17 Ga. St. U.L. Rev. 379, *

Winter, 2000

17 Ga. St. U.L. Rev. 379

ARTICLE: PURVEYORS OF HATE ON THE INTERNET: ARE WE READY FOR HATE SPAM?*

* This article was prepared in conjunction with the Symposium, "Law and the Internet: A Symposium for the New Millennium," hosted by the Georgia State University Law Review on February 8, 2000.

Elizabeth Phillips Marsh+

+ Professor of Law, Quinnipiac College School of Law. I am grateful to my colleagues for their helpful comments and encouragement in response to an oral presentation of this piece, particularly Neal Feigenson, Marilyn Ford, Carrie Kaas, Stan Krauss, Linda Meyer, Steven Latham, Alan Soloway, Mary Moers Wenig, Ann DeVeaux, and Michael Hughes. I also wish to acknowledge the generous aid of the Quinnipiac College School of Law as well as the patience and encouragement from my family.

[*379]

Introduction

The Internet ¹ has opened up new vistas of empowering communicative opportunities. We marvel ² at how this power has made the Internet part of the fabric of our everyday life. It is now entrenched as a means by which to accomplish many common tasks, such as shopping, ³ banking, ⁴ investing, ⁵ [*380] education, ⁶ and simple interpersonal communication. ⁷ For many, the Internet has become our newspaper, our public forum, our mail service, our shopping mall, and our library and research center. It presents a marketplace of ideas ⁸ unlike any we have ever encountered; ⁹ of course this multiplicity of functions defies easy characterization for doctrinal analysis.

[*381]

This technology, however, is not an unalloyed tool for good. ¹⁰ New technology creates opportunities for criminals as well as law-abiding citizens. ¹¹ For every legitimate use of the Internet, there is a potential for criminal abuse. Efforts to purchase goods over the Internet can lead to fraud; therefore, customers are understandably reluctant to transmit credit card information without satisfactory assurances that it will not be misused. ¹² Legitimate banking may exist side by side with illegitimate money laundering over the Internet. ¹³ Legal investing activities can give way to new forms of securities offenses. ¹⁴ Accredited educational endeavors can be diluted or harmed by spurious Internet courses. ¹⁵ Otherwise appropriate communication [*382] channels become gateways for gambling, ¹⁶ pornography, ¹⁷ and hate speech. This should not surprise us. With any new technology, opportunities for misuse will arise.

Just as the Internet has empowered its users with a great communicative capacity for good, it has also created vast opportunities for communication of socially undesirable messages. The Internet empowers not only the groups that society may wish to foster, namely, churches, synagogues, school groups, political organizations, ¹⁸ and public interest groups, but also criminals, hate groups, and groups that seek to impede others in the exercise of their rights. ¹⁹ Many [*383] commentators have examined the problems surrounding pornography on the Web, but as a society, we might need to worry as much, if not more, about the purveyors of hate as we do about the purveyors of flesh. ²⁰ The dangers do not lend themselves to easy answers. Much of this activity is protected by our First Amendment. Much of it is difficult to investigate and enforce given current law enforcement attitudes regarding Web activities. ²¹ Nevertheless, given the dangers posed, there is a point at which the use of the Internet by purveyors of hate must evoke a response from law enforcement authorities and society as a whole. The nature and extent of that response should be evaluated by looking at the entire context of the anti-social message.

In recent years, many have suggested that the glory of the Internet is its lack of regulation. Supporters of this unregulated environment contend that any regulation of the Internet violates the First Amendment ²² or, alternatively, that cyberspace presents a modern libertarian utopia that should be encouraged [*384] not stilted. ²³ This has led some to frame the issue of Internet regulation as a stark dichotomy between anarchy and government control. ²⁴ Others view the debate as one of private ordering versus government regulation. ²⁵ Still others accept regulation, but raise issues of federalism and preemption. ²⁶ Others argue that existing legal doctrines are the only ones needed for cyberspace. ²⁷ There is also a significant question of whether regulation can come solely from government or whether technology imposes its own restraints on the flow of information. ²⁸ "Regulation" can come from self-help as well as from the government; indeed, in a "wild west" environment, self-help is often the only help. Many have commented on the need for private measures in the face of an unwillingness to seek law enforcement assistance and inadequate law enforcement response, thus necessitating a self-help response. This is a fascinating and exciting debate, and while I take no position on these views, I would argue that we do well to consider Lawrence Lessig's comment that "[w]e need to think collectively and sensibly about how this emerging reality will affect our lives." ²⁹

In this spirit, I will venture a few tentative thoughts on how law enforcement can respond to the use of the Internet by purveyors of hate without violating the principles of the First Amendment. ³⁰ Assuming that these messages of hate are [*385] pervasive, self-help will not suffice. From a background in law enforcement, ³¹ I start with a sense that the use of the Internet magnifies the messages of hate in a way that may deserve a heightened governmental response. Quite simply, speech on the Internet is different. Looking at a few instances in which courts have confronted hate on the Internet, the lesson I draw is that even though courts must be careful to safeguard First Amendment values, the vitriolic nature of these communications must be measured, not in isolation, but rather in their broader social context. In short, when applying the Brandenburg test ³² to Internet speech, particularly hate speech, courts should evaluate these communications within a broader social context and look at the pervasiveness of the speech when deciding whether the speech falls within the protection of the First Amendment.

Part I of this Article raises the question of whether Internet speech is different from other speech, thus deserving a different standard of First Amendment protection. Part II presents a hypothetical fact pattern as a vehicle to explore some of the issues. Part III discusses two cases that deal with hate speech on the Internet. Part IV touches on traditional First Amendment theory and asks whether existing First Amendment doctrine precludes any successful regulation of hate messages, or whether criminal law can reach this type of speech when it is pervasive and within a broad social context of violence. Finally, Part V discusses the difference between simple speech and "hate spam," which magnifies the message of hate. I conclude by arguing that in the absence of governmental regulation of privacy data, criminal law should be able to reach hate spam if, when viewed from a broader social context, it encourages violence to an intolerable degree. Thus, courts should consider the pervasiveness of the speech and how the message is amplified on the Internet.

[*386]

I. Is Internet Speech Different from Other Speech?

Several commentators have compared two or more types of media with an eye towards asking whether different First Amendment standards should be applied to each. Professor Rodney Smolla believes "that one First Amendment fits all." ³³ Professor Frederick Schauer notes that the Supreme Court has hinted at developing First Amendment principles "in an institution-specific manner . . . ." ³⁴ While there is a strong argument that speech on the Internet should be subject to the same standards as other speech, I would suggest that speech on the Internet is different.

The United States Supreme Court has treated different types of media differently for purposes of First Amendment analysis. ³⁵ The emphasis on media parameters will change over time, particularly as one technology merges into another. For example, with the growth of broadband technology, some of the borders will blur. ³⁶

Nevertheless, certain characteristics of the Internet are sui generis. ³⁷ First, communication on the Internet has vast scope. ³⁸ [*387] "It is no longer the case that those with the most massive resources will have the biggest audience." ³⁹ A person can send a single communication to millions with just a few clicks of a mouse. ⁴⁰ In addition, Web sites can have mirror sites so that the same site appears in more than one place. Second, the Internet is open to vast numbers of people, and the cost of communication is low, to the point that the expense is virtually nil. ⁴¹ Thus, there is no scarcity of resources or information channels in this context. ⁴² Third, the Internet permits anonymous speech. ⁴³ While anonymous speech often serves desirable goals ⁴⁴ and claims First Amendment protection, ⁴⁵ the [*388] very ease of access described above may encourage criminal activity and hate speech. ⁴⁶ Fourth, communications can come in many different forms at once. In addition to simple person-to-person messages, senders can create a barrage of e- mails reinforced with banners, advertisements, hyperlinks, and Web sites. Messages can also be sent via other Internet-based media such as Usenet and listservs. Fifth, the reach and widespread use of the Internet have serious implications for individual privacy. Many promotional messages come without an invitation, making the recipient a captive audience. Bulk mailers can reach many people at once, and often these messages are mere spam, unbidden messages, usually commercial in nature, but not necessarily. Moreover, simple reading on the Internet becomes interactive because when someone visits a Web site, cookies ⁴⁷ are left with identifying [*389] information. Profiles of personal information on Web users can be compiled from these cookies, often without the user's knowledge. Finally, in the few instances in which regulation has been imposed on the Web, it has been with reluctance. Often, people prefer not to resort to government intervention, and even when the government does intervene, it is often ineffective. For example, Web sites that have been enjoined have appeared again on another Internet service provider (ISP). ⁴⁸

With these differences in mind, let us turn to my hypothetical.

[*390]

II. The Hypothetical

Suppose I form a small group called the Society for Cutting and Raping Every Woman (S.C.R.E.W). ⁴⁹ We meet once a month. Cognizant of our rights of association, we guard our membership lists closely, ⁵⁰ but we want to expand both our group size and the impact of our message. To do this, we might purchase newspaper, radio, or television ads; try to obtain media coverage at no cost; post, distribute, and mail leaflets; make phone calls; send telegrams; put up posters; leave anonymous graffiti messages; and network with potential members. This is all very time-consuming and often expensive. If we choose to send our message anonymously, the recipient may quickly discount it. Moreover, since the message is unpopular or at least recognized as socially unacceptable, very few will likely respond.

Let me shift the hypothetical a bit. Assume that I possess a personal computer and have access to the World Wide Web. Now with very little cost ⁵¹ or delay, I can reach millions of people who can come to me by visiting my Web site, or I can go to them via e-mail. ⁵² The geographic and monetary barriers are lifted. Moreover, notwithstanding the fact that the message is socially unacceptable, more individuals will feel free to respond since they consider themselves unbounded by the limits of identity and social context. Some will respond out of genuine agreement with the message; others will respond out of mere curiosity; and some, who would never show their faces at a live meeting of [*391] S.C.R.E.W., will respond after overcoming social inhibitions that would bind them in real space, but not cyberspace. In addition, I can now send my hate speech directly to the group against whom my organization is aligned. I can now send hate spam to as many members of the "other" group-here, women-as I can find.

Traditional theory suggests that as long as I am simply conveying a message, the government should not intervene to impede the flow of this information, no matter how noxious. We are loath to invoke the criminal law for thoughts alone, ⁵³ and courts have long stated that the remedy for speech is more speech. ⁵⁴ On the Internet, though, the target of such speech may not be able to counter with additional speech due to the pervasiveness of the message. ⁵⁵ Moreover, there is a body of thought that any regulation at all may curtail the benefits of the Internet. ⁵⁶ Once I shift to cyberspace to spread my hatred, though, this message becomes amplified by the Internet and may become pervasive well beyond the effects of ordinary speech. Beyond the obvious use of Web sites and e-mail, other ways that I can use the Web to spread my message of hatred include: (a) invading chat rooms and discussion groups; (b) usurping domain names; ⁵⁷ (c) using misleading meta-tags; ⁵⁸ [*392] and (d) launching hate attacks in avatar space. ⁵⁹ Obviously, this covers a lot of ground. Even so, the list probably just scratches the surface. ⁶⁰ These combined techniques can magnify my message, creating a "shouting" on the Web that is far louder than any anonymous pamphlet handed out on a street corner. ⁶¹

As a S.C.R.E.W. member, I might use e-mail in three possible scenarios. First, I mail messages to my fellow members of S.C.R.E.W. to discuss topics of mutual interest. ⁶² Second, I send [*393] e-mail messages to non-members in an effort to get them to join my group. Third, I send messages to non-members who are the targets of my hatred in an effort to intimidate them and to obtain publicity for my cause. This could be a few isolated messages or a broad-scale flood in the form of what I will call "hate spam." ⁶³ This latter scenario is of greatest concern. To illustrate, let us suppose that, as president of S.C.R.E.W., I purchase a mailing list of individuals with personal information profiles. I then formulate a message that contains a vitriolic tirade of hate. ⁶⁴ Perhaps I add graphic pictures or even audio and video embellishments. Using a bulk remailer, ⁶⁵ I send my e-mail message to millions of women, and after they hear the cheery message, "You've got mail," they are treated to my presentation calling for the rape, mutilation, and murder of all women.

We can explore some of the questions posed by my S.C.R.E.W. hypothetical by first examining two cases that deal with hate messages on the Internet. The first case involves anti-abortion [*394] "wanted posters" and "Nuremberg files" on the Internet. ⁶⁶ The second deals with an e-mail discussion of the torture, rape, and debasement of a woman identified by name. ⁶⁷ These two cases present fact situations that cover an array of First Amendment issues. In the first case, the court permanently enjoined a Web site commonly known as the "Nuremberg files." ⁶⁸ In the second, the court quashed a criminal indictment based on ominous misogynist e-mail messages. ⁶⁹

III. Two Cases of Judicial Response to Purveyors of Hate on the Internet

A. The "Nuremberg Files" Case ⁷⁰

The debate over abortion rights speech has become extreme and increasingly violent. ⁷¹ The American Coalition of Life Activists (ACLA), an anti- abortion group that advocates the use of force in their efforts to curtail abortions, ⁷² released dramatic "wanted posters," along with the so-called "Nuremberg files." ⁷³ The "wanted posters" included a "Dirty Dozen list" that named a number of doctors who performed abortions, including four of [*395] the named plaintiffs in the lawsuit, ⁷⁴ along with their home addresses and, in some instances, their home phone numbers. ⁷⁵ These "posters" offered "a $ 5,000 [r]eward for information leading to arrest, conviction and revocation of license to practice medicine" of the named individuals. ⁷⁶ The "Nuremberg files" poster listed more doctors and health care providers with names, addresses, and physical descriptions, as persons and clinics wanted "for crimes against humanity." ⁷⁷ These hard copy posters were widely circulated at anti- abortion meetings and on television. ⁷⁸ Ultimately, they were posted on the Internet. ⁷⁹ In later manifestations of the posters, the names of doctors, clinic workers, and security personnel killed during attacks on abortion clinics were listed with strikes through their names; those wounded had their names shaded in gray. ⁸⁰

Plaintiffs, consisting of Planned Parenthood of Columbia/Willamette and the Portland Feminist Women's Health Center, as well as individual doctors named in the posters, sued, alleging violations of the Freedom of Access to Clinic Entrances Act of 1994 (FACE), ⁸¹ the Racketeer Influenced and Corrupt Organizations Act (RICO), ⁸² the state RICO, ⁸³ and state tort law. ⁸⁴ Plaintiffs sought to enjoin these and similar posters and asked for damages. The defendants argued that the posters were protected by the First Amendment. ⁸⁵

The district court surveyed the surrounding context of abortion violence. In a chilling litany of 453 findings of fact, the court laid out a series of events wherein doctor after doctor named in the posters was murdered, some at close range, others by snipers. ⁸⁶ When new doctors, such as Dr. Bernard Slepian, [*396] were murdered, their names were then crossed out on the "Nuremberg poster." ⁸⁷ Accordingly, the court held that in this case the words constituted "true threats" under the applicable test of "whether a reasonable person would foresee that the statement would be interpreted by those to whom the maker communicates the statement as a serious expression of intent to harm or assault." ⁸⁸ Thus, the court did not analyze the posters under the "incitement to imminent violence" exception or the "fighting words" exception. ⁸⁹ Instead, the court recognized that speech could fall outside the protection of the First Amendment if "the target of the speaker reasonably believes that the speaker has the ability to act him or herself or to influence others to act at a level less than incitement-it is the perception of a reasonable person that is dispositive, not the actual intent of the speaker." ⁹⁰

Had the court used the Brandenburg test, it is conceivable that the posters would have been protected under the First Amendment since there was advocacy, but no direct link between the speaker and ultimate action. Instead, however, the court looked to the broader social context, notwithstanding the fact that the plaintiffs had not proven that any of the defendants had engaged in any direct violence. The court granted the injunction, and the jury awarded damages. ⁹¹ In the final opinion, the First Amendment arguments were relegated to a footnote and rejected. ⁹²

B. The Jake Baker Case ⁹³

In 1995, Jake Baker, also known as Abraham Jacob Alkhabaz, was a student at the University of Michigan. ⁹⁴ He posted a story on an Internet forum describing the rape, torture, and murder [*397] of a woman with the name of one of his female classmates. ⁹⁵ When college security officers searched his computer files, ⁹⁶ they found a similar story that was unpublished, but which contained the college and home addresses of the female classmate and identified her as his classmate. ⁹⁷ They also found a series of e-mails exchanged over approximately three months between Baker and a Canadian individual named Gonda. ⁹⁸ In these e-mail messages, Baker and Gonda detailed their plans to attack young women. ⁹⁹ Baker wrote, "Just thinking about it anymore doesn't do the trick . . . I need TO DO IT." ¹⁰⁰ Other e-mails regarding the female classmate referred to the posted story. ¹⁰¹ It is clear that these stories and e-mails all involved "a sexual interest in violence against women and girls." ¹⁰² Ultimately, Baker was prosecuted under 18 U.S.C. § 875(c) ¹⁰³ for transmitting threats to injure or kidnap another in e-mail messages transmitted via the Internet. ¹⁰⁴

The court noted that "coercive or extortionate threats are paradigmatic subjects of a prosecution under 18 U.S.C. § 875(c)" ¹⁰⁵ and identified three elements for the offense: "(1) a transmission in interstate [or foreign] commerce; (2) a communication containing a threat; and (3) the threat must be a threat to injure [or kidnap] the person of another." ¹⁰⁶ The court also analyzed the crime as one of general intent. ¹⁰⁷

In order to secure a conviction that will withstand First Amendment challenge, the court ruled that the prosecution [*398] must show a "true threat." ¹⁰⁸ Determining whether a communication is indeed a true threat, is, however, somewhat problematic. At the district court level, the threat analysis was bound to the First Amendment analysis. First, the threat must be "particularly likely to be intimately bound up with proscribed activity." ¹⁰⁹ The court may also consider the recipient of the threat to determine whether a "'reasonable person would foresee that the statement would be interpreted by those to whom the maker communicates the statement as a serious expression of an intent to inflict bodily harm' or kidnap a person." ¹¹⁰ Despite the language of these communications, the district court was skeptical about the decision to prosecute. ¹¹¹ Relying on a psychological evaluation that found no evidence to indicate that Baker was a danger to himself or others, the court granted the defendant's motion to quash the indictment.

On appeal, the Sixth Circuit Court of Appeals upheld the dismissal, but limited its discussion to the statutory interpretation of 18 U.S.C. § 875(c), under which Baker was charged. ¹¹² The court, however, used an analysis reminiscent of Brandenburg and held, over a strong dissent, that "a threat" for purposes of the statute must reveal an additional "purpose of furthering some goal through the use of intimidation." ¹¹³ Thus the advocacy of these sexual attacks without action to back them up could not be prosecuted. ¹¹⁴

[*399]

The dissent, in contrast, found these e-mails to be highly threatening. ¹¹⁵ As such, a "rational jury" could find that they constituted "threats" under the statute. ¹¹⁶ In the context of pervasive violence against women, this finding would be defensible.

IV. Traditional First Amendment Theory: The Standard for Speech That Constitutes Imminent Incitement

A. The Categorization Approach

Basic First Amendment theory provides that in order to be prosecuted for the content of speech, the speech must fall outside the circle of First Amendment protection. First Amendment protections are not absolute. ¹¹⁷ While content-neutral time, place, and manner restrictions may be imposed, when the government seeks to regulate in a way that may implicate content, it may only do so when the speech in question falls within specific categories of unprotected speech. ¹¹⁸ Thus, under this categorization approach, advocacy of an imminent illegal act (incitement beyond mere advocacy), ¹¹⁹ fighting words, ¹²⁰ obscenity, ¹²¹ and defamation ¹²² fall outside the constitutional pale. Thus, there is a First Amendment right to express unpopular views. Courts have allowed groups such as [*400] the Ku Klux Klan mandatory access to broadcast media. ¹²³ The Supreme Court upheld the right of Neo-Nazis to march in full regalia through a predominantly Jewish neighborhood in Skokie, Illinois. ¹²⁴ Once the speech crosses the line into fighting words, or to the extent that it advocates and has the impact of inciting violence as part of a pervasive campaign of hatred, it should be curtailed.

When we try to draw the line between these two ends of a continuum, we must take care not to hamper or chill legitimate speech. ¹²⁵ This effort, if it is to be undertaken at all, requires a scalpel, not a meat cleaver. Attempts to ban the language of hatred must not chill legitimate speech. ¹²⁶

B. Theorists Who Question the Categorization Approach

In a thoughtful and ambitious treatment of First Amendment theory, Professor Steven Heyman has traced the underpinnings of our First Amendment jurisprudence. ¹²⁷ At the risk of doing him the disservice of oversimplifying his theory, he argues that "free speech is a right that is limited by the fundamental rights of other individuals and the community as a whole." ¹²⁸ Drawing on the natural rights tradition, he suggests that free speech is a right that must sometimes, but certainly not always, bend to the rights of others, particularly when the rights of others implicate similar values and principles such as "personal security, privacy, reputation, and citizenship." ¹²⁹ He concedes that "hate speech may have some political value, [but] that value is distinctly limited." ¹³⁰ While recognizing institutional reasons as [*401] to why regulation of hate speech may be inappropriate, he worries that "[t]he effect of protecting hate speech on these grounds would be to leave target-group members without legal protection against such injuries." ¹³¹ In short, hate speech is protected at the cost of dignity, equality, security, and freedom from emotional distress.

In another attempt to redefine the lines of First Amendment protection, this time with an expanded definition of existing constructs, Professor David Crump explored the dynamics of "camouflaged speech." ¹³²

[I]n spite of the deference that we grant to speech falling short of actual incitement to crime, and in spite of our recognition that there are prohibited utterances that cross the line, a borderland remains in which clever speakers can hide, with form, the substance of what they say. In short, Mark Antony's speech is an example of the phenomenon that [I] refer to as camouflaged incitement. ¹³³

He describes well-established First Amendment principles, but then attempts to adjust them a quarter turn to avoid injustice. ¹³⁴ In discussing Brandenburg, Professor Crump notes that the categories are not particularly helpful in close cases. ¹³⁵ Brandenburg requires a link between advocacy and action. ¹³⁶ If that link is broken, powerful speech may lead a third party to act, but leave the speaker unaccountable. ¹³⁷ Professor Crump further states:

Of course, there also was a more ominous side to Brandenburg. The possibility remained that a person of borderline mentality attending this rally might have taken [*402] the Klan leader's words to heart and acted upon the indirect advocacy they contained. For example, this hypothetical psychopathic Klansman might have ambushed and killed a randomly selected African-American citizen for the purpose of enhancing his own reputation and exacting the "revengeance" that his mentor had mentioned in front of the burning cross. In that case, the Supreme Court's opinion still would exonerate Mr. Brandenburg himself; in effect, it would say to the victim, "Too bad. That's the price of freedom of speech." This tragic possibility exists because law is not a perfect instrument of social regulation-and because of the preferred position of the First Amendment freedoms. Still, our courts should take the potential for tragedy seriously; they should strive to make the best accommodation possible between these competing values, rather than cavalierly writing off the victims of camouflaged incitement. ¹³⁸

He advocates a broader array of possible government intervention with camouflaged speech. ¹³⁹

Using Brandenburg as his starting point, Professor Crump suggests that we should be able to distinguish between mere advocacy and incitement, but that the categorization approach does not achieve this easily. ¹⁴⁰ He suggests, instead, an eight-part balancing test ¹⁴¹ consisting of the following factors: (1) the express words or symbols uttered; ¹⁴² (2) the pattern of the utterance; ¹⁴³ (3) the context; ¹⁴⁴ (4) the predictability and anticipated seriousness of unlawful results and whether they actually occurred; ¹⁴⁵ (5) the extent of the speaker's knowledge or reckless disregard of the likelihood of violent results; ¹⁴⁶ (6) the availability of alternative means of expressing a similar [*403] message, without encouragement of violence; ¹⁴⁷ (7) the inclusion of disclaimers; ¹⁴⁸ and (8) the possibility that the utterance has "serious literary, artistic, political, or scientific value." ¹⁴⁹

In some respects, this test is the type of analysis used by the court in the Planned Parenthood cases. Upon looking at the surrounding context-the explicit threats, the reckless disregard that a member of the group would (and obviously someone did) act on the call to action to kill or maim, and the request to disable abortion doctors-the district court permanently enjoined the "Dirty Dozen posters" and the "Nuremberg files" posted on the Internet by the American Coalition for Life Activists.

In contrast, the court in Alkhabaz did not look at context. If measured against the broader context of violence against women, although not necessarily committed by this speaker, the court might have found, as the dissent suggested, that the e-mail messages were threats and incitements. ¹⁵⁰ A similar result might occur using Professor Crump's eight-part test. Under either approach, both of which are ad hoc balancing tests, the outcome is uncertain, but at least the targets of the speech would have a chance to vindicate their claims upon making the appropriate factual showing.

Using the fact-determinative, ad hoc balancing approach to broaden the window of what constitutes threats or incitement to imminent action is less invasive of First Amendment freedoms than other approaches. It does not cut so broad a swath as legislative efforts such as Georgia's attempt to ban all Internet transmissions which falsely identify the sender, including anonymous transmissions. ¹⁵¹ Moreover, the sending of e-mail messages as "hate spam" would be more readily reachable under this more broadly defined standard.

[*404]

V. Hate Spam

"Spam" ¹⁵² is usually characterized as "unsolicited commercial e- mail." ¹⁵³ Efforts to fashion legal theories for recovery for unauthorized spamming have not been very successful. ¹⁵⁴ As commercial speech, however, spam enjoys less First Amendment protection than other speech. ¹⁵⁵ What happens if a hate group uses spam to target members of the "other" group as the recipients of vituperative e-mail?

How would hate groups get lists of names to whom they would direct their messages? The amount of personal data currently compiled on individuals is considerable. ¹⁵⁶ This personal data is readily available for sale to any third party willing to pay for it. ¹⁵⁷ Spam can be targeted by means of [*405] personal profile data gathered from sources such as cookies. This data can profile an individual in alarming ways, ¹⁵⁸ and use will only increase. ¹⁵⁹ To date, voluntary restraint has been the only safeguard. Unlike Europe, the United States does not have pervasive restrictions on the sale of this data. ¹⁶⁰ There are only a few restrictions under U.S. law. ¹⁶¹ At times, Internet Service Providers might voluntarily remove Web sites of hate groups ¹⁶² or data left by cookies, but there is nothing that requires it.

Unsolicited e-mail is not regulated. Instead, regulation of spam is left primarily to aggressive self-help ¹⁶³ or the rules of [*406] "Netiquette." ¹⁶⁴ Some Web sites self-regulate, albeit often disingenuously. At least one anti-gay site warns persons poised to visit of the nature of their message. ¹⁶⁵ Similarly, several white supremacist sites warn visitors that they should be eighteen before entering, and some indicate that the content of their site should be viewed as parody or humor. ¹⁶⁶

If a violent group with a hate message targeted a group, for example, if the Ku Klux Klan sent a message of hate to African Americans or Jews using hate spam, there would be several remedies. Tort remedies, such as intentional infliction of emotional harm or invasion of privacy, might lie, but would not have the effect of successfully impeding such hate spam. However, under the broader view of threats and the Brandenburg test that is discussed in this Article, I would argue that the government should and could seek broader equitable and criminal remedies when hate spam is sent in a broader social perspective of pervasive violence.

Conclusion

Given the popularity and power of the Internet, speech on the Internet is different. When purveyors of hate use the Internet to send e-mail threats, distribute hate spam, or establish hate-motivated Web sites, these insidious communications are magnified in a way that deserves heightened governmental response. While courts must be careful to safeguard First Amendment values, they should consider Internet hate speech in a broader social context. Under the balancing test advocated by Professor Crump, and seemingly applied in the Planned Parenthood cases, courts should use evidentiary factors to [*407] distinguish between mere advocacy and incitement, according to Brandenburg. When viewed in a broader social context of pervasive violence against the target group, if the hate speech is the spark in the tinderbox and encourages violence to an intolerable degree, it could be enjoined without violating the First Amendment. If the violence against the target group is less pervasive, however, the hate speech should be tolerated. This ad hoc balancing approach, rather than a hands-off acquiescence occasioned by reading Brandenburg too narrowly, would leave courts free to curtail the true threats to the targets of hate speech rather than making the targets of hate speech bear the entire brunt and cost of protecting our constitutional freedoms.

FOOTNOTES:
n1 While the Internet has been defined many ways, at least one court has described it as "a decentralized, global communications medium that links people, institutions, corporations and governments around the world." Cyberspace, Comm., Inc. v. Engler, 55 F. Supp. 2d 737, 741 (E.D. Mich. 1999) (citing ACLU v. Reno, 929 F. Supp. 824, 831 (E.D. Pa. 1996), aff'd, 521 U.S. 844 (1997)); ACLU v. Johnson, 4 F. Supp. 2d 1029, 1031 (D.N.M. 1998), aff'd, 194 F.3d 1149 (10th Cir. 1999); see also Am. Libraries Ass'n v. Pataki, 969 F. Supp. 160, 164 (S.D.N.Y. 1997).

n2 President Clinton began his Millennium address with a paean to the leaps in communication brought about by the Internet:On this day 200 years ago, in 1799, our second President welcomed the 19th century. It then took six weeks by boat to get news from Europe. On this day 100 years ago, when President William McKinley marked the start of the 20th century, it took six seconds to send a text by telegraph. Today, satellites and the Internet carry our voices and images instantaneously all around the world.President William J. Clinton, Millennium Around the World Address (Dec. 31, 1999), available at LEXIS, US News Library, Combined File.

n3 The size of the retail markets available on the Internet is difficult to quantify. Most would agree, however, that it can be measured in billions of dollars and that use is increasing. See, e.g., Steven Bonisteel, 'Internet Economy' Weighs in at $ 507 Billion, Newsbytes News Network, at http://www.newsbytes.com (Oct. 27, 1999).

n4 See Mark E. Budnitz, Stored Value Cards and the Consumer: The Need for Regulation, 46 Am. U. L. Rev. 1027 (1997); Simon L. Lelieveldt, How To Regulate Electronic Cash: An Overview of Regulatory Issues and Strategies, 46 Am. U. L. Rev. 1163 (1997); David G. Oedel, Why Regulate Cybermoney? 46 Am. U. L. Rev. 1075 (1997); Catherine Lee Wilson, Banking on the Net: Extending Bank Regulation to Electronic Money and Beyond, 30 Creighton L. Rev. 671 (1997); Kimbrelly N. Kegler, Note, Electronic Banking: Security, Privacy, and CRA Compliance, 2 N.C. Banking Inst. 426 (1998).

n5 See Tamar Frankel, The Internet, Securities Regulation and the Theory of Law, 73 Chi.-Kent L. Rev. 1319 (1998); Robert Norman Sobol, Intelligent Agents and Futures Shock: Regulatory Challenges of the Internet, 25 Iowa J. Corp. L. 103 (1999) (reviewing Howard M. Friedman, Securities Regulation in Cyberspace (1999)); David M. Cielusniak, Note, You Cannot Fight What You Cannot See: Securities Regulation on the Internet, 22 Fordham Int'l L.J. 612 (1998).

n6 See Michael A. Geist, Where Can You Go Today?: The Computerization of Legal Education from Workbooks to the Web, 11 Harv. J.L. & Tech. 141 (1997); Nat'l Ctr. for Educ. Stat., Distance Education at Postsecondary Education Institutions: 1997- 98, at http://nces.ed.gov/pubs98/distance (last visited Aug. 26, 2000) (noting that seventy-nine percent of the surveyed institutions planned increased use of Internet courses, a trend that is not without its critics).

n7 One of the primary uses of the Internet is e-mail, although the Internet provides other opportunities for interactive communication such as online discussion groups and chat rooms. The ease of interactive communication has special advantages to many who find the Internet helps them combat social and economic isolation stemming from many causes, including age or disability. Also, with the many recent mergers of Internet Service Providers (ISPs) with other media sector players, it seems clear that we will depend increasingly on the Internet for our news services.

n8 This term harkens to one of our axiomatic models of the First Amendment put forth by Justice Holmes. See Abrams v. United States, 250 U.S. 616, 624 (1919) (Holmes, J., dissenting). It is not, of course, the only model. See Cass R. Sunstein, Emerging Media Technology and the First Amendment: The First Amendment in Cyberspace, 104 Yale L.J. 1757 (1995).

n9 Our current uses will seem rudimentary when we encounter later developments of new webs and interactive use. For example, our "virtual reality" will become increasingly complex. See Jerry Kang, Remarks at the American Association of Law Schools, Section on Privacy (Jan. 2000) (on file with author) (discussing some possible permutations of racial and gender identity in avatar space); see also infra text accompanying note 59.

n10 The "dark side" of the Internet has been an image used many times elsewhere. See, e.g., Michael Rustad, Legal Resources for Lawyers Lost in Cyberspace, 30 Suffolk U. L. Rev. 317, 341-42 (1996). "Despite its wonders, it seems the Internet suffers from a split-personality. The Internet's dark side is the home of "Bad Guys"-hackers, crackers, snackers, stalkers, phone freaks, and other creepy Web crawlers." Id. (citing Maggie Canon, Life in the Big City: Internet Concerns, MacUser, May 1995, at 17).

n11 See Scott Charney & Kent Alexander, Computer Crime, 45 Emory L.J. 931 (1996).Why the great concern about computer crime? First, history teaches that criminals will frequently abuse new technologies to benefit themselves or injure others. Automobiles are an apt example. Designed to provide transportation for law-abiding individuals, the automobile soon became a target (e.g., car theft, carjacking), a tool (e.g., the getaway car in a bank robbery), and a weapon (e.g., hit-and-run). Clearly, computers are following the same route.Id. at 934. This metaphor has been suggested for privacy concerns as well. See Glen O. Robinson, The Electronic First Amendment: An Essay for the New Age, 47 Duke L.J. 899, 900 (1998).

n12 See Dee Pridgen, How Will Consumers Be Protected on the Information Superhighway?, 32 Land & Water L. Rev. 237 (1997).

n13 See Ronald K. Noble & Coura E. Golumbic, A New Anti-Crime Framework for the World: Merging the Objective and Subjective Models for Fighting Money Laundering, 30 N.Y.U. J. Int'l L. & Pol. 79 (1997); Sarah N. Welling & Andy G. Rickman, Cyberlaundering: The Risks, The Responses, 50 Fla. L. Rev. 295 (1998); Jonathan Gaskin, Note, Policing the Global Marketplace: Wielding a Knife in a Gunfight, 38 Colum. J. Transnat'l L. 191 (1999).

n14 See Robert A. Prentice, The Future of Corporate Disclosure: The Internet, Securities Fraud, and Rule 10b-5, 47 Emory L.J. 1 (1998); Kevin Mason, Comment, Securities Fraud over the Internet: The Flies in the Ointment and a Hope of Fly Paper, 30 Case W. Res. J. Int'l L. 489 (1998).

n15 See Tony Mauro, Justice Scrutinizes Long-Distance Learning; Ruth Bader Ginsburg Comments on Internet Law School, Tex. Law., Sept. 27, 1999, at 43. Justice Ginsburg spoke at the dedication of the Rutgers Center for Law and Justice, where she stated:I am uneasy about classes in which students learn entirely from home, in front of a computer screen, with no face-to-face interaction with other students or instructors . . . . So much of legal education-and legal practice-is a shared enterprise, a genuinely interactive endeavor. The process inevitably loses something vital when students learn in isolation, even if they can engage in virtual interaction with peers and teachers . . . . I am troubled by ventures like Concord, where a student can get a J.D.-though the school is still unaccredited-without ever laying eyes on a fellow student or professor. We should strive to ensure that the Internet remains a device for bringing people together and does not become a force for isolation.Id.

n16 See Stevie A. Kish, Note, Betting on the Net: An Analysis of the Government's Role in Addressing Internet Gambling, 51 Fed. Comm. L.J. 449 (1999).

n17 Pornography is prevalent on the Web. Congressional efforts to ban child pornography in the Communications Decency Act of 1996 were struck down as unconstitutional in Reno v. ACLU, 521 U.S. 844 (1997). Congress followed soon thereafter with a new statute, the Child Online Protection Act (COPA). See 47 U.S.C. § 231 (1998). COPA bans the "[knowing transmission] in interstate or foreign commerce by means of the World Wide Web [of] any communication for commercial purposes that is available to any minor and . . . is harmful to minors . . . ." Id. at § 231(a)(1). At least one commentator predicted that "[l]ike its predecessor, the COPA will soon be subject to intense judicial scrutiny. Unlike the [Communications Decency Act of 1996], however, the COPA's narrowed scope of application and more precise definition of prohibited content are likely to withstand constitutional scrutiny [as part of the Omnibus Appropriations Act]." Matthew Baughman, Recent Legislation Regulating the Internet, 36 Harv. J. on Legis. 230, 230 (1999) (footnotes omitted). However, the District Court for the Eastern District of Pennsylvania recently enjoined the enforcement of COPA. See ACLU v. Reno, 31 F. Supp. 2d 473, 498-99 (E.D. Pa. 1999), aff'd, 217 F.3d 162 (3d Cir. 2000).

n18 See, e.g., New York Times Co. v. Sullivan, 376 U.S. 254, 270 (1964) (noting "a profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open").

n19 I include all of these groups under the rather anodyne heading of "anti-social groups." There obviously is a vast continuum of viewpoints, and the place where any particular group falls along that continuum will shift radically depending on the observer's attitudes. We could look to the militia movements as an example. Some would place them as our latter-day true patriots, and thus socially desirable challengers to repressive government. See, e.g., Wilson Huhn, Political Alienation in America and the Legal Premises of the Patriot Movement, 34 Gonz. L. Rev. 417, 429 (1998) (noting that "the Militia Movement compares itself to the Minutemen of the Revolutionary War . . . ."). Others would argue that militia movements represent vigilantism and often ill-cloaked racial hatred. See, e.g., Richard Delgado, Rodrigo's Fourteenth Chronicle: American Apocalypse, 32 Harv. C.R.-C.L. L. Rev. 275 (1997).

n20 Some would argue that the line between purveyors of hate and purveyors of flesh is a thin one indeed. See Catharine A. MacKinnon, Only Words 67 (1993); Andrea Dworkin, Against the Male Flood: Censorship, Pornography, and Equality, 8 Harv. Women's L.J. 1, 9 (1985) (arguing that the debasement of women that ensues from pornography increases both sexual harassment and violence against women); see also Catharine A. MacKinnon, Vindication and Resistance: A Response to the Carnegie Mellon Study of Pornography in Cyberspace, 83 Geo. L.J. 1959, 1963-64 (1995) (noting that "when men make new communities, they bring their pornography with them").

n21 There are many impediments to prosecution, often including considerable jurisdictional barriers. See Michael L. Siegel, Comment, Hate Speech, Civil Rights, and the Internet: The Jurisdictional and Human Rights Nightmare, 9 Alb. L.J. Sci. & Tech. 375, 382 (1999). For purposes of this discussion, however, I put jurisdictional issues aside. Another impediment to prosecution is the fact that many law enforcement agencies have neither the resources nor the trained personnel to pursue investigations of this type. See Marc D. Goodman, Why the Police Don't Care About Computer Crime, 10 Harv. J.L. & Tech. 465 (1997).

n22 The Electronic Frontier Foundation, for instance, calls for "broader" access and freedom of expression with regard to online materials. See Elec. Frontier Found., Mission Statement, at http://www.eff.org/about eff.html (last visited Aug. 26, 2000). The American Civil Liberties Union (ACLU) also argues against Internet regulation. See ACLU, Cyber-Liberties, at http://www.aclu.org/issues/cyber/hmcl.html (last visited Aug. 26, 2000).

n23 See, e.g., Walter A. Effross, High-Tech Heroes, Virtual Villains, and Jacked-In Justice: Visions of Law and Lawyers in Cyberpunk Science Fiction, 45 Buff. L. Rev. 931 (1997).

n24 See, e.g., Jo-Ann M. Adams, Controlling Cyberspace: Applying the Computer Fraud and Abuse Act to the Internet, 12 Santa Clara Computer & High Tech. L.J. 403, 405 (1996) (discussing "the current struggle between two increasingly polarized camps: those who combat crime and those who defend individual freedom").

n25 See, e.g., Margaret Jane Radin & R. Polk Wagner, The Myth of Private Ordering: Rediscovering Legal Realism in Cyberspace, 73 Chi.-Kent L. Rev. 1295 (1998).

n26 See, e.g., Steven R. Salbu, Who Should Govern the Internet?: Monitoring and Supporting a New Frontier, 11 Harv. J.L. & Tech. 429 (1998).

n27 See, e.g., Sean Adam Shiff, The Good, the Bad and the Ugly: Criminal Liability for Obscene and Indecent Speech on the Internet, 22 Wm. Mitchell L. Rev. 731 (1996).

n28 See Joel R. Reidenberg, Lex Informatica: The Formulation of Information Policy Rules Through Technology, 76 Tex. L. Rev. 553 (1998).

n29 Lawrence Lessig, Code and Other Laws of Cyberspace 233 (1999).

n30 "Congress shall make no law . . . abridging the freedom of speech, or of the press." U.S. Const. amend. I. This is the first provision of the Bill of Rights made applicable to the states through the Fourteenth Amendment Due Process Incorporation Doctrine. See Gitlow v. New York, 268 U.S. 652 (1925).

n31 I entered law school teaching after a career as an Assistant District Attorney in the Office of Robert Morgenthau, District Attorney, New York County, New York. This left me with a perception (perhaps a self-delusion) that my efforts to theorize retain a practical framework.

n32 See Brandenburg v. Ohio, 395 U.S. 444 (1969) (holding that advocacy of a crime alone is protected unless it directly incites imminent lawless action).

n33 Rodney A. Smolla, Will Tabloid Journalism Ruin the First Amendment for the Rest of Us?, 9 DePaul-LCA J. Art & Ent. L. & Pol'y 1, 34 (1998) (comparing "tabloid journalism" with "serious journalism").

n34 See Frederick Schauer, Principles, Institutions, and the First Amendment, 112 Harv. L. Rev. 84 (1998).

n35 See Robinson, supra note 11, at 967. "The Court has long been in the habit of saying that each medium of mass expression raises particular First Amendment problems." Id. (citing Reno v. ACLU, 521 U.S. 844, 868 (1997); Turner Broad. Sys. v. FCC, 512 U.S. 622, 657 (1994); Southeastern Promotions Ltd. v. Conrad, 420 U.S. 546, 557 (1975)).

n36 See Robinson, supra note 11, at 967.[I]t is not clear that there are important differences between electronic media as far as indecency, for example, is concerned. In Denver Area Educ. Telecomm. Consortium, Inc. v. FCC, 116 S. Ct. 2347 (1996), the Court approved in principle that limited control of indecent content could be extended to cable television, its premise being that there is no relevant distinction between the words and pictures delivered over the air and those delivered by coaxial cable. By the same token, there is no relevant distinction between words and pictures delivered by coaxial cable and those distributed by the Internet. Indeed, with the adoption of broadband delivery media such as cable modems and satellite broadcasting, the Internet may soon become an alternative mode of delivery for full motion video that is identical to standard cable television programming.Id. (footnotes omitted).

n37 Efforts to identify this uniqueness may be a fool's errand. See, e.g., id. at 966.On the other hand, although the Court in Reno did move the electronic First Amendment a little closer to the print model, it would be a mistake to think that the Court has simply applied the print First Amendment simpliciter. In fact, "simpliciter" is the last word one should apply to our modern First Amendment. The modern First Amendment has become a confusing, complex assemblage of doctrines, considerations, and ad hoc factors. Some of the Court's modern opinions read like the Restatement of Torts, with its characteristic laundry list of factors to be weighed in some unspecified fashion to determine liability.Id. (footnotes omitted).

n38 I do not include the theory that the World Wide Web is innately a hostile environment for women and minority groups in my litany of reasons why speech on the Internet is different. Some have noted that the Information Highway is so replete with pornography and chauvinism that it makes some groups less willing to participate in the Internet than others. See Keth A. Ditthavong, Paving the Way for Women on the Information Superhighway: Curbing Sexism Not Freedoms, 4 Am. U. J. Gender Soc. Pol'y & L. 455, 510 (1996) (concluding that the feminist movement should use the Web to organize and voice women's issues on a global scale and that "[i]t would not behoove the female populace for one sector of feminists to purport the silence of other women under the guise of censoring 'indecent materials'"); see also Margaret Chon, Radical Plural Democracy and the Internet, 33 Cal. W. L. Rev. 143 (1997).

n39 James Katz, Struggle in Cyberspace: Fact and Friction on the World Wide Web, 560 Annals Am. Acad. Pol. & Soc. Sci. 194, 194 (1998).

n40 In various efforts to quantify the number of people who have access to the Web, the general consensus is that we are talking in terms of millions, and this suffices for my purpose. See Eugene Volokh, Cheap Speech and What It Will Do, 104 Yale L.J. 1805 (1995).

n41 See id. at 1806 n.3.

n42 See Robinson, supra note 11.

n43 See A. Michael Froomkin, Flood Control on the Information Ocean: Living with Anonymity, Digital Cash, and Distributed Databases, 15 J.L. & Com. 395, 405 (1996).

n44 Anonymity allows full expression of ideas and can shield the speaker from opprobrium and reprisal. See id. at 409.

n45 See, e.g., id.In the U.S., anonymous speech may be guaranteed by the First Amendment or whatever right to privacy exists in the Constitution. In the U.S., anonymous speech also benefits from its association with well-remembered incidents in which political actors holding unpopular views that many now accept benefitted from the ability to hide their identity. The Federalist Papers, the nation's most influential political tracts, were published pseudonymously under the name "Publius." More recently, the Supreme Court held the guarantee of free speech in the Constitution protects a right of anonymous association and that a state therefore lacked the power to compel a local chapter of the NAACP to disclose the names of its members. In so doing, the Court protected the NAACP members from danger at the hands of bigots who would have had access to their identities if the state had prevailed.Id.

n46 See George P. Long, III, Comment, Who Are You?: Identity and Anonymity in Cyberspace, 55 U. Pitt. L. Rev. 1177, 1205 (1994); see also Anne Wells Branscomb, Anonymity, Autonomy, and Accountability: Challenges to the First Amendment in Cyberspace, 104 Yale L.J. 1639, 1675 (1995) (noting that if law enforcement authorities are precluded from obtaining the identities of anonymous users, illegal activities will proliferate).

n47 For a definition of "cookies" and how they work, see Viktor Mayer-Sch nberger, The Internet and Privacy Legislation: Cookies for a Treat?, 1 W. Va. J.L. & Tech. 1, 5-9 (1997), at http://www.wvu.edu/wvjolt/Arch/Mayer/Mayer.htm. The WWW [World Wide Web] is built on a very simple, but powerful premise. All material on the Web is formatted in a general, uniform format called HTML (Hypertext Markup Language), and all information requests and responses conform to a similarly standard protocol. When someone accesses a service provider on the Web, such as the Library of Congress, the user's Web browser will send an information request to the Library of Congress' computer. This computer is called a Web server. The Web server will respond to the request by transmitting the desired information to the user's computer. There, the user's browser will display the received information on the user's screen. Cookies are pieces of information generated by a Web server and stored in the user's computer, ready for future access. Cookies are embedded in the HTML information flowing back and forth between the user's computer and the servers. Cookies were implemented to allow user-side customization of Web information. For example, cookies are used to personalize Web search engines, to allow users to participate in WWW-wide contests (but only once!), and to store shopping lists of items a user has selected while browsing through a virtual shopping mall. Essentially, cookies make use of user-specific information transmitted by the Web server onto the user's computer so that the information might be available for later access by itself or other servers. In most cases, not only does the storage of personal information into a cookie go unnoticed, so does access to it. Web servers automatically gain access to relevant cookies whenever the user establishes a connection to them, usually in the form of Web requests. Cookies are based on a two-stage process. First the cookie is stored in the user's computer without her consent or knowledge. For example, with customizable Web search engines like My Yahoo!, a user selects categories of interest from the Web page. The Web server then creates a specific cookie, which is essentially a tagged string of text containing the user's preferences, and it transmits this cookie to the user's computer. The user's Web browser, if cookie-savvy, receives the cookie and stores it in a special file called a cookie list. This happens without any notification or user consent. As a result, personal information (in this case the user's category preferences) is formatted by the Web server, transmitted, and saved by the user's computer. During the second stage, the cookie is clandestinely and automatically transferred from the user's machine to a Web server. Whenever a user directs her Web browser to display a certain Web page from the server, the browser will, without the user's knowledge, transmit the cookie containing personal information to the Web server.Id.

n48 See, e.g., Scott Hogenson, Nuremberg Files Returns to Internet, Conservative News Serv., at http://www.conservativenews.org/InDepth/archive/199902/IND1999022 4d.html (Feb. 24, 1999).

n49 This acronym and the goals expressed therein are meant to be fictitious. My research has not revealed any such group, and it is my sincere hope that there has not been and never will be such a group. There was, at one point in time, a group that called itself S.C.U.M., the Society for Cutting Up Men. See Sisterhood Is Powerful: An Anthology of Writings from the Women's Liberation Movement (Robin Morgan ed., 1970).

n50 See, e.g., NAACP v. Alabama ex rel. Patterson, 357 U.S. 449 (1958) (holding the privacy of such lists protected by the right of association). For a discussion of rights of association in cyberspace, see Jay Krasovec, Comment, Cyberspace: The Final Frontier, for Regulation?, 31 Akron L. Rev. 101, 135 (1997).

n51 See Volokh, supra note 40, at 1833 (arguing that Internet speech will both democratize and diversify speech, while intermediary control will be lessened).

n52 See Siegel, supra note 21, at 377. "The Internet presents a forum in which racists can take their messages and transmit them to individuals around the world with little effort." Id. (citing Sally Greenberg, Threats, Harassment and Hate On-Line: Recent Developments, 6 B.U. Pub. Int. L.J. 673, 673 (1997)).

n53 See Joshua Dressler, Understanding Criminal Law § 9.01[B], at 70 (2d ed. 1995).

n54 The preferred First Amendment remedy for advocacy of violence is "more speech, not enforced silence." Whitney v. California, 274 U.S. 357, 377 (1927).

n55 As an analogy, corporations have decried their inability to track down and defuse rumors about their companies. For years, Proctor & Gamble has fought rumors that their logo revealed "satanic links." In 1999, Proctor & Gamble unsuccessfully sued Amway, trying to stem these rumors. See Procter & Gamble v. Amway, 80 F. Supp. 2d 639 (S.D. Tex. 1999). This case did not involve rumors on the Internet, but Proctor & Gamble has complained elsewhere that their company has been unfairly linked to "satanic groups" on the Internet as well. See, e.g., John Lang, Internet Shows Dark Side in Liz, Tommy Rumors, San Diego Union-Trib., Mar. 23, 1999, at 8.

n56 See Froomkin, supra note 43.

n57 It may be that the law of intellectual property is protecting us from deceptive messages better than other areas of the law. Planned Parenthood Federation of America successfully enjoined an anti-abortion individual, Richard Bucci, doing business as Catholic Radio, from using the domain name "plannedparenthood.com," and from identifying his Web site on the Internet under the name "www.plannedparenthood.com." See Planned Parenthood Fed'n of Am. v. Bucci, No. 97CIV.0629, 1997 U.S. Dist. LEXIS 3338 (S.D.N.Y. Mar. 19, 1997), aff'd, 1998 U.S. App. LEXIS 22179 (2d Cir. Feb. 9, 1998). The defendant raised a First Amendment challenge, which the court denied, finding that "because defendant's use of the term 'planned parenthood' is not part of a communicative message, his infringement on plaintiff's mark is not protected by the First Amendment." Id. at *35-36.

n58 "Metatags are key words used primarily by search engines to index, identify, and document contents of a web page. Metatags are invisible to computer users." Jeffrey J. Look, The Virtual Wild, Wild West (WWW): Intellectual Property Issues in Cyberspace-Trademarks, Service Marks, Copyrights, and Domain Names, 22 U. Ark. Little Rock L. Rev. 49, 79 n.175 (1999). Search engines select a Web site depending on the meta-tags. If a meta- tag is deceptive, it can draw an unwilling audience. For example, an anti-abortion group could insert meta-tags in such a way as to make it appear that their site embodied a pro-choice outlook. See Ira S. Nathenson, Internet, Infoglut, and Invisible Ink: Spamdexing Search Engines with Meta Tags, 12 Harv. J.L. & Tech. 43, 90 (1998) (discussing the negative consequences of misuse of meta-tags); Barbara Anna McCoy, Comment, An Invisible Mark: A Meta-Tag Controversy, 2 J. Small & Emerging Bus. L. 377 (1998) (discussing the use of another company's trademark with its meta- tag).

n59 Avatar space or virtual space has grown out of text-based virtual worlds known as "MUD" or "MOO" space. See Lessig, supra note 29, at 11. Avatar space is similar to cartoons on a monitor. As I understand it, it is as if you could create a character for yourself and inject it into a video game where the other characters were controlled by others. In avatar space, you control your character and can have it interact with others. One can imagine an individual creating a persona and entering avatar space to assault, rape, lynch, and otherwise cause mayhem motivated by racial, gender, or religious hatred.

n60 Other means might include the following: implanting false hyperlinks in a competitor's Web site, capturing Web surfers so that they cannot leave the Web site, and cyberterrorism techniques of sending worms and viruses over the Web.

n61 This is not to downplay the importance of pamphlets, and one might challenge whether the din of a broadly distributed message over the Internet is commensurate with the impact of the pamphlet's message. A quiet pamphlet can have quite an effect. Thomas Paine's The Rights of Man, for example, changed history. See Grant Wood, The Creation of the American Republic, 1776-1787, at 93-94 (1969).

n62 In evaluating e-mail as a basis of government regulation, particularly in the criminal context, other constitutional rights become implicated, such as the Fourth Amendment. Government surveillance of e-mail is governed by a mesh of legislative provisions, some of which are not necessarily consistent. The Electronic Communications Privacy Act of 1986 (ECPA) amended Title III of the Omnibus Crime Control and Safe Streets Act of 1968 to include e-mail and other forms of electronic communications. Title I of the ECPA ("Federal Wiretap Act") governs the interception of electronic communications and requires a court order for interception. Title II of the ECPA ("Stored Communications Act") prohibits the unauthorized access to and disclosure of stored communications without a valid search warrant. See Electronic Communications Privacy Act of 1986, Pub. L. No. 99-508, 100 Stat. 1848 (codified as amended at 18 U.S.C. § 2510 (1994)); see also Steve Jackson Games v. U.S. Secret Serv., 36 F.3d 457 (5th Cir. 1994). The Communications Assistance for Law Enforcement Act (CALEA), 47 U.S.C. § 1001 (1994), requires service carriers to cooperate with law enforcement and also extends the protections under the ECPA to cordless technologies. The Federal Communications Commission (FCC) adopted an interim order covering "packet-mode communications"-voice communications sent over the Internet. Under the order, access to packet-mode communications is not scheduled to take effect until September 30, 2001. See FCC Gives Law Enforcement Six of Nine Electronic Surveillance Capabilities Sought, 65 Crim. L. Rep (BNA) 590 (Sept. 15, 1999). A summary of the order can be found at Communications Assistance for Law Enforcement Act, 64 Fed. Reg. 51,710 (Sept. 24, 1999). The Electronic Privacy Information Center (EPIC), the American Civil Liberties Union, the Electronic Frontier Foundation, the Cellular Telecommunications Industry Association, and the Center for Democracy and Technology have sued to invalidate this interim FCC order, noting broadly that "[p]acket-mode communication is the transmission technology of the Internet." Brief for EPIC, United States Telecom Ass'n v. FCC (CALEA case), 227 F.3d 450 ( D . C . C i r . 2 0 0 0 ) , a v a i l a b l e a t http://www.techlawjournal.com/courts/ustavfcc/20000120.htm.

n63 Hate spam refers to unsolicited e-mails with a message, not of commercial speech, but of hate speech.

n64 After reading the Jake Baker case, one can easily imagine a message that advocates that all women should be tied up, raped, tortured, and injured. See United States v. Baker, 890 F. Supp. 1375 (E.D. Mich. 1995), aff'd sub nom. United States v. Alkhabaz, 104 F.3d 1492 (6th Cir. 1997).

n65 "Remailers generally delete identifying information about incoming e-mails, place a header naming the remailer as the sender or using another identity such as nobody@nowhere." Froomkin, supra note 43, at 415-16.

n66 See Planned Parenthood v. Am. Coalition of Life Activists, 41 F. Supp. 2d 1130 (D. Or. 1999) [hereinafter Planned Parenthood III].

n67 See United States v. Baker, 890 F. Supp. 1375 (E.D. Mich. 1995).

n68 See Planned Parenthood III, 41 F. Supp. 2d at 1131.

n69 See Baker, 890 F. Supp. at 1375-76.

n70 See Planned Parenthood III, 41 F. Supp. 2d 1130; Planned Parenthood v. Am. Coalition of Life Activists, 23 F. Supp. 2d 1182 (D. Or. 1998) [hereinafter Planned Parenthood II]; Planned Parenthood v. Am. Coalition of Life Activists, 945 F. Supp. 1355 (D. Or. 1996) [hereinafter Planned Parenthood I].

n71 See Eric Schaff, Redefining Violence Against Women: The Campaign of Violence and the Delay of RU486, 8 Temp. Pol. & Civ. Rts. L. Rev. 311 (1999).

n72 The American Coalition of Life Activists (ACLA) formed in 1994 when it split off from the anti-abortion movement of Operation Rescue. ACLA advocates the use of force to prevent abortions. Advocates for Life Ministries (ALM), a group closely aligned with ACLA, was active in the formation of ACLA and helps ACLA in its campaign of intimidation. The court in Planned Parenthood III cited testimony of ACLA co-founder and regional director Andrew Burnett in which he stated, "[I]f someone was to condemn any violence against abortion, they probably wouldn't have felt comfortable working with us." 41 F. Supp. 2d at 1136. The court further noted that many of the defendants had "signed the Defensive Action petition approving the murder of Dr. Gunn, [and] refus[ing] to commit to non-violence. Because they advocated the use of 'force' and justifiable homicide, they were no longer allowed to be leaders of Operation Rescue and therefore agreed to form a new organization that became ACLA." Id.

n73 For a full description, see Planned Parenthood I, 945 F. Supp. at 1362-63.

n74 The named plaintiffs were Dr. Robert Crist, Dr. Warren Hern, Dr. Elizabeth Newhall, and Dr. James Newhall. See Planned Parenthood III, 41 F. Supp. 2d at 1131.

n75 See Planned Parenthood I, 945 F. Supp. at 1362.

n76 Id.

n77 Id.

n78 See id.; Planned Parenthood III, 41 F. Supp. 2d at 1132-34.

n79 See Planned Parenthood III, 41 F. Supp. 2d at 1134.

n80 See id. at 1133.

n81 18 U.S.C. § 248 (1994).

n82 18 U.S.C. § 1962 (1994).

n83 Or. Rev. Stat. § 166.720 (1995).

n84 See Planned Parenthood I, 945 F. Supp. at 1385-87.

n85 See id. at 1370.

n86 See Planned Parenthood III, 41 F. Supp. 2d at 1131- 53.

n87 See id. at 1136.

n88 Planned Parenthood I, 945 F. Supp. at 1371 (quoting United States v. Orozco-Santillan, 903 F.2d 1262 (9th Cir. 1990)).

n89 Id. at 1370-72.

n90 Id. at 1372.

n91 See Planned Parenthood III, 41 F. Supp. 2d at 1155- 56.

n92 See id. at 1155 n.1.

n93 See United States v. Baker, 890 F. Supp. 1375 (E.D. Mich. 1995).

n94 See David C. Potter, Note, The Jake Baker Case: True Threats and New Technology, 79 B.U. L. Rev. 779, 779 n.1 (1999).

n95 See Baker, 890 F. Supp. at 1379.

n96 The security staff searched Baker's computer with his consent, thus obviating any of the difficulties discussed in note 62, supra. See id. at 1379 n.4.

n97 See United States v. Alkhabaz, 104 F.3d 1492, 1498 (6th Cir. 1997) (Krupansky, J., dissenting).

n98 See Baker, 890 F. Supp. at 1379.

n99 See id.

n100 Id. at 1389.

n101 See id. at 1387 n.18.

n102 Id. at 1379.

n103 This statute states: "Whoever transmits in interstate or foreign commerce any communication containing any threat to kidnap any person or any threat to injure the person of another, shall be fined under this title or imprisoned not more than five years, or both." 18 U.S.C. § 875(c) (1994).

n104 See Baker, 890 F. Supp. at 1378.

n105 Id. at 1384.

n106 Id. at 1380 (citing United States v. DeAndino, 958 F.2d 146, 148 (6th Cir. 1992)).

n107 See id.

n108 See id. at 1381.That the phrase "true threat" has been used to describe both the statutory intent requirement and the constitutional "unconditional, unequivocal, immediate and specific" requirement does not imply that the two requirements are identical, or that any statement which meets the intent requirement may be prosecuted under § 875(c) without running afoul of the First Amendment. Typically, in the cases [focusing] on the intent requirement, there is no dispute that the statement satisfies the constitutional standard, and the defendant seeks dismissal or reversal of his conviction on the ground that he or she lacked the requisite intent.Id. at 1383.

n109 Id. at 1384.

n110 Id. (quoting United States v. Lincoln, 462 F.2d 1368, 1368 (6th Cir. 1972)).

n111 "What evaluation, if any, was performed by the Washtenaw County Prosecutor, the logical prosecuting authority, is unknown." Id. at 1379 n.5.

n112 See United States v. Alkhabaz, 104 F.3d 1492 (6th Cir. 1997).

n113 Id. at 1495.

n114 See id. at 1496.

n115 See id. at 1502 (Krupansky, J., dissenting).

n116 See id. at 1504.

n117 See Konigsberg v. State Bar of Cal., 366 U.S. 36, 49 (1961).

n118 See id. at 49-50; see also Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (discussing incitement).

n119 Words that constitute a crime, that is, a contract or agreement to buy illegal drugs, blow up a building, or carry out a murder, would not be protected by the First Amendment, nor would words constituting a criminal conspiracy or solicitation to commit a crime. Advocacy of a crime alone, however, is protected unless it directly incites imminent lawless action. See Brandenburg, 395 U.S. at 447. Brandenburg involved the attempted prosecution of a Ku Klux Klan leader for a rally at which he burned a cross, made racial and religious derogatory comments, and stated in the first of two films that there "might have to be some revengeance taken." Id. at 446. Since the Ohio law was overbroad and punished mere advocacy, it was invalidated. See id. at 449.

n120 See Chaplinsky v. New Hampshire, 315 U.S. 568, 571-72 (1942).

n121 See Miller v. California, 413 U.S. 15 (1973).

n122 See Gertz v. Robert Welch, Inc., 418 U.S. 323 (1974); Steven M. Cordero, Comment, Damnum Absque Injuria: Zeran v. AOL and Cyperspace Defamation Law, 9 Fordham Intell. Prop. Media & Ent. L.J. 775 (1999).

n123 See Missouri Knights of the KKK v. Kansas City, 723 F. Supp. 1347 (W.D. Mo. 1989).

n124 See Collin v. Smith, 578 F.2d 1197 (7th Cir. 1978).

n125 Some have complained that the reaction to abortion violence has led to a curtailment of legitimate debate about abortion rights. See Lynn D. Wardle, The Quandary of Pro-Life Free Speech: A Lesson from the Abolitionists, 62 Alb. L. Rev. 853, 882 (1999).

n126 See, e.g., Reno v. ACLU, 521 U.S. 844 (1997) (invalidating the Communications Decency Act); R.A.V. v. City of St. Paul, 505 U.S. 377 (1992) (invalidating a city ordinance regulating bias-motivated speech).

n127 See Steven J. Heyman, Righting the Balance: An Inquiry into the Foundations and Limits of Freedom of Expression, 78 B.U. L. Rev. 1275 (1998).

n128 Id. at 1279.

n129 Id. at 1280.

n130 Id. at 1388.

n131 Id. at 1389.

n132 See David Crump, Camouflaged Incitement: Freedom of Speech, Communicative Torts, and the Borderland of the Brandenburg Test, 29 Ga. L. Rev. 1 (1994).

n133 Id. at 1-2 (footnote omitted).

n134 See id. at 6.

n135 According to Professor Crump,We should start, as before, with Brandenburg. That decision adopts a categorical approach, in which an utterance either is an unprotected incitement to crime or protected expression. That standard may serve us well when the distinction is clear, but like many seminal Supreme Court opinions, it begs the question in close cases.Id. at 45.

n136 See id. at 4-5, 12-13.

n137 See id. at 13.

n138 Id. at 13-14 (footnotes omitted).

n139 See id. at 71-73.

n140 See id. at 45, 73-74.

n141 Professor Crump's evidentiary factors are to be used to determine "whether an utterance is incitement, according to Brandenburg. None of the factors should be considered determinative; that is the essence of a case-by-case approach, and a contrary approach might blind the reviewing court to either camouflaged incitement or protected expression." Id. at 52-53.

n142 See id. at 54.

n143 See id. at 55.

n144 See id. at 56.

n145 See id. at 57.

n146 See id. at 63.

n147 See id. at 65.

n148 See id. at 66.

n149 Id. at 67.

n150 See United States v. Alkhabaz, 104 F.3d 1492, 1496 (6th Cir. 1997) (Krupansky, J., dissenting).

n151 See ACLU of Georgia v. Miller, 977 F. Supp. 1228 (N.D. Ga. 1997). Some aspects of anonymous speech have been regulated. See Froomkin, supra note 43, at 434 (discussing the Child Protection and Obscenity Enforcement Act of 1988, as amended by the Child Protection Restoration and Penalties Enhancement Act of 1990, which has been upheld by the D.C. Circuit).

n152 "SPAM" connotes a packaged meat product so ubiquitous in World War II. See Official SPAM Home Page, at http://www.spam.com (last visited Feb. 19, 2001). However, in the information age, "spam" has taken on new meanings. See E-mail from wsmith@wordsmith.org (Sept. 12, 2000) (copy on file with author). Many believe that the term "spam" originated in a skit from Monty Python's Flying Circus. See Anne E. Hawley, Comment, Taking Spam Out of Your Cyberspace Diet: Common Law Applied to Bulk Unsolicited Advertising Via Electronic Mail, 66 UMKC L. Rev. 381, 381 n.3 (1997) (reproducing the text of the skit). "Spam is unsolicited advertisements that are broadcast to hundreds or thousands of discussion groups, listservs, or individual e-mail accounts at one time." John M. Kuttler, Actual and Potential Attorney Use of the Internet, 46 Am. U. L. Rev. 328, 350 n.94 (1996) (citing Allison G. Burgess, The Lawyer's Guide to the Internet 338 (1995)).

n153 Ray Everett-Church, Written Testimony on Behalf of the Coalition Against Unsolicited Commercial Email (June 17, 1 9 9 8 ) , a v a i l a b l e a t http://www.cauce.org/testimony/senate testimony.html (arguing that spam threatens the future of online commerce by knocking out systems, shifting tremendous costs onto recipients, and encouraging massive abuse).

n154 See Seidl v. Greentree Mortgage Co., 30 F. Supp. 2d 1292 (D. Colo. 1998).

n155 See Joshua A. Marcus, Note, Commercial Speech on the Internet: Spam and the First Amendment, 16 Cardozo Arts & Ent. L.J. 245 (1998).

n156 See Simpson Garfinkel, Database Nation (1999).

n157 See Froomkin, supra note 43, at 479-81.Both public and private organizations are acquiring unprecedented abilities to build, sell, and use consumer profile data. Every transaction on the World Wide Web, for example, from catalog sales to information acquisition, can be recorded and archived by either party to it. As a result, the Internet could become the mother lode of consumer profile information; parallel developments in the public sphere make it increasingly feasible to monitor what citizens do and where they go. Combine the two, and there is little privacy left. Databases erode the citizen's control over her personal information in several ways. Computerized records allow a firm to form a consumer profile based on the a [sic] customer's transactions with that company. At a slightly more complex level, firms sell customer lists to each other, which may result in junk mail or increased information to the consumer, depending on one's perspective or good fortune. Meanwhile, in the U.S., social security numbers and driver's license numbers (often the same) have become de facto national ID numbers. The most important part of the emerging database phenomenon, however, arises from the combination of the growth in computer processing power with the likelihood that routine personal data collection will soon become nearly ubiquitous.Id. at 479-80 (footnotes omitted).

n158 See id. at 482-83.[A] chilling example of this data linkage is the sale by Farrell's Ice Cream Parlor of the names of those claiming free sundaes on their birthdays. The list was purchased by a marketing firm, which in turn sold them to the Selective Service System. Some of the ice-cream eaters soon found draft registration warnings in their mail.Id. at 482 (footnotes omitted).

n159 See id. at 483. "The ice cream example is trivial compared to what is ahead, given the likely omnipresence of data collection. Data collection will grow in at least five areas: medical history, government records, personal movements, transactions, and reading and viewing habits. Between them these five areas cover most of modern life." Id.

n160 See Peter Swire & Robert Litan, None of Your Business: World Data Flows, Electronic Commerce and the European Privacy Directive (1998).

n161 See Froomkin, supra note 43, at 489-90.The existence of large, and linked, databases is potentially alarming in the United States because the U.S. has relatively few data protection statutes along the lines of the European and Canadian models. U.S. data protection laws place some limits on the use of government databases. They also give consumers the right to correct erroneous entries that may be kept in their files by private credit bureaus.Id. (footnotes omitted); see also Paul Schwartz, Data Processing and Government Administration: The Failure of the American Legal Response to the Computer, 43 Hastings L.J. 1321, 1324 (1992) (stating that from an international perspective, the American legislative response to computer processing of personal data is incomplete).

n162 For example, the Southern Poverty Law Center, the Anti-Defamation League, and the Simon Weisenthal Center successfully persuaded the ISP GeoCities and several other ISPs to remove the KKK Web site from their servers. See Siegel, supra note 21, at 382.

n163 See, e.g., Cyber Promotions, Inc. v. Apex Global Info. Serv., No. 97-5931, 1997 U.S. Dist. LEXIS 15344, at *1 (E.D. Pa. Sept. 30, 1997) (discussing ping attacks in retaliation for SPAM).

n164 Marcus, supra note 155, at 246 (footnote omitted).

n165 See Westboro Baptist Church, Warning Page, at http://www.godhatesfags.com (last visited Aug. 26, 2000).

n166 For example, on the first page of the Web site found at http://www.whitesonly.net, the viewer encounters a page that reads "Nigger Jokes." The warning reads: "Enter only if you realize this site is meant as a Joke with material about niggers on it. Adults only." At another part of this site that leads to an activity entitled "pin the noose on the nigger," a Microsoft Internet Explorer warning reads: "This page is meant as a joke only! We do not condone illegal activity. Do not enter if you are u n d e r 1 8 y e a r s o f a g e . " S e e http://www.whitesonly.net/hanganigger.html (last visited Feb. 15, 2001).