July 3, 2003

Copyright (c) 2003 Law and Contemporary Problems
Law and Contemporary Problems

Winter / Spring, 2003

66 Law & Contemp. Prob. 173

LENGTH: 30749 words


Yochai Benkler*

* Professor of Law, New York University

... It was a federal judge in Atlanta who told Randall that she could not write her critique in the words she wanted to use - a judge enforcing copyright law. ... As a matter of political morality, copyright supports democracy by grounding some types of expression in the market, independent of government patronage. ... '" This does not mean that in every case the specific use by the user and the specific exclusive right claimed must be considered afresh, as demonstrated by Harper & Row. ... At their core, Article I and the First Amendment claim that both democracy and autonomy are better served by an information production and exchange system built around a robust public domain, rather than one built around extensive regulation of the use of information and cultural materials through the creation and enforcement of exclusive private rights. ... As with the case of regulation that may instrumentally limit speech in ways that limit or enhance democracy, so too regulation of corporations may do so in ways that limit or enhance the autonomy of individuals. ... Moreover, enforcement of mass-market shrinkwrap licenses to prohibit users from using information in ways permissible under copyright law is generally a suppression of speech that must be reviewed under Turner. ... Decisions like eBay create an exclusive private right in information under the common law. ...  


I Introduction

A. The White Rabbit
Alice Randall, an African-American woman, was ordered by a government official not to publish her criticism of the romanticization of the Old South, at least not in the words she wanted to use. The official was not one of the many in Congress and the Administration who share a romantic view of the Confederacy. It was a federal judge in Atlanta who told Randall that she could not write her critique in the words she wanted to use - a judge enforcing copyright law. Randall is the author of a book called The Wind Done Gone. n1 In it, she tells a story that takes off from Gone with the Wind n2 from the perspective of Scarlet O'Hara's mulatto half-sister. In 2001, more than fifty years after Margaret Mitchell died, and years after the original copyright for the book would have expired under the law in effect when Mitchell wrote it, a federal district judge ordered Randall's publisher not to publish The Wind Done Gone. The Court of Appeals then overturned the injunction as a prior restraint. n3

B. Off with His Head!
Dmitry Sklyarov is a Russian programmer who faced the prospect of an American jail because he wrote software that enables people to read books that they are not allowed to read. Adobe Systems convinced the government to prosecute Sklyarov because he made it possible for people to read documents that Adobe had encrypted so that they could only be read with its eBook reader program. The Digital Millennium Copyright Act ("DMCA") makes it a criminal offense to provide software that lets people read digitized books with  [*174]  equipment that is not licensed to decrypt them. n4 The DMCA does not exempt people who write software that readers can use to read books that they are perfectly privileged to read. It does not matter that the person who wants to read the book owns a copy of it. It does not matter that the person who wants to quote from a book or a DVD wants to do so in a manner that is permitted under copyright law, such as under the fair use doctrine. It does not even matter if the encrypted materials are in the public domain - like Alice in Wonderland. n5

Adobe wanted to demonstrate how useful its eBook reader would be to publishers who wanted to distribute their books digitally. It used a digital book that it could get cheaply - Alice in Wonderland. The original text was in the public domain. Alice had already been digitized and proofread by volunteers as part of Project Gutenberg. n6 All that Adobe needed to do was take this free text, wrap it in its digital code, and presto - Adobe had a cheap and effective demonstration of how its technology could help copyright owners. The cover sheet of the closed edition of Alice is immensely instructive. It explains to the readers that they may not give, lend, quote, or print out a copy of this public domain work. n7

C. What, Exactly, Is Your Problem?
Edward Felten is a computer scientist at Princeton. As he was preparing to publish a paper on encryption, he received a threatening letter from the Recording Industry Association of America ("RIAA"), telling him that publication of the paper constituted a violation of the DMCA.

The music industry had spent substantial sums developing encryption for digital music distribution. In order to test the system before it actually entrusted music with this wrapper, the industry issued a public challenge, inviting all cryptographers to try to break the code. Felten succeeded in doing so, but he did not continue to test his solutions because the industry required that, to continue testing, he sign a nondisclosure agreement. Felten is an academic, not a businessperson. He works to make knowledge public, not to keep it secret. He refused to sign the nondisclosure agreement and prepared to publish his findings. As he did so, he received the RIAA's threatening letter. In response, he asked a federal district court to declare that publication of his findings was not a violation of the DMCA. The RIAA suddenly realized that trying to silence academic publication of a criticism of the weakness of its  [*175]  approach to encryption was not the best litigation stance, and it moved to dismiss the case on the understanding that it would never dream of bringing suit. n8

This paper does three things. First, it outlines the general framework of the relationship between two constitutional provisions - Article I, Section 8, Clause 8 and the First Amendment - and Congress's power to regulate the use of information and cultural resources through the institutional form of exclusive private rights. Second, this paper explains why it is appropriate, as a normative matter, to require close judicial scrutiny of congressional use of this particular form of regulation. Third, it identifies six specific pressure points currently bearing on this framework.

II General Framework
Copyright law is defined by constant tensions between exclusive private rights on the one hand and the freedom to read and express oneself as one wishes on the other hand. As a matter of economics, copyright represents a tension between the advantages of market-based production of information and cultural goods on the one hand, and the intrinsic limitations of property rights as institutional solutions to the public goods problem of information production on the other hand. As a matter of political morality, copyright supports democracy by grounding some types of expression in the market, independent of government patronage. But in doing so, copyright imposes substantial risks of harm to democracy and individual autonomy. n9

These tensions are mediated by two constitutional provisions: Article I, Section 8, Clause 8 of the Constitution (the "Exclusive Rights Clause"), n10 and the  [*176]  First Amendment's Speech Clause. n11 Because of physical and economic characteristics, existing information and culture are, absent law to the contrary, largely available for anyone to use as and when they please. For good reason and within the constitutional constraints imposed by the two provisions, Congress may deviate from this baseline. No one expressed this more poetically than Thomas Jefferson:

That ideas should freely spread from one to another over the globe, for the moral and mutual instruction of man, and improvement of his condition, seems to have been peculiarly and benevolently designed by nature, when she made them, like fire, expansible over all space, without lessening their density in any point, and like the air in which we breathe, move, and have our physical being, incapable of confinement or exclusive appropriation. Inventions then cannot, in nature, be a subject of property. Society may give an exclusive right to the profits arising from them, as an encouragement to men to pursue ideas which may produce utility, but this may or may not be done, according to the will and convenience of the society, without claim or complaint from anybody. n12
The role of the Constitution is to delimit the contours of the grant of exclusive rights and to constrain Congress's power to regulate, in pursuit of "the will and convenience of society," the use of information and cultural materials.

The Exclusive Rights Clause operates as a threshold filter on congressional attempts to create exclusive private rights in information. Uncharacteristically for Article I, much of its text is involved not in granting power, but in delimiting it. The preamble takes a specific stand on the theory underlying the American approach towards exclusive rights in information. Congress is empowered to grant exclusive rights "To promote the Progress of Science and the useful Arts." Ours is a self-consciously utilitarian, not moral, theory of such rights, consistent with the sentiments expressed in Jefferson's letters and those more generally prevalent at the time. n13 The Supreme Court too has consistently interpreted the Exclusive Rights Clause in a utilitarian way, holding that "the primary objective  [*177]  of copyright is not to reward the labor of authors, but "to promote the Progress of Science and useful Arts.'" n14

For over a century, unanimous Supreme Court opinions have interpreted the textual framework of the Exclusive Rights Clause as imposing a set of threshold constraints on legislation it enables. A congressional law creating exclusive private rights in information must comply with these constraints to fall within the grant of Article I, Section 8, Clause 8. n15 The Trade-mark Cases n16 were the first instance in which the Court held that the clause imposes affirmative constraints on Congress's power to create exclusive rights, but the Court most clearly outlined the content of these constraints in Graham v. John Deere Co.: n17

The Congress in the exercise of the patent power may not overreach the restraints imposed by the stated constitutional purpose. Nor may it enlarge the patent monopoly without regard to the innovation, advancement or social benefit gained thereby. Moreover, Congress may not authorize the issuance of patents whose effects are to remove existent knowledge from the public domain, or to restrict free access to materials already available.

Innovation, advancement, and things which add to the sum of useful knowledge are inherent requisites in a patent system which by constitutional command must "promote the Progress of ... useful Arts." This is the standard expressed in the Constitution and it may not be ignored. n18
This reading of the Clause - that it requires an exclusive right to encourage at least some information production and that it restrains such rights from removing or burdening free access to materials already in the public domain - is one that the Court has unanimously reaffirmed. n19

 [*178]  Furthermore, this constraint is substantive, not formal. It applies whenever Congress attempts to enact exclusive rights in information, regardless of whether it formally invokes the Exclusive Rights Clause. n20 While Congress may regulate information markets under the Commerce Clause, it may not do so by creating exclusive private rights in information in a way that circumvents the substantive limitations placed on its power by the Exclusive Rights Clause. This limitation requires that regulations enacted under the commerce power be different in kind, not only in subject matter and degree, from exclusive property rights - like rights that are the subject of the Exclusive Rights Clause. n21 This does not mean, however, that every technical aspect of the Exclusive Rights Clause - like the fixation requirement - constrains Congress when it acts under the Commerce Clause. Still, plainly "the Commerce Clause cannot be used [by Congress] to eradicate a limitation placed upon Congressional power in another grant of power," n22 - such as the Exclusive Rights Clause.

The First Amendment operates as an overlay. It provides a second level of analysis for laws that pass the initial threshold filter imposed by the Exclusive Rights Clause. This First Amendment analysis is more context-sensitive. It requires that the government justify its regulation of the use of information and cultural goods in terms that are largely understood to be the "intermediate" level of scrutiny applied in Turner Broadcasting System v. FCC. n23 As was the case with the must-carry rules at issue in Turner, copyright or similar laws are usually benign in intent. Their purpose is to enhance, rather than constrain, speech. They are, in many applications, content-neutral. First, they mark information or cultural materials for regulation on the basis of the history of their origination - the fact that they were authored. Second, the form of the regulation is usually a prohibition on use of the protected materials in a pre-stated variety of ways, absent permission of an identified party, the right holder. Not all applications of copyright law are content-neutral, however. One of the current pressures on this general framework comes from the anti-  [*179]  device provision of the DMCA n24 - under which a number of current cases, including the Sklyarov, n25 Felten n26, and Reimerdes/Corely n27 cases, are being decided. The anti-device provision, as applied to software programmers, may better be described as taking the form of a content-based prohibition on the use of a particular professional language or code to describe acts that the government deems particularly harmful - namely, decryption of specific kinds of encryption intended to protect copyrighted materials. For the moment, I will put this problem aside. It is a thorny one, and I will return to it in the third part of this paper.

To pass muster under the Turner standard, the government must show the statute: (1) serves an important government interest, and (2) does so in a manner no more restrictive than is necessary. n28 To fulfill the first prong of the test, it must be shown "that the recited harms are real, not merely conjectural, and that the regulation will in fact alleviate these harms in a direct and material way." n29 The second prong requires that "the means chosen do not "burden substantially more speech than is necessary to further the government's legitimate interests.'" n30 This does not mean that in every case the specific use by the user and the specific exclusive right claimed must be considered afresh, as demonstrated by Harper & Row. n31 It also does not mean that the contours of copyright law - the exclusive rights and the various detailed political deals between stakeholders - are categorically exempt from First Amendment review, as recently held by the Court of Appeals for the District of Columbia in Eldred v. Reno. n32 Categorical exemption is untenable. All one needs to do is imagine Congress amending the Copyright Act so that the definition of "a work" will include the sentence, "except that nothing shall be deemed a work of authorship if the author is at the time of authorship, or was within the preceding ten years, a Republican member of Congress." If the viewpoint bias makes the analogy too easy, imagine the same appellate court being confronted with a copyright law that adds to Section 107 of the Copyright Act: "Any use of a work is categorically a fair use if the copyright in the work is held, directly or indirectly, by any person who holds the copyrights in more than 10% of the works in the same market as the work used." n33 The Supreme Court in Harper & Row refused to create a special First Amendment defense in copyright, "in view of the First Amendment's protections already embodied in the Copyright Act's distinction between copyrightable expression and uncopyrightable facts and ideas, and the latitude for scholarship and comment traditionally afforded by  [*180]  fair use ... ." n34 Nothing in that opinion suggests that if Congress were to abolish the idea/expression distinction, or severely constrain "the latitude for scholarship and comment traditionally afforded by fair use," that its decision to do so would be exempt from First Amendment review.

The contours of copyright law and the shape of the prohibitions must be justifiable under the Turner standard as a proper regulation of the use of the information or cultural materials at stake. This test is not only applicable to changes in the law but also to existing rights as they are applied to categories of cases. Whether sampling for a few seconds n35 or using part of a poster for a number of seconds in a television show n36 is or is not a fair use, at least when a category of cases is being developed in case-law, requires an analysis of whether fair use law, thus interpreted, would cohere with the First Amendment.

III Political Theory of the Constitutional Limitations on Exclusive Rights in Information
What, one might ask, justifies such extensive judicial review of the creation and definition of exclusive rights in information? At some level, the answer is provided by the stories at the beginning of this paper. Alice Randall's freedom to shatter a cultural icon that she sees as profoundly offensive, in the words and means she deems most effective, is no less compelling than Paul Cohen's right to criticize the draft using his particular locution. n37 Dmitry Sklyarov's physical freedom is at stake. It has been threatened so that no one develops software that allows users and courts, rather than vendors, to decide when reading, quoting, and sharing of knowledge is privileged and fair. These freedoms, like Edward Felten's freedom to publish an academic paper, are all central to our understandings of freedom of expression. Randall, Sklyarov, and Felten provide crisp instances of the pressure that exclusive rights in information place on the freedom to read and express oneself. The next few pages explain more abstractly how one might understand the implications of exclusive private rights in information on democracy and autonomy - values at the core not only of the U.S. constitutional protection of freedom of expression, but of expressive freedom in modern liberal democracies more generally.

A. Democracy
Democracy is a concept with many conceptions. It would be irresponsible to argue that there is one well-defined set of policies that best serves democracy. Nonetheless, it is possible to outline the general direction in which strong or weak exclusive rights are likely to lead the organizational structure of information  [*181]  production and exchange. From this, one can develop a sense of whether one's conception of democracy leads to a preference for stronger or weaker protection on the basis that it serves, or disserves, democracy.

The most important effect of exclusive rights, in terms of democracy, is that they make institutional conditions more conducive for some approaches to information production than to other approaches. n38 Individuals with a commercial focus to their work and, more significantly, commercial organizations that build their business model around selling information and culture as finished goods benefit from strong protection. These rights are particularly helpful to organizations that own large inventories of existing information and cultural goods and that integrate new production with inventory management. Strong intellectual property rights are particularly harmful to organizations and individuals who produce information without intending to sell their output as a good. This includes non-profit organizations, such as universities, various public interest organizations, the government, and individuals who communicate with each other either as "amateurs" or as professionals driven by internal motivations, not by a profit motive. Less obviously, it also includes commercial organizations and individual professionals that operate on a service model that provides free access to information around which the service is rendered, rather than the sale of information as a product.

The conflict seen today regarding the scope of the public domain and the extent of exclusive private rights in information is a battle over the shape of the institutional ecology in which two very different modes of information production compete. The first mode is the increasingly industrial model of production that one sees in mass-mediated culture. Disney, AOL Time-Warner, Viacom, and News Corporation are some of the most visible examples. These are increasingly large organizations that control ever-larger inventories and integrate new production with reutilization and recycling of inventory. They aim to capture ever-larger audiences both nationally and internationally. Less well known, and quite poorly understood, is the emergence on the Internet of nonproprietary production as an increasingly important source of information and cultural materials. Over the past decade, with the widescale use of the Internet, the reach and scope of nonproprietary information production by nonprofit actors or by individuals - both amateur and professional - has expanded dramatically.

At a simple level, the Internet has made it possible for traditional non-profit social organizations to extend their reach and scope. In the past, the greatest constraints on non-profit production were the capital cost of production and the cost of communication, or distribution. The declining costs of making and distributing a high quality video or of collecting information and publishing reports exemplify the changes. These changes mean that information production on the public radio model, as well as production by other non-profit groups, can  [*182]  become increasingly salient in the information environment. This possibility must be central to our evaluation of the implications of strong rights to democracy.

More obscure, but potentially much more radical, is the emergence of nonproprietary peer production of information and culture. Thanks to the success of open source development projects like GNU/Linux or Apache, n39 this model of production is increasingly recognized in the area of software development. It is less widely appreciated that this highly decentralized model of non-proprietary peer production is expanding into other areas of information and culture more generally. n40 Tens of thousands of volunteers are mapping Mars craters faster than images of the planet's surface are produced. n41 Almost 40,000 volunteers participate in creating the open directory project, the most comprehensive and highest quality human judgment-based directory of the Web. n42 Thousands of peer reviewers and posters make contributions and write comments on Slashdot, one of the most sophisticated peer-produced news sites. n43 Through many such groups of individuals, who come together to create and make sense of their information environment, peer-production has emerged as a serious antithesis to the industrial production model that dominated twentieth century mass media.

When we think, therefore, of how strong we want copyright or other exclusive rights to information to be, in terms of democratic theory, the most important question is: What can democratic theory tell us about the choice between commercial proprietary production, increasingly organized in large enterprises and whose products are sold as finished goods, and nonproprietary production, both peer-production and professional production on a traditional nonprofit model?

The strongest democratic justifications of highly protective copyright serve what Baker has described as the elitist conception of democracy n44 and one version of what might be thought of as a republican conception of democracy. Strong protection is least attractive when measured by its effect on liberal conceptions of democracy - whether one holds some version of a pluralist conception, or a liberal, rather than republican, discourse-centered conception of democracy. n45 Again, it is unlikely that conceptions of democracy can be  [*183]  described in pure uni-dimensional terms. One's position, however, on the relationship of exclusive private rights in information to democracy ought to be informed by the extent to which one considers one or another of these ideal-type conceptions of democracy as playing a greater role in one's understanding of how democracy is best conceived.

The baseline argument from democracy in favor of strong rights is that dependence on government largesse or patronage tends to require authors, media, and others to be solicitous to those who pay the piper. Exclusive rights create a basis in the market for expression so that professional writers and creators can be sustained, based not on the decisions of government officials or patrons of the arts, but on the popular support of any non-trivial segment of the consumer market for information and cultural products. Copyright also provides incentives for creators to write more and thereby enrich the information universe within which a democracy functions. n46

There are two weaknesses to this argument for strong rights. First, markets for media products and information are rife with market failures that mean that they will not actually reflect the political views of the constituency. n47 Second, strong exclusive rights tend to commercialize, concentrate, and, to an extent, homogenize the information produced. n48 They aid information-as-goods vendors by raising the costs of, and potentially squelching, noncommercial discourse.

The argument for the importance of market-based production must therefore not only prefer market-based production to government-sponsored production. This is perhaps an easy choice for some - though, comparing network television news and magazines, whether national or local, to public television and radio should make one less than sanguine even about this claim. The argument, however, must also prefer market-based production to peer-production and independent nonprofit production. This requires the argument to take on a more explicitly elitist conception of democracy. "True" democracy - the participatory debate of all with all - is an illusion. Large companies, government officials, and other repeat players largely dominate the polity. The role of copyright is to create and sustain a large, powerful, and well-funded watchdog in this system, one enabled by market power to participate in this democracy of titans as an equal, criticizing the excesses of both government and the corporate world. n49 This conception has a good pedigree in the Press Clause focus on the fourth estate, and it is not implausible if it accurately describes the polity. The elitist argument has two primary weaknesses. First, it is unattractive to anyone who holds a more substantive, less cynical conception of democracy.  [*184]  If democracy means something more than an oligarchy of large market actors interacting with government bureaucrats who are watched by a large commercial press with occasional elections in which the masses select from among the elites who will run the government, then this argument in favor of strong rights is insufficient to justify a preference for strong exclusive rights in information. Second, the argument fails to consider the rise of peer production of information as a result of the economies created by low-cost computers and fast, ubiquitous network connections. The capacity of constituents to talk to each other as opposed to receiving the wisdom of those anointed by the television networks is profoundly important if one sees democracy as being about active engagement in discourse, rather than about passive selection from an exogenously-defined slate of candidates. One needs only to run a search request on Google n50 and then follow the variety of individuals and organizations making relevant information available, whether for free or for a fee, to see that a conception that focuses purely on large scale industrial production of information is too heavily committed to the twentieth century model of production by a few corporations transmitting to millions of passive consumers. Even without resorting to a more aspiring conception of democracy, it is not at all clear that thousands or millions of networked peers will not do as well or better than a few hundred professional reporters or commentators can do. An efficient system that allows individuals to report what they see, to blow the whistle, and to collect and comment on information globally, supplemented by a combination of publicly funded and advertising-supported (rather than property-based) professional media may well be superior to a more concentrated and commercial system that is occupied by a small number of media companies whose business model is based on the sale of access to their products.

This, however, raises a second line of defense for concentrated commercial production, which is rooted in a particular, nation-state focused version of a "republican" conception of democracy. Such a defense values polity-wide common discourse toward reaching a conception of the common good as the primary modality of democracy. This is only a somewhat unfair and oversimplified description of Cass Sunstein's primary complaint in n51 Sunstein warns us of the dangers of having too widely distributed an information production system - one where every person can find whatever information he or she wants and filter out everything else. n52 This, he argues, leads to a loss of common culture and with it, a loss of the possibility of a common discourse necessary for a polity to function well. n53 Moreover, Sunstein argues, individuals who interact only with like-minded people tend to reach more extreme positions  [*185]  and discount counter-arguments. n54 Thus, deliberation toward the common good may be harder to achieve.

The aspect of Sunstein's argument that is most relevant to the comparison of widely distributed information production versus concentrated commercial production is his claim that "general interest intermediaries" form the common experiences and knowledge necessary for engaged discourse. n55 Sunstein's argument has a number of limitations, which are both internal to "republican" conceptions and also a function of the limitations of that conception as a description of what one would consider valuable in a democracy.

Internally, there are two important limitations to the argument based on maintaining a common culture and discourse. First, the argument assumes that the relevant political community is the same community whose contours are defined by the media market. This assumes that there are national media outlets and that the most relevant polity is the nation-state. It is not implausible, however, that the engaged polity is more important than the state within republican conceptions. If, for example, the city rather than the nation is the level at which engaged politics can occur, then some republican theorists would locate power where the political action is, rather than attempt to structure engagement so that it fits some other stated level of governance. n56 More generally, if the Internet permits engaged politics in civic groups, interest groups, and communities of interest, rather than national boundaries, then supporting that engagement at the local level should be a concern of republican democracy. In some sense, the attempts to make the Internet Corporation for Assigned Names and Numbers ("ICANN") n57 into a representative, open, deliberative Internet governance forum is the kind of thing one might imagine developing out of the a republican view of the discourse actually occurring on the Internet. It is not at all clear that a commitment to engaged politics requires burdening local discourses so as to secure the revenue streams of national media giants, on which one relies to sustain nationwide common discourse.

This leads to the second internal criticism of this republican defense of strong copyright. Relying on a set of actors to define the common agenda and culture is only acceptable if these actors are "virtuous" in the republican sense - that is, if they set the agenda and the common culture with reference to the common good. Exclusive-rights-based commercial producers are unlikely to provide such a virtuous common culture for two reasons.

First, within the republican conception, common culture only benefits a democracy if it is made of materials conducive to creating informed and engaged citizens. It is crucial to see what the large speakers are saying before one can tell whether they help or harm discourse. There are two distinct types  [*186]  of concern in this category. First, there is what one might call the Berlusconi effect. Owners of large media that occupy center-stage in creating the common culture can capture a disproportionate share of political power and undermine, rather than serve, democracy. Whether this effect is worth the creation of "common culture" is questionable, unless one has a particularly na<um i>ve view of the benign nature of media barons. Programs like the Jerry Springer Show, Baywatch, or even cheap network television news capture the second type of difficulty. Market actors are supported by advertising and direct revenues. To gain these revenues, market actors will offer materials that the largest number of people, on a national and international level, will buy. There is no reason to think that this fare will be particularly conducive to an engaged polity; quite the contrary. Moreover, an information good is particularly valuable as a good requiring property protection if it is not ephemeral. Programming that is most valuable to democratic discourse - ephemeral news, analysis of current affairs - is the type of information that least benefits from exclusive rights, even when produced commercially.

Second, Sunstein's conception of the relationships of individuals to the information production and exchange system is too passive and too limited in its view of individuals as consumers in the market. He does not consider the productive, rather than the consumption, role that individuals can play. Diversity on the Internet does not exist primarily because someone decided to sell consumers something in a finely sliced market segment - the segmentation Sunstein sees as occurring and embodied in Negroponte's trope, the Daily Me. n58 Instead, the diversity exists because individuals themselves participate in developing and expressing a viewpoint. This is precisely the type of productive relationship to the information flow in one's polity that peer-production facilitates and proprietary production squelches. It is in these yeoman speakers that republican democracy thrives, and squelching their speech to maintain universal attention to synthetic discourse packaged for mass consumption is perverse.

Widely distributed and noncommercial production strategies - those best characterized by the emerging properties of peer production on the Internet - are particularly attractive from the perspective of liberal conceptions of democracy. This includes both pluralist conceptions and discourse-focused substantive conceptions that root the value of discourse in individuals rather than an independent valuation of "the common good." In this context, "pluralist" conceptions means those conceptions that accept as given the possibility that some people's interests will be fundamentally and irreconcilably opposed. They are conceptions aimed at a polity that will allow people to express their interests, fight over them politically, and implement them in law. The implementation, however, occurs subject to constraints that allow the losers in the political process sufficient freedom from coercion to remain committed to the democratic polity rather than trying to resist it. n59 Liberal discourse-oriented conceptions of  [*187]  democracy, as opposed to republican or communitarian conceptions of democracy, largely focus on the equivalent notion that democracy entails engaged discourse by individuals. Such discourse-oriented conceptions draw the legitimacy of sovereignty from maintaining appropriate conditions to allow people to engage in fair and open discourse. n60 It differs from the pluralist conception in that it assumes the possibility of the emergence of common values through dialogue among participants and sees such discourse, rather than the clearance of pre-defined conflicting values in peaceful co-existence, as the core of democracy. Both these conceptions, however, value the possibility of discourse not only, or even primarily, on a society-wide basis but also, importantly, in smaller settings, even among like-minded people, working to clarify and develop a full conception of the values they hold as well as the ability to express these views. The capacity of individuals in small and large groups to come together to discuss their interests is central to a well-functioning democracy under both these liberal conceptions. Also, the capacity of non-proprietary production or service-based provision of the platforms over which individuals and groups "meet" to engage in dialogue and collective self-definition becomes central to the democratic enterprise. It, rather than the generation of a "common culture" to be received by individuals, becomes the most important mode of information production. An important caveat here is that if, as Sunstein claims, individuals become more polarized and less willing to listen and engage in discourse when they talk to like-minded people, n61 then this conception of discourse-centered democracy, not only the nationwide republican version, will be undermined by discourse occurring solely at the level of small groups.

The point of this discussion is not, of course, to offer an exhaustive democratic theory of communications. The point is to outline the considerations and concerns regarding democracy that can be seen as involved in decisions about stronger or weaker exclusive private rights. To the extent one values active, engaged individual participation in defining and expressing political values in a polity as central to the democratic enterprise, one should strengthen peer-based models of information production and exchange, even if this requires policies that weaken proprietary production based on a sale of goods model.

B. Autonomy
Autonomy is also a concept admitting of many conceptions, some quite vigorously opposed to others. In general, conceptions of autonomy can be divided into two camps: formal - or as Richard Fallon described them, ascriptive - and substantive. n62 "Ascriptive" refers to those conceptions that view autonomy as a capacity ascribed to human subjects as an assumption of liberal law making. As  [*188]  Robert Post explains, the nature of "structures of social authority" will depend on whether they treat the object of regulation as autonomous individuals. n63 "From the point of view of the designer of the structure, therefore, the presence or absence of autonomy functions as an axiomatic and foundational principle." n64 Those who adhere to an ascriptive conception assume the existence of autonomy as a foundational principle, not as a proper subject for institutional manipulation. Therefore, autonomy is also "formal" in that it operates as an analytic constraint on law making, rather than substantive, or concerned with identifying and affecting the actual condition of autonomy of individuals. Substantive, or descriptive conceptions of autonomy, locate individual autonomy within the actual constraints, both internal and external, with which real human beings live. Under the substantive conception, autonomy is a capacity and a condition admitting of degree, which is partly a function of, rather than an assumption underlying, institutional structures. It functions not as a formal analytic constraint on policymaking, but as a substantive goal to be attained by policy. In this framework, achieving, preserving, and improving the autonomy of actual individuals is a proper, for some the primary, object of liberal institutional design.

Elsewhere I have attempted to offer a specific set of practical guidelines for autonomy-loving institutional design in the area of information policy that is neutral as among these competing conceptions. n65 While I derive the policy guidelines from a substantive conception of autonomy, the operative design principles are two that cannot offend even those who hold a purely ascriptive or formal conception of autonomy.

First, law or policy that systematically and drastically reduces the information available to large numbers of, or defined classes of, individuals in a society, undermines autonomy. "Reduces" has both a quantitative and a qualitative dimension. The quantitative dimension is largely irrelevant in a modern liberal state and is more relevant to describe totalitarian efforts to control people's lives by severely restricting information flows to them. The qualitative dimension is still very much relevant even in the most liberal of democracies and involves the availability of information about non-mainstream or critical life options. For an individual to be autonomous, she must play a substantial role in defining her life plan. Where the range of life options perceived is largely congruent with the range of behaviors followed by everyone else in the mainstream of her society, the opportunity to make that life her own, as opposed to simply the life she leads as part of a herd, is diminished. The presence of knowledge about and opportunity for unconventional life choices enable an individual to make any life plan her autonomously selected own.

 [*189]  Second, a law that systematically gives one person or class of people control over the information flows that make up the information environment within which others live is a law that suffers an autonomy deficit. To the extent that a law increases the opportunities for one person to manipulate the information another person receives, so as to make that other hew more closely to a life plan set by the manipulator, the law violates the autonomy of the person whose information is so manipulated.

Both of these design principles are decidedly agnostic about whether the immediate source of constraint on an individual's autonomy is a government agent. The primary focus here is on the relative role an individual plays in defining and pursuing his own life plan, irrespective of the source of constraint. Whether the manipulator is the state, or whether it is state law or corporate policy that restricts information about the range of available life options, is irrelevant from the perspective of autonomy. Some of the most fruitful discussions of autonomy have, for example, occurred with regard to the physician-patient relationship, where the government is absent. This is because autonomy is a distinctly personal and humanistic value. n66 Autonomy relates to a capacity or condition of an individual qua individual, not as a citizen of a state or constituent of a community. All constraints placed on that individual affect autonomy, not purely, and not always most importantly, those placed by the state. The state-centered focus of legal analysis enters only at the level of selecting for inquiry from among the many potential actions that affect autonomy. A legal or constitutional inquiry, as opposed to general moral inquiry, into autonomy is concerned with laws that affect the relative role individuals have in selecting and pursuing a life plan. But whether those laws directly restrict autonomy, or indirectly do so by, for example, placing one non-government person in the position to control the life of another, is irrelevant to the question of whether those laws restrict autonomy.

The practical implication of all this is that there is an area of overlap between the concerns of democracy - at least a wide range of liberal versions of democracy - and those of autonomy. A widely dispersed system of information production, which produces a wide range of diverse information about and representations of how life can be, serves autonomy in the first dimension just as it serves robust democratic discourse. Furthermore, large-scale commercial media that occupy most of the channels of communication and control most of the cultural raw materials from which expression is made have substantial power to shape the perception of alternative life choices available to many individuals. An Internet-like system of news and cultural production, operating with weak property rights, dissipates this power and structures the interaction of individuals with the information flows around them in a less externally controlled pattern.

 [*190]  In addition, some aspects of autonomy are directly tied to the emergence of peer-production. However, these aspects are more closely associated with a substantive, or descriptive, rather than formal or ascriptive, conception of autonomy. The emergence of peer-production as an economic - and ultimately social - transformation represents, most importantly, a change in the menu of options for productivity in the information economy. In the "atoms economy," n67 we settled more or less on two modes of making production decisions: the market and hierarchy (whether in managerial firms or government-owned enterprise). Coase's Nature of the Firm explains that markets best coordinate some economic activities, while managers better organize others. n68 Accordingly, most individuals live their productive life as part of corporate organizations, with relatively limited control over how, what, or when they produce. These organizations, in turn, interact with each other through a combination of markets and hierarchy. Consumption is strictly separated from production for most people and is largely devoted to receipt of finished goods, not to creative utilization of materials to shape one's own environment. What emerges in the information economy is a model of peer-production - where individuals communicate with each other about what projects are worth pursuing, and who might want to take them up and share their products in an economy of gifts, reputation, and relationally-based rewards. Consumption and production are integrated, not separated, so that each individual is a user, rather than either purely a producer or a consumer.

From the perspective of autonomy, there are two enormously important chunks of life during which individuals can play a larger role in defining how their time is spent: production and consumption. The production dimension of autonomy occupied much of the "third way" literature that emerged in the 1980s, from Piore and Sable's Second Industrial Divide n69 to Unger's False Necessity. n70 It is mirrored on the consumption side by the shift to "user" from consumer. Because this shift is an attribute of autonomy that is not necessarily shared across conceptions of autonomy, I have not included it as a core consideration in thinking about autonomy and information policy for purposes of analyzing the liberal constitutional constraints on the creation of exclusive private rights in information and cultural materials. Yet as background, for those who do believe that autonomy is affected by how much of one's day is, and what range of activities are, more or less valuable under an individual's control, peer production should be a valued social and economic phenomenon; a phenomenon that policymakers concerned with autonomy should facilitate.

 [*191]  There are two important concerns regarding the claim that autonomy is served by weak exclusive rights that lead to decentralized information production. First, there is the information overload complaint. Wide distribution of the capacity to produce information leads to the generation of a tremendous amount of information of varying degrees of quality. An agent who constantly needs to sift through mounds of data to understand the world is left exhausted and incapable of making a decision, rather than empowered to control his own life.

There are two reasons why the concern with information overload does not suggest a retreat from the limited design principles proposed here. First, although a widely distributed information production system causes potential information overload, a concentrated information production system may still not be preferable from the perspective of autonomy. If the "filter" that reduces the amount of information available to an individual is perfectly aligned with the individual's interest in being autonomous, a concentrated system may succeed from an autonomy perspective. Unfortunately, deviations from this ideal state create a situation where the filter controls the life of the recipient. The more that the mechanism of reducing the amount of information and its organization through a relevance and accreditation algorithm is controlled by a third party with an agenda of its own, the more the price paid for reducing information by concentrating the information production function becomes unacceptable. Second, increasingly sophisticated mechanisms allow individuals to reduce the amount of information they receive without abdicating control of one's life to Berlusconi or Murdoch. An individual may utilize technical mechanisms to customize the information environment encountered according to parameters the individual sets. These are some of the same mechanisms that Sunstein decries as harming the public sphere. Furthermore, relevance and accreditation themselves are being produced on a peer-production model, as one sees in a wide range of sites, including the Open Directory Project, n71 Slashdot, n72 some aspects of, n73, n74 and many others. n75 As these mechanisms for common, non-proprietary production of the relevance and accreditation function improve, individuals will be able to manage the universe of diverse information available to them without subjecting themselves to the power of others to the point of surrendering their autonomy.

Second, specifically with regard to exclusive private rights in information, there is an argument that autonomy or some conception of individual freedom actually supports, rather than criticizes, exclusive private rights. This is largely the argument underlying European moral rights. Perhaps, goes this argument, there is some "cost" in terms of autonomy from the creation of exclusive rights  [*192]  to control information and cultural materials, but there are also strong autonomy-based claims to such exclusive rights. This article is mine in the deep moral sense that it is an external reflection of my intellectual will and that it, in some sense, is an externalization of my persona, my identity. It is much more me than many physical things that belong to me, and at least on a Hegelian conception of property, I have a very strong claim to exclusive control over it as the embodiment of my will and the creation of my personality. n76

Initially, it should be understood that such an argument can be based only on a substantive or descriptive conception of autonomy, not on a formal or ascriptive conception of autonomy. Under a formal conception, law treats individuals as having the capacity for autonomy independent of or prior to law, and designs law accordingly. Autonomy under this conception cannot be a function of the law that gives exclusive rights. People are autonomous regardless of whether you grant them such exclusive rights. Furthermore, nothing about formal recognition of individuals' capacity for autonomy requires that (a.) law prohibit some people from reading, viewing, or reworking cultural materials that (b.) they, as a practical matter, can read, view, or rework, so that (c.) other people can successfully pursue their plans to make money from selling permission to read, view, or rework. n77 An autonomy-based argument in favor of exclusive rights must therefore rely on some substantive conception of autonomy, one that sees autonomy as in some important measure a function of and dependent on the legal framework within which individuals live, rather than as an axiomatic presupposition to its design. This is not to say that such an argument is improper, in particular given that aspects of my own argument rely on a substantive conception of autonomy. But it does weaken the appeal of this argument as a criticism of autonomy-oriented policy design principles that are intentionally limited to those that can be defended on grounds neutral as between these two basic conceptions of autonomy.

Substantively, there are two types of answers to the argument from the moral claims of the author. The first, to which I will return, is that this argument is irrelevant whenever the owner of the copyright is a corporation. Works  [*193]  for hire, complex multi-participant works such as films that movie studios produce, works otherwise assigned to corporations are no longer the expression of autonomy. Rather, they represent an alienation of the connection between the author and the work. It is precisely for this reason that countries with strong moral rights traditions do not recognize work-for-hire or alienation of the moral rights, as opposed to commercial exploitation rights. Treating corporations as legal persons may be a very useful legal construct for organizing human production, but that does not make corporations into agents with independent moral claims. Since in many cases the claimant is a corporate owner of rights, not an individual author, the moral claims in favor of exclusive rights are weak.

The second answer is that autonomy does not support the actual exclusive rights Americans have - which are almost completely focused on commodification and alienation. If at all, autonomy supports rights that are not recognized in the United States - inalienable rights to attribution and control over integrity of the work. The Hegelian defense of intellectual property derives from a notion that individual will expresses itself and embodies the person in things. n78 The strongest rights derived from this notion are rights to be associated with the expression and rights to maintain the expression's integrity in the form in which a person has expressed their personality. The former requirement, attribution, has to do with the sense that the value of the thing as an expression of self is in its being an expression of a particular self - its author - and not a fungible expression of human creativity as a general category. The latter requirement, integrity, has to do with respect for the thing as a unique expression of self, and it requires that the individual be able to control that expression, because its alteration, and in particular its alteration inconsistent with the will of the author, alienates the expression from the person and undermines the capacity of the person to be expressed and embodied in the intellectual creation. This is why European "moral rights" include rights of attribution and integrity that resist removal of the actual author from future control over his or her work. n79 In the United States, quite to the contrary, such practices as work-for-hire and relinquishment of authorial control are perfectly common. American exclusive rights are focused on enabling and facilitating the alienation and economic exploitation of covered works, not on giving authors inalienable rights to attribution and control over the integrity of the work. The moment of alienation is the weakest point of the Hegelian defense of property. At that point, one person parts with the thing that previously embodied his will, handing it over to the property of another, who acquires it as part of her plans to embody her will in the thing. Continued control after alienation would subject the latter to the will of the former. Hegelian intellectual property theory then has to rely on some instrumental explanation to sustain support for alienability of the  [*194]  commodity aspects of the work, while requiring inalienability of the moral claims.

Commodification and alienation of my work - the treatment of it as a thing alienable from me and exchangeable for money through a market n80 - is an act of separation of the thing from what is uniquely mine and its instantiation in a fungible commodity. That is not to say that it is morally wrong of me to alienate my work. But when I do so, I am treating those aspects of the work that I am selling as a commodity, whose fungible remuneration is useful to me, but is not an expression of me. Certainly, when the entire right is alienated, as when an author sells all rights to the work to the publisher, what has happened is alienation, not embodiment. And where copies are sold and the right protecting the market in copies is a prohibition on selling copies, each copy is a thing that does not embody the self and which is instead a thing through which the reader is attempting to embody him or herself. Justin Hughes argued in favor of treating the right to control copying as nonetheless a personality-based right. Hughes claimed that by paying me money readers recognize my personality, while those who do not pay for copies recognize the personality embodied in the work no more than a trespasser recognizes the property of a landowner. n81 Yet a company that invests some of its excess cash in buying art based on professional advice as to what will appreciate in value is not "recognizing the personality of the artist." Additionally, fans passing bootleg tapes are expressing a complete association between the artistic expression and the artist and are expressing adoration of the artist's personality as embodied in the art. To hold otherwise is to be a "fetishist of little green paper." n82 Failure to pay for the commodity or for its utilization largely in the form in which it was commoditized is, at most, an interference with the artist's success in commoditizing her work. If commoditizing her work is her life plan, a refusal to pay for the commodity interferes with her wellbeing as an autonomous agent, constituted by her success in pursuing her autonomously chosen life plan. n83 But they are not interferences with the thing as an expression of her personality and are not interferences with her autonomy - her capacity to express herself in the world through the manipulation of ideas and symbolic representations. On the contrary, it is when one interjects a right to commoditize against someone else who uses materials one authored as part of the materials with which they choose to express their autonomous view of the world, that autonomy is offended.

A perfect example of this is open-source licensing or free software licenses. An open source license will often secure the original author's rights of attribution  [*195]  and integrity, n84 while eschewing commercial exploitation rights. This type of license, therefore, is an implementation of the autonomy of the authors that largely preserves the expressive autonomy of users. The license enhances autonomy on both sides. As a result, open source licenses are directly inverse to more traditional licenses that serve the commercial exploitation interests of the original authors (but not their personality interests) while restricting the expressive autonomy of users.

Consider also the Wind Done Gone case. n85 There, SunTrust Bank, administrator of Margaret Mitchell's estate, sued Houghton Mifflin, the publishers of Alice Randall's book, to prevent its publication. The book told a story that clearly borrowed from Margaret Mitchell's Gone with the Wind. First, if SunTrust Bank were the owner of the copyright, standing in its own shoes because it had hired Margaret Mitchell and owned the copyright, it would have had commercial but not moral claims to the book. Second, assuming SunTrust Bank was simply asserting Mitchell's rights as administrator of the Mitchell estate, the claims must be considered as though they were made by Mitchell.

Imagine that Alice Randall and the bank had conducted negotiations in which the bank required Randall to pay the bank 25% of the revenues from The Wind Done Gone and omit any direct or implicit reference to interracial sex. n86 It is intuitively simple to see this request for payment as one for commercial exploitation of the work. It may be perfectly sensible for the author to demand payment. It may also be perfectly sensible, on a utilitarian calculus, for society to pass a law requiring Randall to either pay what the bank asks or forego use of the story. Still, it is not a violation of Mitchell's autonomy for Randall to refuse to pay. If, however, the romantic view of the Old South was a central component of Mitchell's world view, and its expression in Gone with the Wind is an expression of this central aspect of her self, then use of her book, through a critical prism, to expose the ubiquity of interracial sex in the Old South is something that goes to the integrity of Mitchell's work as an expression of her self. Yet it is precisely this transformative utilization of work to subvert and mock the very message of the original author that is the quintessential case of permitted fair use in U.S. copyright law. In terms of autonomy and democracy, this is not a surprising calculus. Randall's retelling of this story, given the central role it plays in American culture by masking what she perceives to be a central truth in her experience as an African American woman and the central role it played in her childhood in defining her alienation from American society, is central to her autonomy. Moreover, Mitchell's autonomy is directly pitted against the democratic commitment to airing the widest range of views  [*196]  expressed in the terms that their speakers find most effective to convey their message.

C. The Court as an Institutional Counterbalance
The preceding two subparts outlined substantive support for the robust constraints Article I of the U.S. Constitution and the First Amendment impose on the creation and definition of exclusive private rights in information. At their core, Article I and the First Amendment claim that both democracy and autonomy are better served by an information production and exchange system built around a robust public domain, rather than one built around extensive regulation of the use of information and cultural materials through the creation and enforcement of exclusive private rights. This subpart discusses the institutional reason, rather than the substantive reason, to introduce close judicial scrutiny of legislation that expands exclusive private rights at the expense of the public domain.

Our legislative process demonstrates a systematic imbalance in favor of the expansion and deepening of exclusive rights to information at the expense of the public domain. The imbalance exists because the benefits of such rights are clearly seen by, and expressed by, well-defined interest holders that exist at the time the legislation is passed. In contrast, most of the social costs - which are economic, social, political, and moral - are diffuse and likely to be experienced in the future by parties not yet aware of the fact that they will be affected by the extension of rights. For example, the Estate of Margaret Mitchell knew that an extension of copyright, negotiated in the 1960s, then in the 1970s, and again in the 1990s, would increase revenues to the estate by postponing the date when the book falls into the public domain. Alice Randall, however, was too young in the 1960s or 1970s to participate in the debate. Similarly, it is unlikely Randall focused on the effect the Sonny Bono Copyright Term Extension Act of 1998 (the "Copyright Term Extension Act"), n87 would have on her book when it was under debate. Likewise, Ed Felten in his worst nightmares probably could not have imagined that his academic paper on the weaknesses of the Secure Digital Music Initiative would subject him to civil suit, and, if Dmitry Sklyarov's experience is a predictor, perhaps even criminal liability. The recording industry, however, when it was lobbying for the Digital Millennium Copyright Act had a clear sense that this legislation would enable them to sell encrypted music in a more tightly controlled fashion.

In contrast, it is never the case that the diffuse and future users will band together to expand fair use. Even if they were to band together, it is impossible that copyright owners would remain unaware of the initiative and fail to offer substantial opposition in the legislative process. The legislative process, therefore, has a systematic bias toward more extensive exclusive, private rights at the expense of the public domain. This bias explains the constitutional framework  [*197]  developed by the Supreme Court over the past century whereby courts provide a filter to limit Congress's power to expand private information rights. Both the threshold inquiry of Article I, and the Turner standard for First Amendment review, treat expansion of rights as subject to constitutional scrutiny. In contrast, expansion of the public domain, or elimination of exclusive rights does not require the same scrutiny. Expansion removes constraints on the use of cultural materials and information, making more modes of expression available. Thus, it deregulates, rather than regulates, the use of cultural and information materials. The justification of this one-way ratchet is precisely the systematic imbalance in the other lawmaking branch - the legislature. Since too extensive a definition of rights is economically inefficient and harms both democracy and autonomy, but is directly in the interest of a powerful lobby, it is the role of courts to prevent the systematic and excessive expansion of exclusive rights by serving as a backstop against this political economy.

Moreover, it is important to underscore that it is even more justified for courts to act as a moderating force on politics in the area of intellectual property rights than in media regulation - a parallel area of regulation of information production and exchange where courts have taken a very active role. When the broadcast and cable industries battle over "must carry" rules, each industry is well represented in Congress, and all of the potentially regulated parties are at the table. Power plays and successful capture of an agency or a legislature sometimes determine the outcome of this process. Often, however, negotiated deals determine the outcome and are later challenged by actors in the negotiation who try to make their deal sweeter. Laws regulating large players whose interests are well defined and understood at the time of the regulation - like broadcast, cable, or telephone regulation - are therefore substantially less in need of judicial review from an institutional perspective. In contrast, with copyright and similar exclusive rights judges must be attentive to prevent legislatures from selling the store in the absence of the parties likely to bear the primary brunt of the regulation.

IV Pressure Points - The State of Play Today
Thus far I have covered two issues. First, I have delineated in doctrinal terms the constitutional framework within which Congress can regulate the use of information and cultural materials through the creation and definition of exclusive private rights. In this context, Article I of the Constitution operates as a threshold constraint, requiring Congress to show in creating such rights: (1) that the exclusive control the rights grant can plausibly increase information and cultural production, (2) that the rights be given only to those who add original contributions, and (3) that the rights not prevent use of materials in the public domain. The First Amendment then requires Congress to show that any given right meets, at least, the Turner standard of intermediate scrutiny. In subpart III, I explain why such a vigorous system of judicial review is normatively  [*198]  required. My argument is that both democracy and autonomy are served by a robust public domain and harmed by exclusive rights that excessively regulate the use of information, knowledge, and cultural materials. Moreover, the political economy of legislation is such that the legislature will systematically tend to create exclusive rights beyond what is economically efficient, as well as beyond what gives proper consideration to the implications of these rights for democracy and autonomy.

This section briefly outlines six particularly salient pressure points on this constitutional framework.

A. That's Not What the Law Says at All
The description I have given of the law is declarative - as though it is an uncontroversial view of the constitutional framework that constrains Congress from regulating information and cultural production too greatly. It is not. It is a decidedly contested view, though I believe it is the best reading of the Supreme Court's precedent. It is also true, however, that the Supreme Court largely has been more solicitous of user privileges and less enamored with "intellectual property" than the lower courts. In addition, not all the lower courts have accepted the spirit of the constitutional framework delimited by the Supreme Court's jurisprudence in this area. The lower courts' rejection of the Supreme Court's view that either Article I or the First Amendment substantially constrain Congress is most completely and starkly seen in Eldred v. Reno. n88

Eldred v. Reno reviews Congress' adoption of the Copyright Term Extension Act in 1998. Through this Act, Congress extended the copyright protection term from life of the author plus fifty years to life of the author plus seventy years. The Act extended the seventy-five year term of protection for works initially owned by a corporation to a ninety-five year term. n89 These extensions were retroactively applied to existing works. n90 Eric Eldred, a retired programmer who scans out-of-print books and produces online editions for free distribution, challenged the Copyright Term Extension Act because it prohibited him from scanning and making available works that would have fallen into the public domain but for the extension. The D.C. Circuit rejected Eldred's claims that the Copyright Term Extension Act violated the "limited times" constraint imposed by the Exclusive Rights Clause and the First Amendment. n91 It held that the Preamble provides no constraint on the power of Congress and that when Congress creates the limited monopolies that it is empowered to create under Article I, it is subject to rationality review. n92 The Circuit Court seemed  [*199]  little bothered by the fact that its holding directly contradicts the Supreme Court's decision in Graham v. John Deere Co. In Graham, the Court stated:

At the outset it must be remembered that the federal patent power stems from a specific constitutional provision which authorizes the Congress "To promote the Progress of ... useful Arts, by securing for limited Times to ... Inventors the exclusive Right to their ... Discoveries." Art. I, 8, cl. 8... . The Congress in the exercise of the patent power may not overreach the restraints imposed by the stated constitutional purpose. n93
The Circuit Court brushed away this specific requirement that copyright be for "limited times." The court claimed that "limited times" included very long, but not perpetual, grants, as well as retroactive extensions of time. Its argument with regard to the First Amendment claim was the implausible argument that laws Congress calls "copyright" are "categorically immune" from First Amendment scrutiny. n94

The Copyright Term Extension Act of 1998 - although not the decision in Eldred v. Reno - provides an excellent ground for explaining how the two constitutional constraints should interact. First, there is the retroactive application aspect of the Act. According to Graham, Congress can only create exclusive rights if the new rights could plausibly stimulate creation and if they do not remove anything from the public domain. n95 The adoption of a retroactive extension for works that have already been created fails both tests set forth in Graham. First, retroactive extension cannot plausibly promote the progress of science and the useful arts. n96 It cannot create incentives. Margaret Mitchell will not rewrite Gone with the Wind in 1998 no matter how many more years of copyright protection the book receives. Similarly, new books will not be written because old books are retroactively protected for a longer period than their authors were promised when they wrote them. For any future action to be affected by retroactive extension, a current author would have to think that a project would not justify itself under the present term of copyright but would nonetheless be worth the effort if copyright were extended to an additional given term, discounted by the probability that copyright will in fact be extended in the future to that necessary length. The implausibility of this argument can readily be captured by conjuring up the image of a movie studio executive pitching a project to investors by saying, "we won't make money within the seventy-five years that copyright law currently gives us, but Congress has traditionally extended rights over time, and if Congress extends copyright to ninety-five years, we'll make a killing on this one!"

Second, once a work is created, there are aspects of it that are enclosed, and aspects of it that are in the public domain. When Margaret Mitchell wrote Gone with the Wind in 1936, the longest term of protection possible for the book was fifty-six years. In 1936, then, the state of the world was that the  [*200]  enclosed domain included the book Gone with the Wind for fifty-six years. The public domain included the book Gone with the Wind in the year 1992. This is not a word game. It is an economically accurate description of the state of the resource called Gone with the Wind as perceived at the time of its creation by rational economic actors. If someone were to value the rights to make a movie from the book, they would have to price the license necessary. Imagine that the price is P. Whether it is worth paying that price is partly a function of the cost of waiting for the expiration of the copyright - P discounted to present value from the date of expiration to the date of use. That calculus is likely to have little effect on a decision made in 1938, but likely to have tremendous effect on a decision made in 1991. Gone with the Wind is "in the public domain" - on a deferred basis - at the moment of its creation. The precise value of its presence in the public domain depends on the time remaining until expiration. To extend the term of copyright for Gone with the Wind after it has been created is to remove its public domain aspect from the public domain.

Where, as in the case of retroactive term extension, Congress passes a law that removes material from the public domain with no plausible claim to increase incentives for creation, the law must fail under the limitations of the Exclusive Rights Clause. The Eldred court's failure to discipline its rationality review by focusing on the specific elements required by Article I, Section 8, Clause 8 of the Constitution resulted in practically no review of the Copyright Term Extension Act.

But what becomes of the prospective application of the Act? Extending protection from seventy-five to ninety-five years for works that do not yet exist removes nothing from the public domain. The story of the movie executive is slightly different. Here the pitch must be, "we won't make money within the first seventy-five years, but just you wait for years seventy-six through ninety-five!" It is possible that this aspect too will not pass the "promoting progress" threshold test. n97 If, however, it does pass the threshold, the prospective applications should not pass the Turner standard. The extension substantially burdens present expressive interests - interests of those people like Eldred who wish to use old materials expressively. The government's legitimate goal of giving sufficient incentives to authors is only furthered very weakly, if at all, and certainly there is no evidence, empirical or theoretical, suggesting that a less restrictive means - such as life of the author plus fifty years, or a term of fifty-six or twenty-eight years, is insufficient.

Considering this background, the D.C. Circuit's decision in Eldred v. Reno represents a lack of understanding of the constitutional dimensions of copyright law and the role of the judiciary in ensuring that the use of information and cultural materials by everyone is regulated no more than necessary. The decision  [*201]  neglects the important role courts have to play as an institutional counterbalance to a systematically imbalanced and overprotective legislative process. The court entirely misses the regulatory aspects of exclusive private rights and the necessity of a significant level of judicial First Amendment scrutiny to protect against over-regulation of access to, and use of, the cultural and information environment by individuals regardless of whether under that standard, it finds the DMCA unconstitutional. Needless to say, how the Supreme Court will resolve Eldred will largely settle the core question of the relationship between the constitution and the line demarcating the boundary between the public and proprietary domains.

B. Neo-Lochnerism and the Moral Inversion of the First Amendment
For over a century, lawyers have been treating corporations as legal "persons." With habit, this useful legal fiction has taken on a natural aura, as though General Motors and Joe's Trucking were just like individuals for all purposes. This fiction was not particularly important from a First Amendment perspective as long as free speech law was largely occupied with working out the relationship of the First Amendment to direct censorship, as it was throughout the first two-thirds of the twentieth century. As the information economy and society have moved to center stage, the First Amendment is increasingly used to impose judicial review on all regulation of this sphere of social and economic life.

Starting with cases like First National Bank of Boston v. Bellotti, n98 which protected a corporation's right to speak on matters of public concern as part of public discourse, and continuing with cases like Central Hudson Gas & Electric Corp. v. Public Service Commission of New York, n99 which began to protect corporate speech aimed for commercial purposes, First Amendment jurisprudence has gradually shifted to treating corporate speakers engaged in any form of information production and exchange as though they were no different than individual speakers engaged in political discourse or personal expression. This is an exaggerated characterization of the law in this area that I will not defend. Rather I will use it to identify a potential source of instability in First Amendment law.

The trend towards judicial review of the regulation of how corporations make money from information production, exchange, and carriage as though the regulation was aimed at individual rights, not economic regulation, has a particularly alarming trajectory in an information economy. The normatization of this position is nowhere more starkly evident than in Chief Justice Rehnquist's joining the majority in the United Foods case. n100 In Central Hudson, it was then Justice Rehnquist who accused the Court of returning to the  [*202]  Lochner v. New York n101 era. n102 In United Foods, the Chief Justice joined the Court in finding unconstitutional a regulation that required mushroom growers to contribute to a general fund to support advertising mushrooms. n103 Justice Breyer responded, "I do not believe the First Amendment seeks to limit the Government's economic regulatory choices in this way - any more than does the Due Process Clause." n104 His invocation of the specter of Lochner in the context of First Amendment law was the first since Justice Rehnquist used it for the same purpose two decades earlier.

The video dial tone cases starkly exemplified the implications of this First Amendment interpretation trend in an economy where information production, dissemination, and carriage are central. n105 In the early 1990s, the FCC sought to allow telephone carriers into the video delivery market, with the caveat that the telephone carriers do so as common carriers. The FCC's video dial tone order required carriers to carry video signals on a nondiscriminatory basis, just as the FCC had previously required for telephone and data signals. The appellate courts treated this imposition as a violation of the telephone companies' editorial rights - the right to determine what information they would carry over their systems. n106 This, in turn, led these courts to scrutinize the economic rationale of the common carriage requirement - which was based on the FCC's experience with the telephone carriers as providers of competitive services dependent on their platform. n107 Rejecting the economic rationale offered by the FCC, the courts overturned the video dial tone order as violating the First Amendment. n108 The Supreme Court eventually vacated the case as moot, because the passage of the Telecommunications Act of 1996 n109 disposed of the video dial tone model. n110 But the problem persists. n111

Perhaps the most extreme version of this use of the First Amendment is Time Warner Entertainment Co. v. FCC. n112 In this case, the D.C. Circuit invalidated the FCC's limits on vertical and horizontal integration of cable carriers. n113 The FCC limits were intended to maintain a sufficient number of outlets to assure that any programmer not owned by a cable operator would have access  [*203]  to forty percent of the viewers in the total U.S. market. n114 The FCC decided that a minimum of four competitors was required to achieve this goal. It reasoned that two competitors could collaborate, and if together they controlled more than sixty percent of the viewers, the entities would be able to prevent programmers unaffiliated with them from surviving. n115 The FCC therefore capped the total number of viewers to whom any single cable operator could offer cable services at thirty percent. n116 To the court, there was no question that this cap on the potential number of subscribers violated the cable company's First Amendment rights. n117 All that was left for the court to do was to supplant its own economic judgment that there was not a high probability of collusion between two giant competitors to shut out unaffiliated programmers for the expert agency it was reviewing. As a result, the D.C. Circuit found the FCC's regulation unconstitutional under the Turner standard, because the regulation was not narrowly tailored. n118

Now, one might think that this represents a general rise in courts' solicitude towards First Amendment claims against government regulation. However, we have already seen that in Eldred v. Reno, n119 which was decided within weeks of Time Warner Entertainment Co. v. FCC, n120 another panel of the same court reached a diametrically opposed result. By holding that copyright was categorically exempt from First Amendment review, the court upheld a regulation sustained by an economic rationale that is laughable under the Time Warner standards. The stark difference between the two cases is instructive: the first reverses a plausible, if unpersuasive to the judges, economic theory regarding concentration and collusion, and the second affirms an impossible economic theory about the effects of retroactive extension of an already ridiculously long copyright protection period.

Clearly, one aspect of this dissonance is the fact that copyright is considered a "private rights" concern, while horizontal ownership rules are considered a "public regulation" concern. Eldred and Time Warner, however, along with other copyright cases, evidence a deeper problem of moral inversion. For example, in the Los Angeles Times v. Free Republic n121 case the court saw the newspapers as the First Amendment rights bearers, while the individual users who would cut and paste stories as a basis for their political commentary and discourse with their fellow forum participants represented a lesser interest. The court did not consider that the clash was between political discourse on one hand and the right to charge for online access to archival materials (already  [*204]  fully remunerated through advertising) on the other hand. The criminalization of noncommercial use and exchange of digitized materials that one sees with the introduction of the NET Act n122 and the DMCA similarly fails to appreciate the chilling effect on real people - like Ed Felten - of a criminal prosecution against Dmitry Sklyarov, where the interest on the other side of the equation is the protection of a business model of a recording company or a movie studio.

This trend in First Amendment law of the information economy - towards protecting corporations broadly, even at the expense of very real and immediate constraints on the expressive autonomy and democratic speech of individuals - is unstable because it represents a moral inversion of the First Amendment. The First Amendment is not a technical rule of law stating that government shall not make any law regulating information flows regardless of their source or nature. Rather, it is a constitutional provision central to the functioning of our democracy and the security of our individual autonomy.

Our democracy does not generally treat corporations as citizens. They do not vote. They cannot be elected. Their contributions to political discourse are valued instrumentally, not intrinsically. Their claims to speech rights derive from the fact that corporations sometimes act as the best vehicles to bring useful information to the attention of a democratic polity. n123 From a First Amendment perspective, that is the corporation's only role. This use may entail quite substantial rights on the part of corporations, in particular news media. In the clash, however, between instrumental reasons to protect corporate speech and instrumental reasons to restrict it, for example, to permit others to enrich democratic discourse over the telephone company's network, the calculus is ultimately instrumental and subject to fairly flexible balancing. This could be a systematic justification for a new balancing approach that Justice Breyer seems to be developing in this area - for which he has received support from Justices O'Connor and Ginsburg in some cases. n124

Furthermore, one cannot in any coherent fashion treat corporations as the bearers of moral claims to autonomy. In some fashion or another, claims to autonomy are based either in some dignitary interest or in some respect for rational beings or will. They are, in any event, entirely unavailable as intrinsic claims for corporations or other corporate bodies. n125 Organizations or corporations can serve or defeat autonomy. Their regulation, similarly, may serve or defeat autonomy. When it does defeat autonomy, such as when a regulation requires a corporation to prevent users from receiving information they wish to  [*205]  use or from disseminating expressions individual users wish to disseminate, the organizations may have an autonomy-based First Amendment claim. This claim, however, is autonomy-based only derivatively, not directly. As with the case of regulation that may instrumentally limit speech in ways that limit or enhance democracy, so too regulation of corporations may do so in ways that limit or enhance the autonomy of individuals.

In any event, when the person regulated is a human being, a citizen of a democracy, a direct claimant of autonomy, that person's claims are translated into an intrinsic First Amendment claim against instrumental regulation. In contrast, corporations' claims under the same constitutional provision are weaker, pitting one instrumental claim against another. Particularly when the government regulates corporations so as to serve the real parties in interest (the constituents and individuals) in an effort to enhance speech and increase diversity in discourse, the First Amendment largely devolves into assessing the instrumental approaches to serve the same intrinsic claims. In some sense, the corporations' First Amendment claims pit a rule-utilitarian instrumental approach (it on the whole serves this desideratum to follow this rule), while the regulation tries to serve the good more directly.

The instability comes from the flow of cases that do exactly the opposite. Exclusive private rights in information, as they are increasingly applied to constrain individual use of information and cultural resources, are a form of regulation of individual expressive freedom. Very often corporations are the beneficiaries of this regulation. These regulations are subject to little or no First Amendment scrutiny. On the other hand, rules that limit business practices or corporate management in an effort to give individuals greater access to information or greater access to avenues of expression are routinely struck down.

The result is a restriction of individual expressive autonomy and democratic discourse by and among the actual constituents of democracy. The First Amendment is interpreted to permit government to restrict individuals' access to and use of information and cultural materials, but prohibits the government from aiding individuals by similarly restricting corporations. The result is a world in which both government and corporations can restrict individuals while public corrective action to increase individual freedom is constrained. This inversion is partly sustained by characterizing copyright-like laws as "private ordering" or "property laws," and media access rules as "public laws" or "regulations." Partly, this approach relies for its stability and coherence on a workmanlike attention to doctrinal moves to the exclusion of principle, but these strategies cannot be stable in the long run. The moral inversion of the First Amendment puts too much pressure on the particular form of First Amendment law that has developed for the information economy.

C. Private Ordering
The fact that the regulations in question are usually conceptualized as property or contract rights, rather than direct command and control regulations is  [*206]  another element that is likely to play an important role in framing the constitutional understanding of exclusive rights in information and culture. As property or contract rights, regulations partake of the aura of "private actions," an area where government action is not generally subject to constitutional review (Shelley v. Kraemer n126 aside). The operation of this construct, in its simplest form, is seen in the D.C. Circuit's statement in Eldred v. Reno, "plaintiffs lack any cognizable First Amendment right to exploit the copyrighted works of others." n127 Here the court conceptualizes the conflict as involving two private parties - one that owns a work, another that wants to use it. The First Amendment has no traction.

The simple fact that the last person triggering a prohibition on a particular use is not a government actor does not, however, render the prohibition immune to First Amendment scrutiny. Much depends, instead, on the source of power that enables the prohibition. Imagine that Congress created and auctioned off rights to prohibit Republicans from appearing on television and sold the rights to private parties. What if courts are then asked to enforce the prohibitions by the individuals? Whether a particular Republican would or would not be allowed on television would not be a government decision, but rather a decision of the private license owner. The source of prohibition, however, would be a law that discriminates against speech based on viewpoint, and thus almost without question would be struck down. Or would it? The baseline rule as to cable access requirements became privatization of the censorship decision. Nonetheless, the privatization of the decision confounds the issues. Denver Area Educational Telecommunications Consortium n128 is in many senses functionally similar. There, the Court in fact upheld a content-based regulation against protected speech, because the law delegated the final censorship decision to cable operators. The statute reviewed in that case n129 related to cable channels on which cable carriers were generally required to carry all programming on a nondiscriminatory basis, much as would a common carrier. The statute challenged in that case excluded from this general nondiscrimination requirement only smut that Congress disapproved of but could not, under the court's First Amendment jurisprudence, directly prohibit. While the plurality did not find private party control dispositive, it did weigh in favor of finding constitutionality. n130 As a practical matter, the Court indeed upheld a content-discriminating regulatory burden of protected speech that it would never have upheld had it not been structured as a privatization of the utility censorship act. Private party control, moreover, was the deciding factor for Justices Thomas, Rehnquist, and Scalia. n131

 [*207]  Clearly, however, cases exist where a private person can rely on, for example, property rights, to prevent someone from speaking, and can do so on a viewpoint-based or any other grounds. Indeed, such cases are regarded as the "normal" case, while situations like the company town in Marsh v. Alabama n132 are the exception. What is the difference between exclusive private rights in information - like copyright - and a private homeowner's request of a guest not to annoy everyone with praise for the Taliban? The point is that understanding copyright within the framework of First Amendment law does not require one to consider whether a copyright owner is more like a company town or a mall. In those cases, the owner of physical property would use a law that was not directed at speech and did not use legal power to exclude the speaker. The speaker, in reliance on the First Amendment, demanded an exception to the general property law. The court either accepted or rejected this claim. The relationship of the First Amendment to copyright is entirely different. Users are not confronted with an inability to speak, print, or play music without access to resources - like streets, parks, or transmitters - governed by a general law like property. They have the means to do so under their control. They are confronted with a legal rule - the law of copyright - that in the pursuit of the public interest prohibits them from speaking, printing, or playing music in ways that they otherwise could. The law of copyright is the only thing that stands between the user and the user's capacity to speak as she pleases. It is a public law, enacted by the legislature, to benefit public interests, that takes the form of telling many people that they are prohibited from printing certain words that they want to print or publicly perform. It may be a perfectly justifiable law, but to say that a person has no cognizable First Amendment interest in not being forbidden by law from printing something he or she wishes to print or say publicly, and can as a practical matter print or say, is implausible.

This answer depends, however, on the theory one has regarding the source of exclusive private rights in information. If one takes some form of natural rights approach - along the Lockean or Hegelian lines - then the source of the property right is pre-legal. In that case the law is simply enforcing pre-political rights, whose contours therefore are not themselves reviewable "government action." Then, the source of the prohibition that the owner places on the user is not itself a reviewable law, though obviously when the Sheriff comes to enforce the prohibition there is a government action, just as in Shelley v. Kraemer. n133

If, however, one sees, as U.S. tradition and Supreme Court opinions repeatedly state, exclusive private rights in information as positive law, rooted in a utilitarian calculus, then the law creating the right is itself the "law" that Congress has made abridging the freedom of speech. The meaning of an exclusive private right in information is that Congress has identified a person who, for reasons of the general welfare, receives a right to prevent others from saying certain words. The person stymied by copyright law is just like the person with  [*208]  the right to prevent Republicans from talking on television. The viewpoint neutrality of copyright goes to the likelihood that the law will be upheld, not to the presence of a "cognizable First Amendment interest." This does not mean that every copyright infringement case now has a case-specific First Amendment review component. It does, however, mean that where the law of exclusive private rights draws the boundaries of the right, these boundaries must be subject to First Amendment scrutiny.

While this issue is important in the copyright context, as Eldred indicates, the primary locus for its likely mischief is in confounding the constitutional analysis of institutional mechanisms explicitly intended to facilitate "private ordering" of access to and use of information and cultural materials, rather than to embed public policy choices. This is particularly true with the Uniform Computer Information Transactions Act ("UCITA"), n134 whose most controversial feature is the validation of mass-market licenses.

Consider a term appended to the news reports of a technology news service: "Information contained in this CNET report may not be republished or redistributed without the prior written authority of CNET, Inc." n135 Under copyright law, the information contained in a report, as distinguished from the expressive form that it takes, is not the property of the reporter. There may be a very limited "hot news" exception to this general rule, n136 but certainly nothing that would encompass the broad claim of right expressed in CNET's terms.

Most courts prior to the passage of UCITA did not enforce such broad terms. n137 In refusing to enforce these provisions, some courts relied on state contract law, by finding an absence of sufficient consent n138 or an unenforceable contract of adhesion. n139 Others courts relied on preemption - stating that to the extent state contract law purported to enforce a contract that prohibited fair use or material in the public domain (like the raw information contained in a report), it was preempted by federal copyright law that chose to leave this material in the public domain. n140 While the Seventh Circuit held otherwise, n141 this was the majority position prior to UCITA. UCITA, however, obviates the state contract law bases for refusing to enforce such contracts.

A state's enactment of UCITA does not trigger direct constitutional review under Article I, although it might under the First Amendment. In Bonito  [*209]  Boats, Inc. v. Thunder Craft Boats, Inc., the Court specifically stated that the limitations imposed on Congress by Article I, Section 8, Clause 8 did not directly apply to the states. n142 The Court, however, also held that states could not create rights that were functionally equivalent to rights created by federal statute - like patent law - but did not serve the core policy goals of the federal law in retaining access to the information to which exclusive rights are granted. The correct analysis of UCITA, insofar as Article I is concerned, is to consider whether its blanket validation of mass-market licenses, including those that prohibit users from making uses preserved for them by the Copyright Act, is preempted by that Act, or conflicts with the requirements of Article I as implemented in the Copyright Act. In the case of UCITA, for example, validation of a shrinkwrap license that prohibits copying of public domain materials covered by the contract would be a simple example of state law enforcement of a term in a mass-market contract that could violate Article I as implemented through the Copyright Act. The Copyright Act implements access to public domain materials in a variety of ways. n143 It does so in many cases as a direct implementation of Article I's mandate not to create rights in public domain materials. When UCITA is enforced to circumvent these privileges, it does so in violation of the constitutionally embedded federal policy just as surely as the boat hull design considered in Bonito Boats violated federal patent policy.

Moreover, enforcement of mass-market shrinkwrap licenses to prohibit users from using information in ways permissible under copyright law is generally a suppression of speech that must be reviewed under Turner. That shrinkwrap license enforcement is based in contract law about speech no more insulates it from First Amendment review than the private law basis of defamation law insulates it from complying with a federally imposed, constitutional baseline. When contract law refuses to enforce contracts that are against public policy, it recognizes the irreducible public role in defining the conditions of enforceability. Just as the exclusive rights themselves are reviewable because they seek to achieve a public purpose by prohibiting certain expressive acts, so too contracts about permitted and prohibited expression call upon the state to prohibit certain speech in pursuit of public policy. The decision whether to enforce contracts for gambling, or prostitution, or assassinations under the law, was a public decision, aimed at public goals, with implications as to the enforcement of a legal form of interpersonal agreements. Similarly, the decision about whether to enforce contracts that determine whether one person will say certain things or read certain things will be enforced under the law is a public decision, implemented through public law. That public decision is subject to review under the First Amendment. Again, not every contract is subject to de novo review. Categories of policy decisions, however - such as whether to enforce mass-market licenses that prohibit uses of information otherwise privileged by copyright law - are subject to de novo review.

 [*210]  The "public" character of the license is even less problematic when the "contract" is not a negotiated agreement between equal parties, but a mass-market license that is in effect a privately selected, but publicly enforced, regulation of how certain information is used by wide ranges of the population. When Congress creates a particular set of rules regulating access to information or cultural materials, it determines that the means it adopted adequately serves the goal of promoting incentives for creativity. Given the political economy of congressional legislation on exclusive private right in information, at any given moment the federally enacted baseline is likely to be more protective than optimal. This baseline, the federal law of copyright, always stands as a less restrictive alternative for attaining that goal than the terms a mass-market vendor will add to that baseline. The result is that a state law or court order that prohibits a person from using information as the user is permitted to under the Copyright Act is an overly restrictive regulation of that use. A less restrictive alternative is readily available in the form of the federal baseline.

D. Regulating the Logical Layer: Code and the Constitution
Imagine a critic of Hollywood culture - say, a feminist film critic or a fundamentalist preacher - preparing a presentation about the ills he or she sees in this culture. The most effective means of explaining and communicating this criticism would be a presentation laced with illustrations from actual films. The Copyright Act generally permits such quotations from video. The DMCA however, has created a framework that operates at the logical, or software layer, and effectively prohibits these quotations.

The DMCA prohibits anyone from circumventing a technical measure that controls access to a work. n144 It also prohibits anyone from making or distributing utilities that would help users circumvent protection measures. n145 Neither provision, at least as currently interpreted, is subject to the fair use exception, and the quotations by the feminist critic or the fundamentalist preacher would not likely fall under any exception to the DMCA. If producers encrypt cultural products, as the film industry has done with DVDs and new videocassettes, it becomes illegal under the DMCA to perform the functions at the logical layer that are necessary to quote from them. n146 A bill is presently being contemplated to force manufacturers of both the physical and logical layers of the information environment - the hard drives, computers, screens and the like, as well as software - similarly to design their wares to enforce the licensing practices of the copyright industry. n147

 [*211]  Some technical protection may be necessary to preserve the viability of a commodified, copyright-based business model in cultural production. A pervasive rebuilding of both the infrastructure's physical and logical layers designed to control the way individuals interact with their cultural environment, however, undermines both individual expressive freedom and the richness of political discourse. In particular, these provisions make cultural resources less available and more expensive for the noncommodified sector - like the feminist film critic or the fundamentalist preacher - thus, threatening to impoverish an increasingly important dimension of social discourse.

Three cases that test the constitutionality of the DMCA's anti-device provision have recently been considered. These are Corely, Felten, and Sklyarov. Corely presents the most substantial First Amendment decision. Sklyarov arises in the context of criminalization. Felten is almost too easy a test, presenting the real-world example of what would otherwise be considered a "parade of horribles"-type argument.

Corely involves a suit by the eight Hollywood studios against a hacker magazine, 2600, published by Eric Corely. The plaintiffs sought an injunction prohibiting 2600 from making available, or linking to other sites that make available, a program called "DeCSS." n148 DeCSS is a computer program that circumvents the copy protection scheme, CSS, used to control access to DVDs. CSS prevents copying or any use of DVDs unauthorized by the vendor. DeCSS was written by a fifteen-year-old Norwegian named Jon Johanson, who claims (though the district court discounted his claim) to have written it as part of an effort to create a DVD player for Linux based machines. n149 2600 posted a copy of DeCSS on its site, together with a story about it. The industry obtained an injunction against 2600 prohibiting not only the posting of DeCSS, but also its links to other sites that post the program. n150 The court rejected the defendant's arguments that sought to interpret the DMCA to include a fair use exception, as well as the defendant's constitutional argument that if the DMCA's anti-device provision prohibited DeCSS, then the DMCA was unconstitutional. n151

The defendant presented the Court of Appeals with both an Article I argument and two types of First Amendment arguments against the DMCA. The Article I argument was that the limitations Article I places on Congress's power to grant exclusive private rights in information require, among other things, that a law not remove materials from the public domain and that it provide protection only for limited times. n152 Encryption, however, allows a vendor to encrypt not only copyrighted materials, but also public domain materials never owned by the copyright owner and materials whose term of protection has expired.  [*212]  Thus, a law that prohibits the existence of circumvention devices, even those usable to reach public domain materials, is a law that excludes these materials from the public domain, or indefinitely extends the term of protection, in violation of the constricts of Article I. Furthermore, it is law, rather than technology or private action that excludes access to the materials. This is because counter-technology (circumvention devices like DeCSS) is perfectly capable of enabling users to use the public domain materials. Therefore, only the law prohibiting these devices causes the exclusion. But, according to Article I, a law giving exclusive rights to control access to information or cultural materials may not include a right to indeterminately exclude others from either public domain or copyrighted material. If the court does not accept this argument, or some similar limitation on what Congress can indirectly achieve through regulating the logical layer of the information environment, then, the argument goes, regulations like the DMCA as a practical matter can obviate the Article I protections of the public domain.

The defendants also advanced two First Amendment arguments - one for strict scrutiny, the other for intermediate, or Turner-level scrutiny. n153 Accepting that the purpose of encouraging commercial proprietary production of information through creating exclusive rights is an important government interest, Turner-level review questions whether the specific law burdens speech too much. The internal limitations in copyright law - like the fair use exception or the privileged use of public domain materials - are constitutive mediating devices permitting copyright law to comply with the constraints of the First Amendment. The anti-device provision of the DMCA in effect eliminates these mediating devices by giving vendors of digitized materials complete, rather than limited control over materials they encrypt. The effective elimination of access to video materials for purposes of quotation, or to materials no longer covered by copyright, or to materials otherwise in the public domain imposes too heavy a burden on the speech of users, particularly in the presence of less restrictive alternatives. Less restrictive alternatives include: imposing liability for infringing uses of circumvention, requiring that access-protection devices permit statutorily defined fair uses, or allowing copy protection mechanisms that permit first-, but not second-generation copying. n154 Given the presence of less restrictive alternatives that Congress has used in similar circumstances, the DMCA imposes too great a burden and should be found to violate the First Amendment.

The district court below held that the Turner standard was in fact applicable but that the harms to fair users were too remote and speculative to justify finding the act unconstitutional on its face. n155 In addition, the court found that the  [*213]  specific defendants were not fair users and could not claim therefore to invalidate the act as applied. n156 The Second Circuit Court of Appeals largely upheld the lower court's decision. n157 It, too, applied Turner scrutiny, and found the DMCA to be sufficiently narrowly tailored given the perceived threat to copyright holders. n158 It did not disagree with the arguments that there were less restrictive means available but held that government need not use the least restrictive means.

The lion's share of the briefs and the opinions, both at the appellate and district court level, were devoted to regulation of code as itself the regulation of speech. Following the district court, the court of appeals accepted that code was a language that computer scientists used to communicate with each other, finding that "programmers communicating ideas to one another almost inevitably communicate in code, much as musicians use notes." n159 The court further added in a footnote, "reinforcing the conclusion that software programs qualify as "speech' for First Amendment purposes - even though they instruct computers - is the accelerated blurring of the line between "source code' and conventional "speech.' There already exist programs capable of translating English descriptions of a program into source code." n160 The court then rejected, however, the appellants' view that this speech nature of code required strict scrutiny of the DMCA.

Unlike a blueprint or a recipe, which cannot yield any functional result without human comprehension of its content, human decision-making and human action, computer code can instantly cause a computer to accomplish tasks and instantly render the results of those tasks available throughout the world via the Internet. The only human action required to achieve these results can be as limited and instantaneous as a single click of a mouse. These realities of what code is and what its normal functions are require a First Amendment analysis that treats code as combining nonspeech and speech elements, i.e. functional and expressive elements. n161
The court then quoted extensively from the lower court's opinion, which described how the functionality of DeCSS was to permit free and easy copying of instantaneous perfect worldwide distribution, concluding that this functionality properly affects the scope of First Amendment protection, "the capacity of a decryption program like DeCSS to accomplish unauthorized - indeed, unlawful - access to materials in which the Plaintiffs have intellectual property rights must inform and limit the scope of it First Amendment protection." n162 This logic eventually led the court to apply intermediate scrutiny, under which it found the DMCA's application to DeCSS unconstitutional.

The fundamental concern of the Second Circuit is, in my opinion, the proper one. Computer programs can indeed perform functions by being applied to  [*214]  machines without human cognitive intervention, in a way that instructions of how to build a bomb published in a magazine cannot. Yet if law had said that musicians may not use notes to exchange ideas about militaristic nationalism because such music incites to violence through some primal mechanism not in cognition, there would be no question that the regulation would be subject to strict scrutiny.

If, instead of copyright and computer code, the law had said one may not speak in Russian about overthrowing the U.S. government or in Arabic about Jihad, the need for strict scrutiny would have been obvious. Nothing in either the change of language or the change of topic should result in a different outcome. It is still the case that there is an identifiable group of people for whom use of a specific language is particularly helpful. It is still the case that there are particular topics the government believes are more dangerously spoken about in some languages than in others. This governmental judgment needs to be subjected to strict scrutiny. Perhaps a court today will decide that under the Brandenburg test, prohibiting the sale of flight manuals in Arabic is justified. Yet, in doing so a court would be required to utilize strict scrutiny, not some lesser standard. The repeated references in the Second Circuit's opinion to the threat DeCSS posed to Hollywood were inappropriate in considering the standard of review. They would only have been appropriate to consider whether the DMCA is valid under the higher standard of review.

From the perspective of First Amendment doctrine, the potential difficulty is that no regulation of computer programming can be undertaken unless it complies with the strict scrutiny standard. That is not, in principle, an implausible result. Journalists generally cannot be regulated in terms of the words they use, except under strict scrutiny. Why couldn't the same be true of computer programmers?

As the Second Circuit noted, the most important potential concern with this outcome is that a communication to a machine intended to cause it to perform a function is not a communication that triggers the First Amendment. No one would suggest that it is unconstitutional to regulate communication between a human and a urinal achieved through an infrared port or communication between a human and a remote-control car. The complicating fact with code is that the same words can be uttered, and some or even many human beings will understand them as meaningful human communications, while some machines will understand them as instructions. The difficulty is how to regulate utterances intended to "push buttons" without burdening human communications. This might mean that executing a program must be distinguished from communicating it in a humanly readable form capable of being either read by a human or run by a machine. It might mean that distribution with the intent of running, as opposed to that of communicating, should be more readily regulable. The obvious difficulty, once one presents the distinction in these terms, is the potential chilling effect on human communication. This is exemplified most clearly by David Touretsky's superb website, devoted to prodding the boundaries of  [*215]  this precise problem. n163 Touretsky presents various forms of describing DeCSS, some more readily machine-readable than others. He presents it, for example, in plain English and in Haiku form, as well as in a computer language for which a compiler has not yet been written. Therefore, a machine could theoretically read it, but the computer translation mechanism does not yet exist.

This kind of richly detailed argument is provocative. It does not, however, necessarily imply that software can never be regulated except under conditions that would permit the regulation of a news report. Journalists cannot have an effect on the world through their writing, except through the acts of human readers absorbing their words and opinions. Regulating the words of journalists is therefore regulating directly the human interaction that lies at the core of both the democracy and the autonomy concerns of the First Amendment. In contrast, computer scientists writing code can have an effect on the world even if no human being ever reads their work. It seems plausible that the state should be able to regulate those aspects of code distribution that are intended to operate without operating on the human cognition.

Touretsky's argument is, therefore, in large measure a well-presented, finely detailed, slippery slope argument. n164 It suggests that some cases will be easier, and others harder. Liability for sending an executable file of a virus, intended to function on the computer of a user without communicating to the user, should be relatively straightforward to understand in First Amendment terms as not implicating expressive values. In contrast, liability for publishing an academic paper that includes instructions for how to attain a certain result in computing should be treated as implicating First Amendment interests of the highest order. The presence of formal representations normally used in the discipline, which are machine-readable and sufficient to be automatically compiled into running code, should not change this characterization. This is why the  [*216]  Felten case is so compelling, and why even though it implicated the same threats as Corely, the court plainly would have been required to find the DMCA unconstitutional. This is probably why the recording industry ran away from the field in it and settled the case. In between there are hard cases that need to be resolved on the basis of the extent to which a regulation burdens speech among computer professionals. Thus, a law that prohibits distribution of source code (as defined functionally in the GNU General Public License ("GPL") to be the form most usable by programmers) n165 should be treated as directly burdening speech. If it treats code about different human actions differently because government treats communication in this form about this subject as more dangerous to its interests, as the DMCA anti-device provision does, it should be seen as content-based speech regulation. This is why the DMCA, which treats uncompiled sources of code and object code alike, should have been treated as content-based speech regulation and struck down. A law that prohibits only the distribution of executable files already compiled for a known set of machines, should be seen as incidentally affecting speech, and subject to O'Brien review. n166 I do not pretend that this cursory discussion resolves this issue. Here I only raise this difficult question, and suggest my own intuitions about this thorny question.

E. Criminalization
The DMCA and regulation of the logical layer is also where the story of Dmitry Sklyarov enters and looms large. n167 Encryption of information and cultural materials is intimately involved in the denial of permission to read, view, or quote the encrypted materials. A legal prohibition on decryption - on taking practical steps that would allow one to read, view, or quote the information or cultural materials regardless of the prohibition - directly implicates First Amendment considerations. To impose criminal sanctions in this delicate area is to force innovators and readers to adopt wide margins around the contours of the law. The high risk of criminal liability and its chilling effect on protected activities requires courts to be especially wary of criminal provisions that burden speech. It is one thing for Alice Randall, Edward Felten or 2 Live Crew to test the bounds of the law when the consequences might be an injunction preventing publication or a damage award. It is quite another to ask them to continue to enrich our speech environment at the risk of spending years in a federal prison.

Beginning with the No Electronic Theft Act ("NET Act") n168 and later incorporated into the DMCA, n169 criminal copyright recently has greatly expanded. Prior to passage of the NET Act, only commercial pirates - those that slavishly made thousands or millions of copies of video or audiocassettes and sold them for profit - would have qualified as criminal violators of copyright. With the passage of the NET Act, criminal liability has been expanded to cover private copying and free sharing of copyrighted materials whose cumulative nominal price, irrespective of actual displaced demand, is quite low. As criminal copyright law is currently written, many individuals involved with the over tens of millions of peer-to-peer file sharing systems are felons. It is one thing when in an effort to conform social norms to their business model the recording industry labels tens of millions of individuals in a society as "pirates." It is quite another when the state brands these individuals as felons and fines or imprisons them.

 [*217]  Jessica Litman has offered the most plausible explanation of the criminalization phenomenon. n170 She argues that as networking makes low-cost production and exchange of information and culture easier, large-scale commercial producers are faced with a new source of competition - volunteers, people who provide information and culture for free. n171 As the universe of people who can threaten the industry has grown to encompass almost all potential customers, the plausibility of using civil actions to force individuals to buy rather than share information decreases. Suing all of one's customers is not a sustainable business model. In an effort to maintain their business model, which relies on control over information goods and their sale as products, the copyright industry has enlisted criminal enforcement. By doing so, commercial producers prevent the emergence of such a system of free exchange. In Sklyarov, this is illustrated by the fact that Adobe instigated the prosecution. Even though Adobe later receded in the face of the outrage of many of its customers, it does not change the fundamental fact that it instigated precisely the chain of events that it wished would occur - criminal enforcement without direct involvement by the vendor.

The changes in law, coupled with the Justice Department's current focus on enforcement of intellectual property rights, n172 have lead to a substantial increase in the likelihood of criminal enforcement. The potential stakes of using information and cultural materials have risen. Now users are required to seek a license and pay for use in many more instances than they would have only a few years ago. n173 It is still, as of this writing, a serious constitutional question whether a law like the DMCA, that prohibits Dmitry Sklyarov from writing code that has substantial noninfringing uses - such as letting users read public domain materials like Alice in Wonderland or quote from e-books without permission - is a valid exercise of Congress's power under Article I or is consistent with the First Amendment. Asking programmers to write the software that will make these privileged uses available, so that legal challenges can determine the constitutionality of that prohibition is difficult when the price of a mistake is imprisonment.

The Supreme Court has expressed particular concern with the chilling effect of criminal, as opposed to civil, enforcement of policy. In Reno v. ACLU, the Court specifically stated that "the severity of criminal sanctions may well cause speakers to remain silent rather than communicate even arguably unlawful words, ideas, and images," n174 distinguishing it from the civil measures used to effectuate similar child-protection goals in Denver Area Educational Telecommunications  [*218]  Consortium, Inc. v. FCC. n175 Similarly, in FCC v. Pacifica Foundation the court's approval of the FCC's administrative action specifically noted as one consideration that the sanction imposed was not criminal. n176

The thoughtful critical essay that David Touretsky has published includes materials that could be interpreted as circumvention devices; providing them could be interpreted as trafficking. n177 The problem with the expansive criminalization of copyright is that Touretsky must be a braver man to bring his criticism to the public than he would need to be if the sole remedy were civil - say an injunction requiring him to remove his web page following time-wasting and costly litigation. Some brave critics will continue. Others will be chilled. Discourse, however, will undoubtedly be impoverished.

Given the tremendous expansion of rights in the past few years, and the serious arguments that these expansions have deleterious implications for free speech and are unjustified as a matter of economic theory, the heightened criminalization of copyright should present an immediate target for judicial review. The Sklyarov case presents an excellent opportunity for the judiciary to exercise its moderating power.

F. Raw Information and Information about Information
In 1991, in Feist Publications, Inc. v. Rural Telephone Service Co., the Supreme Court held that raw facts in a compilation or database were not covered by the Copyright Act, and could not be protected consistent with the constraints imposed by the Exclusive Rights Clause. n178 The Court held that the creative element of the compilation - for example, its organization or selectivity - could be protected under copyright law. n179 The facts compiled, however, could not be protected. n180 Copying data from an existing compilation was therefore not "piracy"; it was not unfair or unjust; it was purposefully privileged in order to advance the goals of exclusive rights - the advancement of progress and creative uses of the data. n181

In the years since the Court decided Feist, there have been repeated efforts by the larger players in the database publishing industry to pass legislation that would, as a practical matter, overturn Feist and create exclusive private rights in the raw data in compilations. n182 Since the Court rooted its Feist decision in a robust interpretation of the Exclusive Rights Clause, efforts to protect database providers eventually settled on an unfair competition law. Based in the Commerce Clause, protections under unfair competition law are free and clear of  [*219]  the inconvenient weight of Feist. In reality, however, the primary law that has repeatedly been introduced walks, talks, and looks like a property right. n183 If some version of this law ultimately passes, it will present an important focal point for defining the constitutional status of raw data, under both the Exclusive Rights Clause and the First Amendment.

Even if the congressional law may be stopped, however, other avenues have recently opened to appropriate raw data. In particular, some litigants have turned to state law remedies to protect their data indirectly, by developing a trespass-to-server form of action. The primary instance of this trend is eBay, Inc. v. Bidder's Edge, Inc., a suit by the leading Internet auction site against an aggregator site. n184 Aggregators collect information about what is being auctioned in multiple locations, and make them available in one place so that a user can simultaneously search eBay, n185 Yahoo, n186 and other auction sites. The eventual bidding itself is done on whatever site the item's owner chooses to make his or her item available, under whatever terms are imposed by that site. The court in eBay held that the automated information collection process - the process of running a computer program that continuously requests information from the server about what is listed on it, called a "spider" or a "bot" - was a trespass to chattels. n187 The injunction led to Bidder's Edge closing its doors before the Ninth Circuit had an opportunity to review the decision. n188

Decisions like eBay create an exclusive private right in information under the common law. While, in principle, the information is still free of property rights, reading it mechanically, which is an absolute necessity given the volume of the information and its storage on magnetic media, can be prohibited as trespass. The practical result would be equivalent to some aspects of a federal exclusive private right in raw data, but without the mitigating attributes of any exceptions that would be directly introduced into legislation. To prevent such an eventuality, if these cases cannot be won on state common law grounds, they must be challenged either on preemption grounds - based on the copyright law - or on First Amendment grounds - using the model of New York Times Co. v. Sullivan. n189 The preemption approach could be similar to the model followed by the Second Circuit in NBA v. Motorola. n190 In Motorola the court restricted state misappropriation claims to narrow bounds delimited by the federal policy embedded in the Copyright Act. n191 Perhaps requiring actual proof that the bots have stopped service, or that they threaten the very existence of  [*220]  the service - a requirement imposed, mutatis mutandis in Motorola - would be sustainable under a First Amendment or preemption analysis.

Beyond raw data and the various ways of controlling it, a central question that will have to be addressed is the status of legal control of information about information - like linking, or other statements people make about the availability and valence of some described information. Linking - the mutual pointing of many documents to each other - is the very core idea of the World Wide Web. In a variety of cases, parties have attempted to use law to control the linking practices of others. These efforts have largely been an effort to retain control over the information about which the challenged link provides information. What is common to these cases is that they aim to create an exclusive private right to give people information about where they can find information. The linking aspect of Reimerdes directly raises this exclusive right argument as an interpretation of the anti-device provision of the DMCA. n192

In Reimerdes, the movie industry sought and received an injunction prohibiting defendants from linking their site to places on the Internet where users could access DeCSS. n193 The district court concluded that the defendants were prohibited from telling others where copies of DeCSS were available. n194 It concluded that providing a link to an online distribution point was the virtual equivalent to giving it on a disk. n195 The court argued that a link is just another way to make the prohibited software available to users, thus violating the prohibition on trafficking in circumvention devices. n196 The district court found this analysis straightforward where the link directly began a downloading process. n197 It was not as clear-cut where a general-purpose publication, like a newspaper, linked to a broad site that discussed many issues, and included a link to a copy of the program. The court found its decision harder in instances where there was some material on the end of a link, but not much other than the prohibited circumvention software. n198 Similarly, the court found it important that the person linking had intended to facilitate circumvention by linking. n199 As a result, the court set out a sliding scale of likelihood of finding liability. The scale depended on the intent of the linking and the extent to which a link was close to, or removed from, being a download button in practice. n200 The Court of Appeals largely accepted this analysis, and, indeed, refused to hold that the test might not be even less constraining.

The difficulty with the district court's approach in Reimerdes is that linking is simply a statement about where information can be found. Prohibiting  [*221]  people from telling other people where information can be found, while leaving them free to use that information as they please is very difficult to square with the First Amendment. A list of links is not fundamentally different from a newspaper listing of all the bookstores in town where pornography can be found. The "intent" factor the district court used to mitigate this effect would have allowed imposition of liability against a publication that intended to help its readers obtain pornography, but would have exempted a conservative newspaper that published the list as part of a campaign to boycott the stores. The general-purpose publication versus dedicated distribution site factor may have meant that if a publication had enough other content it would not be liable, but a person handing out handbills with the same exact list would be liable. This degree of ex post judicial judgment as to intent and purpose of the publication would chill speech. Perhaps, under a clear and present danger analysis, some links can indeed be prohibited. For example, a link that, without warning the person clicking, begins to download a virus without the intervention of human cognition is a fairly obvious example of a link that could be prohibited. Such a link plays a purely functional, not communicative role. If, however, you are linked to a site that has such a link, particularly with some statement - "you can find this virus here" - it is then a matter providing information, not of causing harm.

A more subtle regulation of linking occurs when parties seek either to prohibit others from linking to them or to control how others do so. The quintessential case involved a service that Microsoft offered - n201 This service provided access to, among other things, information on events in various cities. If a user wanted a ticket to an event, the sidewalk site linked that user directly to a page on where the user could buy a ticket. Ticketmaster objected to this practice, preferring instead that link to its home page. n202 Linking to the Ticketmaster home page would ensure user exposure to all of the advertising and services Ticketmaster provides, rather than solely to the specific service the user sought after being referred by This case settled. Another similar case, Ticketmaster Corp. v., Inc., was resolved in an unpublished opinion that focused on other aspects of the case. n203

At stake in these linking cases is who will control the context in which certain information is presented. If deep linking is prohibited, Ticketmaster will control the context. They will control the other movies or events available to be seen, their relative prominence, reviews, and similar information. The right to control linking then becomes a right to shape the meaning and relevance of one's statements for others. Although the choice between Ticketmaster and Microsoft as controllers of the context of information may seem of little normative  [*222]  consequence, it is important to recognize that the right to control linking could easily apply to a local library, or church, or a neighbor.

On the Internet, there are a variety of ways that people can provide information about materials elsewhere on the Web. In doing so, they loosen someone else's control over the information - be it the government, a third party interested in limiting access to the information described, or the person offering the information described. In a series of instances we have seen people attempt to limit the ability of others to loosen their control by providing information about the specified information. These are not cases in which a person without access to information seeks affirmative access. Rather, these are cases where someone seeks the aid of law to control what others say to each other about the controlled information. Understood in these terms, the restrictive implications of these moves for free speech become clear. In turn, the need to subject them to First Amendment scrutiny is obvious.

V Conclusion
Exclusive private rights in information exist in tension with individual freedom to read and express oneself. This tension is mediated by constitutional constraints placed on Congress when it enacts such rights, constitutional constraints that in practice some lower courts have relaxed.

The constraints are justified because overly strong exclusive private rights in information entail substantial costs for democracy and autonomy. The driving mechanism from both a democracy and autonomy perspective is that strong exclusive rights increase the importance of large-scale commercial producers of commodified information. This is at the expense of both nonprofit information production and the emergence of nonproprietary, peer production as core elements of our information production system. For democracy, that means that more of the available information and the channels of communication are funneled through a small number of large commercial media companies. This curtails the opportunities for a more diverse universe of content and loci of discourse. For autonomy, it means that we will have a system with substantially less information of critical and fringe possibilities, and greater opportunities for some players - the owners of media and "content" - to structure the information environment of consumers. It also means that opportunities for enhancing personal autonomy in both the productive and consumption aspects of individuals' lives - opportunities made possible by the emergence of peer production - will be more limited.

The constraints are also explained by the fact that the political economy of legislation in this field has a systematic bias towards ever-stronger exclusive rights. The beneficiaries of the rights see private and present gains from strengthening rights. Those who bear the costs are diffuse. Usually, also, the costs are to be incurred in the future, sometimes by generations not yet born or  [*223]  at least not yet able to foresee the effects on them. This leads to a systematic overstatement of the benefits and understatement of the costs of rights.

At present, a number of pressure points are likely to play a central role in defining the relationship between the Constitution and the institution of exclusive private rights in information. First, the framework I suggest is controversial. Whether it will end up reflecting the law will largely be determined in a number of cases currently in the courts, in particular Eldred.

Second, the way the First Amendment plays in this debate will be heavily affected by how First Amendment law resolves an internal tension in responding to the rise of the information economy. The First Amendment's gradual extension of rights to corporations, and of the status of speech to what are essentially the commercial operations of firms in the information economy, pushes towards a new-Lochnerism for the information economy. The simultaneous failure to protect individuals from commercial overreaching that impoverishes and controls individuals' information environment introduces a moral inversion in First Amendment law. The First Amendment has come to protect as central rights that can only be justified instrumentally. This is often at the expense of individuals who are the real bearers of intrinsic claims to freedom of speech and expression.

Third, this is an area in which the form of many of the regulations has the look and feel of private ordering, a characteristic that tends to confound First Amendment law. Nonetheless, the design of exclusive private rights should be treated as any other law whose operative characteristic is the prohibition of speech.

Fourth, how we resolve the First Amendment status of the regulation of code will be immensely important. It will determine how the logical layer of the information environment is made to comply with, or resist, enclosure of the public domain. The increasing sophistication of computers and the ease of translation from human to machine languages significantly complicates this problem.

Fifth, heightened criminalization raises the stakes of the constitutional debate. It presents one of the most important needs for immediate resolution.

Sixth, a variety of mechanisms being developed to give some people power to control the information that other people give about particular information is an area where the First Amendment has an important role to play. It should stem the expanding range of rights to control information that are intended to sustain the business model of selling information as goods.

What is up for grabs in these debates is the way that information and culture is produced in the pervasively networked society. How information is produced and used, who engages in information production and exchange, with what motives and with what degree of control over what others see and speak in society, will have significant implications for democracy and freedom. The Constitution cannot be silent or neutral in answering these questions. It places  [*224]  its thumb on the scales of freedom on the side of a robust democratic discourse, of diversity of antagonistic voices, and of individual expressive autonomy.

n1. Alice Randall, The Wind Done Gone (2001).

n2. Margaret Mitchell, Gone with the Wind (1939).

n3. This subpart describes the facts of SunTrust Bank v. Houghton Mifflin Co., 268 F.3d 1257 (11th Cir. 2001).

n4. 17 U.S.C. 1201 (1998).

n5. Lewis Carroll, Alice in Wonderland (1865).

n6. Project Gutenberg is an effort involving hundreds of volunteers who find books no longer covered by copyright, scan them, and proofread them so as to make them freely available on the web. See Project Gutenberg at (last visited Sept. 9, 2002).

n7. This subpart describes the facts of United States v. Elcom, Ltd., 203 F. Supp. 2d. 1111 (N.D. Ca. 2002). See also Robert Menta, Read an E-book to Your Child, Go to Jail?, Dec. 26, 2002, at

n8. Felten v. RIAA, No. CV-01-2660 (GEB) (D.N.J. June 26, 2001) available at<uscore>v<uscore>RIAA/.

n9. The past decade has seen a growing literature critical of exclusive private rights along these dimensions. See Lawrence Lessig, Code and Other Laws of Cyberspace (1999); James Boyle, A Politics of Intellectual Property: Environmentalism For the Net?, 47 Duke L.J. 87 (1997); Julie E. Cohen, Examined Lives: Informational Privacy and the Subject as Object, 52 Stan. L. Rev. 1373 (2000); Rosemary J. Coombe, Objects of Property and Subjects of Politics: Intellectual Property Laws and Democratic Dialogue, 69 Tex. L. Rev. 1853 (1991); Niva Elkin-Koren, Cyberlaw and Social Change: A Democratic Approach to Copyright Law in Cyberspace, 14 Cardozo Arts & Ent. L.J. 215 (1996); William W. Fisher, III, Reconstructing the Fair Use Doctrine, 101 Harv. L. Rev. 1661 (1988); Jessica Litman, Copyright Owner's Rights and User's Privileges on the Internet: Reforming Information Law in Copyright's Image, 22 U. Dayton L. Rev. 587 (1997); Michael Madow, Private Ownership of Public Image: Popular Culture and Publicity Rights, 81 Cal. L. Rev. 125 (1993); Eben Moglen, Anarchism Triumphant: Free Software and the Death of Copyright, 4 First Monday 8 (Aug. 2, 1999) available at<uscore>8/moglen/index.html; Pamela Samuelson, Is Information Property? 34 Communications of the ACM 15 (1991); David Lange, Cyberspace and its (Dis)Contents: The Future of an Illusion (published as part of the symposium collection prepared for Yale Law School Symposium on Internet Policy Choices), available at (April 1999).

n10. U.S. Const. art. I, 8, cl. 8. This is also known as "the Patents and Copyrights Clause" or the "Intellectual Property Clause." I think "the Exclusive Rights Clause" is better for two reasons, one trivial and one significant. The trivial point is that neither the historical terms-of-art "patents" and "copyrights" - well known at the time of the Constitution's drafting - nor the anachronistic term "intellectual property" are used in the clause. The clause actually refers to granting "exclusive rights." There are two functional reasons for shifting our usage to the original term from either of the more common terms. First, it is broader than the technical terms, and it implies that the clause is intended to govern all exclusive rights in information, not only those that formally are called patents or copyrights. This is clearly the correct understanding of the scope of the limitations the clause places on Congress when it enacts exclusive rights in information. Second, it is more neutral about the desirability of the rights enacted than the term "intellectual property." As we stand at the beginning of the twenty-first century, "property" has come to be seen almost universally as a beneficial institutional device to manage physical resources. By calling the cluster of regulations of information production and exchange that give private parties exclusive rights to control information "property," we endow it with an inevitability and desirability that are not justified by either economic theory or empirical evidence. It is better to use the descriptive term - exclusive rights - and subject any given set of exclusive rights to neutral inquiry as to whether these particular rights are in fact beneficial.

n11. U.S. Const., amend. I.

n12. Letter from Thomas Jefferson to Isaac McPherson, quoted in Graham v. John Deere Co., 383 U.S. 1, 8 n.2 (1966).

n13. Edward C. Walterscheid, To Promote the Progress of Science and Useful Arts: The Background and Origin of the Intellectual Property Clause of the United States Constitution, 2 J. Intell. Prop. L. 1, 33-36 (1996).

n14. As the Court said in Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340 (1991):

It may seem unfair that much of the fruit of the compiler's labor may be used by others without compensation. As Justice Brennan has correctly observed, however, this is not "some unforeseen byproduct of a statutory scheme." Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S., 539, 589 (Brennan J., dissenting). It is, rather, "the essence of copyright," ibid., and a constitutional requirement. The primary objective of copyright is not to reward the labor of authors, but "to promote the Progress of Science and useful Arts." Art. I, 8, cl. 8... . To this end, copyright assures authors the right to their original expression, but encourages others to build freely upon the ideas and information conveyed by a work.
Id. at 349-50.

n15. See Feist, 499 U.S. at 346-47; Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 146 (1989); Sears, Roebuck & Co. v. Stiffel Co., 376 U.S. 225, 230-31 (1964); Compco Corp. v. Day-Brite Lighting, Inc., 376 U.S. 234, 237-38 (1964); Trade-Mark Cases, 100 U.S. 82 (1879). For discussions of the limitations imposed by Article I, Section 8, Clause 8 of the Constitution, see Julie E. Cohen, Copyright and the Jurisprudence of Self-Help, 13 Berkeley Tech. L.J. 1089, 1130-35 (1998); Theodore H. Davis, Jr., Copying in the Shadow of the Constitution: The Rational Limits of Trade Dress Protection, 80 Minn. L. Rev. 595, 596 (1996); David L. Lange, The Intellectual Property Clause in Contemporary Trademark Law: An Appreciation of Two Recent Essays and Some Thoughts About Why We Ought to Care, 59 Law & Contemp. Probs. 213, 232 (1996); Malla Pollack, The Right to Know?: Delimiting Database Protection at the Juncture of the Commerce Clause, the Intellectual Property Clause and the First Amendment, 17 Cardozo Arts & Ent. L.J. 47, 50-4 (1999); Malla Pollack, Unconstitutional Incontestability?: The Intersection of the Intellectual Property and Commerce Clauses of the Constitution: Beyond a Critique of Shakespeare Co. v. Silstar Corp., 18 Seattle U. L. Rev. 259, 260 (1995).

n16. 100 U.S. 82 (1879).

n17. 383 U.S. 1 (1966).

n18. Id. at 5-6 (emphasis in original).

n19. See Feist, 499 U.S. at 346-47; Bonito Boats, 489 U.S. at 146.

n20. See Yochai Benkler, Constitutional Bounds of Database Protection, 15 Berkley Tech. L.J. 535, 549 (2000). The core of the argument is that what was left of the Sears and Compco cases by Bonito Boats was a commitment to subject all legislation that is functionally equivalent to exclusive rights to the limitations of the Exclusive Rights Clause.

n21. This difference in kind is nowhere more clearly articulated than in the differentiation provided by the Supreme Court between what kind of laws regulate commerce, and what kinds of laws are in effect exclusive private rights. Bonito Boats, 489 U.S. at 157-59.

n22. United States v. Moghadam, 175 F.3d 1269, 1279-80 (11th Cir. 1999) (treating the fixation requirement as not central to the Exclusive Rights Clause, and holding that the anti-bootlegging statute at issue there, which did not comply with the fixation requirement but was enacted under the Commerce Clause, was valid, and implying that were there another, central requirement of the Exclusive Rights Clause, like originality, implicated, the outcome would have been different).

n23. 512 U.S. 622 (1994).

n24. 17 U.S.C. 1201(a).

n25. 203 F. Supp. 2d 1111 (N.D. Ca. 2002).

n26. Felten v. RIAA, No. CV-01-2660 (GEB) (D.N.J. June 26, 2001) available at<uscore>v<uscore>RIAA/.

n27. Universal City Studios v. Reimerdes, 111 F. Supp. 2d 294 (S.D.N.Y. 2000), aff'd sub nom., Universal City Studios, Inc. v. Corely, 273 F.3d 429 (2d Cir. N.Y. 2001).

n28. See Turner, 512 U.S. at 662.

n29. Id. at 664.

n30. Id. at 662 (citations omitted).

n31. Harper & Row, Publishers, Inc., v. Nation Enters., 471 U.S. 539 (1985).

n32. 239 F.3d 372, 375-76 (D.C. Cir. 2001), cert. granted sub nom., Eldred v. Ashcroft, 534 U.S. 1062, and cert. amended, 534 U.S. 1160 (2002).

n33. Compare the holding of another panel of the same court, within weeks of Eldred, in Time Warner Entertainment Co., L.P. v. FCC, 240 F.3d 1126 (D.C. Cir. 2001), of which more will be said below.

n34. Harper & Row, 471 U.S. at 560.

n35. See, e.g., Jarvis v. A&M Records, 827 F. Supp. 282 (D.N.J. 1993).

n36. See, e.g., Ringgold v. Black Entm't TV, 126 F.3d 70 (2d Cir. 1997).

n37. Cohen v. California, 403 U.S. 15 (1971).

n38. See Yochai Benkler, Intellectual Property and the Organization of Information Production, 22 Int'l Rev. L & Ec. 81 (2002) [hereinafter Benkler, Intellectual Property].

n39. See generally Glyn Moody, Rebel Code: Linux and the Open Source Revolution (2001).

n40. For an inspired statement, see Moglen, supra note 9. For a pedestrian economic explanation see Yochai Benkler, Coase's Penguin, or Linux and the Nature of the Firm, 102 Yale L.J. 369 (2002).

n41. For more information about NASA's Clickworkers experiment, see (last visited Oct. 7, 2002).

n42. (last visited Sept. 17, 2002).

n43. (last visited Sept. 17, 2002).

n44. C. Edwin Baker, The Media that Citizens Need, 147 U. Pa. L. Rev. 317 (1998).

n45. This latter is a conception Baker called "complex democracy." Id. at 319. One might think of it as vaguely Habermasian, in the sense that it is discourse-centered but fundamentally liberal - centered on the participating agents - rather than communitarian, or republican, and centered on claims of the collectivity to independent weight in political morality.

n46. See Paul Goldstein, Copyright's Highway: From Gutenberg to the Celestial Jukebox 11 (1994); Neil Weinstock Netanel, Copyright and a Democratic Civil Society, 106 Yale L.J. 283, 288 (1996).

n47. C. Edwin Baker, Giving the Audience What It Wants, 58 Ohio St. L.J. 311 (1997).

n48. Benkler, Intellectual Property, supra note 38.

n49. See Neil Weinstock Netanel, Market Hierarchy and Copyright in our System of Free Expression, 53 Vand. L. Rev. 1879, 1919-20 (2000).

n50. (last visited Sept. 17, 2002).

n51. Cass Sunstein, (2001).

n52. Id. at 55-56.

n53. Id.

n54. Id. at 76-77.

n55. Id. at 36.

n56. See Gerald E. Frug, The City As a Legal Concept, 93 Harv. L. Rev. 1057, 1128 (1980).

n57. ICANN is the non-profit organization that supervises the management of Internet domain names. See (last visited Sept. 17, 2002).

n58. Nicholas Negroponte, Being Digital (1995).

n59. See John Rawls, Political Liberalism 103-04 (1993).

n60. See, e.g., J<um u>rgen Habermas, Between Facts & Norms: Contributions to a Discourse Theory of Law and Democracy (1996); Bruce A. Ackerman, Social Justice in its Liberal State (1980).

n61. Sunstein, supra note 51, at 76-77.

n62. Richard H. Fallon, Jr., Two Senses of Autonomy, 46 Stan. L. Rev. 875, 875 (1994).

n63. Robert Post, Meiklejohn's Mistake: Individual Autonomy and the Reform of Public Discourse, 64 U. Colo. L. Rev. 1109, 1131 (1993).

n64. Id.

n65. Yochai Benkler, Siren Songs and Amish Children: Autonomy, Information, and Law, 76 N.Y.U. L. Rev. 23, 109-113 (2001).

n66. By humanistic I mean to emphasize that it is a value that places the human being at the center of moral gravity, rather than any external supra-individual being, such as God or community.

n67. Nicholas Negraponte divided the world of business into "atoms" such as manufacturing businesses, and "bits," such as online digital-based business. See Nicholas Negraponte, Being Digital 1 (1995) available at (last visited Oct. 8, 2002).

n68. Ronald Coase, The Nature of the Firm, 4 Economica 386 (1937).

n69. Michael Piore & Charles Sable, The Second Industrial Divide (1982).

n70. Roberto Mangaberia Unger, False Necessity: Anti-Necessitarian Social Theory in the Service of Radical Democracy (1987).

n71. See (last visited Sept. 17, 2002).

n72. See (last visited Sept. 17, 2002).

n73. See (last visited Sept. 17, 2002).

n74. See (last visited Sept. 17, 2002).

n75. See Benkler, Intellectual Property, supra note 38.

n76. For the most complete statement of the Hegelian property theory underlying intellectual property, see Justin Hughes, The Philosophy of Intellectual Property, 77 Geo. L.J. 287 (1988).

n77. This description does not undermine the acceptability of my design principles to a formal conception of autonomy. The formal conception will not accept law designed to enhance or facilitate autonomy of persons. Indeed, it does not acknowledge the coherence of such an enterprise. A formal conception, however, has no objections to a legal design principle that constrains law from being shaped so as to limit the information flows to already autonomous individuals in a way that either limits the options they see too much, or that gives the controller of information flows power to manipulate the autonomous subject by selectively revealing and concealing information. A substantive conception of autonomy would treat such laws as "reducing" the autonomy of the agent, while a formal conception might treat them as "violating" the agent's autonomy. The congruence between these conceptions in their treatment of the policy design principles is not analytical, but practical. I am not claiming to have found a philosopher's stone; only the analysis of specific policy proposals can adopt - as a self-constraining device - a focus on finding policies that are "autonomy loving" under either conception. The reason to do so is that those policies are an area of practical agreement among autonomy lovers and can be adopted to enhance autonomy without a need to solve more or less immutable disputes about the nature of human agency.

n78. See Hughes, supra note 76 at 344-45.

n79. See Michael B. Gunlicks, A Balance of Interest: The Concordance of Copyright Law and Moral Rights in the Worldwide Economy, 11 Fordham Intell. Prop. Media & Ent. J. 601, 653-54 (2001) (detailing the concept of moral rights in each of the European nations).

n80. I rely here on Radin's definition. See Margaret Jane Radin, Contested Commodities 118-19 (1990).

n81. Hughes, supra note 76 at 349.

n82. Ronald Dworkin, A Matter of Principle 246 (1985).

n83. Raz defines the wellbeing of an autonomous agent in terms of successful pursuit of an autonomously chosen life. Joseph Raz, The Morality of Freedom 390-95 (1986).

n84. This is seen most expressly in the Open Source Definition, Version 1.9 Section 4, available at<uscore>plain.html; see also GNU GPL 2(a).

n85. SunTrust Bank v. Houghton Mifflin Co., 268 F.3d 1257 (11th Cir. 2001).

n86. See David D. Kirkpatrick, A Writer's Tough Lesson in Birthin' a Parody, N.Y. Times, April 26, 2001, at E1.

n87. Pub. L. No. 105-298 (1998).

n88. 239 F.3d 372 (D.C. Cir. 2001), cert. granted sub nom, Eldred v. Ashcroft, 534 U.S. 1126, and cert. amended, 534 U.S. 1160 (2002).

n89. Id. at 373.

n90. Id.

n91. See id. at 374.

n92. Id. at 380.

n93. Graham v. John Deere Co., 383 U.S. 1, 5-6 (1966).

n94. Eldred, 239 F.3d at 375-76.

n95. Graham, 383 U.S. at 6.

n96. See William M. Landes & Richard Posner, An Economic Analysis of Copyright Law, 18 J. Legal Stud. 325, 326 (1989).

n97. Both Stephen G. Breyer, The Uneasy Case for Copyright: A Study of Copyright in Books, Photocopies, and Computer Programs, 84 Harv. L. Rev. 281, 323-28 (1970), and Landes & Posner, supra note 96, express substantial doubts about the marginal impact of term lengths so distant into the future.

n98. 435 U.S. 765 (1978).

n99. 447 U.S. 557 (1980).

n100. United States Dep't of Agric. v. United Foods, Inc., 533 U.S. 405, 429 (2001).

n101. 198 U.S. 45 (1905)

n102. Central Hudson, 447 U.S. at 589.

n103. United Foods, 533 U.S. at 429.

n104. Id.

n105. Chesapeake & Potomac Tel. Co. v. United States, 42 F.3d 181 (4th Cir. 1994); see also In re Telephone Company-Cable Television Cross-Ownership Rules, Further Notice of Proposed Rulemaking, First Report and Order and Second Further Notice of Inquiry, 7 F.C.C.R. 300, 309 (Nov. 22, 1991).

n106. See Chesapeake & Potomac, 42 F.3d at 203.

n107. See id. at 202-03.

n108. Id. at 185.

n109. Pub. L. No. 104-104 (1996).

n110. United States v. Chesapeake & Potomac Tel. Co., 516 U.S. 415, 416 (1996).

n111. See Comcast Cablevision, Inc. v. Broward County, 124 F. Supp. 2d 685 (S.D. Fla. 2000).

n112. 240 F.3d 1126 (D.C. Cir. 2001).

n113. Id. at 1144.

n114. Id. at 1137.

n115. Id. at 1132-33.

n116. Id.

n117. Id. at 1129.

n118. Id. at 1136.

n119. 239 F.3d. 372 (D.C. Cir.) en banc denied sub nom. Eldred v. Ashcroft, 255 F.3d 849 (D.C. Cir. 2001), cert. granted, 534 U.S. 1126, and cert. amended, 534 U.S. 1160 (2002).

n120. 240 F.3d. 1126 (D.C. Cir. 2001).

n121. No. 98-7840, 2000 WL 1863566 (C.D.Ca. Nov. 16, 2000).

n122. Pub. L. No. 105-147 (1997).

n123. C. Edwin Baker, Human Liberty and Freedom of Speech 220-21 (1989). This is the crux of Baker's focus on treating the rights of corporations as always based in what he sees as the more instrumental press clause, and not in the speech clause that he largely reserves for protection of liberty interests in expressive autonomy.

n124. See U.S. Dep't of Agric. v. United Foods, Inc., 533 U.S. 405, 425-28 (2001); Nixon v. Shrink Missouri Gov't Pac, 528 U.S. 377, 400-03 (2000); Turner Broad. Sys., Inc. v. FCC, 520 U.S. 180, 225-28 (1997).

n125. See Meir Dan-Cohen, Rights, Persons, and Organizations: A Legal Theory for Bureacratic Society 55-119 (1986).

n126. 334 U.S. 1 (1948).

n127. 239 F.3d 372, 376 (D.C. Cir. 2001).

n128. Denver Area Educ. Telecomm. Consortium v. FCC, 518 U.S. 727 (1996).

n129. Id. at 735.

n130. See id. at 768.

n131. Id. at 812 (Thomas, J., dissenting).

n132. 326 U.S. 501 (1946).

n133. 334 U.S. 1 (1948).

n134. Unif. Computer Info. Transactions Act (UCITA) (2001), available at

n135. See (last visited Sept. 17, 2002).

n136. See Int'l News Serv. v. Associated Press, 248 U.S. 215 (1918); National Basketball Ass'n v. Motorola, 105 F.3d 841, 853 (2d Cir. 1997).

n137. Mark A. Lemley, Intellectual Property and Shrinkwrap Licenses, 68 S. Cal. L. Rev. 1239, 1248-53 (1995).

n138. Step-Saver Data Sys., Inc. v. Wyse Tech., 939 F.2d 91, 99-100 (3d Cir. 1991).

n139. Vault Corp. v. Quaid Software Ltd., 655 F. Supp. 750, 761 (E.D. La. 1987), aff'd, 847 F.2d 255 (5th Cir. 1988).

n140. Vault Corp. v. Quaid Software Ltd., 847 F.2d 255, 270 (5th Cir. 1988).

n141. See Hill v. Gateway 2000, Inc., 105 F.3d 1147 (7th Cir. 1997); ProCD, Inc. v. Zeidenberg, 86 F.3d 1447 (7th Cir. 1996).

n142. 489 U.S. 141, 165 (1989).

n143. Jessica Litman, The Public Domain, 39 Emory L.J. 965, 992-94 (1990).

n144. 17 U.S.C. 1201(a)(1).

n145. 17 U.S.C. 1201(a)(2).

n146. Universal City Studios, Inc. v. Reimerdes, 111 F. Supp. 2d 294, 322 (S.D.N.Y. 2000), aff'd sub nom., Universal City Studios, Inc. v. Corely, 273 F.3d 429 (2d Cir. N.Y. 2001) (note that this case is referred to as Reimerdes at the district court level and Corely on appeal).

n147. Security Systems Standards and Certification Act [Draft], available at htm (Aug. 6, 2001).

n148. Reimerdes, 111 F. Supp. 2d at 308-09.

n149. Id. at 311.

n150. Id. at 312.

n151. Id. at 332-33.

n152. Julie Cohen, Brief of Amici IP Law Professors, at<uscore>DVD<uscore>cases/20010126<uscore>ny<uscore>lawprofs<uscore>amicus.html (Jan. 26, 2001).

n153. Reimerdes, 111 F. Supp. 2d. at 327-28.

n154. See ACLU et al., Brief of Amici Curiae, at (Jan. 26, 2001); Yochai Benkler & Lawrence Lessig, Brief of Amici, at<uscore>DVD<uscore>cases/20010126<uscore>ny<uscore>2profs<uscore>amicus.html (Jan. 26, 2001); EFF Supplemental Letter Brief, at<uscore>DVD<uscore>cases/20010530<uscore>ny<uscore>eff<uscore>supl<uscore>brief.html (May 30, 2001).

n155. See Reimerdes, 111 F. Supp. 2d at 327-37.

n156. See id. at 323.

n157. Corely, 273 F.3d at 429.

n158. Id. at 454-55.

n159. Id. at 448.

n160. Id. at n.23.

n161. Id.

n162. Id.

n163. David Touretsky, Gallery of CSS Descramblers, at<diff>dst/DeCSS/Gallery/index.html (last modified Aug. 29, 2002). In principle, the logic of linking liability in the DeCSS case could treat this footnote as an act of trafficking (though this is unlikely).

n164. Id.

n165. GNU GPL 3, available at (last visited Oct. 25, 2002). The source code for a work means the preferred form of the work for making modifications to it.

n166. See United States v. O'Brien, 391 U.S. 367, 376-77 (1968).

n167. See supra, Part I.B.

n168. Pub. L. No. 105-147 (1997).

n169. 17 U.S.C. 204.

n170. See Jessica Litman, Electronic Commerce and Free Speech, 1 J. Eth. & Info. Tech. 213 (1999), available at

n171. Id.

n172. See DOJ Intellectual Property Policy and Programs, at (last modified July 20, 2001).

n173. See supra Part IV.C.

n174. 521 U.S. 844, 872 (1997).

n175. 518 U.S. 727 (1996).

n176. 438 U.S. 726, 747 (1978).

n177. Touretsky, supra note 163.

n178. 499 U.S. at 349-50.

n179. Id. at 350-51; 17 U.S.C. 103.

n180. Feist, 499 U.S. at 351.

n181. Id. at 349-50.

n182. The Collections of Information Anti-Piracy Act (CIAA) H.R. 354, 106th Cong (1999).

n183. Id.

n184. 100 F. Supp. 2d 1058 (N.D. Cal. 2000).

n185. (last visited Sept. 17, 2002).

n186. (last visited Sept. 17, 2002).

n187. Ebay, 100 F. Supp. 2d at 1069.

n188. Peg Brickley, Now-Defunct Bidder's Edge Settles Online Dispute, Corporate Legal Times, July 2001, at 70.

n189. 376 U.S. 254, 266 (1964).

n190. 105 F.3d 841 (2d Cir. 1997).

n191. Id. at 850.

n192. See Universal Studios v. Reimerdes, 111 F. Supp. 2d 294, 339-41(S.D.N.Y. 2000)

n193. Id. at 341.

n194. Id.

n195. Id. at 324.

n196. Id.

n197. Id. at 325.

n198. Id.

n199. Id. at 324-25.

n200. Id. at 340-41.

n201. See Michael Newman, Ticketmaster vs. Microsoft is Bully Battle, Pittsburgh Post-Gazette, May 18, 1997, at C-3.

n202. (last visited Sept. 17, 2002).

n203. No. CV99-7654-HLH (BQRx) 2000 U.S. Dist. LEXIS 12987 (C.D. Cal. Aug. 10, 2000).

Prepared: July 3, 2003 - 5:02:29 PM
Edited and Updated, July 4, 2003

Back to
Ammad Bahalim
Free Software Page