• 9 •
CLINTON FEIN and the ACLU

Clinton Fein felt betrayed. He thought he had been promised freedom of expression. Instead he got the Communications Decency Act.

Born in South Africa, Fein grew up under Apartheid. As he graduated from the University of Witwatersrand, he was interested in journalism. But the South Africa of the time was not a promising environment for young journalists. Under the repressive censorship regime, one could be imprisoned for quoting Nelson Mandela, who was then in prison for revolutionary activities. Fein emigrated to the United States, studied for citizenship, learned all about the Constitution, and was naturalized in 1994. When he took the oath, he was bemused by the seeming contradiction in having to swear to God to defend a constitution that prohibited any religious coercion. But his primary allegiance, and hope, lay with the First Amendment’s free-speech clause. Then Congress enacted, and President Bill Clinton signed, the Communications Decency Act of 1996. As a result, because his speech was online and subject to the Act, Fein would have had freer speech in South Africa, which by that time was post-Apartheid, than he had in the United States. Fein’s sense of betrayal propelled him to the federal courthouse.

The act, called the CDA, was the most sweeping restriction of the speech of ordinary citizens that Congress had ever attempted. It was more threatening to the average person than the Sedition Act of 1798, in that it was global in its reach and less clear about what was prohibited. It had two main provisions. One made it a federal crime to communicate anything “indecent” online, knowing that the communication was to a minor (17 or under). The term “indecent” was not defined at all. The other made it a crime to “display” on the Internet anything “patently offensive” if it was “available” to a minor (as virtually everything is). Offensiveness would be judged by contemporary “community standards” similar to the way the FCC defined indecency for broadcast purposes. Violation of either provision was punishable by two years in prison and a fine.

The stated purpose of the CDA was to shield children from exposure to indecent online material. Every politician wants to protect America’s children from pornography. Few politicians are willing to invoke the First Amendment to defend smut peddlers. The CDA was an act of legislative cowardice. Any modestly intelligent Congressperson had to know that the clumsily drafted CDA was unconstitutional. Passing the law meant dumping responsibility for saying so on the courts.

The CDA was promptly challenged by the ACLU and other organizations. The federal district court in Philadelphia issued a preliminary injunction preventing the government from enforcing the law during the lawsuit. The court thoroughly examined the nature of the Internet and online communications, and found that the CDA would criminalize a vast range of speech that Americans, at least adults, had a right to engage in. For example, if “indecent” means the same thing as it does in the FCC broadcast regulations (Chapter 8), using any of George Carlin’s dirty words in an email to a 17-year-old or on any Web site “available” to a minor would be a felony. Posting online the Supreme Court’s decision in the Pacifica case would be a felony because the Court appended a transcript of the Carlin monologue to it. “Displaying” online great works of art such as Tony Kushner’s Pulitzer Prize–winning play Angels in America, which has plenty of rough language and sex scenes, would be criminal. So would posting the Attorney General’s 1986 Report on Pornography, which described pornographic materials in some detail. Magazines like Wired that appeared both in print and online and used some of Carlin’s words would be protected on the newsstand but subject to criminal prosecution when editors pressed the button to post an article online. And because offensiveness would be judged by “community” standards, only material bland enough to pass muster in the most conservative, least tolerant community could safely be communicated.

At the time the CDA was enacted, self-taught computer whiz Clinton Fein’s day job was constructing and maintaining Web sites. But he also had begun publishing edgy, provocative material that attracted the government’s attention. He made a CD-ROM of Randy Shilts’s book Conduct Unbecoming: Gays and Lesbians in the U.S. Military. It included material about a sailor pictured in a Navy recruiting poster who turned out to be gay and was discharged by the Navy. The Navy claimed that Fein could not legally use the picture and said it was protected by trademark law. Fein managed to find help from Michael Traynor, a prominent San Francisco lawyer, who wrote to the Navy that it could not prevent Fein from using the image on trademark or any other legal ground, and the Navy gave up. At the time, Traynor and I were representing Wired magazine in various First Amendment–related matters, and we were concerned about the CDA’s impact. We met with Fein, who wanted to challenge the CDA in any way he could. So we helped him have his day in court.

Our main attack was on a provision of the CDA different from the ones in the ACLU case. It was an update of the old federal “obscene telephone call” statute. As part of the CDA, Congress amended the law to make it a crime, punishable by two years in prison and a fine, to say anything “obscene, lewd, lascivious, filthy, or indecent” using a telecommunications device (defined to include both telephones and computer modems) with intent to “annoy” another person. Fein himself was annoyed, and wanted to annoy the politicians and the prudes who egged them on. So he launched a Web site called annoy.com, which he announced was its own CDA: “Created and Designed to Annoy.”

Annoy.com had several innovative features, using cuttingedge technology. One was “heckle,” which enabled site visitors to construct and send email messages, anonymously, to various public officials and public figures. It operated like the popular MadLibs game and invited visitors to criticize the recipients in the most vulgar (and hilarious) terms. Most of the messages predictably would “annoy” the recipients. For example, one could compose a letter to speaker of the house Newt Gingrich. It might say “Dear Speaker Gingrich, Your Contract with America is [brilliant/a pile of shit]. It will [save/ruin] America. It will benefit [all Americans/only the rich]. You are truly [a genius/an asshole].”

Another feature was “censure,” which allowed visitors to send digital postcards via the Internet. Hallmark greetings these were not. No floral designs or cuddly animals. Some of the cards were sexually explicit and all were provocative. “Gibe” was a message board on which visitors could post messages anonymously. Fein and other writers contributed essays containing language and ideas that might well be considered “indecent” in some communities. The issues they discussed included same-sex marriage, gays in the military, health care, and the environment. Fein was not shy about announcing that the site made it possible for visitors, using “whatever language or imagery seems to them appropriate” to “annoy” public officials and public figures “by getting their attention, upsetting them, and making them understand the depth of displeasure with their acts or political positions.” For example, Fein added, “Some of us wish to criticize President Clinton, Speaker Newt Gingrich, Senators Dianne Feinstein, James Exon, and Jesse Helms,” and others “because of their role in proposing, enacting, and approving the Communications Decency Act.”

The problem with annoy.com was that, if the CDA was constitutional, Fein was subject to criminal prosecution for violating both the statutory prohibition on posting “indecency” and the online communication of “indecent” material “with intent to annoy.” So we mounted a two-pronged attack. First, we filed a brief amicus curiae in the Supreme Court in the ACLU’s suit. There, we attacked the constitutionality and, indeed, the rationality of the CDA. We made some arguments not made by the ACLU and other parties. We questioned, for example, whether the federal government had the authority under the Constitution to regulate content on the Internet at all and whether the government had a “compelling” interest in meddling in what is first and foremost the domain of parents, not government: what speech their children would have access to. Second, we filed our own suit against attorney general Janet Reno in federal court in San Francisco, seeking to enjoin her from enforcing the “intent to annoy” law. (Reno was my law school classmate; I find it hard to believe that she could endorse the arguments that her underlings made on her behalf to support the CDA.)

Our case was much stronger than the ACLU’s because the statute we were challenging had nothing to do with children. The government could not defend it on the ground that it was needed to protect children from indecency. The statute was a flat-out ban on speech that adults have every right to communicate. The fact that a communication was made with intent “to annoy” the recipient could not save the law. As Hustler v. Falwell and the Jehovah’s Witnesses cases amply demonstrate, government cannot ban speech on the ground that it is offensive, motivated by hatred, or upsetting to the recipient.

The government repeatedly requested more time to respond to our suit and pointed out that the ACLU’s case, then pending in the Supreme Court, would likely affect the issues in our case. The government insisted, however, on its right to prosecute Fein for any violations of the CDA provision criminalizing communications with intent to “annoy” that might be committed while the case was delayed. The three-judge court in San Francisco deferred any ruling until Reno v. ACLU came down, leaving Fein at risk if he continued to operate annoy.com. He continued.

On June 26, 1997, the Supreme Court decided the ACLU case. The decision was wonderful, a true landmark of First Amendment freedom. The opinion was written by Justice John Paul Stevens, who thereby redeemed himself from his Pacifica blunder discussed in Chapter 8. He now can be considered for candidacy in the pantheon of First Amendment heroes.

Reno v. ACLU was the Court’s first Internet case. This was a new medium of communication, and no one could know how the Court might apply First Amendment rules to it. In the past, when confronted with having to decide how the First Amendment applies to a given medium, the Court had sometimes fumbled. For example, when movies were invented and became commercial, the Court in 1915 allowed cities to censor them at will, reasoning that movies were just a business like any other, subject to local regulation, and were not protected by speech or press guarantees. (That ruling was not overruled until 1952.) Also, when the Court had to decide the extent to which cable television could be regulated, the justices were unable to agree on the proper First Amendment analysis.

Stevens got it right in the ACLU case. His opinion first laid out in detail how the Internet works and the various ways in which people can communicate on it. Seventy-seven years old at the time, Stevens casually noted that “navigating the Web is relatively straightforward,” and then proceeded to explain it in commonsense terms. He had to confront the government’s argument that the Internet should be treated like broadcast, subject to government regulation similar to the FCC’s regulation of radio and broadcast television. Indeed, the government urged that “the indecency problem” on the Internet was “much more pronounced than it is on broadcast stations.” This was because, Attorney General Reno’s lieutenants said, the Internet operates “without the intervention of editors, network censors, or market disincentives.” In other words, the government argued that because ordinary citizens could communicate with each other directly, not as the passive recipients of programming directed at them by powerful corporations and their commercial sponsors, the government must step in to police “indecent” material. (The perverse result would be that the more democratic the medium—free of corporate censors and commercial “disincentives”—the greater the government’s right to regulate the medium.)

Stevens demolished the government’s argument, pointing out that the broadcast decisions, including his own Pacifica case, relied on three unique characteristics of the broadcast medium: the scarcity of frequencies on the electromagnetic spectrum, the long history of government regulation in allocating frequencies and ensuring that licensees acted in the public interest, and the “invasive” nature of the medium. “Those factors are not present in cyberspace,” Stevens said. Agreeing with the lower court that “content on the Internet is as diverse as human thought,” Stevens concluded that there is “no basis for qualifying the level of First Amendment scrutiny that should be applied to this medium.” Thus, Internet speech is at least as free as speech in newspapers, in books, or on soapboxes. For government to restrict the content of online communication, it would have to prove that the restriction serves a “compelling” government interest and is narrowly tailored to serve that interest so that no less restrictive means would do.

Justice Stevens then turned to the ambiguities in the CDA statute and how they endangered free speech. He pointed to the use of “indecent” in one provision and “patently offensive” in the other, noting the lack of definition for either. These ambiguities, Stevens said, would

provoke uncertainty among speakers about how the two standards relate to each other and just what they mean. Could a speaker confidently assume that a serious discussion about birth control practices, homosexuality, the First Amendment issues raised by the Appendix to our Pacifica opinion, or the consequences of prison rape would not violate the CDA? This uncertainty undermines the likelihood that the CDA has been carefully tailored to the congressional goal of protecting minors from potentially harmful materials.

The law’s vagueness was troublesome for First Amendment purposes not only because the restrictions on speech were content-based but also because the CDA was a criminal law and, Stevens asserted, “The severity of criminal sanctions may well cause speakers to remain silent rather than communicate even arguably unlawful words, ideas, and images.”

Justice Stevens’s opinion for the Court concluded that “the CDA lacks the precision that the First Amendment requires when a statute regulates the content of speech.” In trying to protect children from exposure to indecent material, the justice explained, “the CDA effectively suppresses a large amount of speech that adults have a constitutional right to receive and to address to one another.” He added that government may not reduce the adult population to “only what is fit for children” and the level of Internet discourse “cannot be limited to that suitable for a sandbox.”

The Court also noted that if “community standards” are used to judge offensiveness, as they are in broadcast and obscenity cases, any Internet communication, available to a nationwide audience, “will be judged by the standards of the community most likely to be offended by the message.” Material at risk would include not only any of the seven dirty words used by George Carlin but also “discussions of prison rape or safe sexual practices, artistic images that include nude subjects, and arguably the card catalogue of the Carnegie Library. Also, a parent allowing her 17-year-old to use the family computer to obtain information that the parent deems appropriate, or a parent who emailed his 17-year-old college freshman information on birth control, would be subject to a lengthy prison term.” Since there are less restrictive alternatives to criminal prosecution for shielding children from exposure to unwanted Internet material (such as parents installing filters, Web sites requiring credit cards for blatant sexual material, etc.), the government failed to justify the CDA’s ham-handed suppression of the Internet speech of ordinary citizens.

Finally, Justice Stevens made short shrift of the government’s last-ditch, desperate contention that, apart from the interest in protecting children, it had an “equally significant” interest in “fostering the growth of the Internet” as a communications medium. The government argued that the easy availability of “indecent” material on the Internet was “driving countless citizens away from the medium because of the risk of exposing themselves or their children to harmful material.”

Justice Stevens responded: “The dramatic expansion of this new marketplace of ideas contradicts the factual basis of this contention ... [as] the growth of the Internet has been and continues to be phenomenal.” More importantly, as a First Amendment matter, he continued: “We presume that governmental regulation of the content of speech is more likely to interfere with the free exchange of ideas than to encourage it. The interest in encouraging freedom of expression in a democratic society outweighs any theoretical but unproven benefit of censorship.” The CDA was thus swiftly interred, and the most democratic means of communication ever invented was allowed to flourish.

In light of Reno v. ACLU, Clinton Fein’s case looked like a slam dunk. The Supreme Court had proclaimed that “indecent” online speech could not be made criminal, and in our case the government did not even have the supposed interest in protecting children to rely on. Instead of graciously conceding that the provision of the CDA outlawing indecent communications made with intent to annoy was unconstitutional, however, Attorney General Reno’s underlings came up with a shameless “interpretation” of the law that would rescue it. For the first time in the litigation, the government said that the law did not mean what it said. It did not in fact outlaw “indecent” communications at all. The government asserted that the provision prohibiting “obscene, lewd, lascivious, filthy, or indecent” communications was limited to “obscene” communications. Therefore, the government contended, because Clinton Fein was not in the business of putting obscene material online, the law did not apply to him, and his case should be dismissed.

The government’s supposed “interpretation” of the statute was ridiculous. It robbed “indecent” and all the other terms of independent meaning. It was as if Congress had enacted the Communications Obscenity Act instead of the Communications Decency Act. It ignored the fact that the CDA used “indecent” as contrasted with “obscene” in other provisions of the same law (the provisions struck down in the ACLU case). As in Pacifica and many other decisions a clear distinction had always been made between “obscene” and “indecent” speech, and they had different meanings. Further, the government’s creative exercise in statutory interpretation made the law redundant of all the other federal obscenity laws. Obscenity already was prohibited, in any medium, so the government’s interpretation would render the statute meaningless.

Alas, two of the three district court judges agreed with the government. They reasoned that a court should construe a law in a way to avoid declaring it unconstitutional (a judicial restraint principle rejected in the 2010 Citizens United case). They referred to a few older cases that construed a “string of words” similar to those in the statute in question to mean “obscene,” and they said cases like the ACLU case and Pacifica were somehow different because they involved attempts to protect minors. Their reasoning did not make any sense, but Reno’s interpretation gave the judges a convenient way to avoid sticking their necks out and forthrightly declaring the law unconstitutional. The judges concluded that because Fein did not challenge the prohibition of obscenity, his case would be dismissed.

Judge Susan Illston dissented. She pointed out that the law applied to the Internet, a communications medium with hundreds of millions of users, and a criminal law like this “should mean exactly what it says, so that users will know what the rules are.” She believed that, following the decision in ACLU, the criminal prohibition of “indecent” speech violated the First Amendment.

Since the district court decision applied only in northern California, and Fein’s communications were available throughout the nation, he remained at risk of prosecution in other districts, where courts might read the law to mean what it said, covering nonobscene material. Fein and all Internet users would live under the uncertain cloud of a law that on its face made “indecent” annoying communications a felony. So we took the case on a direct appeal to the Supreme Court. On April 19, 1999, the Court handed down what had to be the shortest First Amendment decision ever made: “The judgment is affirmed.” No explanation. This outcome was terribly disappointing and a waste of several years of litigation during which the government had never suggested construing the law to prohibit only obscenity.

The decision had a sliver of a silver lining. The Supreme Court’s action had the effect of making the district court’s interpretation binding nationally. It was now the law of the land. In practical terms, therefore, the litigation gave Clinton Fein what he sought: the ability to communicate nonobscene material that might be considered indecent in some communities and to annoy the recipients without fear of criminal prosecution. He deserved but did not get a ringing affirmation that the CDA provision that threatened annoy.com violated the First Amendment.

Congress could not resist the temptation to exploit the CDA’s demise politically by trying again to shield America’s children from sexual material on the Internet. It promptly passed a new law, which we called “Son of CDA.” The law’s proper name was the Child Online Protection Act, or COPA. Congress attempted to patch up some of the constitutional deficiencies that Justice Stevens had identified in the CDA. It made three main changes from the CDA: COPA outlawed only communications made “for commercial purposes”; it was limited to communications on the World Wide Web, as opposed to the entire Internet including email; and instead of proscribing “indecent” material, it criminalized material that is “harmful to minors.” Included was a laughable definition of “harmful to minors.” In a parody of the three-part Miller definition of obscenity, COPA simply tacked on to each of the three parts something about minors. So the material must appeal to the prurient interest “with respect to minors,” must be patently offensive under community standards “with respect to minors,” and must lack literary or other societal value “for minors.” If the Miller obscenity definition was subjective and vague, the COPA statute compounded these vices. Adding in all the references to minors really would have reduced Internet discourse to the lowest common denominator, because if material might not be suitable for a toddler—even though unremarkable for a sophisticated 17-year-old—it would be illegal, since both are minors.

Once again the ACLU sued the very day the law was to take effect. Once again the lower courts condemned the law. The reach of COPA admittedly was narrower than the CDA. Congress may have been targeting commercial pornographers, but the law it wrote was much broader. It still prohibited speech that is protected among adults, and it still provided no clear guidance to Web communicators who wished to avoid crossing the invisible line into criminal behavior. The court of appeals was especially concerned about COPA’s use of “community standards” for judging the offensiveness of material harmful to minors, which was lifted from the Miller test. The court noted that material posted on the Web “is accessible by all Internet users worldwide,” and Web publishers cannot restrict access to their sites based on where the users are. Therefore, the court said, COPA requires every Web publisher to “abide by the most restrictive and conservative state’s community standards” to avoid criminal liability. The court of appeals threw out COPA on that ground alone.

The government of course appealed to the Supreme Court. On May 13, 2002, the Court vacated the court of appeals’ decision. The Court’s opinion was written by Justice Clarence Thomas, no friend of the First Amendment. The Court decided that the use of community standards to identify the prohibited material did not by itself make COPA unconstitutional. Justice Thomas pointed out that COPA applied to less material than the CDA and its definition of the forbidden material was narrower because it required the government to prove appeal to the prurient interest and no redeeming social value for minors. Responding to the contention that Web publishers could not control where their material might be accessed, Justice Thomas gave them some blunt advice: “If a publisher wishes for its material to be judged only by the standards of particular communities, then it need only take the simple step of utilizing a medium that enables it to target the release of its material into those communities.” In other words, Web publishers shouldn’t be Web publishers; they should go into the newspaper, magazine, or radio business. (This suggestion contradicts what the Court held in the CDA case when the government made a similar argument and Justice Stevens responded that this was “equivalent to arguing that a statute could ban leaflets on certain subjects as long as individuals are free to publish books”; where a restriction on speech is content-based, it is no answer to say that some other medium might be used.)

Justice Thomas candidly confessed the real reason for vacating the lower court’s condemnation of COPA: “If we were to hold COPA unconstitutional because of its use of community standards, federal obscenity statutes would likely also be unconstitutional as applied to the Web.” In other words, obscenity would be allowed on the Web but not in any other medium, and the Court could not countenance that result. Having so concluded, the Court sent the case back to the court of appeals for another look, to consider whether COPA was unconstitutional for some reason other than its invocation of community standards.

The court of appeals got the message and found that COPA violated the First Amendment because of its vagueness and overbreadth. The government of course appealed again. On June 29, 2004, the Supreme Court upheld the lower court’s grant of a temporary injunction against enforcing COPA. This time the opinion was authored by Justice Anthony Kennedy, who perhaps surprisingly is quite consistently sensitive to First Amendment values. He began by noting that “content-based prohibitions, enforced by severe criminal penalties, have the constant potential to be a repressive force in the lives and thoughts of a free people,” that these restrictions on speech are “presumed invalid,” and the government has the burden of demonstrating their constitutionality. Justice Kennedy concluded that the government had not shown that less restrictive alternatives to criminal prosecution—like filtering software installed by concerned parents—could be as effective in protecting children from exposure to potentially harmful material. The court remanded the case to the lower courts for further “proceedings” to determine whether the temporary injunction should be made permanent.

Justice Stevens concurred, reaffirming his belief that “Government may not penalize speakers for making available to the general World Wide Web audience that which the least tolerant communities in America deem unfit for their children’s consumption.” Having struggled with restrictions on sexual speech in several cases over the years, he also took the occasion to venture his conviction that criminal prosecutions are an inappropriate means of regulating even obscene material, as the line between communications that “offend” and those that do not “is too blurred to identify criminal conduct.”

Justice Stephen Breyer, who perhaps surprisingly is too apt to compromise on First Amendment issues, dissented. He thought filtering software was faulty and allowed some pornographic material to pass through without hindrance: “The software alone cannot distinguish between the most obscene pictorial image and the Venus de Milo.” He also despaired that “After eight years of legislative effort, two statutes, and three Supreme Court cases, the Court sends this case back to the district court for further proceedings.” Breyer thought Congress had tried hard enough to repair the defects in the CDA, and the COPA was good enough. He threw up his hands and said, literally, “What else was Congress supposed to do?”

Back in the district court in Pennsylvania, the judge conducted a trial, giving the government the opportunity to show that there were no less restrictive alternatives to criminal prosecution under COPA that could protect children. Judge Lowell Reed found, however, that filters were more effective than COPA’s criminal prohibitions. He pointed out that perhaps half of sexually explicit Web sites are foreign in origin and that it is practically impossible to prosecute them in the United States, while filters can screen at least some of their material. He found COPA both under-inclusive (in not being able to do anything about foreign Web sites) and over-inclusive (in deeming material unlawful if it might be harmful for a five-year-old even though it would be unobjectionable for a sophisticated teenager), and therefore not “narrowly tailored” to serve the interest in protecting children. Judge Reed concluded that “I may not turn a blind eye to the law in order to satisfy my urge to protect this nation’s youth by upholding a flawed statute.... Perhaps we do the minors of this country harm if First Amendment protections, which they will with age inherit fully, are chipped away in the name of their protection.”

The government of course appealed again. It argued that the First Amendment “does not prevent Congress from adopting a ‘belt-and-suspenders’ approach ... with filters acting as the ‘belt’ and COPA [criminal prosecution] as the ‘suspenders.’” The court responded: “Under the First Amendment, if the belt works at least as effectively as the suspenders, then the government cannot prosecute people for not wearing suspenders.”

By this time the government was bring represented by attorney general Michael Mukasey (President George W. Bush’s successor to Alberto Gonzales), who tried to take the case to the Supreme Court for the third time. On January 21, 2009, the Court denied review. COPA was dead. It, like CDA, never took effect.

Congress made another attempt to restrict sexual content on the Internet. The Child Pornography Prevention Act of 1996 (CPPA) amended the federal child pornography law to make it a crime to possess or distribute pictures not just of real children but of what “appear to be” minors in sexual poses. Congress said that computer-generated images, even those that do not involve real children at all in their production, should be prohibited because such pictures might be used to “whet the appetite” of pedophiles or be used by child abusers to seduce their victims. This was a new rationale for outlawing child pornography. The Supreme Court had approved laws outlawing pictures of minors performing sex acts because of the child abuse inherent in the production of the pictures, without requiring that the pictures themselves be legally obscene.

The CPPA was challenged by a group of “adult” entertainment businesses, a publisher of nudist books, a painter of nudes, and a photographer specializing in erotic images, who called themselves the Free Speech Coalition. They won in the lower court, and of course the government took the case to the Supreme Court. I wrote an amicus curiae brief for the ACLU and several other organizations. Justice Kennedy wrote a courageous opinion, reaffirming vital First Amendment principles in an unsavory context. One of the problems with CPPA, Justice Kennedy found, was that it outlawed not just computer-generated images but also “a renaissance painting depicting a scene from classical mythology” as well as Hollywood movies like Lolita, Traffic, and American Beauty, in which young-looking adult actors (who appear to be minors) are seen engaging in sexual acts. It also covered pictures in a psychology manual and a documentary “depicting the horrors of sexual abuse.” In other words, the CPPA outlawed a range of images of significant societal value even though no real children were used in their production.

Justice Kennedy acknowledged that “sexual abuse of a child is a most serious crime,” and Congress has the power to protect children from abuse, but he contended, “The prospect of crime ... by itself does not justify laws suppressing protected speech.” Repeating a lesson learned in Hustler v. Falwell, Kennedy said: “Speech may not be prohibited because it concerns subjects offending our sensibilities.” The justice was especially troubled that, in the absence of the Miller obscenity requirement—material must lack literary, artistic, or other value to be deemed obscene—works of art and movies could be condemned “without inquiry into the work’s redeeming value.” Moreover, he said, the CPPA prohibits “the visual depiction of an idea—that of teenagers engaging in sexual activity—that is a fact of modern society and has been a theme of art and literature throughout the ages.”

As for the government’s argument that even virtual images could lead to actual child abuse, Justice Kennedy first noted that “the causal link is contingent and indirect.” That is, the connection between viewing child porn and actually molesting kids is not clear. (Indeed, the social science research does not establish a causal link between viewing any kind of pornography and sexual crime.) Also, cartoons, video games, and candy might be used to seduce children, yet Kennedy points out: “We would not expect those to be prohibited because they can be misused.” More fundamentally, the government’s “whet-the-appetite” contention runs afoul of the First Amendment principle that “the mere tendency of speech to encourage unlawful acts is not a sufficient reason for banning it.” The CPPA was basically aimed at the evil of planting bad ideas and impure thoughts, but government is not allowed to control thought. Justice Kennedy reminded everyone of Brandenburg v. Ohio: even “advocacy” of illegal conduct and hateful ideas is protected unless it amounts to incitement of imminent lawless acts. The CPPA, like the CDA and the COPA, was dead.

Congress finally found limited success in attempting to regulate sexual content on the Internet when, on June 23, 2003, the Supreme Court upheld a fourth law, the Children’s Internet Protection Act (CIPA). The law denies federal funding to public libraries unless they have a “policy of Internet safety for minors” that uses software filters to prevent access by anyone to “visual depictions” that are “harmful to minors.” The statute, sponsored by Senator John McCain, does not define any of these terms.

Chief Justice Rehnquist wrote the Court’s opinion in the ominously named case, United States v. American Library Association. The Court emphasized that this law, unlike CDA and COPA, did not impose any criminal sanctions and, importantly: “Congress has wide latitude to attach conditions to the receipt of federal assistance.” In other words, the government is not required to fund libraries and can put strings on the subsidies it chooses to make. Rehnquist stated that libraries are free to offer unfiltered access “without federal assistance” and CIPA “is a valid exercise of Congress’s spending power.”

The Court understood that filters notoriously over-block, screening out material that is completely innocuous as well as valuable, such as the Carlin monologue or the syllabus for a university course on sex education. Asked at the oral argument whether the law would prevent adult users of library computers from access to such material, the government lawyer told the Court that a librarian would unblock any given site or disable the filter at the request of an adult user. On the basis of this unrealistically optimistic representation, the Court concluded that adult access to the full resources of the Internet would not be unduly hindered. Two justices concurred in the decision only because of this representation, making a majority upholding the law. Justice Stevens, in dissent, was unpersuaded by the government’s assurance and said the law was tantamount to requiring a library’s materials to be kept in “unmarked, locked rooms or cabinets, which could be opened only in response to specific requests.” More broadly, Stevens reasoned that a law’s abridgement of speech “is equally obnoxious” whether it is enforced by “a threat of penalties or by a threat to withhold a benefit.”

Whether software filters solve First Amendment Internet problems is an amusing diversion. In the library case, the government urged that filters are effective in protecting children from exposure to harmful material, while civil libertarians argued that they over-block and under-block, and that government has no business policing libraries anyway. In the CDA and COPA cases, the roles were reversed. The ACLU argued that filters were effective, and that children could be protected by less restrictive means than criminal prosecutions, while the government urged that filters were ineffective and the criminal sanction was needed. The seeming inconsistency was nicely resolved by Ann Beeson of the ACLU, who said the central issue is whose finger is on the mouse, the government’s or the parent’s: “When a parent installs a filter that keeps a kid from seeing a bunch of sites that may or may not be pornography, that’s parenting. When a government forces all adults and minors to use filters, that’s censorship.”

The broader issue in all this is whether the courts should accept the government’s assertion that it has a “compelling” interest in protecting minors from being exposed to material that it deems harmful. So far, the courts have uncritically accepted this proposition in case after case and then have wrestled with whether the particular speech restriction is “narrowly tailored” to serve the asserted interest. They have done so without any evidence that the supposedly harmful material actually causes any harm. That is, the government has not been required to prove that minors are in fact injured in any way by hearing dirty words or seeing sexual images. This has allowed politicians to enact laws that restrict everyone’s speech in the name of protecting children. This cause is seductive and dangerous because the principle has no limits. If we uncritically accept that the government has a compelling interest in shielding children from various kinds of speech that it deems harmful, there is nothing to prevent legislators from enacting a Newspaper Decency Act, a Literature Decency Act, or laws banning profanity, Facebook, or bad grammar.

Having survived the CDA and the COPA, Clinton Fein continued to annoy various people with his digital muckraking and provocative images. Annoy.com had another run-in with the Justice Department. Unbeknownst to Fein, someone used annoy.com’s postcard service anonymously to send a threatening message to a university administrator in Houston. Out of the blue, without notice or opportunity to be heard, Fein was served with a court order issued by a federal magistrate in Houston, commanding the site to disclose the identity of the user. The order also prohibited Fein from disclosing the existence of the order to the user “or to any other person.” Fein responded by informing the government that he had no information about the identity of the anonymous user and by moving to quash the “gag” provision as an unconstitutional prior restraint on his free speech. The lower court refused to lift the nondisclosure order so that Fein could discuss publicly the propriety of the government’s obtaining secret orders demanding that Web sites surrender user information. We had to appeal to the Fifth Circuit Court of Appeals. That court noted that the law the government had invoked authorized nondisclosure orders only to protect life or physical safety and only for a limited period, and it sent the case back to the lower court to consider the “substantial constitutional questions” raised by its order. The lower court then vacated the order and, of course, annoy.com covered the controversy.

Clinton Fein’s other brush with the legal system found the ACLU, long the guardian of First Amendment freedom, not in a heroic role. In an employment discrimination suit against Avis Rent a Car at the San Francisco airport, three Latino workers proved that they had been subjected to verbal harassment by one supervisor. By way of relief, a California Court of Appeal directed that the trial court compile a list of “prohibited epithets,” the use of which the company would be enjoined from allowing in the workplace. The record in the case did not disclose what disparaging names the workers had been called or the context in which they were used, but the court required the promulgation of a list of government-forbidden words that could not be uttered in the workplace.

Momentarily blinded by some notion of political correctness, the ACLU supported the order and argued to the California Supreme Court that it should be affirmed. It said that a proven pattern of workplace hostility justified an injunction against using government-prescribed bad words in the future. That is, a competing value—the need to remedy employment discrimination—trumped free-speech considerations. Clinton Fein, ever alert to diminution of First Amendment freedoms, disagreed. So we tendered an amicus curiae brief to the California Supreme Court.

We argued that “The First Amendment does not say government ‘shall make no law abridging the freedom of speech, except to protect certain persons from ridicule and insult in the workplace.’” We urged that words in the abstract cannot be made unlawful or enjoined. Meaning depends on context. For example, to take the most famous racial slur, uttering the word “nigger” may be a sign of friendship and intimacy, or of bigotry and hate, depending on the context—to whom it is said, by whom, in what tone of voice, and in what circumstances. As Supreme Court Justice John Marshall Harlan once said: “One man’s vulgarity is another’s lyric.” The same is true of whatever epithets would be on the court’s list. If “spic” were on the list, the injunction would prohibit both screaming at a Latino subordinate: “You fucking lazy spic!” and saying discreetly to a peer: “I believe it is a violation of the First Amendment for a court to issue an injunction forbidding me from saying ‘spic’ to anyone.” To say that there is no legal difference between these two is to say that context is irrelevant and that government can outlaw the mere utterance of disfavored words.

Even though courts may award damages to victims of workplace discrimination who have been subjected to abusive language, that does not mean, we argued, that a court can issue an injunction against future offensive speech. A government order that specific speech may not be uttered would be an impermissible prior restraint on speech. Orders prohibiting speech in advance have always been constitutionally suspect. The point is that speech is protected until it can be shown to have done actual harm. One of the chief purposes of the First Amendment is to prevent prior restraints. They can be granted only in the most exceptional circumstances, such as immediate threats to the national security. In the Pentagon Papers case, the U.S. Supreme Court found that the government had failed to prove sufficient harm to national security to justify a prior restraint.

Alas, the California Supreme Court, doubtless lulled by assurances from none other than the First Amendment champion ACLU that it should not worry about free-speech considerations in remedying employment discrimination, and quoting the ACLU brief, approved the concept of an injunction against uttering any “prohibited epithets.” However, the court ducked the issue of whether courts should compile lists of forbidden words, saying that since the parties had not “sought review” of that issue, the court would express “no opinion” on it. The court left it to the lower court to fashion an order telling employees what they are permitted to say about other employees.

Undaunted by his encounters with the legal system, Clinton Fein continues to rail against politicians and powerful corporations, using both annoy.com and art. He has become a digital artist of some renown and is a direct beneficiary of Hustler v. Falwell and its protection for offensive satire. Working with a computer, Photoshop software, and a digital camera, Fein has created political “cartoons,” collages, and other images that offend their targets and many others. His work has been widely published and has appeared in art galleries, but not without controversy. The headline of a review in the San Francisco Chronicle states plainly: “His art is not subtle. It can be hard to take. But Clinton Fein is not afraid to make a statement.” Shortly after the 9/11 attacks, Fein had a show at a San Francisco gallery. The magazine Artforum refused to run an advertisement for the show. The ad, one of Fein’s images, showed former New York mayor Rudy Giuliani sitting naked in a urine-filled glass (reminiscent of a Damien Hirst work), referencing Giuliani’s retaliatory withholding of funding for the Brooklyn Museum.

Fein savagely attacked the George W. Bush administration at every opportunity, especially the invasion of Iraq and the Abu Ghraib torture. One of his images shows the president nailed to the cross crucifixion-style, with an erection popping out of his loincloth in the shape of a rocket, under the banner “Who Would Jesus Torture?” Another looks to a casual viewer like an American flag, but in the background of each stripe is the text of the Taguba Report on the Abu Ghraib torture, and the “stars” are whitened images of the iconic photograph of the hooded and wired torturee. For an art gallery show, Fein sent the images to a large format printer in Palo Alto, Zazzle. The printer found the images offensive and destroyed them. This sort of censorship does not violate the First Amendment, of course, because only the government is bound by and can violate the Constitution.

Nor does it dampen Fein’s willingness fully to exercise the freedom of expression that brought him to America. His imagery is sometimes sexual and includes anuses, erect penises, and Carlinesque words describing them. He says his mother “wishes [he] would do watercolors.” But his work is not pornographic and certainly not obscene, even though right-wing groups call him a “noted homosexual pornographer.” His work does not appeal to the prurient interest, and it clearly has political and artistic value. A First Amendment true believer, he indefatigably tries to expose hypocrisy, skewer the mainstream media for laziness and cowardice, and provoke debate over accepted wisdom. But this defender of online freedom also laments young people’s willingness to give up any semblance of privacy on Facebook and other social networking sites, so that “Nothing [is] left for anyone to discover about you. Government doesn’t have to spy; they can just join Facebook.”