Content Regulations Where the Message Directly Causes Harm

The content of messages received can directly cause many types of harm that government may legitimately try to prevent. The content of a message may be a confidence between an attorney and his client, a physician and her patient, or a priest and his penitent.

Now if content regulation of this sort - the sort involved in invoking con­fidentiality and secrecy restrictions, the privacy tort, intellectual property law, contractual obligations, laws banning offensive speech, the tort of infliction of emotional distress, gag orders on lawyers and closed judicial proceedings, the crime of uttering threats, and even the rules governing the admissibility of evidence^{[98] [99] - is to be deemed to implicate freedom of expression, there are three possible courses open. First, we might hold that all such laws violate the right of freedom of expression. People should be free to send and receive messages the content of which is confidential, private, created by another, emotionally dis­tressing, and so forth. This would, of course, have very profound implications and would represent a major change in the jurisprudence of all those countries that recognize a legal right of freedom of expression.}

The second alternative would be to say that although such regulations im­plicate freedom of expression, they do not necessarily violate freedom of ex­pression. Whether a content-regulation of the sort we are considering violates freedom of expression will depend upon the results of weighing the interest in freedom of expression against the government’s interest in preventing harm to confidentiality, privacy, ownership and creativity, emotional peace, and so forth.

If we take this second, less radical course, however, then we are back in the same briar patch we entered with Track Two regulations. For how can we construct a metric for weighing the interests in receiving certain messages against the harms those messages cause unless we assign a value to the ideas or information those messages contain? If we assume ignorance of what ideas and information will be affected, we have an obvious problem in assigning weight to the ideas and information. On the other hand, if we assess the ideas and information ex post, with knowledge of what they are, we still must assign them value to be weighed against the harms they threaten; and how can we do that except by reference to a framework that will differ from person to person and that will be partisan and controversial through and through?^[100]

Let me illustrate this dilemma through some examples. First, assume that a client reveals to his attorney, in confidence, that he committed a serious crime - perhaps perjury, fraud, or even murder - in his past. Later, the client is a candidate for high office - the Senate or the presidency - and because of public-spiritedness, the attorney desperately wants to reveal what he knows about the candidate. Finally, assume there are no exceptions in the laws requiring attorneys to keep clients’ confidences that would permit the attorney to reveal those confidences in cases like the one we are imagining. Thus, if the attorney discloses what he knows about the candidate, the attorney may be disbarred and/or fined.

Second, assume a newspaper prints a “human interest” story that discloses that a child math prodigy now works at a menial job,^[102] or that a current mem­ber of respectable society used to be a prostitute.^[103] The subject of such a story sues the newspaper based on the state tort of disclosure of embarrassing pri­vate facts, and the newspaper invokes freedom of expression as its defense. Is freedom of expression more or less important than the state’s interest in pro­tecting privacy?^[104]

Third, assume that a photographer copyrights his exclusive photographs of the assassination of a president, that certain facts about the assassination are a matter of widespread interest and concern, and that the photographs might shed light on those facts. And suppose that a magazine publishes the photographs without obtaining the photographer’s permission and invokes freedom of ex­pression as a defense to the subsequent copyright infringement suit.⁸ Is freedom of expression more or less important than protecting artistic, scholarly, and lit­erary creativity? In this case? In a sufficient number of cases to warrant a broad rule that covers this case? What if the copyrighted matter were a human cannon­ball act and the infringement its unauthorized broadcast on the eleven o’clock news?⁹

Fourth, assume that a defendant utters hateful epithets (or makes otherwise offensive comments) to the plaintiff, knowing that the plaintiff will be trauma­tized, and is sued for the tort of intentional infliction of emotional distress, or is upheld any such restriction. In every case presenting the issue, the Court has found that the information was in the public domain because it had been lawfully obtained from public sources. See, e.g., Cox Broadcasting Corp. v. Cohn, 420 U.S. 469 (1975); Smith v. Daily Mail Pub.

The courts in other countries that recognize freedom of expression have generally allowed interests in privacy and confidentiality to trump it. See, e.g., Re C, 10 B.H.R.C. 131 (1991) (protecting privacy over freedom to publish celebrity photos in Germany); Douglas v. Hello! Ltd. (No.1) [2001] Q.B. 967 (protecting interests in privacy and confidentiality in England).

For an attempt to justify excluding some speech about private matters from freedom of expression protection, and an attempt to define such speech, see Daniel J. Solove, “The Virtues of Knowing Less: Justifying Privacy Protection Against Disclosure,” 53 Duke L. J. 967 (2003).

⁸ See Time, Inc. v. Bernard Geis Associates, 293 F. Supp. 130 (S.D.N.Y. 1968).

⁹ See Zacchini v. Scripps-Howard Broadcasting Co., 433U.S. 562 (1977) (upholding claim against unauthorized broadcast of a performance).

Edwin Baker takes the position that copyright laws do violate the right of freedom of ex­pression in cases where the infringer is neither acting for commercial gain at the expense of the copyright holder nor attempting merely to destroy the value of the copyright.

And as my final example, assume that a defendant, for valuable considera­tion, agrees not to disclose publicly certain information that the plaintiff does not wish disclosed, and that defendant later breaches that agreement and in­vokes freedom of expression as a defense to the breach of contract suit.¹¹ Does freedom of expression trump the state’s interest in enforcing otherwise valid contracts? In all cases? If only in some, which ones?

My point with these examples is not that we, individually, cannot do the weighing. We can, or at least we do, just as the legislature has done whenever it decides that the tort, contract action, or crime covers the act of communicating a particular message. My point is, rather, that the results we come up with will be a function of what we believe is true and valuable, and that each of us will differ along these lines.

The Supreme Court generally has upheld regulations of speech when the content of the speech could be deemed to be “owned by another,”¹² or “subject to another’s contractual right,”¹³ and so forth. What is not appreciated is how radically these decisions undermine principle (5). In all these cases the govern­ment wishes to prevent receipt of a message because of its content. Of course, the laws under which the government acts are framed in a content-neutral way in terms of the harms the government is trying to prevent - use of another’s property; breach of contract; disclosure of confidences, secrets, or private facts; infliction of emotional distress; and so on. But in every case implicating free­dom of expression, government acts to prevent harms that can be described in content-neutral ways. The fact that the ultimate purposes of governmental regulations of the content of expression are the prevention of harms that them­selves can be described without reference to the content of messages cannot exempt those regulations from freedom of expression concerns without reduc­ing the latter to a nul set.¹⁴ So if government’s purpose is to prevent harm by

¹² See Harper & Row Publishers, Inc. v. Nation Enterprises, 471 U.S. 539 (1985) (upholding claim against unauthorized use of copyrighted materials); Zacchini v. Scripps-Howard Broadcasting Co., supra note 9.

¹³ See Cohen v. Cowles Media Co., supra note 11.

¹⁴ Perhaps were a government to ban speech solely on the ground that the speech is false, and not because its being false would lead to further harms (i.e., to some type of detrimental reliance), we would have an example of a regulation of messages that could not be recharacterized as the prevention of a harm describable in a content-neutral way. But if freedom of expression applied only in that circumstance, it would have virtually no practical implications. Governments almost always ban messages because of fear of harms that those messages might cause. If freedom of expression only requires that government be able to cite to some harms that it may legitimately prevent and to offer sufficient evidence to support its fear that the messages it seeks to suppress will bring about those harms, freedom of expression would be an almost completely inconse­quential right. Moreover, because any restriction of liberty by government should be supported by a showing that the exercise of the liberty will cause harm to a legitimately protectable interest, freedom of expression would be merely an instance of a general right to liberty.

For an example of the view of content neutrality against which I am arguing, see Jeremy J. Ofseyer, “Speech or Opinion? Two Objects of First Amendment Immunity,” 2002 Utah L. Rev. 843, 903-6.

The United States Supreme Court occasionally ignores this point and treats message reg­ulation as content-neutral because government’s concern is with the harms the receipt of the message brings about, and not with the message per se. The most notorious examples of this are the Court’s “secondary effects” cases, in which it upheld restrictive zoning of adult movie theaters and bookstores because of the neighborhood blight their presence produces. See Young v. American Mini-Theatres, 427 U.S. 50(1976); City of Renton v. Playtime Theatres, 475U.S. 41 (1986). In such cases, however, it is the “adult” message that draws undesirables into the neighborhood and causes desirable uses to exit.

Likewise, in some cases that the Court has treated as involving only the regulation of media and not of messages - in particular, the public nudity cases such as Barnes v. Glen Theatre, Inc., 501 U.S. 560 (1991), and Erie v. Pap’s A. M., 529 U.S. 277 (2000) - the harm that the govern­ment is trying to avert arguably is a product of the ideas formed in the minds of the audience for the proscribed conduct (sexual ideas and fantasies, which might lead to unwanted conduct), suppressing or altering messages, its regulation falls within principle (5) and thus within the scope of freedom of expression.

Finally, one might argue that content regulations designed to prevent one- step harms violate freedom of expression unless prevention of those harms is a “compelling interest.” According to this argument, we need not balance expression interests against harm-prevention interests. Instead, we need only assign all expression that is potentially suppressed a constant value equivalent to whatever governmental interests in harm-prevention we deem “compelling.”

The problem with this argument is that it is essentially ad hoc. Presumably, it represents the conclusion of a wholesale balancing of the interest in expression against other interests. How that balancing is achieved, however, is not revealed. Moreover, neither is it revealed how one decides which kinds of harms are sufficiently serious to render preventing them through suppressing expression “compelling.” Is erosion of lawyer-client confidentiality such a harm? The loss of privacy? Infringement of copyright? Breach of contract?

The third alternative would be to hold that content regulations of the sort that we are considering never violate the right of freedom of expression, the same conclusion reached with respect to Track Two regulations. The reason would be the same reason for excluding Track Two regulations from the scope of freedom of expression: We cannot have a nonpartisan assignment of weight to the value of the messages suppressed that can be compared to the legitimate harms the government wishes to prevent.

Jed Rubenfeld, like me, identifies the scope of freedom of expression with principle (5).¹⁵ And, again like me, he regards evaluative neutrality as at the core of freedom of expression and therefore, vehemently rejects balancing.¹⁶ However, when it comes to the laws of interest here - the laws protecting against harms caused by messages directly - Rubenfeld is selectively silent. One might expect that because Rubenfeld rejects balancing and accepts princi­ple (5), he would find laws protecting confidences, secrets, privacy, contracts, emotional wellbeing, fair trials, and so on to be per se violations of freedom of expression. That would be a quite radical conclusion, but one that would follow from Rubenfeld’s premises. If, however, one’s views of freedom of expression entailed the illegitimacy of all sorts of commonplace laws, one would expect some mention and indeed a vigorous defense of this result. But Rubenfeld says nothing about most laws of this type, much less that they are all illegitimate. He seems to be too preoccupied with defending principle (5) and evaluative neutrality from balancers to see the radical implications of his views.

or perhaps formed in the minds of others merely thinking about the conduct and its audience (offense). Indeed, it is difficult to understand how the harms that nude dancing supposedly causes could occur other than through the production of antisocial or offensive ideas.

¹⁵ See Rubenfeld, supra note 1; Rubenfeld, supra note 3.

¹⁶ See note 3 supra and accompanying text.

I said that Rubenfeld is selectively silent. The one area where Rubenfeld does analyze laws that fall under principle (5) and are designed to prevent a one-step harm is that of copyright laws.^[107] And Rubenfeld concludes, unsurprisingly, that because copyright laws are aimed at affecting messages - specifically, they are meant to suppress messages that are “owned” by another - they violate freedom of expression. As Rubenfeld puts it

In some parts of the world, you can go to jail for reciting a poem in public without permission from state-licensed authorities. Where is this true? One place is the United States of America.

Copyright law is a kind of giant First Amendment duty-free zone. It flouts basic free speech obligations and standards of review. It routinely produces results that, outside copyright’s domain, would be viewed as gross First Amendment violations.^[108]

Rubenfeld argues that the fact that copyright law allows others to express the copyright holder’s ideas as long as they do not employ the holder’s particular form of expressing those ideas, or that copyright law provides various excep­tions under the rubric of “fair use,” does not eliminate the basic freedom of expression violation that copyright protection constitutes. The idea/expression distinction is unavailing because form and substance, medium and message, are inextricable one from another.^[109] Each form of expression affects the idea expressed thereby in subtle and not-so-subtle ways. And as for the fair use exceptions, which favor, for example, parodic and critical treatments of copy­righted works over other treatments, Rubenfeld points out that these amount to favoritism toward certain viewpoints (those critical of the work), a violation of the evaluative neutrality that lies at the core of freedom of expression.^[110] Fair use, in other words, rather than salvaging the legitimacy of copyright pro­tection, “renders copyright law viewpoint-discriminatory, which... amounts almost everywhere else in free speech law to virtually a per se constitutional violation.”^[111]

Rubenfeld also rejects as a sufficient justification for copyright protection that it incentivizes the production of copyrightable works - that is, that it will lead to the production of more expression than will occur in its absence.^[112] Even if this is true, he argues, this would not be sufficient. Bans on other types of speech - the uncivil or the offensive, for example - might lead more people to engage in public dialogue rather than to shy away from it. Yet those bans are violations of freedom of expression notwithstanding their possibly speech- productive effects.

Finally, and perhaps most importantly, Rubenfeld rejects the notion that by dint of effort and creativity, one can acquire property in forms of expression:

The property intuition says that copyright claims are immune from First Amendment scrutiny because there is no FirstAmendmentrighttotrespass on someone else’s “legally recognized intellectual property.” The proposition proves far too much. If Congress could act with First Amendment impunity whenever it turned speech into property, the freedom of speech would turn out to mean a freedom to speak only at the sufferance of federally designated individuals or corporations. If there is no First Amendment right to “trammel on legally recognized intellectual property rights,” then there would be no First Amendment problem with, say, a statute granting Microsoft the exclusive right to use English on the Internet.^[113]

Having taken a strong position against the constitutionality of copyright protection, however, Rubenfeld then proceeds to weaken it considerably. First, he would not protect the simple piracy of another’s work, for piracy involves no new creative act - no “reimagining” - but is purely mechanical.^[114] Indeed, Rubenfeld argues that because piracy is mechanical, laws against it are not content-based. One need not understand the messages to ascertain whether straight copying has occurred.^[115] And Rubenfeld would not protect work that makes only trivial or obvious modifications of the copyrighted work.^[116] All other work that derives from the copyrighted work would be protected, even if it were substantially similar to the copyrighted work, although Rubenfeld would require commercial derivations to share profits with the copyright holder.^[117]

In the end, therefore, Rubenfeld’s critique of copyright laws is less radical than it first appears. Copyright holders do get a property right over the expression of an idea, a property right that enjoins verbatim copying and trivial changes, and that provides the basis for a claim to share in the profits of derivative works.

Why does Rubenfeld not go all the way and repudiate any property in messages? It is true that in one sense, pirates of copyrighted works do not exercise their imagination when they copy. But so what? If creators cannot own their messages, then pirates take nothing that belongs to the creators.

In another sense, moreover, pirates do exercise their imaginations. Suppose that after I write this book, Cambridge University Press markets it for fifty dollars. Someone who likes this book and thinks its message important might “imagine” that its message gets disseminated widely, including to those who would not buy or could not afford to buy it from Cambridge. And in pursuance of that “imagining,” this admirer might place his purchased copy of the book on a copying machine and sell the reproductions for a few dollars each, the cost of the copying. Rubenfeld would not protect the pirate, but why is his “imagining” - “Alexander’s work on freedom of expression widely disseminated” - less privi­leged than the imagining of one who borrows liberally from this work to produce a competing product?

Of course, Rubenfeld protects the latter only to the extent of (1) preventing me from enjoining him and (2) allowing him to keep whatever profits are traceable not to my efforts but to his alone. That is why Rubenfeld’s proposal is much less radical than he implies. But given his initial position, why does Rubenfeld allow me to enjoin the pirate and monopolize my particular message? Why is my imagination in creation privileged over his imagination in distribution?

Or consider the example of the photographer who copyrights his exclusive photographs of a presidential assassination. Suppose another person “imagines” those photographs widely and freely distributed in order to inform the public about the assassination, and she proceeds to so distribute them without autho­rization from the photographer. She has merely copied the photographs. On the other hand, she has “imagined” them in a way different from how the photogra­pher “imagines” them - restrictively available for a high fee. And she has made no profit from their use, though she has destroyed any profit the photographer hoped to make. Would Rubenfeld allow a court to enjoin her from distributing the photographs? Would he allow her to be subject to damages liability? If so, then again why is the photographer’s imagination in creating privileged over her imagination in distributing?

Suppose for a moment that copyright protection were a freedom of expres­sion violation even as to pirating. We might assume that the following would occur: Whenever one purchased a book or magazine, watched a movie or game on television, attended a live performance, and so forth, one had to sign an agreement not to “copy” it and distribute the copy in various ways. And if one violated the agreement, one would be liable in damages and be enjoinable from future violations. The question would then be, is enforcing agreements not to convey certain messages a violation of freedom of expression? If not, then copyright protection, banished through the front door of property, would enter through the back door of contract. (No one denies that the item obtained that is subject to the contract - for example, the tangible book that is then copied - is property in the ordinary sense.)

Indeed, if enforcement of contracts regarding messages is consistent with freedom of expression - and the United States Supreme Court has indicated that it is^[118] - some of the one-step harms caused by messages other than harms to ownership could be regulated by contract. Protection of confidential com­munications and many secrets could be effected by contract even if they could not be protected in the absence of contract.

I cannot see, however, why resorting to contract should be necessary. It makes more sense to say that if government can ban messages that disclose confidences, secrets, or copyrighted materials as long as there is an agreement not to disclose, it can also ban such messages if its reasons for doing so are as important as those for enforcing contracts. In other words, although enforcement of contracts may be an interest sufficiently weighty to justify suppressing messages, so too might be protecting confidences and secrecy.^[119]

In the end, Rubenfeld does not deem illegitimate all government actions that implicate principle (5) by suppressing messages that cause one-step harms. His indictment of copyright is marginal. And he is silent about the legitimacy of protecting confidentiality, secrecy, privacy, emotional well-being, contrac­tual obligations, fair trade, and other one-step harms through suppression of messages.^[120]

Many will find it implausible that government can never forbid breaches of confidentiality, revelations of private or secret matters, unauthorized publication of owned content, and so forth. At the very least they would maintain, the value of speech that imposes such harms must be balanced against, and may be outweighed by, the value of preventing those harms. But we must keep in mind that at the core of freedom of expression lies the principle of evaluative neutrality. And that principle deprives us of a scale according to which we can conclude that the harm, say, to client confidentiality is or is not outweighed by the attorney’s revelation of the client’s confidences. Unless government can take a position on what is true, right, and good, it cannot perform such a balancing of interests, either on a retail (case by case) or wholesale basis. Evaluative neutrality, however, precludes its taking such a position.

My conclusion, then, is that content regulations premised on the direct harms that specific messages cause present us with a stark trilemma: Either we must deem the laws protecting confidential communications, privacy, copyright, and so forth always or never to be violative of freedom of expression, or we must permit government officials to declare what is true and valuable in the realm of ideas and information. None of the choices seems acceptable, but no fourth way is apparent.

II.