A. Access for expressive activity

The first time the Court addressed the question of the First Amendment rights of those who wish to engage in expressive activity against the wishes of a private property owner was in the 1946 case of Marsh v. Alabama.⁵⁸

Marsh suggested that the Court was inclined to be receptive to the First Amendment claim, for it reversed the trespass conviction of a Jehovah's Witness who had attempted to distribute religious literature on the streets of a privately owned company town. Defending the conviction, the state argued that the owner corporation's “right to control the inhabitants of Chickasaw is coextensive with the right of a homeowner to regulate the conduct of his guests.”⁵⁹ Though it implicitly acknowledged that a guest in a private home would have no First Amendment right to stay and speak if the owner wished to exclude him, the Court nonetheless held that

In other words, in Marsh the Court attributed the private owner's decision to exclude to the state, and held that the state was not free to make such a decision consistent with the First Amendment.

Marsh has never been overruled, but it has had virtually no impact as a First Amendment precedent. During the Civil Rights movement of the 1960s, for example, the Court was frequently asked—and just as frequently refused—to hold that the First Amendment protects protestors and demonstrators on private property that the owner has opened to the public for purposes of commerce.⁶¹

Marsh seemed to have suggested an affirmative answer to the question, but the Court's refusal to rely on it in the sit-in cases—indeed, its positive disinclination even to confront the issue that the cases presented—strongly implies that the Court intended to confine Marsh strictly to its own particular facts. The inference that this continues to be the Court's intention is supported by the fact that the Court first backtracked from and then explicitly overruled the only case in which it had relied on Marsh, namely its 1968 holding in Amalgamated Food Employees Union Local 590 v. Logan Valley Plaza.⁶² The Court in Logan Valley held that, in a shopping center that was “the functional equivalent of the business district of Chickasaw involved in Marsh,” the state could not enforce its trespass laws to support the decision of the owner to exclude “members of the public wishing to exercise their First Amendment rights.”⁶³ However, just four years later, in Lloyd Corporation v. Tanner⁶⁴ the Court sustained a private shopping center's ban on the distribution of handbills. The center had invoked its ban to exclude antiwar leafleteers. Though purporting to distinguish Logan Valley on the ground that it involved First Amendment activity that was related to the shopping center's operations, Justice Lewis Powell's opinion in Lloyd expressed a view quite at odds both with Logan Valley and with the impulses that had led the Court to decide as it had in Marsh:

And four years after Lloyd, in Hudgens v. NLRB (1976),⁶⁶

the Court expressly overruled Logan Valley (though it left Marsh at least formally intact). As had Logan Valley, Hudgens involved labor picketers. But, wrote Justice Potter Stewart unequivocally, “the constitutional guarantee of free expression has no part to play in a case such as this.”⁶⁷ He noted, “if the respondents in the Lloyd case did not have a First Amendment right to enter that shopping center to distribute handbills concerning Vietnam, then the pickets in the present case did not have a First Amendment right to enter this shopping center for the purpose of advertising their strike.”⁶⁸

The upshot of these cases, and in particular of Hudgens since it is both the most recent and the most unequivocal, is that the enforcement of trespass laws at the behest of private property owners who wish to exclude persons who want to engage in expressive activity on the owner's property does not violate the First Amendment. Even if the access seekers could not constitutionally be excluded from otherwise similar but publicly owned property, state law may permit a private property owner to enlist the aid of a court to exclude them.

Notice what this means in institutional terms. It means that the First Amendment permits the decision whether to exclude or not to rest with the property owner. It means that courts may enforce that decision without regard either to the message the access seekers wish to convey or to the property owner's reasons for wishing to exclude. It means, in other words, that insofar as the content of the speech is concerned—its ideas or the message it conveys—the courts (i.e., the government) will not take sides.

In several cases the Court has analyzed claims of compelled access to private property for speech by asking whether being required to grant access violated the property owner's First Amendment rights as opposed to the owner's property rights. In most of these cases the access seekers did not claim that the First Amendment of its own force guaranteed them a right of access. Instead, they invoked one or another statutory or regulatory scheme that purported to grant them access to the property of others. The results of these cases are somewhat evenly divided between those sustaining the compelled access claim and those sustaining the First Amendment challenge to it. However, the facts, as well as the rhetoric and tone, of two of the most significant opinions sustaining compelled access suggest that the Court regarded them as exceptional cases that required departures from the mainstream of First Amendment doctrine. The 1969 case of Red Lion Broadcasting Co. v. FCC,⁶⁹

for example, sustained the FCC's ‘fairness doctrine’, which required licensed broadcast stations to present discussions of both sides of public issues and to provide free reply time to those seeking to respond to personal attacks and political editorials. The holding followed from the Court's belief that “differences in the characteristics of new media justify differences in the First Amendment standards applied to them.”⁷⁰ In particular, the opinion stressed the “scarcity of broadcast frequencies, the Government's role in allocating those frequencies, and the legitimate claims of those unable without government assistance to acquire access to those frequencies for expression of their views.”⁷¹ Commentators have long challenged both the conceptual and the empirical validity of the ‘spectrum scarcity’ rationale that provided the underpinning for Red Lion;⁷² and the Supreme Court, acknowledging the criticism, has hinted that it might be prepared to reconsider the rationale if it were directly challenged in a relevant case.⁷³ Perhaps more interesting still is that the FCC itself repealed the fairness doctrine in 1987 because it found that compelled access in practice worked not to enhance diversity but to inhibit the presentation of controversial issues of public importance.⁷⁴

The other case in which the Court departed from the mainstream of First Amendment doctrine to sustain a compelled access claim was the 1980 case of PruneYard Shopping Center v. Robins,⁷⁵

in which high school students sought access to a private shopping center for the purpose of protesting a UN resolution against Zionism. The California Supreme Court interpreted the California state constitution to guarantee access for speech to privately owned shopping centers. The U.S. Supreme Court concluded that “state constitutional provisions, which permit individuals to exercise free speech and expression rights on the property of a privately owned shopping center to which the public is invited” do not violate the center owner's “First Amendment right not to be forced by the state to use his property as a forum for the speech of others.”⁷⁶ As the Court described its view, three key facts saved the access requirement from constitutional invalidity: because of the public availability of the center the views expressed would not be identified with the owner; the state did not dictate a specific message; and the owner could expressly disavow any connection with the petitioners.

The only case in which the Court has sustained compelled access using its customary First Amendment analytical tools is Turner Broadcasting System v. FCC (1994) (Turner I).⁷⁷

Turner I involved the “must-carry” provisions of the Cable TV Consumer Protection and Competition Act of 1992, which require cable operators to carry the signals of a certain number of local broadcast stations. Cable operators challenged the provisions on First Amendment grounds, claiming that they were content-based regulations that could not survive strict scrutiny. Justice Anthony Kennedy's opinion for the Court acknowledged that at the First Amendment's core “lies the principle that each person should decide for him or herself the ideas and beliefs deserving of expression, consideration, and adherence.”⁷⁸ In addition, he conceded that “[g]overnment action … that requires the utterance of a particular message favored by the [g]overnment, contravenes this essential right.”⁷⁹ Nevertheless, the Court concluded that the must-carry provisions were “content-neutral,” not “content-based,” and thus they were subject to only intermediate scrutiny.⁸⁰ On rehearing the case in 1997, the Court ultimately concluded that the provisions passed constitutional muster because they furthered important governmental interests in competition and diversity.⁸¹

Two cases squarely rejected the government's effort to compel access for speech purposes. In Miami Herald Publishing Co. v. Tornillo,⁸²

decided in 1974, the Court invalidated Florida's right of reply law, which required newspapers to give political candidates a right to equal space to reply to editorial criticism. The opinion vividly illustrates the difference between the First Amendment rights of broadcasters, who lost their challenge to right of reply requirements in Red Lion, and print publishers, who resoundingly won theirs. It also illustrates how the First Amendment, like private property rights, functions to allocate to private property owners and not to government officials decision-making authority about what resources shall be used to communicate what messages. Whereas the upshot of Red Lion, PruneYard, and the Turner rulings was to lodge decision-making authority in government actors—courts, legislatures, the FCC—the upshot of Tornillo was to lodge it in private owners and editors.

The 1986 case of Pacific Gas & Electric Co. v. Public Utilities Commission of California⁸³

is to the same effect, though the context was different. The California Public Utilities Commission adopted a requirement that PG&E allow a private advocacy group to use the “extra space”⁸⁴ in its billing envelopes four times a year. Relying heavily on Tornillo, and distinguishing PruneYard, the Court sustained PG&E's challenge to the requirement on the ground that being compelled to include in its billing envelope speech of a third party with which it disagreed violated its First Amendment rights. Again, the institutional upshot of the holding is that decision-making authority regarding access to private property for speech purposes resides in the hands of private owners.

B. Access for communication with the property owner

The Supreme Court has been steadfast if occasionally somewhat oblique in recognizing property owners' right to decline to grant access to those who wish to communicate with them. The flagship case, decided in 1970, is Rowan v. United States Post Office Dept.,⁸⁵

in which the Court upheld a federal statute that permitted a person who “in his sole discretion” has determined that certain advertisements are “erotically arousing or sexually provocative” to insulate himself from such advertisements by requesting the Postmaster General to order the sender to refrain from further mailings. The Court interpreted the relevant statutory provision to remove “'the right of the Government to involve itself in any determination of the content and nature of these objectionable materials.' ”⁸⁶ It then addressed the plaintiff mail order businesses' argument that the statute violated their constitutional right to communicate. The plaintiffs' brief had broadly asserted that “'[t]he freedom to communicate orally and by written word and, indeed, in every manner whatsoever is imperative to a free and sane society.”⁸⁷ The Court conceded that “[t]o make the householder the exclusive and final judge of what will cross his threshold undoubtedly has the effect of impeding the flow of ideas, information, and argument that, ideally, he should receive and consider.”⁸⁸ It concluded, nevertheless, that “[n]othing in the Constitution compels us to listen to or view any unwanted communication, whatever its merit.”⁸⁹

Rowan relied on Martin v. City of Struthers,⁹⁰

a 1943 case that invalidated an ordinance prohibiting door-to-door solicitation, drawing a distinction between the city government's right to protect homeowners from unwanted solicitations and the enforceable right of an individual homeowner to protect himself. This distinction explains the results in a number of cases in which legislatively enacted bans on or permit requirements for certain types of communication have been invalidated even though a majority of the jurisdiction's homeowners probably would not have wanted to receive or be bothered by the material.⁹¹ The distinction warrants emphasis. It helps to bring the “who decides” aspect of the First Amendment access issue into focus. In Rowan the Court sustained the statute because it put the decision whether to reject mail in the hands of the individual citizen. The Postmaster General's job was merely to enforce the private decision. In Martin, by contrast, the town did more than simply enforce individual owners' decisions to exclude. The ordinance that town officials passed attempted “to make this decision for all [the town's] inhabitants,”⁹² but the Constitution does not generally allocate decision-making authority regarding access for purposes of communication with private property owners to local legislators or other government officials. Instead it allocates it to the private property owner.

It remains to justify these outcomes, a task made more difficult by the fact that the Court itself has done little to explain them. In addition, at first glance the decisions seem to run counter to the frequently invoked rhetorical commitment to the “principle that debate on public issues should be uninhibited, robust and wide-open.”⁹³

Those who seek access to any place, public or privately owned, for the purpose of engaging in First Amendment activity invoke this principle as a sword to compel the property's unwilling proprietor either to let them in or to require him to satisfy a court that his reasons for not doing so have merit. The access seekers in effect make the claim that mandating property owners to grant them access will result in more speech, and more speech is presumptively better. They assert that compelled access achieves the purpose of the Free Speech Clause, which they imply is to assure the widest possible debate about matters of concern to the community.⁹⁴ But this view, as Charles Fried has pointed out, “mistakes an effect of the principle for the principle itself. The First Amendment protects a liberty—liberty of expression—and it is an effect of this liberty that there is wide and uninhibited discussion of political matters.”⁹⁵ In other words, the fact that wide and uninhibited discussion occurs when private citizens have liberty to speak should not be taken to imply that the Constitution directly guarantees that the widest possible debate must occur. It would be impossible for the Court to administer such a guarantee, if for no other reason than that the Court would have to make factual determinations in particular cases that it would be impossible for it honestly or confidently to make. The Court could not possibly hope to discern how much access for what kinds of speech activity on what kinds of property would provide the “widest possible debate,” and it is certain that the Constitution contains no criteria in terms of which the Court could evaluate when granting a particular claim of access would be required in order for there to be “enough” speech.⁹⁶

There are at least two additional explanations for the Court's refusal to grant First Amendment rights of access to private property in addition to the fact that the Constitution does not actually purport to assure that debate will be uninhibited, robust and wide open. One is that the refusal to grant access rights is grounded in the Court's understanding that the Constitution itself has very little—indeed, almost nothing—to say about either the substance of laws that resolve disputes between private individuals or about the reasons why private actors invoke the aid of courts. The Constitution takes as given that states supply certain generally applicable background rules—the rules of property, contract, and tort, for example—for the resolution of private disputes. The extent to which legislatures or courts are constitutionally free to change those rules, and, in particular, to modify them so as to transfer decision-making authority from private to public actors, is the subject of intense debate both on the Court⁹⁷

and off.⁹⁸ With regard to access, for example, the Supreme Court has squarely held that legislatures may not authorize an unwanted “permanent physical occupation” of private property without paying just compensation.⁹⁹ On other occasions, the Court has sustained collective modifications of the property owner's right to exclude in certain circumstances: the public accommodations title of the Civil Rights Act of 1964,¹⁰⁰ for one example, and PruneYard for another. And, though the decision has not gone unchallenged by commentators,¹⁰¹ the New Jersey Supreme Court has interpreted its state common law and the “public trust doctrine” to grant to the general public a right to use privately owned land for recreational purposes.¹⁰² But so long as the background rule is that the property owner has the right to exclude, and the state merely enforces the property owner's decision to exercise that right, the Constitution is indifferent either to the reason access is desired or to the reason it is denied.

A second justification for the refusal to read the First Amendment as a guarantee of access to private property is that enforcing the right to exclude does not entail any risk of deliberate government discrimination among viewpoints. That trespass laws may, in particular cases and from time to time, fall more heavily on particular viewpoints does not signal deliberate viewpoint discrimination. Trespass laws are neutral as to the points of view of both the access seeker and the property owner. Enforcing trespass laws does not put judges in a position to censor or punish particular speakers, or to manipulate public debate by selectively denying speech opportunities to disfavored views or granting them to favored ones. As the Court itself noted in Rowan, vesting decision-making authority in individuals with regard to what communications come onto their private property is a way to avoid “vesting the power to make any discretionary evaluation of the material in a government official,”¹⁰³

including in a judge.

A. Copyright and the First Amendment

This subsection considers whether, as several academic commentators have recently argued, copyright law in principle is in significant tension with the First Amendment. For example, Mark Lemley and Eugene Volokh regard copyright as facially problematic because “[c]opyright law restricts speech: it restricts you from writing, painting, publicly performing or otherwise communicating what you please. If your speech copies ours, and if the copying uses our ‘expression,’ not merely our ideas or facts that we have uncovered, your speech can be enjoined and punished ….”¹⁰⁴

For another example, Jed Rubenfeld claims that copyright ignores three basic principles of free speech jurisprudence, namely, the principles that content-based restrictions of speech command strict scrutiny, that prior restraints on speech (which are routinely granted in copyright infringement cases) are deeply disfavored, and that speech restrictions that are based on the speaker's viewpoint are practically unconstitutional per se.¹⁰⁵ For a third example, Yochai Benkler maintains that, because they prevent some people from using or communicating information, “all property rights in information conflict with the ‘make no law’ injunction of the First Amendment.”¹⁰⁶

The Supreme Court has directly confronted the question of copyright's compatibility with the First Amendment on two occasions in the last two decades. In neither case was the Court receptive to the idea that copyright infringers have serious claims to First Amendment sympathy. In both cases, the Court focused almost exclusively on copyright's incentive effects and seemingly regarded the restrictions it places on the speech of infringers as for all practical purposes irrelevant to the First Amendment analysis. In 1985, in Harper & Row Publishers v. Nation Enterprises,¹⁰⁷

the magazine The Nation sought to excuse on First Amendment grounds its use of verbatim excerpts from President Gerald Ford's copyrighted but not-yet-published memoirs. The Nation argued that the ordinary rule, which is to the effect that “the author's right to control the first public appearance of his undisseminated expression will outweigh a claim of fair use,”¹⁰⁸ should be relaxed in view of “the substantial public import of the subject matter of the Ford memoirs.”¹⁰⁹ Deploying what Melville B. Nimmer had described in an influential law review article fifteen years earlier¹¹⁰ as a “definitional balancing” methodology, the Court firmly rejected the magazine's claim. The methodology relied for the accommodation of First Amendment values on doctrines internal to the Copyright Act, namely, the “Act's distinction between copyrightable expression and uncopyrightable facts and ideas, and the latitude for scholarship and comment traditionally afforded by fair use.”¹¹¹ More importantly, the Court did not regard the Copyright Act as in any way fundamentally inconsistent with the First Amendment. To the contrary, the Court said, by “supplying the economic incentive to create and disseminate” works, “the Framers intended copyright itself to be the engine of free expression.”¹¹²

More recently, in the 2003 decision of Eldred v. Ashcroft,¹¹³

the Court rejected a facial First Amendment challenge to the Copyright Term Extension Act (CTEA).¹¹⁴ Despite the recent outpouring of academic commentary and several amicus briefs urging the Court to reconsider, the Court showed no hesitation in continuing to embrace Harper & Row's analytical underpinnings. “Copyright's purpose is to promote the creation and publication of free expression,”¹¹⁵ the Court said. The Eldred challengers had argued that, because it “bars unauthorized use or dissemination of copyrighted works, copyright law restricts speech.”¹¹⁶ They claimed that congressionally enacted copyright law “must be justified under intermediate review.”¹¹⁷ The Court disagreed, affirming that “uncommonly strict scrutiny” is not necessary for “a copyright scheme that incorporates its own speech-protective purposes and safeguards.”¹¹⁸ Distinguishing CTEA from the Telecommunications Act's must-carry provisions, to which the challengers had analogized it and to which the Court did give intermediate scrutiny in Turner I (1994),¹¹⁹ Justice Ruth Bader Ginsburg's opinion concluded that restricting the speech of copyright infringers is generally not to be regarded as an “abridgement” for First Amendment purposes:

Eldred is striking in several respects. Most significant was the Court's affirmation of Harper & Row's focus on copyright's incentive effects on producers of original works of authorship in the face of the increasing scholarly attention to the fact that the law restricts the speech of infringers. The Court was brusque in dismissing the CTEA challengers' First Amendment arguments, and it paid the academic literature virtually no heed whatsoever. While its decision to regard copyright as generally not problematic for First Amendment purposes, and its determination to focus almost exclusively on copyright's incentive effects is clear, its failure seriously to engage the arguments that emphasize copyright's speech restrictive effects is unfortunate.

From the point of view of preserving its institutional capital, the Court's indifference to the commentators' arguments is understandable. Had the arguments been found persuasive, the Court would have had not only to redraw a long-standing First Amendment boundary, but also to adumbrate its parameters in completely uncharted territory. The latter task might well have seemed particularly daunting. It is difficult to envision how heightened First Amendment scrutiny in copyright cases would or should or could actually work in practice, and equally hard to imagine how it would be possible to keep the doctrine from becoming just one more multifactored but essentially rootless, unprincipled pocket of First Amendment jurisprudence.¹²¹

Still, the arguments in favor of heightened scrutiny have sufficient prima facie appeal to merit a more considered rebuttal than the cold shoulder they received from the Court. In what follows, I shall try to put the case that the result in Eldred stands on firmer ground than the Court's rather offhand analysis might imply.

The first strand in the argument points to the awkward fact that copyright and the First Amendment have coexisted practically since the United States' founding,¹²²

and until fairly recently the alleged First Amendment anomaly of copyright law struck few people as problematic. Before the recent avalanche of scholarly interest, there was little important scholarly literature that even raised the issue. The literature that did exist neither discerned profound conflict between copyright and First Amendment values nor regarded the speech of copyright infringers as having significant First Amendment worth.¹²³ Ironically, the conclusion of one of the few early commentators, Melville B. Nimmer, that copyright law itself contained the appropriate “definitional balance” between copyright owners and First Amendment values, provided the touchstone for the Court's analysis in Harper & Row and was reaffirmed in Eldred, though without reference to the fact that Nimmer himself had thought that the “reasons … justifying first amendment subordination to copyright [do not] justify [an] extension of an existing copyright term.”¹²⁴ And there exist few cases in which First Amendment claims have been seriously considered.¹²⁵

The historical pedigree and stubborn persistence of copyright's immunity to a successful First Amendment assault seems something of a puzzle today. Many scholars argue that the immunity is anomalous. They do not seem to think there exists a satisfactory explanation for it, and they certainly do not regard its longevity as posing a legitimate impediment to change.

Some commentators contend that developments in copyright law and the profound technological innovations of recent years have brought about a copyright regime that differs in kind and not just in degree of protection from that which obtained earlier. At the time of the founding, only literal copying was prohibited and printing and copying were in their infancy. During the mid-twentieth century, the decades in which First Amendment doctrine as we know it today was being forged, technological change had not yet become transformative.¹²⁶

The claim appears to be that copyright has become a fundamentally different legal animal, with fundamentally different consequences, from what it was at the time of the founding. Those who make this claim regard the seemingly endless outward thrust of copyright protection—which has come to include derivative works, “substantially similar” as well as literal copies, and anticircumvention rights—as significant because it has expanded the extent of the copyright owner's control and thus put in private hands power over more speech than formerly. They also regard technological innovations as significant because digital technology makes it so easy and so cheap for individuals to make perfect copies of works. Anyone with a computer can do it, and just about everyone does. Computers make more copying possible than ever before, critics claim, and, therefore, enforcing copyright restricts more speech than ever before.

Commentators are correct to point out that Congress has responded to technological innovations—in particular to the ever-advancing technology of copying—and to changes in the market for copyrighted works by continually adjusting, indeed by steadily expanding, copyright owners' exclusive rights. Commentators are also correct to note that the anticircumvention provisions of the DMCA enhanced copyright owners' legal control over easily copied works in digital form. Refuting the inferences they draw from these facts will bring us to the second strand in the argument to demonstrate that the result in Eldred stands on firm ground, at least insofar as it denies heightened First Amendment scrutiny in run of the mill copyright cases: the continual adjustments of copyright owners' rights do not alter a previously immutable copyright balance, for there never was such a thing. In fact, as Jane Ginsburg has demonstrated, the history of copyright's evolution has been one in which each significant development of technology—the printing press, movable type, photography, motion pictures, audio recording, Xerography, digitization and digital networks—has willy-nilly altered the balance of control between authors and users, and each has eventually prompted “a new legal calibration” of the proper balance.¹²⁷

Moreover,

More to the point from this paper's perspective, the expansions of copyright owners' rights do not so alter the nature of copyright as a restriction on speech as to render the founders' understanding of the First Amendment implications of copyright no longer pertinent. The reason is that the Copyright Clause of the Constitution authorizes Congress “to secur[e] to authors … the exclusive right to their … writings.¹²⁹

Since the very definition of a property right is that it is an “exclusive right,” the Copyright Clause explicitly contemplates propertization. Beginning with the first copyright statute, and continuing to the present day, Congress has granted what are and always have been the functional equivalent of property rights (regardless of what they were called) to authors of original works. Though their scope has been altered and expanded from time to time, and the degree of protection they afford has been adjusted to reflect both significant developments in intellectual property markets and revolutionary changes in the technology of copying, these property rights have since the founding given copyright owners the power to call upon the courts for enforcement by restricting expression that infringes. Since Congress passed the first copyright statute granting property rights in 1790, and the Bill of Rights was proposed by Congress in 1789 and ratified in 1791, and since the fundamental legal nature of copyright has remained unchanged, it seems that Justice Ginsburg was correct in Eldred when she concluded that the Bill of Rights' proximity in time to the first copyright statute “'indicates that, in the Framers' view, copyright's limited monopolies [i.e., property rights] are compatible with free speech principles.”¹³⁰

Some scholars who support expanded First Amendment review in copyright cases seem to suggest that, whatever the historical reality, expanded review would correct what has become an obvious and unjustified doctrinal error.¹³¹

They maintain that copyright as it is presently enforced is simply not compatible with First Amendment doctrine as it is currently enunciated.¹³² Their analysis implies (and sometimes it explicitly avers) that the success of a copyright plaintiff involves a government decision to restrict speech. They do not specify whether the reason they characterize it as the government's decision is because Congress passed the Copyright Act or because copyright owners invoke the aid of courts for the enforcement of their rights. Viewing the matter as a “who decides” question, however, the point is of no particular importance. We can assume arguendo that Congress acted when it passed successive versions of the Copyright Act, and that when courts rule in plaintiffs' favor in copyright cases they do so in their capacity as state actors. The assumption does not compel the conclusion that a judgment for a plaintiff in a copyright case represents the government's decision to restrict speech. Just as property rights in tangible things reflect a collective decision to allocate enforceable decision-making authority regarding use, possession, and disposition to private owners, property rights in original works of authorship reflect a collective decision to allocate decision-making authority regarding reproduction, derivative works, distribution, and the like, to private owners. In other words, when Congress decided to grant property rights in authors' original works, and from time to time revised the boundaries of and expanded the available enforcement mechanisms for such rights, Congress in effect allocated to copyright owners the decision whether to “restrict speech.” As we have seen, in making this allocation, Congress acted consistently with the First Amendment and pursuant to the Copyright Clause of the Constitution. When a judge enjoins or awards damages for an infringement of copyright, though she is a government actor acting in her official capacity, she is not “deciding to restrict speech.” Rather, pursuant to Congress's direction to put the decision whether to restrict speech in the hands of the private copyright owner, she is merely ascertaining the boundaries of the plaintiff's property right, determining that the defendant has trespassed upon it, and, enforcing the copyright owner's “decision to restrict speech.”¹³³

From an institutional perspective, there is a close analogy between what a court does in a copyright case and what a judge does who enforces a trespass law at the behest of an owner of real property who seeks to exclude someone who would use the owner's property for First Amendment—speech—purposes. In the trespass case, of course the government is “acting,” but for constitutional purposes its action is reflected in its anterior decision to enforce private owners' decisions to exclude. Having decided to enforce such decisions via the law of trespass, the government cannot meaningfully be said to have endorsed the particular (speech-restrictive) consequences of any particular owner's decision. By a parity of reasoning, the government “acts” in a copyright case too, but the constitutional significance of its action is contained in its anterior decision to enforce private copyright owners' decisions to exclude infringers who wish to use any particular owner's property—his “original work of authorship”—for speech purposes. True, the trespass cause of action will lie regardless of whether the trespasser was speaking and without regard to what he wanted to say, whereas deciding that there has been a copyright infringement will require the fact-finder to look at the content of both the plaintiff's and the defendant's speech. Because the inquiry into content is limited to determining the boundaries of the plaintiff's property right and whether the defendant's speech infringes, it raises none of the risks that the prohibition on restricting speech “because of its content” is designed to forestall. The judge in the copyright case will not inquire, for example, into either the infringer's or the copyright owner's point of view, nor will she ask about the effects their substantive messages are likely to have on the audience. I develop these points in the analysis that follows.

Jed Rubenfeld is among the commentators who reject the copyright-trespass to real property analogy, and his argument provides a good platform from which to assess more thoroughly the analogy's soundness. Rubenfeld presents three main objections. First, he observes, the law of trespass to real property does not make speaking an element of the offense. Copyright law, in contrast, makes people liable not only because they are speaking but also because of what they say. Second, he claims that the relevant comparison is not between copyright and the First Amendment-trespass case of Hudgens v. NLRB, but between copyright and the First Amendment-libel case of New York Times v. Sullivan.¹³⁴

Third, he claims that the reasons why state enforcement of trespass laws do not violate the First Amendment “disappear when we turn to copyright.”¹³⁵

Consider the first objection, which is based on Rubenfeld's observation that copyright law makes people liable not only because they are speaking but also because of what they say, whereas trespass does not make speaking an element of the offense. The observation seems to imply that copyright law necessarily conflicts with the canonical doctrine, enunciated in dictum in the 1972 case of Police Department of Chicago v. Mosley,¹³⁶

that “above all else, the First Amendment means that government has no power to restrict expression because of its message, its ideas, its subject matter, or its content.”¹³⁷ In response, it is important to emphasize that, because it aims at speech that infringes, and restricts a defendant's expression only if it reproduces the plaintiff's, copyright liability does not represent a genuine threat to the implementation of the First Amendment strategies embodied in the Mosley dictum.

In fact, it is plausible to argue that copyright infringement liability neither punishes nor deters speech on account of its content. Instead it aims at conduct that infringes: unauthorized reproduction, preparation of derivative works, distribution, public performance, and the like. Viewing copyright infringement as behavior that combines speech and nonspeech elements would require the Court to subject copyright law to First Amendment review, but it would be the lenient scrutiny prescribed by the Court in the draft-card burning case of United States v. O'Brien.¹³⁸

In O'Brien the Court held that a regulation of behavior that combines speech and nonspeech elements will be sustained if it is

The most significant part of the O'Brien test is the determination whether the government's interest in the regulation is “unrelated to the suppression of freedom of expression,” because the answer is likely to be outcome determinative. If the Court determines that the government's interest is unrelated to the suppression of freedom of expression, then it is likely in practice to subject the regulation, both on its face (as discerned from the terms of the regulation itself) and as applied (as evidenced from the particular application at issue), to extremely lenient review, which means that the regulation is very unlikely to be found to violate the First Amendment. Though the cases applying the test for whether a regulation is “unrelated to suppression of free expression” are not entirely uniform, it is possible to discern a consistent theme: where the harm that the statute (or its application) is seeking to avert is one that is independent of the message being regulated—that is, the harm arises from something other than a fear of how people will react to what the speaker is saying—the government's interest is “unrelated to suppression.”¹⁴⁰

Application of this test to copyright law on its face, or as it is applied in an infringement judgment in any particular case, yields the conclusion that copyright is regulation unrelated to suppression of freedom of expression. The reason is that the harm the copyright statute (or its application in a particular case) seeks to avert—the harm to the copyright owner's property that unauthorized reproduction, preparation of derivative works, distribution, and the like create—is independent of the infringer's message that is being regulated and arises from something quite other than a fear of how people will react to the ideas the infringing speaker is attempting to convey.

Rubenfeld's implication that copyright law is at odds with the First Amendment's aversion to content regulation takes no account of a very important fact: even if the question of whether the defendant has infringed turns in every case on the “content” of both the plaintiff's and the defendant's work, the nature of the inquiry into content will not bring into play any of the reasons for the aversion to content regulation. The Court's express distrust of content-based regulations is more than a convenient knee-jerk doctrinal guideline. It reflects a number of concerns that emerge from consideration of the First Amendment's animating values.¹⁴¹

For example, it reflects a concern for equality—that particular categories of speech be treated equally so as to deflect the government from attempting to manipulate public discourse. Second, it reflects a concern to prevent government from regulating speech based on its communicative impact, thus forestalling restrictions of speech that surreptitiously rely on constitutionally disfavored justifications. It reflects a concern to foreclose the possibility that government will deliberately distort debate by preventing the communication of particular disfavored ideas, viewpoints, or items of information. Finally, it reflects a concern with improper legislative motivation. But a judgment that a copyright defendant's work reproduces the plaintiff's, or is based upon it, or distributes or publicly performs it, though it will of course be based on the content of both the plaintiff's and the defendant's speech, simply raises none of these concerns. It does not jeopardize the equality of treatment of particular categories of speech so as to enable government to manipulate discussion. It does not turn on the speech's communicative impact. And it does not raise the risk of an improper government motivation to prevent the dissemination of particular ideas, viewpoints, or items of information that the government disfavors. In fact, the predicate for copyright liability is not the communicativeness of an infringer's speech at all. It is, rather, its similarity to the plaintiff's—and even then it is only its similarity to the plaintiff's expression, not to the ideas or facts therein.

Rubenfeld analogizes the Copyright Act to the flag-burning statute that prohibited “defacing or destroying an American flag in a fashion that intentionally communicates disrespect” and that the Court declared unconstitutional in 1989 in Texas v. Johnson.¹⁴²

In fact, the two statutes are very unalike. The flag-burning statute was held unconstitutional, to be sure, but not as Rubenfeld claims “because it [made] the communicativeness of the defendant's actions an element of the offense.”¹⁴³ Instead, the problem with the statute was that defendant's liability “depended on the likely communicative impact of his conduct.¹⁴⁴ More particularly, liability depended on the fact that the message he communicated “would cause serious offense to others.”¹⁴⁵ A copyright defendant's liability does not turn on the communicative impact of his conduct, nor on the fact that his message might cause offense to his audience. It turns only on the form his expression takes and on whether it copies the plaintiff's.¹⁴⁶

Next consider Rubenfeld's second objection to the copyright-trespass analogy, which is based on the claim that the appropriate comparison is not between copyright and Hudgens v. NLRB but between copyright and New York Times v. Sullivan,¹⁴⁷

the 1964 case in which the Supreme Court for the first time subjected libel law to First Amendment scrutiny. Rubenfeld is not alone in regarding Sullivan as analogous,¹⁴⁸ but he puts the argument most succinctly: “Copyright stands to property law as libel stands to tort law; copyright makes speech property, as libel makes speech a tort.”¹⁴⁹ The conclusion that since libel must conform to First Amendment constraints, so must copyright, is said to follow. The conclusion does not, however, necessarily follow, though the reasons are complicated and thoroughly to explore them is beyond the scope of this paper, as they raise surprisingly profound issues of First Amendment theory. I shall attempt, however, to give the reasons in broad outline.

My reasons begin with a restatement of the First Amendment implications of the fact that “copyright makes speech property.” As I have demonstrated, the argument is straightforward (though, as I hope I have made clear, I do not claim it is uncontroversial): the Copyright Act confers the functional equivalent of property rights on authors, and infringement of copyright is thus the functional equivalent of trespass. According to Hudgens, there is no First Amendment right to trespass on tangible property, and by a parity of reasoning there is no First Amendment right to “trespass” on intangible property, that is, there is no First Amendment right to infringe copyright. This, in a nutshell, is why the copyright-trespass analogy is apt. Rubenfeld rejects this analogy in favor of the copyright-libel analogy, but he does so not because copyrights are not property rights. Rather, he thinks that what matters is that copyrights differ from “ordinary” property rights in the crucial respect that copyrights, like libel laws, render “speaking an element of the offense.”¹⁵⁰

And he asserts that it was “precisely because libel … makes speech as such illegal [that] courts were eventually obliged to address and redress its First Amendment consequences.”¹⁵¹ This assertion, in my view, mischaracterizes the reason why courts—in particular, the Supreme Court—undertook to constitutionalize the law of libel, and it is this mischaracterization that renders Rubenfeld's equation of copyright and libel problematic. It was not a concern with “speech as such” that moved the Court in Sullivan. It was, instead, the concern to prevent the punishment of seditious libel, to protect those who criticize government and the conduct of government officials. On account of this focus, and not on account of the supposed importance of “speech as such”, the Sullivan case was hailed for having discerned the “central meaning” of the First Amendment.¹⁵² True enough, defamation jurisprudence in the Supreme Court soon detached itself from its seditious libel roots,¹⁵³ and by now the First Amendment has been held to apply to false statements of fact about private figures whether or not the speech is even about matters of public concern.¹⁵⁴ In addition, recent case law provides some support for the claim that the Court may have become persuaded that “speech as such” is worth protecting,¹⁵⁵ but there is case law that points in quite the opposite direction.¹⁵⁶ Thus, the fact that copyright, like libel, “renders speaking an element of the offense” does not provide unequivocal support for the conclusion that the First Amendment constrains copyright, as it constrains libel. Nor is it reason enough to reject the conclusion that, as the First Amendment does not constrain the law of trespass, so it does not constrain the law of copyright.

Consider next Rubenfeld's third objection to the copyright-trespass analogy, which he discusses under the heading “Private Power over Public Speech.” He claims that the reasons why state enforcement of trespass laws do not violate the First Amendment “disappear when we turn to copyright.”¹⁵⁷

One of those reasons, he says, is that property owners lack general power to block the speech of the public at large, whereas a “copyright owner's power applies to the public at large, anywhere and everywhere.”¹⁵⁸ The argument fails to account for the fact that a copyright owner's power over the speech of the public at large exists only with regard to particular expression and not with regard to ideas. Thus, the important implication of this aspect of copyright law, which is known as the ‘idea-expression dichotomy’, is that, though it is notoriously difficult to apply in practice, in principle it embodies a significant limitation on a copyright owner's power to “block the speech of the public at large.” Just as trespassers who are denied access to a particular parcel of real property may communicate their ideas or messages in an almost infinite variety of alternative forums, so, too, may infringers who are denied the right to copy particular expressions communicate the same ideas or messages in an almost infinite variety of alternative formulations. The continual outpouring of new expressive works—books, movies, songs, and plays—that do not infringe existing works would seem to testify to the narrowness of a copyright owner's rights to control the speech of the public at large, not to its breadth.

Rubenfeld thinks it matters that when ordinary property owners invoke their right to exclude, we “tend to” regard them as “exercising First Amendment rights of their own,”¹⁵⁹

but when copyright owners exercise their statutory rights to prevent infringement, we do not so regard them. To be sure, there are cases where compelled access to property for the purposes of speech has been held to have violated the property owners' First Amendment rights,¹⁶⁰ but these holdings have not applied in the cases in which First Amendment claims of access to private property have failed on property rights grounds, and the soundness of the copyright-trespass analogy therefore does not depend on them. Hudgens v. NLRB¹⁶¹ is, of course, the principal case in which the Court held that there is no First Amendment right to trespass on real property. Its holding rests unequivocally on the owner's property right to exclude. The opinion makes no mention of the owner's First Amendment rights. Rubenfeld points out that “no one has a First Amendment right to be the only speaker of certain words,”¹⁶² but the relevance of this fact to the copyright-trespass analogy is not as obvious as he thinks. Copyright does not grant authors exclusive rights in “certain words.” Rather, it grants them rights in their expressions, that is, in words arranged in particular sequence, conveying ideas in particular ways. Moreover, and more importantly, copyright ownership does not purport to give copyright owners First Amendment rights to prevent infringement; instead, it gives them property rights in their original expressions, rights that are purely creatures of statute—products of Congress's collective judgment to lodge decision-making power regarding the exploitation of original works of authorship in the hands of those who create them.

B. Trespass in virtual space and the First Amendment

This subsection considers the conformity to the First Amendment of the anticircumvention provisions of the DMCA, such as those that were litigated in Universal City Studios, Inc. v. Corley,¹⁶³

as well as the trespass claim that was sustained by the California district court of appeal decision in Intel Corp. v. Hamidi.¹⁶⁴ In institutional, “who decides,” terms, the effect of the anticircumvention provisions is to lodge control over access to copyrighted works in the hands of copyright owners and the effect of the court decision sustaining the trespass claim is to lodge control over communication with persons on proprietary networks in the hands of their private proprietors. Thus, the way to frame the inquiry pursued in this section, consistent with this paper's institutional perspective, is to ask whether the Constitution forbids Congress or state common law rules to vest in private hands substantively unreviewable authority to control access. The more conventional way to frame the inquiry, and to make it more specific, is to ask whether access seekers have First Amendment rights to compel private owners to let them in, either for purposes of using, copying, or disseminating content on a Web site or for the purpose of communicating content to those on an e-mail network. If the question is framed in this conventional way, and the conclusion is that access seekers do have First Amendment rights, this means that the Constitution lodges in government decision-makers—namely judges—the ultimate power to make substantive judgments about when, and on what terms, and for what purposes access shall be granted to privately owned Web sites or e-mail networks. The anticircumvention provisions present the question whether compelled access for the purpose of digitally gathering content is constitutionally required, while Intel raises the question whether the Constitution mandates access for the purpose of communicating.

Given the First Amendment results canvassed in Section IV, if the trespass to tangible property analogy developed in Section III is sound, then the answer to the question posed in this section is virtually a foregone conclusion: the First Amendment neither requires that access be granted to digital content nor commands that proprietary networks receive communications they do not welcome. It is worthwhile, nevertheless, to summarize the cases that have addressed the issues.

Take first Universal City Studios, Inc. v. Corley,¹⁶⁵

in which the U.S. Court of Appeals for the Second Circuit dismissed a First Amendment challenge to the anticircumvention provisions of the DMCA. The defendant had posted a decryption computer program on his Web site. The program circumvented encryption technology that motion picture studios use to prevent unauthorized viewing and copying of their movies. Eight studios sought injunctive relief against the defendant, relief that the district court granted and the Second Circuit affirmed. The defendant argued that, because computer code is “speech,” and the DMCA regulates it, the DMCA must be subjected to strict scrutiny. Judge Jon O. Newman's opinion for the Second Circuit first concluded that, because of what computer code is and what its normal functions are, it should be treated as combining nonspeech and speech elements, and that “the causal link between the dissemination of circumvention computer programs and their improper use is more than sufficiently close to warrant selection of a level of constitutional scrutiny based on the programs' functionality.”¹⁶⁶ The court analogized the decryption code to “a skeleton key that can open a locked door, a combination that can open a safe, or a device that can neutralize the security device attached to a store's products,”¹⁶⁷ but because computer code is a form of communication and thus has a claim to being “speech,” the court declined to hold that “Congress has as much authority to regulate the distribution of computer code … as it has to regulate distribution of skeleton keys.”¹⁶⁸ Therefore, the court applied the intermediate scrutiny that the Turner I case prescribes for content-neutral regulation, and concluded that the DMCA and the provisions of the injunction that prohibited the defendant from posting the decryption code on the Internet or from linking to other sites that contained it were targeted at the nonspeech element of computer code, namely, at the code's capacity to instruct a computer to decrypt the studios' encryption code and at the capacity of linking “instantly to enable anyone anywhere to gain unauthorized access to copyrighted movies.”¹⁶⁹ Accordingly, the court held that these provisions of the injunction survived scrutiny.

In Intel Corp. v. Hamidi,¹⁷⁰

the defendant, a former employee whom Intel had fired, flooded Intel's internal e-mail system on a number of occasions with grievances about the company. He sent thousands of copies of the same message on six occasions over a period of twenty-one months, ignored Intel's requests to stop, and was able to evade its efforts to block his messages. Intel brought a trespass to chattel claim, the California trial court granted an injunction, and the intermediate appellate court affirmed. After canvassing treatises and cases regarding the necessity of showing actual damage in order to recover damages for trespass to chattels, the court determined that the availability of damages was irrelevant: “Intel seeks no damages.”¹⁷¹ The important fact was that “Hamidi's conduct was trespassory,”¹⁷² and “[j]udicial enforcement of neutral trespass laws has been held not to constitute state action.”¹⁷³ Though the court addressed the issue principally as though the result were a function of the formal absence of “state action,” it relied equally on the more substantively realistic claim that it was not the absence of state action that justified the result, but, rather, the fact that the Supreme Court “has never held that a trespasser or an uninvited guest may exercise general rights of free speech on property privately owned and used nondiscriminatorily for private purposes only.”¹⁷⁴

Thus, both the U.S. Court of Appeal for the Second Circuit and the California Court of Appeal for the Third Appellate District implicitly embraced the trespass rationale for resolving access claims in the digital world. The courts did not reason in terms of the institutional analysis offered here, and they did not frame their conclusions as having been driven by “who decides” considerations. Nevertheless, it is possible to understand both cases as applications of the principle that the Constitution permits the government to allocate enforceable decision-making authority about production and use of resources, including intangibles and including the resources' use for speech purposes, to private actors.