This spring, talk-show host G. Gordon Liddy, speaking on the radio to millions of people, explained how to shoot agents of the Bureau of Alcohol, Tobacco, and Firearms: “Head shots, head shots… Kill the sons of bitches.” Later he said, “Shoot twice to the belly and if that does not work, shoot to the groin area.”
On March 23 the full text of the Terrorist’s Handbook was posted on the Internet, including instructions on how to make a bomb (the same bomb, as it happens, that was used in Oklahoma City). By the time of the Oklahoma bombing on April 19, three more people had posted bomb-making instructions, which could also be found on the Internet in the Anarchist’s Cookbook. On the National Rifle Association’s Internet “Bullet ‘N’ Board,” someone calling himself “Warmaster” explained how to make bombs using baby-food jars. Warmaster wrote, “These simple, powerful bombs are not very well known, even though all the materials can be easily obtained by anyone (including minors).” After the Oklahoma bombing, an anonymous notice was posted to dozens of Usenet news groups, listing all the materials in the Oklahoma City bomb, explaining why the bomb allegedly did not fully explode, and exploring how to improve future bombs.
Fifty hate groups are reported to be communicating on the Internet, sometimes about conspiracies and (by now this will come as no surprise) formulas for making bombs. On shortwave radio, people talk about bizarre United Nations plots and urge that “the American people ought to go there bodily, rip down the United Nations building and kick those bastards right off our soil.” A few months ago Rush Limbaugh, who does not advocate violence, said to his audience, “The second violent American revolution is just about, I got my fingers about a fourth of a inch apart, is just about that far away. Because these people are sick and tired of a bunch of bureaucrats in Washington driving into town and telling them what they can and can’t do.”
In the wake of the tragedy in Oklahoma City, a national debate has erupted about speech counseling violence or inciting hatred of public officials. Of course, we do not know whether such speech had any causal role in the Oklahoma City bombing. But new technologies have put the problem of incendiary speech into sharp relief. It is likely, perhaps inevitable, that hateful and violent messages carried over the airwaves and the Internet will someday, somewhere, be responsible for acts of violence. This is simply a statement of probability; it is not an excuse for violence. Is that probability grounds for restricting such speech? Would restrictions on speech advocating violence or showing how to engage in violent acts be acceptable under the First Amendment? Aside from legal restrictions, what measures are available to the nation’s leaders and private citizens to discourage incendiary hate and promote the interests of mutual respect and civility?
THE LIMITS OF PROTECTED SPEECH
Recent events should not be a pretext for allowing the government to control political dissent, including extremist speech and legitimate hyperbole. But narrow restrictions on speech that expressly advocates illegal, murderous violence in messages to mass audiences probably should not be taken to offend the First Amendment.
For most of American history, the courts held that no one has a right to advocate violations of the law. They ruled that advocacy of crime is wholly outside of the First Amendment–akin to a criminal attempt and punishable as such. Indeed, many of the judges revered as the strongest champions of free speech believed that express advocacy of crime was punishable. Judge Learned Hand, in his great 1917 opinion in Masses v. United States, established himself as a true hero of free speech by saying that even dangerous dissident speech was generally protected against government regulation. But Hand himself conceded that government could regulate any speaker who would “counsel or advise a man” to commit an unlawful act.
In the same period the Supreme Court concluded that government could punish all speech, including advocacy of illegality, that had a “tendency” to encourage illegality. Justices Holmes and Brandeis, the dissenters from this pro-censorship conclusion, took a different approach, saying that speech could be subjected to regulation only if it was likely to produce imminent harm; thus they originated the famous “clear and present danger” test. But even Holmes and Brandeis suggested that the government could punish speakers who had the explicit intention of encouraging crime.
For many years thereafter, the Supreme Court tried to distinguish between speech that was meant as a contribution to democratic deliberation and speech that was designed to encourage illegality. The former was protected; the latter was not. In 1951 the Court concluded in Dennis v. United States that a danger need not be so “clear and present” if the ultimate harm was very grave.
The great break came in the Court’s 1969 decision in Brandenburg v. Ohio. There the Court said the government could not take action against a member of the Ku Klux Klan, who said, among other things, “We’re not a revengent organization, but if our President, our Congress, our Supreme Court, continues to suppress the white, Caucasian race, it’s possible that there might have to be some revengence taken.” The speaker did not explicitly advocate illegal acts or illegal violence. But in its decision, the Court announced a broad principle, ruling that the right to free speech does “not permit a State to forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action.”
Offering extraordinarily broad protection to political dissent, the Court required the government to meet three different criteria to regulate speech. First, the speaker must promote not just any lawless action but “imminent” lawless action. Second, the imminent lawless action must be “likely” to occur. Third, the speaker must intend to produce imminent lawless action (“directed to inciting or producing imminent lawless action”). The Brandenburg test borrows something from Hand and something from Holmes and produces a standard even more protective of speech than either of theirs.
OLD STANDARDS, NEW TECHNOLOGY
Applied straightforwardly, the Brandenburg test seems to protect most speech that can be heard on the airwaves or found on the Internet, and properly so. Remarks like those quoted from Rush Limbaugh unquestionably qualify for protection; such remarks are not likely to incite imminent lawless action, and in any case they are not “directed to” producing such action. They should also qualify as legitimate hyperbole, a category recognized in a 1969 decision allowing a war protester to say, “If they ever make me carry a rifle the first man I want to get in my sights is LBJ.” Even Liddy’s irresponsible statements might receive protection insofar as they could be viewed as unlikely to produce imminent illegality. A high degree of protection and breathing space makes a great deal of sense whenever the speech at issue is political protest, which lies at the core of the First Amendment.
But there is some ambiguity in the Brandenburg test, especially in the context of modern technologies. Suppose that an incendiary speech, expressly advocating illegal violence, is not likely to produce lawlessness in any particular listener or viewer. But of the millions of listeners, one or two, or ten, may well be provoked to act, and perhaps to imminent, illegal violence. Might government ban advocacy of criminal violence in mass communications when it is reasonable to think that one person, or a few, will take action? Brandenburg made a great deal of sense for the somewhat vague speech in question, which was made in a setting where relatively few people were in earshot. But the case offers unclear guidance on the express advocacy of criminal violence via the airwaves or the Internet.
When messages advocating murderous violence flow to large numbers of people, the calculus changes: Government probably should have the authority to stop speakers from expressly advocating the illegal use of force to kill people. There is little democratic value in protecting counsels of murder, and the ordinary Brandenburg requirements might be loosened where the risks are so great. Congress has made it a crime to threaten to assassinate the president, and the Court has cast no doubt on that restriction of speech. It would be a short step, not threatening legitimate public dissent, for the Federal Communications Commission to impose civil sanctions on those who expressly advocate illegal, violent acts aimed at killing people. Courts might well conclude that the government may use its power over the airwaves to ensure that this sort of advocacy does not occur.
Of course, there are serious problems in drawing the line between counsels of violence that should be subject to regulation and those that should not. I suggest that restrictions be limited to express advocacy of unlawful killing because it is the clearest case.
Authorizing the restriction of any speech, even counsels of violent crime, has risks. Government often overreacts to short-term events, and the Oklahoma City tragedy should not be the occasion for an attack on extremist political dissent. Vigorous, even hateful criticism of government is very much at the heart of the right to free speech. Indeed, advocacy of law violation can be an appropriate part of democratic debate. As the example of Martin Luther King, Jr., testifies, there is an honorable tradition of civil disobedience. We should sharply distinguish, however, King’s form of nonviolent civil disobedience from counsels or acts of murder. The government should avoid regulating political opinions, including the advocacy of illegal acts. That principle need not, however, be interpreted to bar the government from restricting advocacy of unlawful killing on the mass media.
THE WIDER DEFENSE OF CIVILITY
What else might be done? First, nothing that I have said suggests that government lacks the power to limit speech containing instructions on how to build weapons of mass destruction. The Branden burg test was designed to protect unpopular points of view from government controls; it does not protect the publication of bomb manuals. Instructions for building bombs are not a point of view, and if government wants to stop the mass dissemination of this material, it should be allowed to do so. A lower court so ruled in a 1979 case involving an article in the Progressive that described how to make a hydrogen bomb, and the court’s argument is even stronger as applied to the speech on the Internet, where so many people can be reached so easily.
Second, the nation’s leaders can do a good deal short of regulation. The president and other public officials should exercise their own rights of free speech to challenge hateful, incendiary speech. Although public officials could abuse these rights so as to chill legitimate protest, President Clinton’s statements about hatred on the radio and the Internet were entirely on the mark. Public disapproval may ultimately have a salutary effect (as it recently did in the case of violent television shows), even without the force of law.
Third, private institutions, such as broadcasting stations, should think carefully about their own civic responsibilities. An owner of a station or a programming manager is under no constitutional obligation to air speakers who encourage illegal violence. Stations that deny airtime for such views do no harm to the First Amendment but on the contrary exercise their own rights, and in just the right way. In recent months, public and private concern about hate-mongering has encouraged some stations to cancel G. Gordon Liddy’s show; this is not a threat to free speech but an exercise of civic duties. Similarly, private on-line networks, such as Prodigy and America Online, have not only a right but a moral obligation to discourage speech that expressly counsels illegal killing.
The advocacy of murder is an extreme version of a far more widespread social practice: treating political opponents, or large groups of people, as dehumanized objects of hatred and fear. Too often people who disagree are portrayed as if their political disagreement is all that they are–as if they are not real human beings who have hopes, fears, and life histories of their own. Too often the individuality of opponents is hidden behind political abstractions–“the government,” “the bureaucrats,” “the liberals,” “the radical right,” “the counterculture.” The seeds of violence lie in these abstractions.
The communications media sometimes help promote violence by turning people into abstractions, but they can also help to reduce violence by telling the stories of individual people. By focusing the nation on the individuals who happened to be in a federal office building in April, the Oklahoma City tragedy may have helped break through the abstractions that enable government-hating extremists to commit unspeakable acts.