costs of determining the proper boundaries of fair use, one might argue for requiring that those who benefit from copyright protection in the digital intermedia license all uses. Implementing that proposal almost certainly would require new legislation, however, such as that which already requires the compulsory licensing of non-dramatic musical works and the like.[190] The prospect of Congress dictating the price and availability of every copyrighted bit surging through the digital intermedia does not, to say the least, give much comfort.[191] On this point, even the otherwise activist NII White Paper counsels a hands-off approach.[192] The sanest response to fair use's lingering influence, therefore, would tolerate its remaining (if reduced) transaction costs, let courts continue shaping common law solutions, and give information providers and consumers freedom to polish rough spots with contractual devices.[193]
[p. 598/p. 599]
To gauge from Campbell, American Geophysical, and Princeton University Press, defeating the fair use defense requires only that an information provider guarantee that consumers can license copyrighted works at some price--not at any particular price. Future courts might apply this doctrine to find that exorbitant licensing fees, or price discrimination designed to deter objectionable uses, shows the lack of "a ready market or means to pay for the use."[194] Perhaps courts also will review licensing practices for reasonableness.[195] Courts should, at any rate, enforce valid contracts between information providers and consumers.[196] Settling the pricing issue will have to wait until fared use enters actual practice and, inevitably, litigation.
[p. 599/p. 600]
Compared to fair use, the fared use system described here would offer a number of advantages. The discussion above explained the power of fared use to lower transaction costs radically, and the manifold advantages of such savings.[197] Another consequence of streamlining licensing procedures merits note: it encourages public discourse by providing a safe harbor for the reuse of copyrighted material. Critics, satirists, and parodists can thus rely on fared use to immunize them from the infringement claims of irate copyright owners seeking injunctive relief.[198]
Such irate copyright owners hardly would have grounds to complain, however, that fared use treats them roughly. In other times and in other media, the fair use doctrine would have put them at risk of suffering objectionable uses of their copyrighted material and of losing licensing fees. Fared use blunts at least one side of that double-edged sword. It does so, moreover, without undue cost to the public. The licensing arrangements allowed by ARM would, after all, merely ask that those who engage in objectionable uses pay for enjoying a supply of victims.
Methodologically speaking, fared use offers the considerable advantage of relying solely on existing legal doctrines. It thus requires only technical innovation--not legislative or judicial innovation--to give it effect. Of course, in practice it may turn out that fared use has some unforeseen and unpleasant consequences. Legal tinkering might cure these. More likely, however, private parties will contract around any untoward effects caused by fared use and once more render judicial and legislative invention unnecessary. The next Part addresses how such contracts might affect the default rules of a fared use regime.
Copyright owners and consumers probably will want to contract around the spontaneously generated default version of fared use described above. On the one hand, copyright owners will want to bar [p. 600/p. 601] objectionable reuses of the works that they place in digital intermedia. Many consumers would, for the right price, agree to such anti-criticism contracts. On the other hand, some consumers will want, and probably get, blanket licenses to engage in any reuses of protected works, however offensive. This Part considers the likelihood and desirability of such contractual modifications to fared use.
Subpart A considers the utility of two broad classes of contracts modifying fared use: anti-criticism contracts and anti-censorship contracts. It predicts that only a few especially thin-skinned copyright owners will forego licensing fees in favor of anti-criticism contracts. In contrast, the popularity of anti-censorship contracts remains in doubt. Because anti-censorship contracts raise few red flags, Subpart B focuses its enforceability analysis on anti-criticism contracts. Notwithstanding the risk of preemption that this analysis of anti-criticism contracts uncovers, Subpart C argues that we should invite experimentation with this and other ways of structuring rights to information. Subpart C thus suggests, as a compromise, an exit strategy: enforce otherwise preempted contracts on condition that either party permanently forsake appeal to copyright law with regard to the subject matter of the contract.
Information providers and consumers undoubtedly will want to modify fared use's default rule through a variety of contractual devices. Although the details of such contracts necessarily remain beyond our ken at present, two broad types seem likely: anti-criticism contracts and anti-censorship contracts. This Subpart considers the utility of each in turn. Section 1 argues that most copyright owners, on careful analysis, will find trying to prevent objectionable uses too expensive and futile to justify the effort. They will thus hew fairly closely to the default version of fared use described above.[199] Section 2 finds anti-censorship contracts an attractive option, but one most likely limited to narrow markets.
In most cases, a contract that modifies fared use's default rules to allow a copyright owner to bar offensive uses simply will not be cost-effective. The problem, in brief: an anti-criticism contract fails to provide the licensor with any greater monetary benefit than [p. 601/p. 602] unconstrained licensing would provide, and may provide considerably less. Therefore, only a copyright owner with an overriding non-monetary objective, such as a powerful aversion to public criticism, will find anti-criticism contracts worthwhile.
Recall, for example, the Church of Technolism scenario introduced earlier in this Article.[200] By its close, the licensee (you) had violated a contract that gave the licensor (the Church) the right to bar offensive uses of its copyrighted material-even uses that would otherwise fall within the scope of the fair use defense. The Church cannot hope to make money by suing you for breach of contract. Tolerating your offensive use, pocketing the license fee, and avoiding the costs of litigation would prove much more cost-effective.[201] Rather, the Church will bother suing you only if it favors non-monetary goals, such as getting an injunction to end your sacrilegious quotations.[202]
Churches often have very powerful non-monetary goals, of course. But that marks them as unique. Most other copyright owners put finance before honor.[203] If that sounds crass, keep in mind that anti-criticism contracts will not come cheaply. Licensors who want injunctive relief against offensive uses will not only have to forego licensing fees for such uses and (almost certainly) bear litigation costs; they will also have to offer consumers something extra to make such censorious contracts attractive in the first instance. That something extra probably will take the form of lower [p. 602/p. 603] licensing fees for acceptable uses of copyrighted material. In sum, copyright owners so thin-skinned as to demand and enforce anti-criticism contracts had better have fairly thick wallets.
In contrast to anti-criticism contracts, anti-censorship contracts would modify fared use's default rule, the reciprocal quasi-compulsory license, to commit copyright owners to allowing criticism, satire, and other potentially offensive uses. Such access contracts would obviously appeal to reporters, reviewers, academics, and others who regularly engage in biting commentary. With fared use or without, commentators would have to pay licensing fees for what they borrow in excess of fair use's limits. Anti-censorship contracts would guarantee, however, that everyone who pays such fees under fared use would enjoy free access to and reuse of the works that they quote and critique.
The size of the market for censorship-free access remains uncertain. Some copyright owners might demand a steep premium for giving up the right to license their works as they see fit. On the other hand, some copyright owners might charge less for such contracts. What Landes and Posner observe of book reviews may well hold true in general: "Ex ante, publishers are better off if reviewers are free to quote without permission; it makes reviews a credible form of book advertising."[204] Publishers might even come to advertise their anti-censorship agreements as proof of confidence and, thus, credibility. Still other publishers--both amateurs and start-ups employing something like Netscape's marketing strategy--would probably offer censorship-free access at no charge. Some might even pay consumers for their attention.[205]
[p. 603/p. 604]
Here, as elsewhere, the acts of freely contracting parties defy prediction. Regardless of the price and popularity of anti-censorship contracts, however, they at least avoid the questions of enforceability that dog anti-criticism contracts. The next Subpart takes up that question.
Unless Congress or courts forbid them, contracts will undoubtedly change the shape of fared use.[206] At least some copyright owners will want to limit the use of their works in contexts that they regard as critical, blasphemous, satirical, violative of their secrecy, or otherwise offensive.[207] Mere licensing fees will not satisfy such thin-skinned copyright owners, who would happily forego monetary gain for the right to bar offensive uses of their proprietary information.[208] ARM can technically empower them to police contracts reserving their anti-criticism rights. But can copyright [p. 604/p. 605] owners count on courts to enforce such contracts?[209] The following sections consider objections to such agreements based in shrinkwrap licenses, adhesion contracts, § 301 preemption, and Supremacy Clause preemption. Only the last of these appears to have much bite.[210] The following Subpart, C, will suggest a way for mutually consenting parties to save their otherwise preempted agreements by exiting from copyright into contract law.
Any contract that attempts to alter the default rules for using copyrighted works in an automated rights management system will probably resemble the types of agreements that already mediate access to digital intermedia.[211] A contract shaping ARM rights will, in other words, require a consumer to consent to its terms before it grants access to the protected works. Indeed, existing ARM systems already employ contracts conforming to this model.[212]
Commentators have suggested that this type of agreement may qualify as a shrinkwrap license.[213] Insofar as that equation would render ARM invalid per se, however, it errs. Consumers can apprehend the terms of an ARM contract at the point of sale, prior to committing to purchase the information in question.[214] The lack of [p. 605/p. 606] this degree of notice has proved fatal to shrinkwrap licenses in some cases.[215] In other cases, giving consumers prior notice has saved "shrinkwrap" licenses--better called "access contracts"[216]--from invalidation.[217]
Still other factors widen the margin of safety for ARM access contracts that, rightly or wrongly, get equated to "shrinkwrap" licenses. The Seventh Circuit recently enforced a shrinkwrap license despite its reliance on terms disclosed after purchase,[218] and the forthcoming Article 2B of the Uniform Commercial Code almost certainly will guarantee the enforceability of a wide class of "shrinkwrap" licenses.[219] Point-of-sale contracts offered to consumers via automated rights management thus appear likely to escape the risk of invalidation that, according to some authorities, shrinkwrap licenses should face.[220]
[p. 606/p. 607]
Despite its rhetorical sting, the "adhesion contract" label carries surprisingly little legal weight. As one commentator has noted, "[t]he law currently governing contracts of adhesion is a jumble of different lines of analysis, contradictory outcomes, and convoluted expressions."[221] For the most part, courts remain unmoved by adhesion contract claims. Absent extraordinary circumstances,[222] courts traditionally enforce all duly signed contracts unless the drafting party affirmatively misleads the signing party.[223] It thus does not seem likely that the contracts giving rise to fared use routinely will suffer invalidation on grounds of adhesion.[224]
That fared use survives this test should please even those who would have courts take more aggressive steps against adhesion contracts. Fared use does not exist in a vacuum. One can evaluate its merits only by considering the alternative: copyright law's fair use doctrine. Insofar as that doctrine represents a "bargain" between copyright owners and the public--a popular fiction[225]--it epitomizes the type of take-it-or-leave-it offer that foes of adhesion contracts so dislike.[226] Neither authors nor their audiences have much real say in [p. 607/p. 608] how Congress distributes rights to expressive works.[227] It hardly answers this objection to observe that parties can bargain away from fair use to agree to alternative allocations of rights--not, at least, so long as lawmakers stand ready to preempt certain of those agreements. Nor does it do much to observe that special interests can--and indeed do--successfully lobby Congress for favorable treatment under the Copyright Act.[228] For "although the political process itself involves bargaining, government typically offers benefits to individual beneficiaries on a take-it-or-leave-it basis."[229]
Fared use ameliorates the adhesive aspects of this political bargain by empowering mutually consenting parties to experiment with alternative ways of allocating their rights to information. Given the likelihood that fared use will force information providers to compete for consumers by offering them attractive terms of access, and that decreasing communication costs will encourage consumer coordination and self-help, we should not assume that fared use will decrease consumers' bargaining power.[230] Ultimately, however, bargaining power has little to do with the question at hand. As Judge Posner has observed of adhesion contracts, "[t]he problem is monopoly, not bargaining power."[231] Insofar as fared use weakens the State's monopoly on specifying the terms by which consumers access expressive works, therefore, it addresses the very concerns at the heart of the adhesion contract doctrine.
Both § 301 of the Copyright Act[232] and the Constitution's [p. 608/p. 609] Supremacy Clause[233] threaten to preempt the types of agreements that would regulate access to information under fared use. Because § 301 typically fails to interfere with contractual obligations, however, it poses little risk of preempting fared use contracts.
Through § 301, the Copyright Act preempts common law doctrines or state statutes that might otherwise conflict with its provisions. Section 301 specifies that the Copyright Act alone will govern "all legal or equitable rights that are equivalent to any of the exclusive rights" set forth in § 106 of the Act.[234] This terse delineation of the Copyright Act's exclusive domain leaves a good deal uncertain,[235] a problem that reference to the legislative history of § 301 fails to alleviate.[236]
Under the consensus view, however, contractual rights do not qualify as "equivalent to any of the exclusive rights within the general scope of copyright"[237] because enforcing contractual rights calls for proof of an additional legal element: the existence of a contractual obligation.[238] As the Seventh Circuit recently explained,
[c]opyright law forbids duplication, public performance, and so on, unless the person wishing to copy or perform the work gets permission . . . . A copyright is a right against the world. Contracts, by contrast, generally affect only their parties; strangers may do as they please, so contracts do not [p. 609/p. 610] create "exclusive rights."[239]It thus seems highly unlikely that § 301 would preempt a contract that authorizes information providers to forbid those who obtain information under the contract from putting it to offensive uses.[240]
The Supremacy Clause states that the "Constitution, and the Laws of the United States which shall be made in Pursuance thereof . . . shall be the supreme Law of the Land."[241] It of course provides the legal foundation for § 301, which Congress enacted merely to clarify and reaffirm preemption principles originally developed under the Supremacy Clause.[242] Perhaps the broader wording and intent of the Supremacy Clause allows it to preempt fared use contracts that the more specific provisions of § 301 would leave untouched.[243] A paucity of relevant case law[244] and the subtleties inherent to [p. 610/p. 611] Supremacy Clause preemption[245] leave the supposition unresolved, however.[246]
The Supreme Court has interpreted the Supremacy Clause to open several avenues to preemption. First, Congress can preempt state law "by so stating in express terms."[247] Congress has said nothing to bar the sorts of fared use contracts at issue here, though.[248] Second, preemption may take place when "federal regulation is sufficiently comprehensive to make reasonable the inference that Congress 'left no room' for supplementary state regulation."[249] This description hardly fits copyright, however, which continues to enjoy protection--albeit solely for unfixed works-at common law. What about state protection for fixed works? That Congress has not seen fit to attack the ample authority upholding contracts under § 301[250] indicates that here, too, Congress has "left room" for fared use contracts.[251]
Third, federal law may preempt a state law that "actually conflicts" with federal law because the state law "stands 'as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.' "[252] This test arguably frames [p. 611/p. 612] preemption in terms sufficiently broad to pose a threat to anti-criticism contracts enforced by ARM as part of a widespread system of fared use. Suppose that in practice and in the aggregate, such contracts would render the fair use defense largely inapplicable. Would this result thwart the "purposes and objectives" that led Congress to enact § 107 in particular, and the Copyright Act in general?
Here we enter the realm of conjecture. As one court has noted, "[i]t is possible to hypothesize situations where application of particular state rules of [contract] construction would so alter rights granted by the copyright statutes as to invade the scope of copyright law or violate its policies."[253] This hypothetical has, however, remained exactly that: a mere hypothetical. On the other hand, the Supreme Court has made clear that preemption arguments have a high hurdle to cross before they begin to threaten contracts between private parties: "Commercial agreements traditionally are the domain of state law. State law is not displaced merely because the contract relates to intellectual property . . . ."[254]
The Supreme Court has regarded state laws specifically regulating intellectual property with far less solicitude than it has contracts that, backed by state law, affect rights to intellectual property.[255] Some commentators thus have argued that the Supreme [p. 612/p. 613] Court should preempt form contracts that so pervade the market as to amount to "private legislation."[256] The Supreme Court appears highly unlikely to accept this invitation, however. The Court has shown great unwillingness to expand the definition of "state action"--particularly with regard to the private resolution of commercial disputes.[257] It explained that "even if we were inclined to extend the sovereign-function doctrine outside of its present carefully confined bounds, the field of private commercial transactions would be a particularly inappropriate area into which to expand it."[258] Because ARM's power to thwart unauthorized use will often make recourse to litigation unnecessary, moreover, it looks even less likely to qualify as state action.[259]
The question of Supremacy Clause preemption of fared use contracts thus remains unresolved and probably involves too many variables to ever allow an easy answer, much less at this far remove.[260] The burden of proof of course will fall on those who would use preemption to force open access to others' copyright works.[261] [p. 613/p. 614] Nonetheless, for reasons discussed above, there remains a non-negligible chance that courts will interpret the fair use doctrine to require that copyright owners put up with objectionable uses.[262] The next Subpart thus proposes an exit strategy that both protects the integrity of copyright law and liberates mutually consenting adults to give and receive information as they see fit.
The NII White Paper[263] and the legislation it begat[264] have drawn criticism for rushing to impose uniform solutions to problems that remain poorly understood.[265] Even commentators who disagree about the proper role and scope of copyright agree that federal authorities should give private parties time to craft their own solutions.[266] Professor Samuelson has suggested that we "[l]et 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, there will be time enough to adopt appropriate legislation."[267] The U.S. Register of Copyrights likewise [p. 614/p. 615] has argued against premature legislative attempts to update fair use for digital intermedia.[268]
Attempts to abort the development of fared use would merit similar criticism. Private parties must have free rein to explore technological and contractual answers to the problems of protecting information in the new digital intermedia.[269] At this early stage, we can predict neither the solutions that entrepreneurs will discover nor the systemic effects of their various, individual experiments. Few people will object to anti-censorship contracts. Even anti-criticism contracts deserve a presumption of enforceability, however.
Though third-party commentators and lawmakers might frown on some types of fared use contracts, they should refrain from banning them. They should instead require merely that those who would rely on such contracts choose between copyright law and contract law. Suppose, in other words, that the worst-case scenario described above comes to pass,[270] and that contracts under fared use develop so as to leave copyright's traditional concern for the public interest wholly unrepresented in digital intermedia.[271] The proper response would not invalidate such agreements, and thereby punish private parties for quite naturally pursuing their private interests. The proper response would instead withdraw copyright's public benefits. Information providers who prefer contract to copyright would have to rely on ARM to protect their wares, which otherwise would fall into the public domain.[272] Those who prefer copyright [p. 616/p. 617] would have to forego overly restrictive contracts.[273]
In broad terms, this option to exit[274]274 copyright law would apply only to fared use contracts otherwise barred by preemption. The exact details of implementing this program obviously deserve thoughtful input from many points of view.[275] The drafters of the proposed UCC-2B[276] could certainly stir discussion by considering a new § 104(c):[277] "Any party to a contract governed by this title but preempted by copyright law may enforce the contract if that party permanently abandons any rights and remedies arising under copyright law to the subject matter of the contract."[278]
Amending UCC-2B ultimately would prove futile, however, because--even if adopted by each state--the UCC exercises little more than an advisory influence on federal lawmakers. A more complete and effective reform would therefore amend § 301 of the federal Copyright Act[279] by adding something like the following sub-section: "(g) Nothing in this title annuls or limits any rights or remedies that a party enjoys under contract law if said party permanently abandons any rights and remedies arising under copyright law to the subject matter of that contract."[280]
Congress already has made explicit its willingness, in the context of the first sale doctrine, to force copyright owners to decide between contract law and copyright law.[281] The amendment to § 301 proposed here would merely make that Hobson's choice generally available. A measure such as this proposed amendment could open up alternatives to an outright ban on any fared use contract that skirts preemption. Policy makers therefore can--and should--let fared use develop unhindered, interfere with private arrangements only on proof of imminent peril to the public interest, and provide the freedom to exit from a statutory regulatory scheme to contractual ones.
To the extent that information providers and consumers opt to manage expressive works solely by contract, they will deregulate (more aptly, "re-regulate") access to expressive works.[282] Their various private agreements will supplant the allocation of rights that copyright's fair use doctrine formerly required. This prediction raises an interesting choice: should we favor private agreements, or a state law, when it comes to controlling the creation, dissemination, and use of expressive speech? Commentators and courts largely agree on how to answer this question in the First Amendment context. No such consensus exists in the context of copyright, however.[283] Indeed, [p. 617/p. 618] scarcely anyone even asks the question in those terms.[284]
Academics view with horror the prospect that automated rights management may limit copyright's fair use defense. But academics have a peculiar affection for fair use. It gives them the valuable right to use others' copyrighted works generously and free of charge. Furthermore, academics themselves suffer scant harm from fair use because they have few licensing fees at risk. In fact, academics typically prefer that others copy, read, and cite their works freely.[285] Such special incentives do not encourage academics to regard fared use with disinterested eyes. Only a comprehensive and objective view can accurately assess ARM's impact on the fair use defense, copyright law, and the public interest.
This Article has argued that a confluence of new ARM technologies and existing legal doctrines stand ready to radically reduce the scope of the fair use defense--and that, on reflection, this should horrify no one. In the place of fair use, a reciprocal quasi-compulsory license will likely come to define rights to expressive works in the digital intermedia. A balanced accounting reveals that, whether in its default mode or specially modified by contract, this [p. 618/p. 619] system of fared use would generate considerable net benefits and distribute them equitably.
In comparing fair use to fared use we might well ask, "which allows 'freer use'?" Fair use seems freer to academics and others who regularly use copyrighted works because, quite simply, nobody demands payment. But such use comes "for free" only in a very superficial sense. More careful scrutiny demonstrates that fair use imposes costs on consumers via risk of suit, on copyright owners via uncompensated uses, and on society as a whole via transaction costs. The alternative presented by fared use, because it can liberate us from a copyright system rotten with uncertainty, in fact offers freer access to expressive works.
As copyright owners and consumers contract around the default rules for fared use, they will experiment with a wide variety of methods for managing information in the digital intermedia. A full assessment of this exploratory and entrepreneurial process must await our observations of actual results. We can assume, however, that copyright owners and consumers will modify fared use's reciprocal quasi-compulsory license only when they find it mutually beneficial to do so. Their agreements thus deserve a presumption of enforceability. If federal preemption threatens to negate that presumption, lawmakers should protect continued experimentation by opening an exit route from copyright law to contract law.
Neither lawmakers nor academics can expect to dictate the single best means of regulating online access to expressive works. The necessary information--information that would give policy-makers a shortcut to the future--hardly "wants to be free." To the contrary, such information cannot yet be had at any price. It must come from those who actually participate in the market for information, and it will appear in the mosaic of their diverse experiments and agreements. Only by patiently studying their evolved preferences, in the fine and in the aggregate, will we discern which types of rights best suit the new digital intermedia.
[p. 619/p. 620]
[*] Assistant Professor, University of Dayton School of Law, Program in Law and Technology. B.A., University of Kansas; M.A., University of Southern California; J.D., University of Chicago. The author thanks the University of Dayton Law School for supporting his work on this Article through a summer research grant; Francis Conte, Robert A. Kreiss, and David G. Post for early guidance; Robert Kreiss again, as well as Julie E. Cohen and Lydia P. Loren, for detailed comments on a late draft; various participants of the CNI-Copyright, <cni-copyright@cni.org>, and Cyberia-L, <cyberia-l@listserv.aol.com>, email lists for analyses and opinions; Dianne Keppler for research assistance; Chris Schumacher for useful cites; Tom G. Palmer and Walter E. Grinder for encouraging critical scrutiny of copyright; and Donna G. Matias for editing and moral support. The author bears sole responsibility for what he submits for publication, however.
[p. 557/p. 558]
[1]. See David Stipp, The Electric Kool-Aid Management Consultant, FORTUNE, Oct. 16, 1995, at 160, 166 (characterizing "information wants to be free" as the "cyberhacker rallying cry," and attributing it to Stewart Brand). For a recent explication of this claim, see Interview by Frontline with Stewart Brand, founder of the WELL online service, in San Francisco, Cal. (June 15, 1995), available at <http://www2.pbs.org/wgbh/pages/frontline/cyberspace/brand.html> (visited Oct. 3, 1997) ("[T]he net . . . is basically a gift economy. . . . For a company or business to get into the net, they need to join that way of [p. 558/p. 559] doing things and so typically it means giving away software . . . . It means often giving away content.").
[2]. To hold otherwise would require one to embrace a sort of digital animism, and to attribute cognitive states to mere collections of bits, or to rely on an outmoded, scholastic notion of causation. See, e.g., ARISTOTLE, PHYSICS, bk. II, ch. 3, 194a33 (R.P. Hardie & R.K. Gaye trans., c. 335-320 B.C.), reprinted in THE COMPLETE WORKS OF ARISTOTLE 332 (Jonathan Barnes ed., 1984) (describing final causation as "the sense of end or that for which a thing is done").
[3]. This Article uses "digital intermedia" to refer to the Internet, circuit-switched networks, and other interactive channels over which digital information gets distributed and through which automated rights management can function. "Digital intermedia" does not encompass such comparatively non-interactive distribution channels as CDs, CD-ROMs, and digital audio cassettes.
[4]. Stewart Brand actually came fairly close to this sense in an early formulation of his now-widespread aphorism: "Information wants to be free because it has become so cheap to distribute, copy and recombine-too cheap to meter. It wants to be expensive because it can be immeasurably valuable to the recipient." STEWART BRAND, THE MEDIA LAB: INVENTING THE FUTURE AT M.I.T. 202 (1987).
[5]. See infra Part III.A.1.
[6]. See infra Part III.A.2.
[p. 559/p. 560]
[7]. This Article favors "automated rights management" as the most exact label for the processes at issue. Terminology in this new field remains in flux, however. Alternative terms, rejected here as either too broad or narrow, include "trusted systems," see Mark Stefik, Trusted Systems, SCI. AM., Mar. 1997, at 78, 78, available at <http://www.sciam.com/0397issue/0397stefik.html> (visited Nov. 5, 1997); "copyright management," see WORKING GROUP ON INTELLECTUAL PROPERTY RIGHTS, U.S. DEP'T OF COMMERCE, INTELLECTUAL PROPERTY AND THE NATIONAL INFORMATION INFRASTRUCTURE: THE REPORT OF THE WORKING GROUP ON INTELLECTUAL PROPERTY RIGHTS passim (1995) [hereinafter NII WHITE PAPER]; Julie E. Cohen, A Right to Read Anonymously: A Closer Look at "Copyright Management" in Cyberspace, 28 CONN. L. REV. 981, passim (1996); "rights management," see Mary Grace Smith & Robert Weber, A New Set of Rules for Information Commerce-Rights-Protection Technologies and Personalized-Information Commerce Will Affect All Knowledge Workers, COMM. WK., Nov. 6, 1995, at 34, 34; Robert Weber, Digital Rights Management Technologies passim (Oct. 1995) (unpublished report to the International Federation of Reproduction Rights Organizations) (on file with the North Carolina Law Review); "copy-protection," see Technological Solutions Rise to Complement Law's Small Stick Guarding Electronic Works, INFO. L. ALERT, June 16, 1995, at 3, 4 [hereinafter Technological Solutions]; "transcopyright," see Theodor Holm Nelson, Transpublishing and Transcopyright (visited Nov. 5, 1997) <http://www.sfc.keio.ac.ip/˜ted/transpub.transco.html>; and "telerights," see Wade Riddick, From Copyright to Telerights, BYTE, Feb. 1996, at 248, 248.
[8]. See infra Part II.A (describing the functions of ARM). ARM encompasses a variety of technologies, including: encryption, firewalls, and passwords to limit access to information; digital watermarks and stegonography to identify electronic documents; and micropayments and embedded applications to ensure that users pay for protected information.
[9]. See David Post, Battle or Dance?, AM. LAW., Jan./Feb. 1996, at 116 (observing that due to automated rights management, " 'transaction costs'-of negotiating a license fee for each use of a copyrighted work, however trivial and insignificant-are rapidly disappearing," and raising the question: "Once tracking and payment mechanisms of this kind are in place, is there still a place for fair use?"); David G. Post, Controlling Cybercopies; Leaping Before Looking; Proposals Would Make Unsettling Changes, LEGAL TIMES-SPECIAL REPORT; INTELL. PROP., Apr. 8, 1996, at 39, 45 [hereinafter Post, Controlling Cybercopies] ("[A]lthough one may retain the theoretical legal right to make fair use of material, where rights holders are permitted to use powerful technological means to control access to their works, fair use may prove illusory.").
[p. 560/p. 561]
[10]. Broadly speaking, the fair use defense covers certain unauthorized uses of a copyrighted work, such as in commentary or scholarship, that do not displace the copyright owner's potential licensing fees. See infra note 20.
[11]. Very broadly speaking, fared use would require consumers to pay for the right to access and reuse information, rather than appealing to a statutory fair use exception. See infra Part II.A.
[12]. Although some aspects of fared use have yet to come to pass, we have much to gain from proactively preempting the difficult legal problems that originate with the onset of new technologies. See Tom W. Bell, Virtual Trade Dress: A Very Real Problem, 56 MD. L. REV. 384, 388 (1997).
[13]. Such hidden costs include the uncertainty created by the fair use doctrine's uncertain boundaries, the losses passed on to consumers by copyright owners who lose licensing revenue due to fair use, and various other costs. See infra Part III.A.3.
[14]. The NII White Paper has drawn exactly this criticism by calling on lawmakers to, among other things, outlaw devices that could negate copyright protection and penalize tampering with copyright management information. See, e.g., Post, Controlling Cybercopies, supra note 9, at 45; Pamela Samuelson, The Copyright Grab, WIRED, Jan. 1996, at 134, 136 [hereinafter Samuelson, The Copyright Grab]; Pamela Samuelson, Technological Protection for Copyrighted Works 22-27, 33 (Feb. 14, 1996) (unpublished manuscript, on file with the North Carolina Law Review) [hereinafter Samuelson, Technological Protection], available at <http://sims.berkeley.edu/˜pam/ [p. 561/p. 562] courses/cyberlaw/docs/techpro.html> (visited Nov. 5, 1997).
[15]. For recent treatments of this important topic, see generally Cohen, supra note 7 (evaluating the import of digital monitoring of individual reading habits in the context of traditional notions of freedom of thought and expression); A. Michael Froomkin, Flood Control on the Information Ocean: Living with Anonymity, Digital Cash, and Distributed Databases, 15 J.L. & COM. 395, 486-88 (1996) (discussing business profiling of Web users).
[16]. The first sale doctrine, codified at § 109 of the Copyright Act, specifies that copyright law does not prevent the owner of a particular copy or phonorecord to sell or otherwise dispose of it at will. See 17 U.S.C. § 109 (1994). The first sale doctrine does not affect ARM transactions, however, because they more closely resemble licenses than sales. Furthermore, Congress and the courts have made it quite clear that the first sale doctrine does not preempt contracts controlling post-sale use of copyrighted works. See American Int'l Pictures, Inc. v. Foreman, 576 F.2d 661, 664 (5th Cir. 1978) (noting that the first sale doctrine does not bar contract suits); United States v. Wise, 550 F.2d 1180, 1187 (9th Cir. 1977) (same); H.R. REP. NO. 94-1476, at 62 (1976), reprinted in 1976 U.S.C.C.A.N. 5659, 5675-76 (same).
[17]. See supra note 7.
[18]. See, e.g., Post, Controlling Cybercopies, supra note 9, at 45 (criticizing the NII White Paper for proposing potentially revolutionary changes to online information use without providing adequate justification or allowing adequate discussion); Samuelson, Technological Protection, supra note 14, at 14-35 (questioning the wisdom of the NII White Paper's provisions regarding protection of copyright management information).
[19]. For interesting treatments of this issue, see generally OFFICE OF TECH. ASSESSMENT, CONGRESS OF THE U.S., INTELLECTUAL PROPERTY RIGHTS IN AN AGE OF ELECTRONICS AND INFORMATION (Apr. 1986), available at <http://www.wws.princeton.edu/˜ota/ns20/alpha_f.html> (visited Nov. 5, 1997) (questioning the copyright paradigm); Stephen Breyer, The Uneasy Case for Copyright: A Study of Copyright in Books, Photocopies, and Computer Programs, 84 HARV. L. REV. 281 (1970) (giving skeptical consideration of the instrumental need for copyrights); Negativland, Fair Use (visited Oct. 3, 1997) <http://www.eff.org/pub/Intellectual_property/fair_use.article> [p. 562/p. 563] (questioning private ownership of mass culture's elements); Tom G. Palmer, Are Patents and Copyrights Morally Justified? The Philosophy of Property Rights and Ideal Objects, 13 HARV. J.L. & PUB. POL'Y 817 (1990) [hereinafter Palmer, Patents and Copyrights] (arguing on philosophical grounds that copyrights unjustifiably and unnecessarily rely on state sanctions); Tom G. Palmer, Intellectual Property: A Non-Posnerian Law and Economics Approach, 12 HAMLINE L. REV. 261 (1989) [hereinafter Palmer, Intellectual Property] (making a similar argument on historical and economic grounds); John Perry Barlow, The Economy of Ideas: A Framework for Rethinking Patents and Copyrights in the Digital Age, WIRED, Mar. 1994, at 84 (describing alternatives to copyright based on the relationship between information providers and consumers rather than on the possession of information).
[20]. Codifying the common law defense of fair use, § 107 of the Copyright Act provides that
the fair use of a copyrighted work . . . for purposes such as criticism, comment, . . . scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include-17 U.S.C. § 107 (1994). Despite the enumeration of these four factors, however, the fair use defense "is not to be simplified with bright-line rules, for the statute, like the doctrine it recognizes, calls for case-by-case analysis." Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 577 (1994).
(1) the purpose and character of the use . . . ;
(2) the nature of the copyrighted work;
(3) the amount and substantiality of the portion [used] . . . ; and
(4) the effect of the use upon the potential market for or value of the copyrighted work.
[21]. Compare generally, e.g., PAUL GOLDSTEIN, COPYRIGHT'S HIGHWAY (1994) (arguing for giving fair use a narrow scope), with Samuelson, The Copyright Grab, supra note 14 (arguing for a broad definition of fair use). See also, e.g., Coalition for Networked Information, Coalition CNI-COPYRIGHT Forum (visited Nov. 5, 1997) <http://www.cni.org/Hforums/cni-copyright> (advancing various viewpoints on this topic in ongoing discussions).
[p. 563/p. 564]
[22]. See, e.g., Neil Weinstock Netanel, Copyright and a Democratic Civil Society, 106 YALE L.J. 283, 285-87 (1996) (describing the debate between "neoclassicists," who regard expressive works as commodities, and "copyright minimalists," who argue that new technologies render copyright irrelevant, and advocating as an alternative a "democratic paradigm" under which copyright advances political and social discourse). Readers intent on applying Professor Netanel's labels to the current work will find that it borrows "neoclassical" methods to preserve values shared with the "democratic paradigm" while opening a voluntary exit to the "copyright minimalist" world.
[23]. See Christine Hudgins-Bonafield, Selling Knowledge on the Net, NETWORK COMPUTING, June 1, 1995, at 102, 102, available at <http://techweb.cmp.com/nwc/607/607feature2.html> (visited Nov. 6, 1997); Otis Port, Copyright's New Digital Guardians, BUS. WK., May 6, 1996, at 62, 62; Riddick, supra note 7, at 248; Ira Sager, IBM's Tollbooth for the I-Way, BUS. WK., May 13, 1996, at 114, 114; Smith & Weber, supra note 7, at 34; Stefik, supra note 7, at 78-79; Technological Solutions, supra note 7, at 3-4; Weber, supra note 7, at 9.
[24]. It also gives information providers new-found power to "profile" consumer behavior, a development that threatens consumers' privacy. See Cohen, supra note 7, at 985-87.
[p. 564/p. 565]
[25]. See generally Hudgins-Bonafield, supra note 23 (discussing research in ARM); Imprimatur, Issues: Monitoring and Managing Use (visited Nov. 5, 1997) <http://www.imprimatur.alcs.co.uk/imprimatur/IRM/xrmonuse.htm> (discussing various ARM proposals); Smith & Weber, supra note 7, at 34 (discussing research in ARM); Stefik, supra note 7, at 79-80 (same).
[26]. See Port, supra note 23, at 62 (discussing products offered by Digimarc Corp., DICE Co., Electronic Publishing Resources Inc., and Release Software Corp.); Sager, supra note 23, at 114 (discussing IBM's InfoMarket); Smith & Weber, supra note 7, at 36 (discussing Electronic Publishing Resources Inc.'s InterTrust architecture and IBM's InfoMarket); Technological Solutions, supra note 7, at 3-4 (discussing Infosafe Systems' Design Palette and National Semiconductor's iPower unit); Custom Innovative Solutions, Web Copyright Management System (visited Nov. 6, 1997) <http://www.cisc.com/bin/webright.cgi?fle=3Dcprt/main.html> (describing, demonstrating, and offering for sale "WebRight" ARM system).
[27]. See Sager, supra note 23, at 114. For an introduction to IBM's InfoMarket, see InfoMarket (visited Oct. 7, 1997) <http://www.infomarket.ibm.com>.
[28]. See Hudgins-Bonafield, supra note 23, at 108; Technological Solutions, supra note 7, at 4.
[29]. See Sager, supra note 23, at 114; Smith & Weber, supra note 7, at 37; Technological Solutions, supra note 7, at 4. See generally Hudgins-Bonafield, supra note 23 (discussing agreements between ARM providers).
[30]. See Weber, supra note 7, at 9-12, 19-21 (discussing factors influencing the development of a standard for ARM). Some commentators claim that IBM's early entry into the ARM field, and the superiority of its Cryptolope technology, ensure that it will set the standard for later developments. See Sam Albert, Surprise! IBM May Crash the Internet Party, COMPUTERWORLD, June 10, 1996, at 41, 41. Others bet that EPR's comprehensive patents give it an edge. See Technological Solutions, supra note 7, at 3. At any rate, it looks unlikely that federal authorities will mandate a particular type of ARM system. See NII WHITE PAPER, supra note 7, at 233 ("Copyright owners should be free to determine what level or type of protection (if any) is appropriate for their works, taking into consideration cost and security needs, and different consumer and market preferences.").
[31]. See Hudgins-Bonafield, supra note 23, at 105-06 (describing the operation of [p. 565/p. 566] services such as Lexis-Nexis); see also I. Trotter Hardy, Contracts, Copyright and Preemption in a Digital World, 1 RICH. J.L. & TECH. 2, ¶ 9 (1995), available at <http://www.urich.edu/˜jolt/v1i1/hardy.html> (visited Nov. 5, 1997) (discussing satellite signal scramblers); Richard Rapaport, In His Image, WIRED, Nov. 1996, at 172, 175 (describing how Corbis uses passwords and watermarks to sell images through its online digital gallery).
[32]. After all, subscription services rely on automated processes (dedicated lines and passwords, for example) to manage rights to proprietary information.
[33]. See Port, supra note 23, at 62; Smith & Weber, supra note 7, at 34; Technological Solutions, supra note 7, at 3; Weber, supra note 7, at 3-4.
[34]. See Smith & Weber, supra note 7, at 34 ("Once the information is unlocked, customers can do whatever they wish with it, including making digital copies, redistributing it to friends and colleagues, printing many copies and so on."); Weber, supra note 7, at 3-4.
[35]. Using steganography, an information provider can firmly fix its digital signature, sort of an electronic watermark, on the documents it sells. See Port, supra note 23, at 62. Steganography also has cryptographic applications-it can hide "secret writings" in other documents-that relate to ARM only indirectly.
[36]. A micropayment is simply a very small payment, perhaps even on the scale of a fraction of a penny. Retail micropayments appeal to information providers because they lower consumer resistance to paying for information and yet can generate considerable revenues in the aggregate. See Hudgins-Bonafield, supra note 23, at 106; Smith & Weber, supra note 7, at 37; Weber, supra note 7, at 19-20.
[37]. Imbedded applications attach to and travel with the documents protected by ARM. These mini-programs can thus continue to control how consumers use the information that they purchase-even past the point of sale. See Hudgins-Bonafield, supra note 23, at 106-07; Sager, supra note 23, at 114, 116; Smith & Weber, supra note 7, at 36-37; Technological Solutions, supra note 7, at 3; Weber, supra note 7, at 5-6.
[38]. See Hudgins-Bonafield, supra note 23, at 106, 108; Port, supra note 23, at 62; Riddick, supra note 7, at 248; Sager, supra note 23, at 114, 116; Smith & Weber, supra note 7, at 36-37; Stefik, supra note 7, at 81; Technological Solutions, supra note 7, at 3; Weber, supra note 7, at 3-7.
[p. 566/p. 567]
[39]. See Hudgins-Bonafield, supra note 23, at 106, 108; Port, supra note 23, at 62; Riddick, supra note 7, at 248; Sager, supra note 23, at 114, 116; Smith & Weber, supra note 7, at 37; Technological Solutions, supra note 7, at 3; Weber, supra note 7, passim.
[40]. "There are always limitations to rights management and content control technologies. Nothing will prevent someone from scanning a printed page of a protected document (although fingerprinting or digital watermarking techniques may make it easier to trace and prosecute such infringements)." Weber, supra note 7, at 11-12; see also Stefik, supra note 7, at 80 ("A computer user can always print a digital page and then photocopy it. A digital-movie pirate can sit in front of a screen with a camcorder.").
[41]. 60 F.3d 913 (2d Cir. 1994), cert. dismissed, 116 S. Ct. 592 (1995).
[42]. See id. at 930-31 (holding that copying journal articles is not fair use where payment and licensing opportunities existed through Copyright Clearance Center).
[43]. 99 F.3d 1381 (6th Cir. 1996) (en banc), cert. denied, 117 S. Ct. 1336 (1997).
[44]. See id. at 1387 ("The approach followed by Judges Newman and Leval in the American Geophysical litigation is fully consistent with the Supreme Court case law.").
[p. 567/p. 568]
[45]. See American Geophysical, 60 F.3d at 914-15.
[46]. See id.
[47]. See id. at 914.
[48]. See id.
[49]. See id. at 926-32.
[50]. 17 U.S.C. § 107(4) (1994); see also supra note 20 (listing the four factors for fair use established in § 107). The Supreme Court has described this fourth factor as "undoubtedly the single most important element of fair use." Harper & Row, Publishers, Inc. v. Nation Enters., 471 U.S. 539, 566 (1985). "[T]o negate fair use one need only show that if the challenged use 'should become widespread, it would adversely affect the potential market for the copyrighted work.' " Id. at 568 (emphasis added) (quoting Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 451 (1984)). Notably, however, the American Geophysical court seemed more impressed by the Supreme Court's refusal in Campbell v. Acuff-Rose Music, Inc., to reiterate the importance of the fourth factor, quoting that case's instruction that " '[a]ll [four factors] are to be explored, and the results weighed together, in light of the purposes of copyright.' " American Geophysical, 60 F.3d at 926 (quoting Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 578 (1994)). The court in Princeton University Press v. Michigan Document Services, Inc., disagreed with such a reading: "We take it that this factor, 'the effect of the use upon the potential market for or value of the copyrighted work,' is at least primus inter pares . . . ." 99 F.3d 1381, 1385 (6th Cir. 1996), cert. denied, 117 S. Ct. 1336 (1997).
[51]. American Geophysical, 60 F.3d at 930.
[52]. Id. at 930-31.
[p. 568/p. 569]
[53]. Id. at 937 (Jacobs, J., dissenting).
[54]. Id. at 931 (emphasis added). But cf. Williams & Wilkins Co. v. United States, 487 F.2d 1345, 1357 n.19 (Ct. Cl. 1973) (holding that in the absenc