ABSTRACT

John Doe has become a popular defamation defendant as corporations and their officers bring defamation suits for statements made about them in Internet discussion fora. These new suits are not even arguably about recovering money damages but instead are brought for symbolic reasons -- some worthy, some not so worthy. If the only consequence of these suits were that Internet users were held accountable for their speech, the suits would be an unalloyed good. However, these suits threaten to suppress legitimate criticism along with intentional and reckless falsehoods, and existing First Amendment law doctrines are not responsive to the threat these suits pose to Internet discourse. Although the constitutional privilege for opinion holds promise as a solution to this problem, the Supreme Court's jurisprudence provides little assurance that the privilege can protect the "robust, uninhibited, and wide-open nature" of Internet discourse without giving Internet speakers free license to harm the reputation of [*pg 856] others. Therefore, this Article attempts to articulate a theory that justifies protecting John Doe and suggests the steps courts should take to adapt the existing opinion privilege to the unique context of cyberspace.

INTRODUCTION

[T]he poisonous atmosphere of the easy lie can infect and degrade a whole society.¹

[T]he First Amendment . . . presupposes that right conclusions are more likely to be gathered out of a multitude of tongues, than through any kind of authoritative selection. To many this is, and always will be, folly; but we have staked upon it our all.²

John Doe is a day trader.³ Trading stocks is a hobby bordering on an obsession, and, like many other day traders, John Doe likes to exchange information about stocks via online message boards.⁴ But as Doe recently learned, free speech on the Internet may not be as free as he thought. After posting a scathing message accusing Net Company of defrauding its investors and accusing its CEO of being a liar and a cheat, John Doe found himself a defendant in a libel suit.⁵

[*pg 858]

This hypothetical case, Net Company v. John Doe, is typical of a rapidly expanding new category of Internet libel suits.⁶ One of the [*pg 859] most striking features of these new cases is that, unlike most libel suits,⁷ they are not even arguably about recovering money damages, for the typical John Doe has neither deep pockets nor libel insurance from which to satisfy a defamation judgment.⁸ Why, then, do plain- [*pg 860] tiffs, many of whom are wealthy corporations, choose to sue relatively impecunious John Does? The goals of this new breed of libel action are largely symbolic,⁹ the primary goal being to silence John Doe and others like him.¹⁰ This feature, standing alone, does not distinguish these new Internet libel cases from more traditional ones.¹¹ All libel suits are at least partially about silencing the defendant,¹² and from the standpoint of traditional First Amendment law, there is no harm in silencing knowingly or recklessly false statements of fact, for these statements have no value to public discourse.¹³

What is unique about these new Internet suits is the threat they pose to the new realm of discourse that has sprung up on the Internet. The promise of the Internet is empowerment: it empowers ordinary individuals with limited financial resources to "publish" their views on matters of public concern.¹⁴ The Internet is therefore a powerful tool for equalizing imbalances of power by giving voice to the disenfranchised and by allowing more democratic participation in public [*pg 861] discourse.¹⁵ In other words, the Internet allows ordinary John Does to participate as never before in public discourse, and hence, to shape public policy.¹⁶ Yet, suits like the hypothetical John Doe suit discussed above threaten to reestablish existing hierarchies of power, as powerful corporate Goliaths sue their critics for speaking their minds. Defendants like John Doe typically lack the resources necessary to defend against a defamation action, much less the resources to satisfy a judgment. Thus, these Internet defamation actions threaten not only to deter the individual who is sued from speaking out, but also to encourage undue self-censorship among the other John Does who frequent Internet discussion fora.

Although one might intuitively expect the First Amendment¹⁷ to prevent powerful plaintiffs from silencing their critics,¹⁸ the First Amendment extends only limited protections in such circumstances. Beginning with the landmark decision of New York Times Co. v. Sullivan,¹⁹ the Supreme Court grafted numerous constitutional limitations onto the structure of the defamation tort. But Sullivan and much of its progeny involved an individual plaintiff (often a public official or public figure) challenging statements made by a relatively powerful media defendant.²⁰ The First Amendment jurisprudence that devel- [*pg 862] oped was therefore responsive to the culture of the institutional press²¹ and its need to deliver information quickly without risking crippling liability for minor mistakes of fact.²²

It is little wonder that this same jurisprudence may not be responsive to the emerging institutional culture -- or, more appropriately, cultures²³ -- of the Internet. Although Internet communications are almost invariably "written" communications, they lack the formal characteristics of written communications in the "real world."²⁴ In the real world, the author is separated from her audience by both space and time, and this separation interposes a formal distance between author and audience, a distance reinforced by the conventions of written communication. Internet communications lack this formal distance. Because communication can occur almost instantaneously, participants in online discussions place a premium on speed.²⁵ Indeed, in many fora, speed takes precedence over all other values, including [*pg 863] not just accuracy but even grammar, spelling, and punctuation.²⁶ Hyperbole and exaggeration are common, and "venting" is at least as common as careful and considered argumentation.²⁷ The fact that many Internet speakers employ online pseudonyms²⁸ tends to heighten this sense that "anything goes," and some commentators have likened cyberspace to a frontier society free from the conventions and constraints that limit discourse in the real world.²⁹ While this view is undoubtedly overstated, certainly the immediacy and informality of Internet communications may be central to its widespread appeal.

Although Internet communications may have the ephemeral qualities of gossip with regard to accuracy,³⁰ they are communicated through a medium more pervasive than print, and for this reason they have tremendous power to harm reputation.³¹ Once a message enters [*pg 864] cyberspace, millions of people worldwide can gain access to it.³² Even if the message is posted in a discussion forum frequented by only a handful of people, any one of them can republish the message by printing it or, as is more likely, by forwarding it instantly to a different discussion forum. And if the message is sufficiently provocative, it may be republished again and again.³³ The extraordinary capacity of the Internet to replicate almost endlessly any defamatory message lends credence to the notion that "the truth rarely catches up with a lie."³⁴ The problem for libel law,³⁵ then, is how to protect reputation without squelching the potential of the Internet as a medium of public discourse. This Article attempts to solve this problem and, in the pro- [*pg 865] cess, to assess the proper role for defamation law in "civilizing" discourse in cyberspace.

Part I examines a perplexing feature of the new Internet libel suits: why do plaintiffs wish to sue John Doe when they can have no hope of recovery? As this part argues, plaintiffs find bringing suit worthwhile because of the distinctive nature of libel actions, which allows plaintiffs to pursue symbolic goals regardless of any monetary award. Yet, while it might be tempting to characterize all of these new suits as Internet SLAPPs (i.e., Strategic Lawsuits Against Public Participation)³⁶ brought by plaintiffs solely to harass and to silence their critics, this characterization ignores the power that the Internet gives irresponsible speakers to damage the reputations of their targets and underestimates the potential benefits that defamation law may bring to Internet discourse.

Part II argues that, despite the potential benefit, strict application of existing law threatens to chill unduly the ordinary John Does who frequent online discussion fora and, by doing so, threatens the Internet's promise as a medium for revitalizing the "marketplace of ideas"³⁷ metaphor that lies at the heart of First Amendment theory. Part II appraises John Doe's contribution to public discourse and concludes that it warrants substantial, though not unlimited, First Amendment "breathing space" to guarantee its survival.

Existing First Amendment protections, however, are inadequate to prevent corporations from intimidating John Doe into silence, at least so long as these protections do not recognize the unique culture of the Internet message boards. Part III therefore contends that the constitutional privilege for nonfactual expression (the "opinion privilege") may be a viable defense against the use of defamation law to silence John Doe, but only if courts are willing to adapt the privilege to the unique social context of cyberspace. The Article concludes with a plea for the Supreme Court to remedy the deficits in its defamation jurisprudence and gives practical guidance to lower courts that wish to protect John Doe in the likely event that this plea goes unheeded.

[*pg 866]

I. SUING JOHN DOE

It is easy to portray the recent rash of libel actions against Internet users³⁸ as David versus Goliath battles, or even perhaps as Internet SLAPP suits, used by big corporations to intimidate their critics into silence. After all, these suits almost inevitably involve wealthy and powerful plaintiffs suing anonymous individuals for speaking out on matters of public interest. However, upon closer examination, this characterization is too facile to capture the complex issues raised by these new suits.

A. HealthSouth v. Krum and the New Internet Libel Cases

Consider the case of HealthSouth Corp. v. John Doe,³⁹ which later became HealthSouth Corp. v. Krum.⁴⁰ There could hardly be a less sympathetic John Doe defendant than Peter Krum. Peter Krum was a disgruntled former employee of HealthSouth,⁴¹ a publicly traded corporation that operates rehabilitative health care facilities.⁴² Posting under the name "I AM DIRK DIGGLER," a reference to [*pg 867] the well-endowed porn star in the 1997 movie Boogie Nights,⁴³ Krum verbally savaged HealthSouth, its CEO Richard Scrushy, and even Scrushy's wife, Leslie, on a Yahoo! Finance message board dedicated to discussion of the corporation.⁴⁴ Krum's postings ranged from relatively innocuous statements about HealthSouth's "self-serving" management⁴⁵ to accusations that Richard Scrushy was "bilking . . . [M]edicare reimbursement."⁴⁶ The most egregious postings by Krum were those that discussed, in salacious detail, an affair "I AM DIRK DIGGLER" was allegedly having with Leslie Scrushy.⁴⁷ Krum, for example, wrote, "I am dirk diggler and I have what [Richard] Scrushy wants. Too bad I keep giving it to his new wife . . . [and] [a]s for those of you who disapprove of my crowing about sexual liasons [sic] with Dick's wife, lighten up. I'm practicing safe sex."⁴⁸ Krum seemed to relent from his vicious attacks at one point, stating that he would be "'toning down'"⁴⁹ his statements because he "felt blessed" that his wife was expecting.⁵⁰ Yet, just a week later, Krum was back to his old tricks,⁵¹ which ultimately prompted HealthSouth to sue for libel and commercial disparagement and Richard and Leslie Scrushy to sue for libel and intentional infliction of emotional distress.⁵²

There is no real question that the statements at issue in Krum were at least facially defamatory. A defamatory statement is a false statement that harms or tends to harm⁵³ an individual's reputation in [*pg 868] the eyes of his community.⁵⁴ Accusations of dishonesty, criminality, and adultery are prototypical defamatory statements.⁵⁵ The real issue raised by the case is what the plaintiffs hoped to gain by suing "I AM DIRK DIGGLER." HealthSouth operates almost two thousand rehabilitation centers in the United States,⁵⁶ Australia, and the United Kingdom, with recent annual revenues in the billions.⁵⁷ Its CEO, Richard Scrushy, makes $13 million a year.⁵⁸ Even before Yahoo! revealed Krum's identity, the plaintiffs could have had no expectation that "I AM DIRK DIGGLER" would have the resources necessary to satisfy a defamation judgment. As it turned out, Krum was a food-service worker at Penn State University, who lost his job (and his annual salary of $35,000) once the plaintiffs filed their suit.⁵⁹ Thus, even if one concedes, as one must, that Peter Krum was an embittered and malicious former employee engaged in what might be thought of as a new form of industrial sabotage, it is not entirely clear what HealthSouth stood to gain by suing him.

B. The Practical Effect of Section 230 of the Communications Decency Act

Obviously, plaintiffs would prefer to sue defendants with deeper pockets than John Does typically have. Indeed, the first generation of Internet libel actions sought to impose liability on Internet service providers ("ISPs") for defamatory messages posted by their subscribers.⁶⁰ ISPs are logical targets for defamation suits. Suing the ISP frees [*pg 869][*pg 870] plaintiffs from having to discover the identity of the person who posted the message, and ISPs have plenty of money to satisfy defamation judgments.⁶¹ Moreover, ISPs have at least some of the characteristics of a traditional publisher, including, in some cases, the ability to edit content supplied by a third party. Logical or not, Congress largely foreclosed access to these deep-pocket defendants when it passed section 230 of the Communications Decency Act ("CDA").⁶² Subsection 230(c)(1) provides that "[n]o provider or user of an interactive computer service⁶³ shall be treated as the publisher or speaker of any information provided by another information content pro- [*pg 871] vider."⁶⁴ Despite the broad wording of subsection (c)(1), the title of the subsection (c) suggests that its purpose is to protect "Good Samaritans," as the Act dubbed them,⁶⁵ by shielding them from liability when they attempt to censor offensive material⁶⁶ on the Internet.

Both the title of the Act and its legislative history suggest that Congress meant to establish a policy that ISPs would not be subject to liability as "publishers" of content posted by third parties just because they exercise a limited degree of editorial control over content.⁶⁷ But court decisions interpreting subsection 230(c) have broadened its ambit far beyond merely protecting "Good Samaritan" editorial control. As interpreted, section 230 gives ISPs complete immunity from liability for defamatory content initiated by third parties, even if the ISP consciously decides to republish the defamatory content.⁶⁸ The practi- [*pg 872] cal effect of these interpretations of section 230 of the CDA is to leave Internet defamation victims with no deep pocket to sue. The defamed plaintiff can no longer sue the intermediary who republished a defamatory communication. Instead, the plaintiff must go to the source and sue the person who originated the defamatory communication, even if that person is an unknown John Doe.

C. The Nonmonetary Benefits of Libel Actions

Thus, one answer to the question of why sue John Doe is that he is the only person left to sue.⁶⁹ But this is an inadequate answer. Most tort actions will never be pursued if it is clear that the potential defendant lacks the resources to satisfy a judgment.⁷⁰ Most plaintiffs do not have the money to litigate based on principle,⁷¹ and, even in contingency-fee cases most plaintiffs' attorneys realize that 33% (the typical contingency fee) of zero is still zero.⁷²

The sudden surge in John Doe suits stems from the fact that many defamation actions are not really about money.⁷³ If they were, the tremendous obstacles to recovery would almost certainly make plaintiffs (and their lawyers) conclude that the game was not worth the candle. Defamation plaintiffs often find their suits derailed at the very early stages, with never so much as a mention of the First [*pg 873] Amendment.⁷⁴ The common law elements of defamation,⁷⁵ which have been set for centuries,⁷⁶ remain "filled with technicalities and traps for the unwary."⁷⁷ A plaintiff must begin her suit by proving, at a minimum,⁷⁸ that the defendant made a defamatory communication about the plaintiff to a third party.⁷⁹ What she must prove next depends on whether the communication was communicated orally or whether it was written. If the defamation was oral (i.e., slanderous), the plaintiff will have to prove that the defendant's slander caused her economic or pecuniary loss, unless the slanderous allegation falls into one of the four narrowly defined categories of slander per se.⁸⁰ If the defamation was written (i.e., libelous), the plaintiff ordinarily does not have to prove that it caused her any economic loss, unless the defamatory meaning was not obvious from the statement itself.⁸¹ Even if a plain- [*pg 874] tiff is able to prove the elements of her case, she must still overcome a host of common law privileges designed to protect reports of government meetings,⁸² job recommendations,⁸³ and other types of speech deemed worthy of extra protection from defamation liability.⁸⁴

In the unlikely event that the plaintiff makes it past the high barriers posed by tort law, First Amendment obstacles will prove almost insurmountable.⁸⁵ What these obstacles are depends on a number of factors: the identity of the plaintiff, the identity of the defendant, and the type of speech at issue.⁸⁶ Public officials and public figures must prove that the defendant's statements were false and that they were made intentionally or with reckless disregard of their falsity, at least where the speech at issue deals with a matter of public concern.⁸⁷ Private figures face a different set of requirements. Private figures suing for speech on a matter of public concern must prove falsity⁸⁸ and neg- [*pg 875] ligence (at least) on the part of the defendant in order to recover.⁸⁹ Private figures suing for speech on private matters may not face any First Amendment hurdles,⁹⁰ but the Supreme Court's decisions make at least one thing clear: truly private figure/private concern cases are likely to be few and far between.⁹¹ As if these obstacles were not enough, a host of other constitutional privileges for hyperbole,⁹² satire,⁹³ and the like stand as independent obstacles to recovery.⁹⁴

Empirical studies confirm that the practical effect of these labyrinthine doctrines is to make it almost impossible for any plaintiff to succeed in a defamation action.⁹⁵ Statistics show that only 13% of plaintiffs ultimately prevail in libel litigation,⁹⁶ and, as one commentator has observed, the few plaintiffs who do prevail owe more to good fortune than "to their virtue, their skill, or the justice of their cause."⁹⁷ However, even these grim statistics do not indicate that it is always economically irrational to sue for defamation; after all, defamation [*pg 876] verdicts reach millions of dollars,⁹⁸ which may be large enough to make it worthwhile to gamble on a large recovery.⁹⁹

But the slim chance of a damages award does not explain why plaintiffs sue in most defamation cases,¹⁰⁰ and it certainly cannot explain why plaintiffs sue in Internet libel cases. Instead, what the evidence suggests is that libel plaintiffs often sue because they believe that the social and psychological benefits of suing make it worthwhile, even if they never recover a money judgment from the defamer.¹⁰¹ Plaintiffs often seek vindication, and bringing suit provides a means -- perhaps the only means available -- to announce to the world that the defendants' statements were false.¹⁰² On the other hand, plaintiffs may be seeking an even simpler goal: they may just want the defamation to stop, and a defamation suit is the only legal tool available to accomplish this goal. Thus, to say that the new Internet libel suits are brought for symbolic goals is not necessarily pejorative, particularly when a symbolic victory is the only kind available.

D. Why Corporations Sue John Doe

This symbolic aspect of bringing a defamation action is particularly evident in the John Doe cases, although corporate plaintiffs may be seeking a symbolic victory partially for financial reasons.¹⁰³ Many of these actions appear to be brought as merely one tool in a concerted public relations campaign. Corporations often issue press re- [*pg 877] leases announcing their decision to sue those who post on financial bulletin boards,¹⁰⁴ even though doing so gives more widespread publicity to the defendants' remarks than they received at the time they were posted.¹⁰⁵ The tendency to publicize the decision to file suit may be partly a function of the economic motivations that drive corporate plaintiffs to sue. Although corporations that sue John Doe may never recover money damages, they may still deem it economically rational to sue the pseudonymous posters who make negative statements about them on financial message boards. Corporate plaintiffs are at least partly motivated by the fear that negative statements on financial bulletin boards will drive down their stock price.¹⁰⁶ The stock market trades on information, and negative information shifts stock prices very quickly.¹⁰⁷ Hence, corporations must act quickly to offset the potentially negative effects of defamatory messages by offering an alternative version of events.¹⁰⁸ Indeed, failure to respond may itself be deemed an admission that the negative statements are true. Suing John Doe, as it turns out, is good for business to the extent that it [*pg 878] sends a message to shareholders that the corporation is stable and strong.¹⁰⁹

It may be no coincidence, then, that some corporations decide to bring suit for Internet libel when they are already taking a beating in the market for unrelated reasons. The chronology of Hitsgalore.com's decision to file a libel action,¹¹⁰ for example, seems somewhat suspect. Hitsgalore.com, Inc., is an Internet company that provides a "business-to-business Web portal and search engine."¹¹¹ In 1998, Hitsgalore.com had a reported revenue of less than $10,000.¹¹² In March of 1999,¹¹³ Hitsgalore.com entered a "reverse merger"¹¹⁴ with another company to become a publicly traded over-the-counter ("OTC") security on the OTC Bulletin Board.¹¹⁵ The move was a stunning success. Due to investor enchantment with Internet stocks,¹¹⁶ the market [*pg 879] value of Hitsgalore.com leapt from $53 million to $1 billion in less than three months.¹¹⁷ In March the stock was trading at $2.38.¹¹⁸ By May 10, 1999, the stock was trading at $20.69 per share.¹¹⁹ Hoping to capitalize on this tremendous success, the company announced plans to launch a nationwide advertising campaign.¹²⁰ But on May 11, the Cinderella story took an unhappy turn. Bloomberg News reported, apparently truthfully, that a principal shareholder and "founder" of Hitsgalore.com had had a run-in with the Federal Trade Commission ("FTC") over false and deceptive promises he had made to customers of an earlier Internet firm with which he was involved.¹²¹ The report also implied, perhaps incorrectly,¹²² that Hitsgalore.com improperly failed to disclose the FTC case to the Securities and Exchange Commission ("SEC").¹²³

In the wake of the report, Hitsgalore.com's stock price cratered.¹²⁴ The bad press continued, and lawyers began announcing [*pg 880] plans to pursue class action lawsuits against Hitsgalore.com on behalf of angry shareholders.¹²⁵ Discussions of the company on the Raging Bull and Silicon Investor bulletin boards also took a negative turn.¹²⁶ Many message posters saw Hitsgalore.com as a "scam," and many reviled its officers as "crooks," "criminals," and "con men."¹²⁷ One poster, "Mr. Pink," stated that "[t]hese crooks belong in Jail!", and he even dared Hitsgalore.com to sue him: "No disclaimer, this is not opinion but a fact and if company doesn't like it, please sue Him; discovery will be a treat!"¹²⁸

Hitsgalore.com obliged "Mr. Pink" by promptly suing him and one hundred other John Doe defendants (although, notably, the company chose to identify only one defendant by name and only four by their "screen names").¹²⁹ The $20-million suit included claims for libel, tortious interference, and a civil conspiracy to defame the plaintiff in order to unlawfully drive Hitsgalore.com's stock price "into a downward spiral" and thus to enable the defendants to profit by selling the common stock short.¹³⁰ And, in addition to the $20 million that plaintiff asked for in compensation for the 75% drop in its stock price, the plaintiff also sought removal of the allegedly defamatory postings from the message boards and an injunction against further postings of that nature.¹³¹

A company's decision to sue when it finds itself in a position similar to Hitsgalore.com makes some degree of financial sense. By announcing its decision to file suit, the company appears to respond aggressively to the Internet rumor-mongers who revel in reports of its demise. Bringing suit sends a message to shareholders and potential [*pg 881] investors that they should not believe all the negative information they hear about the company; it quells rumors and takes the focus away from the negative press the company has been receiving -- whether true or untrue. Even if the company ultimately decides not to pursue its action past filing a complaint, it may have won a symbolic victory simply by suing John Doe.

Suing John Doe may also be a victory if it silences John Doe.¹³² Although "silencing" ordinarily has a pejorative connotation,¹³³ it is perfectly legitimate for plaintiffs to seek to stop an onslaught of offensive and damaging untruths. Indeed, silencing John Doe may be one of the chief motivations behind the new Internet libel actions. Armed with subpoenas, plaintiffs often are able to discover the real identity of the John Doe who has attacked them in an Internet discussion forum.¹³⁴ The mere fact of being uncovered may itself be enough [*pg 882] to stop the alleged defamer from posting further messages.¹³⁵ Suing may be particularly effective in stopping an onslaught of damaging messages being posted by corporate insiders,¹³⁶ since the fear of losing their jobs may make "inside agitators" tread more cautiously in the online environment.¹³⁷ This strategy may not always be successful, of course. Some John Doe defendants have begun to fight these subpoenas with their own subpoenas seeking to uncover embarrassing information about the company,¹³⁸ and at least one John Doe seems to have escalated his remarks once the plaintiffs tried to uncover his identity.¹³⁹ Nevertheless, removing the cloak of anonymity from John Doe defendants is likely also to remove their sense that anything goes, and the mere threat of being revealed may be enough to force a defendant to temper his remarks in the future.

Even silencing the defendant may not go far enough for some plaintiffs. Corporate plaintiffs, like individual plaintiffs, will sometimes insist that the defendant publicly atone for posting defamatory falsehoods, and, as Krum illustrates, this goal can be attained even without a formal trial. HealthSouth not only sued Peter Krum but also sought criminal harassment and stalking charges against him.¹⁴⁰ According to court records, the abashed Krum apparently had never realized that his words might have consequences.¹⁴¹ Krum made no effort to contest the suit, but instead agreed to sign an apology, pay $50 a month to a charity of the plaintiffs' choice for a period of four years, and to perform three hours of community service every week for two [*pg 883] years.¹⁴² The fact that Richard Scrushy, who makes $13 million a year, deemed it necessary to exact this modest monetary fine on a defendant who now works as a fry cook making $22,000 a year¹⁴³ suggests that the suit was not about the money but about the principle at stake: the plaintiffs must have thought it worthwhile to score at least a symbolic victory over the likes of Peter Krum.

E. David vs. Goliath?

It is tempting to portray the new Internet libel suits as David versus Goliath battles, pitting ordinary John Does against powerful corporate interests out to intimidate their critics into silence. This image is bolstered by the fact that libel suits are hard to win but easy to bring.¹⁴⁴ Many corporate plaintiffs that sue for Internet libel seek to send a message to the public that they will pursue aggressively anyone who criticizes them online, and these plaintiffs seem to be using libel law to squelch not just defamatory falsehoods but legitimate criticism as well (a topic this Article discusses in the next part).¹⁴⁵ But to focus only on the obvious power differential between corporate plaintiffs and John Doe defendants is to ignore the real harm Internet libel can cause.

As a rule, corporate plaintiffs have more power and wealth in the "real world" than any anonymous John Doe could hope to have, and in the real world that power and wealth translates into corporations' having the ability to be heard. Corporations often will be able to conduct expensive media campaigns to influence favorable coverages, [*pg 884] and corporate CEOs can use traditional media outlets to communicate with the public.

But the Internet helps equalize power imbalances in the real world by giving anonymous John Does like "I AM DIRK DIGGLER,"¹⁴⁶ "Mr. Pink,"¹⁴⁷ or "HiSiCat"¹⁴⁸ a meaningful opportunity to be heard. In the online world, every John Doe is potentially a publisher, capable of transmitting messages instantaneously to millions of readers. More significantly, the Internet allows John Doe to target a message to an audience with common interests and concerns, the very audience likely to be most receptive to his comments. Thus, for example, a John Doe who wishes to discuss the management and future prospects of a company in which he has just invested can find a bulletin board dedicated to the use of like-minded shareholders. Moreover, John Doe's online comments can have real-world effects. While the financial bulletin boards ordinarily give notice to subscribers that the messages posted are merely the opinions of the author and that they should not be relied on to trade,¹⁴⁹ people do, of course, use them to trade.¹⁵⁰ If John Doe is a scrupulous critic of a particular corporation and its CEO, the Internet is a powerful tool for him to begin a dialogue about the corporation and to convey his criticisms to a receptive audience.¹⁵¹

If John Doe is unscrupulous or merely reckless, however, he can use the power the Internet gives him to inflict serious harm on the corporation. He can pollute the information stream with defamatory falsehoods, which may in turn influence other investors to question the corporation's credibility or financial health.¹⁵² Moreover, once the defamatory information enters the information stream, it may have a greater impact than if it had appeared in print. Because the defamatory statements can be copied and posted in other Internet discussion [*pg 885] fora, both the potential audience and the subsequent potential for harm are magnified. And, as the persistence of Internet hoaxes demonstrates,¹⁵³ once a rumor takes hold in cyberspace, it may be almost impossible to root out. Thus, to view the rise of John Doe libel suits as merely an attempt by powerful corporations to intimidate their critics into silence is to substitute metaphor for analysis and to suppress the fact that the "speech" of ordinary John Does, both scrupulous and unscrupulous, has more power to affect corporate interests than ever before.

F. Benefits of Civilizing Cyberspace

To view these libel suits as no more than attempts to silence John Doe also ignores the potential contribution of defamation law to cyberspace discourse. The Internet has often been compared to the Wild West, a frontier society free from the stifling conventions of civilization, and some have even argued that defamation law is an unnecessary anachronism in this new society.¹⁵⁴ A strong argument can be made, however, that Internet discourse could benefit from the civilizing influence of defamation law.

"[A] civilized society," as David Anderson has written, "cannot refuse to protect reputation,"¹⁵⁵ but it is worthwhile to explore why this is so. By protecting reputation, defamation law safeguards the dignity of citizens.¹⁵⁶ Defamation law therefore reflects liberal society's "basic concept of the essential dignity and worth of every human being."¹⁵⁷ Even if this were the only contribution of defamation law, policymakers might justifiably conclude that defamation law is necessary to mediate interactions between "netizens" in order to preserve individual dignity.¹⁵⁸ Yet, as Robert Post has shown, defamation law [*pg 886] exists not merely to validate the dignitary interests of individual plaintiffs; defamation law also helps to make meaningful discourse possible.¹⁵⁹ Defamation law has a civilizing influence on public discourse: it gives society a means for announcing that certain speech has crossed the bounds of propriety.¹⁶⁰

This civilizing influence could benefit Internet discourse in at least two ways. First, to the extent that the prospect of being verbally "attacked" deters some citizens from participating in Internet discourse, application of defamation law can help to ensure that Internet discourse remain open to all.¹⁶¹ Second, defamation law might help to cure the largest single threat to meaningful discourse in cyberspace: incoherence. Precisely because the Internet makes every person a publisher, the volume of information available is enormous. Furthermore, since much of the information is "published" by unknown John Does, it is difficult to evaluate its credibility.¹⁶² Even the participants on financial bulletin boards complain of this problem. The people who use the boards not only seek the experience of discussing the stock market or particular companies with like-minded individuals; they also seek information that would enable them to evaluate a particular company. But because most users post messages pseudony- [*pg 887] mously, it becomes difficult (though not impossible) to evaluate the reliability of the information. Thus, participants criticize those who post inane messages or those who they suspect are "bashers" -- people who post negative messages trying to drive the stock price down so that they can profit from selling short. Some bulletin board participants even laud corporations for bringing defamation suits against other bulletin board participants,¹⁶³ indicating their recognition that some curb on abusive or meaningless speech is necessary if meaningful discussion is to occur. One bulletin board service, Raging Bull, offers participants a nonlegal means of ridding the information stream of unwanted flotsam: it offers a button that allows users to block the messages of other users whose comments are inane, useless, or merely annoying.¹⁶⁴

Defamation law has the potential to curb the excesses of Internet discourse and to make Internet discourse not just more civil but more rational as well. It is important to note, moreover, that defamation law can serve these functions regardless of whether plaintiffs actually pursue their lawsuits all the way to judgment. Consider again the case of HealthSouth v. Krum.¹⁶⁵ The result of that case sent a powerful message to other users of the Yahoo! financial message boards on which Krum posted his retraction. That message -- that users should not rely on anonymity to shield them from being sued if they post abusive and untrue messages -- is one that has positive implications for Internet discourse. The quality of speech is improved when speakers realize that their speech has consequences.¹⁶⁶

II. CHILLING JOHN DOE

If the only effect of these new Internet libel suits were to send a message to Internet users that they are accountable for their speech, [*pg 888] these suits might well be an unalloyed good. It might still be worth questioning whether plaintiffs should be allowed to use the legal system to pursue symbolic goals,¹⁶⁷ but the secondary benefits of the suits might be enough to overcome this objection.¹⁶⁸ The problem, however, is that these new suits threaten to make Internet users too accountable for their speech, thereby threatening to suppress legitimate criticism along with intentional falsehoods. Although existing law has a number of doctrines available to combat the chill of defamation law, these doctrines are ill suited to the unique context of cyberspace.

A. The Chill in Cyberspace

Much ink has been spilled on the chilling effect,¹⁶⁹ but the basic idea is a simple one. "The very essence of a chilling effect," as Professor Frederick Schauer has observed, "is an act of deterrence."¹⁷⁰ Defamation law legitimately seeks to deter individuals from communicating defamatory falsehoods; the problem arises, for First Amendment purposes when defamation law "overdeters" -- that is, when it deters speech that is truthful or nondefamatory -- for such speech occupies a "preferred position" in the constitutional hierarchy of values.¹⁷¹ In other words, the chilling effect occurs when defamation law encourages prospective speakers to engage in undue self-censorship to avoid the negative consequences of speaking.¹⁷² Although commentators and the Supreme Court have been preoccupied principally with the chilling effect of defamation law on the mass me- [*pg 889] dia,¹⁷³ chilling-effect arguments have particular resonance in cases involving "nonmedia"¹⁷⁴ defendants like those typically sued in the new Internet libel cases.

First, these new libel suits may chill simply by threatening to reveal the identities of those who speak their minds online. As noted previously, corporate plaintiffs easily can make out a prima facie libel claim any time they receive harsh criticism online. Once a complaint is filed, it is a simple matter to get a subpoena to force the ISP to divulge the anonymous defendant's identity.¹⁷⁵ As more and more suits are filed, many Internet users will come to recognize the ease with which their online anonymity can be stripped simply by the filing of a libel action, and they will censor themselves accordingly. Such self-censorship is salutary to the extent that it makes Internet users more temperate and more cautious about making unsupported factual assertions.¹⁷⁶ Indeed, the widespread use of pseudonyms online is re- [*pg 890] sponsible for many of the abuses perpetrated by Internet speakers.¹⁷⁷ But revelation of identity has negative consequences as well -- it may subject the user to ostracism for expressing unpopular ideas, invite retaliation from those who oppose her ideas or from those whom she criticizes, or simply give unwanted exposure to her mental processes.¹⁷⁸ There is some danger, therefore, that the growing popularity of the new Internet libel suits may chill more than defamatory falsehoods -- it may also chill the use of the Internet as a medium for free-ranging debate and experimentation with unpopular or novel ideas.¹⁷⁹

The high cost of libel litigation is another source of the chilling effect of libel suits. Media defendants identify litigation costs as a primary source of the chilling effect,¹⁸⁰ and these costs will fall even more heavily on the nonmedia defendants¹⁸¹ who are the targets of the [*pg 891] new Internet libel cases. This new class of nonmedia defendants are unlikely to have enough money even to defend against a libel action, much less to satisfy a judgment. Thus, wealthy plaintiffs can successfully use the threat of a libel action to punish the defendant for her speech, regardless of the ultimate outcome of the libel action. As a case in point, consider U-Haul International, Inc. v. Osborne.¹⁸² John Osborne and his roommate, Glenda Woodrum, rented a U-Haul to move from Florida to Georgia.¹⁸³ Not only did the truck break down numerous times, turning an eight-hour trip into a twenty-seven-hour trip,¹⁸⁴ but U-Haul also tried to make them pay the costs of fixing the truck.¹⁸⁵ Whatever the merits of U-Haul's argument, Osborne was understandably upset about this episode. He first complained to U-Haul's customer service department and then to the Better Business Bureau.¹⁸⁶ When neither of these avenues gave him satisfaction, Osborne decided to use the Internet to express his displeasure. Osborne developed a website entitled The U-Hell Website: Misadventures in Moving as a forum to tell his own story and to let other disgruntled U-Haul customers tell theirs.¹⁸⁷

After an initial round of legal correspondence, U-Haul promptly sued Osborne and Woodrum for libel and trademark infringement. U-Haul did not sue them in Georgia, where they lived. U-Haul instead chose to sue them in Arizona,¹⁸⁸ arguing by analogy to tradi- [*pg 892] tional defamation law that Osborne had "published" his website in Arizona because it was available to be downloaded by Internet users there.¹⁸⁹ But suing Osborne and Woodrum in Arizona was also a strategic means of trying to force them to capitulate to U-Haul's demands. As Osborne said, U-Haul knew that "we ha[d] no way to travel [to Arizona] to defend ourselves. [T]he apparent aim there [was] to prevent us from answering their charges, so that they [would] win a default judgment against us. . . . They want[ed] us quiet."¹⁹⁰ Luckily for the defendants in this case, the ACLU of Arizona and the Electronic Frontier Foundation came to their rescue by providing them with free legal assistance.¹⁹¹ As a result, the Arizona suit was dismissed, although U-Haul is currently threatening to pursue the action in Georgia.¹⁹²

Not all defendants are as lucky or as tenacious as Osborne and Woodrum, and many would choose to forego speaking in the future to avoid the hassle and expense of libel litigation. The great jurist Learned Hand once wrote that, "short of sickness and death," he dreaded nothing more than being involved in a lawsuit.¹⁹³ If this is true for the great jurist, how much more dread and anxiety must the prospect of being sued pose for the targets of the new Internet libel actions?¹⁹⁴

B. John Doe and First Amendment Theory

As argued in Section II.E, the new libel suits can be valuable to the extent that they deter malicious falsehoods. Thus, before lamenting the chill that defamation actions will have on the John Does who frequent financial message boards, it is worthwhile to explore whether their speech is worthy of First Amendment protection.

[*pg 893]

It is easy to get lost in the hype surrounding the Internet in answering this question. Scholars have touted the Internet as the living embodiment of the "marketplace of ideas" metaphor that lies at the heart of First Amendment theory. This idealized vision of Internet discourse contrasts rather sharply with the reality of the financial message boards. Discourse on the boards bears more resemblance to informal gossip than to rational deliberation, and the culture of the boards fosters, as one commentator put it, "disinformation, rumors and garbage."¹⁹⁵ Yet, extending "breathing space" to this type of speech may be necessary if the Internet is to fulfill its promise of giving ordinary John Does a meaningful role in public discourse.

1. The New Marketplace of Ideas. From a First Amendment scholar's perspective,¹⁹⁶ the fascination of the Internet lies in its potential for realizing the concept of public discourse at the heart of the Supreme Court's First Amendment jurisprudence.¹⁹⁷ The dominant First Amendment metaphor for describing public discourse is the "marketplace of ideas."¹⁹⁸ The marketplace of ideas is a sphere of discourse in which citizens can come together free from government interference or intervention¹⁹⁹ to discuss a diverse array [*pg 894] of ideas and opinions.²⁰⁰ Ideally, the process of interacting in the marketplace of ideas not only fosters the "search for truth";²⁰¹ it also enables citizens to transcend their differences in order to forge consensus on issues of public concern, or, as Professor Robert C. Post eloquently puts it, "to speak to one another across the boundaries of divergent cultures."²⁰² Public consensus, in turn, is an essential precondition of democratic self-government.²⁰³

This, at least, is the First Amendment ideal. As a practical matter, however, many citizens are barred from meaningful participation in public discourse by financial or status inequalities, and a relatively small number of powerful speakers dominate the marketplace of ideas.²⁰⁴ But the Internet promises to eliminate structural and financial barriers to meaningful public discourse, thereby making public discourse more democratic and inclusive, less subject to the control of powerful speakers, and, at least potentially, richer and more nuanced. It therefore promises to make the marketplace of ideas more than just a hollow aspiration.²⁰⁵

One of the most significant ways in which the Internet promises to change the nature of public discourse is by allowing more participants to engage in public discussion and debate.²⁰⁶ The Internet gives [*pg 895] citizens inexpensive access²⁰⁷ to a medium of mass communication and therefore transforms every citizen into a potential "publisher" of information for First Amendment purposes.²⁰⁸ "[F]reedom of the press," as one court noted, "is [no longer] limited to those who own one."²⁰⁹ The Internet enables speakers to bypass commercial publishers and editors and to speak directly "to an audience larger and more diverse than any the Framers could have imagined."²¹⁰ Although it may be an overstatement to say that the speech of ordinary John Does "compete[s] equally with the speech of mainstream speakers in the marketplace of ideas,"²¹¹ it is certainly true that ordinary John Does need no longer win approval of the mainstream media in order to be heard²¹² and that Internet discourse is more broadly inclusive than real-world discourse.²¹³

But the Internet not only removes barriers to speaking; it also removes barriers to being heard. In a now well-known New Yorker cartoon, a picture of a dog typing at a computer reads, "On the Internet nobody knows you're a dog."²¹⁴ The cartoon pithily encapsulates [*pg 896] one of the chief attractions of online anonymity.²¹⁵ Many participants in cyberspace discussions employ pseudonymous identities, and, even when a speaker chooses to reveal her real name, she may still be anonymous for all practical purposes. For good or ill, therefore, the audience must evaluate the speaker's ideas based on her words alone.²¹⁶ This unique feature of Internet communications promises to make public debate in cyberspace less hierarchical and discriminatory than real-world debate to the extent that it disguises status indicators such as race, class, gender, ethnicity, and age, which allow elite speakers to dominate real-world discourse.²¹⁷

The Internet also challenges existing hierarchies of power that permeate public discourse by making it harder for powerful speakers to control the interpretation of public events.²¹⁸ The mainstream media no longer have the power to exclusively define what is "news,"²¹⁹ [*pg 897] and governments, particularly authoritarian governments, have less power to control public discourse²²⁰ by controlling the information their citizens receive.²²¹ Wealthy and powerful private speakers may similarly find it more difficult to impose their agendas by manipulating media coverage. By changing the locus of control of public discourse, the Internet may therefore contribute to the development of an informed citizenry capable of deciding its collective fate.

A more subtle benefit of the Internet to public discourse is its potential for making public discourse richer and more nuanced. The more speakers who engage in public debate, the more perspectives that will be brought to bear on public problems. And beyond simply allowing more people to speak, the Internet also gives people more topics about which to speak.²²² The Internet allows people to transcend the limits of geography in order to find those with similar interests, and no topic is too obscure to generate Internet discussion. Moreover, speakers are able to experiment with controversial or unpopular ideas online, thereby making Internet debate more free- [*pg 898] ranging and expansive than real-world debate. Thus, the number and type of discussions in which citizens engage are bound to expand, and, as they do, the nature of public discourse is likely to change as well.²²³

2. Chaos in the Marketplace? This high theory contrasts rather markedly with the nature of actual discourse on the boards. A primary justification for protecting the marketplace of ideas from governmental interference is that competition of ideas in the marketplace fosters the search for truth. Indeed, this justification was the underpinning of Justice Holmes's famous defense of the "free trade in ideas," in which he argued that "the best test of truth is the power of the thought to get itself accepted in the competition of the market."²²⁴ If the speech of anonymous John Does in cyberspace fosters the search for truth, it is largely by accident. As anyone who has spent time perusing the financial message boards knows, board discussions are no model of rational deliberation and informed debate. Idle speculation, unintelligible musing, and "off-topic" trivia predominate over serious financial discussion, and many of the participants appear to lack financial sophistication. The operative principle is speed rather than accuracy, and the idea seems to be that users should present whatever tidbits of rumor or speculation they can and the audience can sort it out.²²⁵

Discourse on financial message boards resembles informal gossip²²⁶ more than it does formal written conversation. Indeed, Sissela Bok's observations about the informal character of gossip form a use- [*pg 899] ful point of comparison.²²⁷ Like gossip, discourse on the boards "lacks formal rules setting forth who may speak and in what manner, and with what limitations from the point of view of accuracy and reliability."²²⁸ Discourse on the boards is, like gossip, highly "spontaneous," and it "relies more on humor and guesswork" than it does on rational argumentation. Also like gossip, it is a form of discourse that forces the audience to take an active role in gleaning the valuable bits of information from the great mass of chaff.²²⁹

Nevertheless, the boards are not useless as a source of information.²³⁰ If one "lurks"²³¹ around long enough on a message board, observing the ongoing, rambunctious, and often disjointed dialogue over a period of time, one can become adept at judging the credibility of individual participants based on the quality and tone of a participant's remarks and the interaction between that participant and others.²³² Often, too, one goes to the financial boards to get a "feel" for what is being said about the corporation rather than to gather actual facts; what is important is the general tenor of the discussion, rather than any particular factual assertion.²³³

Thus, even if the financial message boards do not further the search for truth in a direct and efficient manner, they do allow those interested in a particular corporation to gather information that may not be supplied by traditional media outlets.²³⁴ The message boards may even help to remedy the "'chronic tilt [that] distorts mainstream [*pg 900] media coverage of grave, persisting, and pervasive abuses of corporate power.'"²³⁵ This is not to say that the speech of anonymous John Does is as important a source of financial information as, say, the Wall Street Journal or CNBC.²³⁶ The typical John Doe is not a financial journalist, and the boards are often cluttered with pointless ramblings.²³⁷ Yet, to the extent that participants in online discussions learn more information about a corporation through a message board, it is still a valuable, albeit flawed, source of specialized financial information.

From a First Amendment perspective, the financial message boards contribute to the marketplace of ideas by encouraging citizens to participate in public decisionmaking. The financial message boards exercise a powerful democratizing effect on public discourse about the publicly held corporations that shape citizens' daily lives. Publicly held corporations exercise tremendous influence in American society: they influence "what we buy, where we work . . . which diseases we cure, [and] much, much more."²³⁸ At a minimum, therefore, the boards provide an avenue for citizens to converse with one another and to seek consensus about topics that affect their lives. But the boards also serve as a kind of informal education for investors and noninvestors alike about the behavior of a particular corporation, about the workings of the stock market,²³⁹ and about economic matters in general.

By providing a forum for discussion of financial matters and by contributing to citizens' store of knowledge about such matters, the boards help to shape public opinion.²⁴⁰ But the boards also help citi- [*pg 901] zens to transform public opinion into action.²⁴¹ The more citizens know about economic matters, the more likely they are to make informed voting decisions. More significantly, the boards provide an important "back channel" for individual shareholders to convey their feelings to the corporations in a market largely dominated by institutional shareholders,²⁴² especially since, as noted above, many CEOs closely follow what is said about their corporation on the financial message boards.²⁴³ Even if the "buzz" on the boards does not directly influence corporate decisionmaking, it may influence decisions indirectly. And, if the boards have even a tiny impact on corporate behavior, they have broad social ramifications not limited to the participants in online discussions. Indeed, it is the worry that message [*pg 902] boards will transform opinion into action that prompts many companies to hire others to monitor what is being said about them on the web, to hire private investigators to track down online detractors,²⁴⁵ and even to sue anonymous John Does for libel.

Any argument that message boards further First Amendment values must, of necessity, concede that much of the conversation on the boards is cryptic, uninformed, and inane.²⁴⁶ But "uninhibited, robust, and wide-open debate"²⁴⁷ will often be cryptic, uninformed, and inane, especially when such debate is not restricted to educated elites but is expanded to include ordinary John Does. Thus, to concede this point does not detract from the importance of the boards as an outlet for ordinary citizens to discuss corporate behavior, to vent frustrations, and to have some small measure of input into corporate decisions that affect not only their own lives but those of their fellow citizens.

3. The Pitfalls of Participatory Public Discourse. A central premise of First Amendment theory in the twentieth century has been that truth is best gathered "out of a multitude of tongues,"²⁴⁸ and the Internet promises, if nothing else, to put this theory to the test. Already there are indicators that the "new" public discourse will look quite different from the old. As the financial message boards demonstrate, a financial discourse that includes ordinary John Does looks quite different from a discourse dominated by Wall Street analysts.²⁴⁹ Discourse on the boards is lively and engaging, providing "the masses" with a highly accessible source of information and education about the stock market. The boards give ordinary citizens an opportunity to become active participants in an ongoing dialogue about economic affairs and, in the process, allow them to transform the way business is done.²⁵⁰

But fostering a more participatory public discourse may come at a high cost. Speech from a "multitude of tongues" may lead to truth, [*pg 903] but it may also lead to the Tower of Babel.²⁵¹ And the level of discourse on the financial message boards also suggests that fostering unmediated participation may make public discourse not only less rational and less civil; it also runs the risk of making public discourse meaningless. A discourse that has no necessary anchor in truth has no value to anyone but the speaker,²⁵² and the participatory nature of Internet discourse threatens to engulf its value as discourse.

The problem, therefore, is to strike a balance between free speech and the preservation of civility. If the goal of making public discourse more participatory and ultimately more democratic is to be realized, the speech of ordinary John Does merits a very wide expanse of "breathing space,"²⁵³ wider than it currently receives. First Amendment doctrine therefore cannot hold ordinary John Does to the standards of professional journalists with regard to factual accuracy, because part of what gives the Internet such widespread appeal is the fact that it allows ordinary citizens to have informal conversations about issues of public concern. Although any approach to the problems posed by the new Internet libel actions must respond to the unique culture of the message boards, the law cannot allow that culture to degenerate into a realm where anything goes, where any embittered and malicious speaker can lash out randomly at innocent targets. Although many of the new libel plaintiffs are powerful corporate Goliaths suing to punish and to deter their critics, some are not. Some are simply responding in the only way available to prevent aggressively uncivil speech, the sole purpose of which is to cause emotional and financial harm. Hence, any solution to the problems posed by [*pg 904] these new suits must be tuned finely enough to distinguish incivility that must be tolerated for the good of public discourse from incivility that destroys public discourse.²⁵⁴

C. John Doe and First Amendment Doctrine

One might intuitively expect such a solution to come from the First Amendment. Since 1964, the First Amendment has been attuned to the danger that plaintiffs will use the threat of libel lawsuits to intimidate their critics.²⁵⁵ Existing First Amendment doctrine, however, provides only a limited response to the problems posed by the new Internet libel suits.

First Amendment doctrine, for example, has little to say about the chilling effect that the mere threat of being sued imposes on defendants of modest means.²⁵⁶ Although some courts have indicated that the First Amendment requires judges to take an active role in dismissing libel cases at an early stage, it is not a constitutional requirement.²⁵⁷ Indeed, as Professor Susan Gilles has noted, the first set of constitutional protections often will not come into play until the summary judgment stage,²⁵⁸ which may be too late to protect the de- [*pg 905] fendant from incurring substantial costs in defending himself, even if the case is frivolous.

More to the point, the Supreme Court crafted the constitutional privileges largely for the benefit of the institutional media, and these privileges are not entirely responsive to the chilling effect of defamation law on nonmedia defendants of the type targeted by the new Internet libel actions. Indeed, it is not even certain what level of First Amendment protection, if any, the typical John Doe defendant would receive.²⁵⁹ The typical John Doe case involves a publicly held corporation suing a defendant over statements posted on a financial message board dedicated to discussion of the company's prospects and management -- indeed, of anything that might affect its stock price.²⁶⁰ With respect to the level of constitutional protection that John Doe will receive, the Supreme Court's decisions make clear that courts must resolve at least three questions: (1) Is John Doe entitled to the same level of constitutional protection as a media defendant? (2) Is the corporate plaintiff a public figure? and (3) Is the speech at issue of public concern? In the typical case, the answer to all three questions should probably be yes, which means that John Doe should receive the maximum level of First Amendment protection from defamation liability.²⁶¹ But even if John Doe receives the maximum level of First Amendment protection, it is not clear that corporate plaintiffs will be barred from using defamation law as a tool for silencing their critics.

1. The Media/Nonmedia Distinction (or Lack Thereof). Internet users like John Doe defy the traditional distinction between media and nonmedia defendants.²⁶² One of the defining characteristics of the media is their ability to reach a mass audience. Although one might argue, therefore, that the typical John Doe is a media defendant, the [*pg 906] Supreme Court traditionally has referred to only the institutional media -- broadcasters, newspaper publishers, and so forth -- as the "media" for First Amendment purposes.²⁶³ The question, then, is whether nonmedia defendants like John Doe are entitled to the same level of First Amendment protection as media defendants. The Supreme Court has studiously avoided providing a definitive resolution to this question,²⁶⁴ but the "best educated guess"²⁶⁵ of most commentators²⁶⁶ (and, more importantly, of most lower courts²⁶⁷) is that nonmedia de- [*pg 907] fendants should receive the same level of protection from defamation liability as media defendants. A media/nonmedia distinction would draw unsound and unworkable status distinctions among speakers, and, as the Supreme Court has recognized in a poignant but now obsolete metaphor, the "lonely pamphleteer's" contribution to public debate can be just as important as that of the most powerful media speaker.²⁶⁸ Moreover, in Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc.,²⁶⁹ at least six Justices seemed to reject the media/nonmedia distinction,²⁷⁰ although the case was decided on other grounds.²⁷¹ Since that time, the Supreme Court has gone out of its way to state that this issue remains unresolved,²⁷² but it seems fair to conclude (or at least to hope) that the "increasing convergence of what might be labeled 'media' and 'nonmedia'"²⁷³ speakers makes it unlikely that the Supreme Court will extend less First Amendment protection to Internet users than it has extended to traditional media defendants.

2. Corporations as Public Figures. The second determinant of the level of First Amendment protection that John Doe will receive is the public figure status²⁷⁴ of the new class of Internet defamation [*pg 908] plaintiffs.²⁷⁵ The distinction between public and private figures is an important one because a public figure plaintiff must prove that the defendant acted with actual malice in order to recover for defamation.²⁷⁶ Although this distinction is important, the Supreme Court has given only limited guidance on how to distinguish public from private figures, especially when the plaintiff is a corporation.²⁷⁷ A strong argument can be made that the typical corporate plaintiff in a John Doe libel case will be a public figure. The typical plaintiff is a publicly held corporation listed on a national stock exchange; a publicly held corporation by definition avails itself of the capital markets to raise funds from investors.²⁷⁸ Regardless of how such corporations should be treated generally,²⁷⁹ it is fair to conclude that a [*pg 909] corporation should be treated as a public figure when the alleged defamation appears in a forum dedicated to discussion of the corporation's management and operation and is reasonably related to that subject.²⁸⁰

Although courts have developed a variety of tests for determining whether a corporation is a limited-purpose public figure,²⁸¹ the two most important determinants of public figure status are those set forth in the Supreme Court's decision in Gertz v. Robert Welch, Inc.:²⁸² (1) whether the plaintiff has access to channels of effective communication to rebut the defamatory falsehood;²⁸³ and (2) whether the plaintiff voluntarily assumed a prominent role in a public controversy and the attendant risk of enhanced public scrutiny that accompanies it.²⁸⁴ Both of these factors cut in favor of treating the typical corporate plaintiff in the new Internet libel actions as a public figure. The typical publicly held corporation will have a great deal of access to effective channels of communication.²⁸⁵ The corporation can post an authoritative rebuttal to any defamatory message on the financial [*pg 910] message board on which the message first appeared.²⁸⁶ The corporation can also issue press releases to counteract any defamatory statements, and many publicly held corporations can even finance intensive media campaigns to rehabilitate a damaged corporate reputation.

Although the corporation's access to "self-help"²⁸⁷ remedies for defamation suggests that it should be treated as a public figure, it is somewhat harder to argue in the typical John Doe case that the corporate plaintiff "thrust [it]self to the forefront of [a] particular public controvers[y]."²⁸⁸ The "public controversy"²⁸⁹ at issue ordinarily will be a topic that affects the corporation's stock price: it may be as diffuse a subject as the overall financial health of the corporation or as specific as a particular venture on which the company has embarked or a particular corporate decision. The problem with applying this analysis to the John Doe cases is that they do not tend to involve a "controversy"²⁹⁰ but rather a subject of discussion.²⁹¹

[*pg 911]

Despite the lack of a true "controversy," the justifications for treating a publicly held corporation as a public figure for purposes of Internet discussions of its financial well-being are compelling.²⁹² All corporations have an interest in convincing the public of their financial well-being,²⁹³ but publicly held corporations are more likely to "invite attention and comment"²⁹⁴ on their business dealings and affairs by their affirmative acts. Publicly held corporations must make an affirmative decision to "take the company public" in order to solicit money from investors. In making the decision to "go public," the corporation seeks and often obtains national attention.²⁹⁵ The corporation also voluntarily submits itself both to extensive legal regulation (including public disclosure requirements)²⁹⁶ and to a degree of public scrutiny²⁹⁷ in order to obtain "the special benefits conferred by the corporate structure."²⁹⁸ Moreover, corporations have a vested interest [*pg 912] in fostering discussion of their stock valuation amongst investors and potential investors who frequent financial message boards. Many corporations even monitor board discussions, and they may sometimes take an active role in shaping the discussion to their benefit.²⁹⁹ Given that Internet message boards are dedicated to discussion of business activities voluntarily undertaken by corporations, courts should ordinarily treat corporations as having assumed the risk of hard-hitting debate and criticism of their activities.

3. Public Concern and Financial Discourse. The last determinant of the level of protection John Doe's speech will receive is whether the speech deals with a matter of public concern or a matter of only private concern. The Supreme Court grants extensive First Amendment protection from defamation claims only to "speech on matters of public concern."³⁰⁰ If this Article is correct in predicting that many plaintiffs in the new John Doe libel cases will be public figures, it follows almost automatically that the cases will also involve matters of public concern.³⁰¹ Even if the plaintiff is a private figure, [*pg 913] however, defendants like John Doe should still receive a high degree of First Amendment protection because their speech will typically involve matters of public concern.

Discussion of the financial health of a publicly held company listed on a national stock exchange would certainly be a matter of public concern if it appeared in the Wall Street Journal, a Bloomberg News report, or a CNBC broadcast.³⁰² Yet, the very same discussion may not be of public concern when it occurs on a financial bulletin board in cyberspace. Whether speech deals with a matter of public concern is not judged by its subject matter alone, but by its "content, form and context."³⁰³ In Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc.,³⁰⁴ for example, the Supreme Court deemed a report about the financial health of a corporation to be of only private concern,³⁰⁵ although the facts and the reasoning of the case suggest that matters of truly private concern should be rare. Dun & Bradstreet, a credit reporting service, erroneously reported to five of its subscribers that a corporation was bankrupt; however, the five subscribers who received the report were contractually bound not to reveal the contents to anyone else.³⁰⁶ The Court upheld a jury award in favor of the defamed corporation for $50,000 in presumed damages and $300,000 in punitive damages,³⁰⁷ even though the corporation failed to show that Dun & Bradstreet published the report with actual malice.³⁰⁸ Although the Supreme Court did not spell out the practical implications of the decision, the reasoning suggests that the state's strong interest in protecting the reputations of private figures might well justify imposing strict liability on defamation involving issues of private concern, and indeed this interpretation has persuaded many commentators that defendants receive no special constitutional protections in this type of case.

The new John Doe cases, however, are distinguishable from Dun & Bradstreet on a number of grounds. A key factor in the Court's decision in Dun & Bradstreet was that the credit report at issue involved [*pg 914] "speech solely in the individual interest of the speaker and its specific business audience."³⁰⁹ Almost by definition, the credit report could not contribute to "the free flow of commercial information" because the "audience" for the speech consisted of only five paid subscribers who were contractually bound not to disseminate the information more widely.³¹⁰ A second factor that the Court emphasized was that the report merited no "special protection" due to its "hardy" nature.³¹¹ Because credit reporting agencies like Dun & Bradstreet have a powerful market incentive both to continue publishing credit reports and to verify the accuracy of such reports, the Court thought it unlikely that imposing defamation liability for false reports would unduly chill credit reporting agencies from producing them.³¹²

Neither of these factors is present in the typical John Doe case. In contrast to the speech at issue in Dun & Bradstreet, the John Doe cases involve speech that contributes, albeit modestly,³¹³ to the free flow of financial information and that is particularly susceptible to being chilled by libel suits. First, unlike the credit report in Dun & Bradstreet, messages posted on financial bulletin boards reach a large national and even international audience. Of course, one might still argue that, regardless of the size of the audience, most of those who follow the message board for a particular corporation will be investors or potential investors who have self-interested motives for following the discussion. However, this argument overstates the extent to which the message boards involve speech that is "solely in the individual interest of the speaker and [his] specific business audience."³¹⁴ As argued in Section II.C, publicly held corporations play a vital role in shaping both the United States economy and the lives of ordinary citizens.³¹⁵ The financial message boards can be a useful source of information about the financial health of publicly held corporations, and they can provide an informal education to investors and nonin- [*pg 915] vestors alike about financial matters more generally. But most significantly, the boards allow ordinary citizens to participate, however marginally, in corporate decisionmaking, and they enhance the ability of citizens to make informed decisions on matters of economic policy. To the extent, therefore, that discourse on the boards affects corporate decisionmaking or even political decisions on economic matters, it has the potential to affect all citizens. The powerful democratizing effect of the boards gives corporations a reason to fear them, but it also justifies treating discourse on the board as involving a matter of public concern.

The threat libel suits pose to John Does who contribute to financial message boards bolsters the argument for treating such discussions as a matter of public concern. Unlike the credit report at issue in Dun & Bradstreet, the speech of individual posters to Internet discussion boards is not likely to be especially "hardy." Although a shareholder of a corporation may have some financial self-interest in following discussion of the corporation on a financial bulletin board, the shareholder has no financial interest in contributing to the discussion (unless, perhaps, she is planning to try to "hype" the stock to drive up its price, behavior which in some cases is illegal).³¹⁶ Because John Doe has no financial incentive to speak, he likely will find it easier to engage in self-censorship than to risk being sued for libel. Being sued is not a risk John Doe can easily avoid. It is no simple matter for John Doe to "objectively verify" every piece of information about a corporation before publishing it, and, even if he kept a staff of fact-checkers, much of the information published on the boards is inherently speculative in nature, even where it appears to be couched in the form of "facts." John Doe's contribution to public discourse may therefore by silenced by strict application of defamation law. Given this threat, it seems clear that, even in cases involving private figure plaintiffs, John Does should ordinarily receive at least some level of First Amendment protection.

D. The Inadequacies of the Actual Malice Standard

Even if a court were to conclude that the corporate plaintiffs in the new Internet libel actions should be treated as public figures, corporations might still be able to use libel suits to intimidate John Doe [*pg 916] and others like him into silence, since the actual malice standard is not responsive to the distinctive problems raised by these new cases.³¹⁷ In addition to proving the common law elements of a libel action, public figure plaintiffs must also prove that the defendant's statements were false³¹⁸ and that the defendant made them with actual malice -- that is, with knowledge or reckless disregard of their falsity.³¹⁹ The actual malice standard is designed to prevent juries from imposing crippling liability on unpopular defendants for minor mistakes of fact.³²⁰ Negligent, and even grossly negligent, publication of defama- [*pg 917] tory falsehoods is protected, at least in cases involving public officials or public figures; it is only intentional or reckless falsehoods that subject a defendant to liability.

The rationale for protecting negligent falsehoods is not that they are valuable in and of themselves. To the contrary, the Supreme Court has repeatedly said that there is no constitutional value in false statements of fact.³²¹ The reason for protecting negligent falsehoods is to ensure a measure of breathing space around constitutionally valuable speech. As the Supreme Court recognized, "erroneous statement is inevitable in free debate."³²² Without a breathing space around such erroneous statement, the threat of libel suits would cause "would-be critics" to be overly cautious due to "doubt whether [their statements] can be proved true or fear of the expense of having to do so." In New York Times v. Sullivan and its progeny, therefore, the Supreme Court crafted a standard designed to protect the inevitable erroneous statement (i.e., a negligent erroneous statement) from liability. The implicit assumption underlying the standard seems to be that intentional or reckless falsehoods will be rare and that the ordinary case will, like Sullivan itself, involve only negligent falsehoods.³²³

This assumption makes a great deal of sense when applied to those who publish in the institutional media, particularly the institutional press. Reporters have both professional and market incentives to publish accurate information, and the media have institutional mechanisms designed to ensure the accuracy of the information they publish. But it is not entirely clear how the actual malice standard [*pg 918] would apply to the typical John Doe who posts on financial bulletin boards. The Supreme Court's examples of what constitutes actual malice are geared to the investigative practices of the institutional press.³²⁴ In order for a plaintiff to establish actual malice, "[t]here must be sufficient evidence to permit the conclusion that the defendant in fact entertained serious doubts as to the truth of his publication."³²⁵ Actual malice may be found, for example, where the defendant invents a story, bases it on "an unverified anonymous telephone call," publishes information "so inherently improbable that only a reckless man would have put [it] in circulation," or where the defendant publishes despite "obvious reasons to doubt the veracity of [an] informant or the accuracy of his reports."³²⁸

Except for stories based on pure fabrication (like Krum's statements that he was having an affair with the wife of HealthSouth's CEO), these examples have little to do with the process John Doe goes through before posting to a message board. The typical John Doe does not get his information from interviews with informants or even from direct perusal of primary sources.³²⁹ Instead, John Doe's information is likely to be based on secondary sources or on rumors started by other John Does. The typical John Doe will have no special training in determining the credibility of his information sources. Nor will he have a staff to check his facts before publishing, an editor to peruse his post for problems, and a lawyer on retainer to advise him of the complexities of libel law. No professional ethic impels him to be accurate.³³⁰ In fact, as previously argued,³³¹ the culture of the boards encourages just the opposite. The boards prize speed over accuracy [*pg 919] on the assumption that the reader will be able to sort reason from rhetoric and fact from fiction.

What this evidence suggests is that John Doe is more likely than the average media defendant to be reckless -- perhaps indifferent is a better description -- with regard to the accuracy of the facts he publishes on Internet financial bulletin boards. Therefore, it may be relatively easy for plaintiffs to establish actual malice.³³² But an alternative, and more convincing, explanation is possible. Rather than regarding John Does as more apt than their media counterparts to be reckless with regard to facts, it may be that both John Doe and his online audience view most of his commentary as rhetorical hyperbole or subjective speculation rather than a sober recitation of actual facts. This alternative explanation also suggests an alternative avenue of First Amendment protection for John Doe's speech: the opinion privilege.

III. PROTECTING JOHN DOE: ADAPTING THE OPINION PRIVILEGE TO CYBERSPACE

The First Amendment extends a privilege to statements that do not imply an assertion of objective fact, either because such statements "cannot 'reasonably [be] interpreted as stating actual facts'" or because such statements are not provably false.³³³ This privilege, the opinion privilege,³³⁴ has the potential to protect Internet discourse from wealthy and powerful plaintiffs who attempt to use defamation law to intimidate their online critics into silence. Courts can deploy the opinion privilege at the earliest stages of a defamation action,³³⁵ early enough to prevent even impecunious defendants from engaging [*pg 920] in undue self-censorship. In Milkovich v. Lorain Journal,³³⁶ however, the Supreme Court threw the opinion privilege into such disarray that it may lack the sensitivity necessary to protect the type of online commentary at issue in the new Internet libel cases.³³⁷ Even so, Milkovich need not hinder lower courts in taking the necessary steps to adapt the opinion privilege to cyberspace, a task for which this part gives practical guidance.

A. The Opinion Privilege in the Lower Courts, Post-Gertz and Pre-Milkovich

In Gertz v. Robert Welch,³³⁸ the Supreme Court observed in a now famous dictum, "Under the First Amendment there is no such thing as a false idea. However pernicious an opinion may seem, we depend for its correction not on the conscience of judges and juries but on the competition of other ideas."³³⁹ Lower courts seized this dictum as evidence that opinion, traditionally protected by the common law privilege of fair comment,³⁴⁰ deserved expansive First Amendment protection.³⁴¹ The constitutional privilege for opinion soon came to be a powerful defense to counter the chilling effects of libel ac- [*pg 921] tions. Unlike other constitutional privileges, the opinion privilege could be deployed as early as the filing of a motion to dismiss,³⁴² thereby safeguarding even impecunious defendants from much of the chilling effect of being sued for defamation.

"Opinion," in the ordinary, nonlegal sense of the term, includes statements couched in loose, figurative, or vituperative language, statements that are purely subjective expressions of the speaker's point of view, and statements that contain "deductions from known data or personal observation."³⁴³ Although a prior Supreme Court decision gave lower courts a mandate to protect hyperbole and vituperation,³⁴⁴ the Supreme Court's enigmatic dictum in Gertz did not indicate whether all types of opinion are equally deserving of First Amendment protection. Moreover, the Gertz dictum did not give lower courts any guidance in separating fact from opinion,³⁴⁵ a challenging task under the best of circumstances.³⁴⁶

[*pg 922]

Left to their own devices, the lower courts developed various approaches to this critical task.³⁴⁷ The most influential approach was that put forth in Ollman v. Evans,³⁴⁸ an en banc decision of the U.S. Circuit Court of Appeals for the District of Columbia. The Ollman decision was a thoughtful exposition of the difficulties of distinguishing fact from opinion. Writing for the court,³⁴⁹ then-Judge Kenneth Starr designated four factors that, when considered along with the "totality of the circumstances,"³⁵⁰ should guide courts in distinguishing fact from opinion: (1) "the common usage or meaning of the specific language of the challenged statements itself"; (2) "the statement's verifiability";³⁵¹ (3) the linguistic context of the statement;³⁵² and (4) "the broader [social] context or setting in which the statement appears."³⁵³ Although the Ollman multifactor approach did not obviate the need for courts to grapple with complex interpretive issues, the factors at least guided courts in making rational judgments separating fact from opinion.³⁵⁴ Yet, just when courts appeared to be reaching consensus [*pg 923] on a method for applying a constitutional opinion privilege, the Supreme Court stepped in and threw the entire area into disarray.³⁵⁵

B. Milkovich Throws the Opinion Privilege into Disarray

After leaving the lower courts to grapple with formulating a constitutional opinion privilege for sixteen years, the Supreme Court finally took up the issue in Milkovich v. Lorain Journal Co.³⁵⁶ The plaintiff in Milkovich was a high school wrestling coach from Maple Heights, Ohio, whose team was involved in a fight that broke out during a wrestling match.³⁵⁷ After a hearing in which Coach Milkovich and his superintendent testified, the Ohio High School Athletic Association ("OHSAA") censured Milkovich, placed his team on probation and declared the team ineligible for the state tournament because of their role in the altercation.³⁵⁸ Subsequently, several wrestlers and their parents sued, claiming that the OHSAA proceeding violated due process.³⁵⁹ The judge agreed and, after a hearing in which both Milkovich and his superintendent again testified, overturned the OHSAA probation and ineligibility orders.³⁶⁰

The next day, the defendant, Thomas Diadiun, a sports columnist, harshly criticized Milkovich's role in the altercation and his testimony in the court proceeding.³⁶¹ The heading for Diadiun's "TD Says" column that day read, "Maple beat the law with the 'big lie.'"³⁶² In the body of the column, Diadiun made it clear that he had attended both the match at which the brawl had taken place and the OHSAA hearing.³⁶³ Diadiun then wrote:

Anyone who attended the meet, whether he be from Maple Heights, Mentor [the rival team], or impartial observer, knows in his heart that Milkovich and [Superintendent] Scott lied at the [court] hearing after each having given his solemn oath to tell the truth.

[*pg 924]

But they got away with it.

Is that the kind of lesson we want our young people learning from their high school administrators and coaches?

I think not.³⁶⁴

Milkovich and Scott both sued for defamation, alleging that Diadiun's column accused them of perjury.³⁶⁵ While Milkovich's suit was ongoing, the Ohio Supreme Court upheld a grant of summary judgment against Scott.³⁶⁶ Applying Ollman's four-factor analysis, the Ohio Supreme Court determined that the context of the column negated the impression that Diadiun was asserting that Scott had actually committed perjury; therefore, the column was deemed "constitutionally protected opinion."³⁶⁷ Subsequently, the Ohio Court of Appeals upheld a grant of summary judgment against Milkovich, concluding that it was bound by the Ohio Supreme Court's analysis in Scott's case.³⁶⁸

In a stinging rejection of the Ollman approach, the United States Supreme Court reversed.³⁶⁹ The Supreme Court chastised the lower courts for misreading Gertz "to create a wholesale defamation exemption for anything that might be labeled 'opinion.'"³⁷⁰ For the Supreme Court, the operative distinction was not that drawn by the lower courts between fact and opinion; instead, the operative distinction was between statements that imply an assertion of objective facts and those that do not.³⁷¹ The Court rejected the fact/opinion distinction as "artificial" because a statement of opinion will often imply the existence of objective facts and such implicit assertions of fact can harm reputation just as much as explicit assertions of fact. Hence, the [*pg 925] Court declared that the statement "'In my opinion Jones is a liar'" can be just as damaging to the reputation of Jones as the statement "'Jones is a liar.'"³⁷² The first statement may imply a false assertion of fact because it invites the audience to assume that unstated defamatory facts undergird the author's assertion. Moreover, even if the author states the underlying facts on which her conclusion is based, she should still be liable if the underlying facts are incorrect or incomplete, or if she draws erroneous conclusions from them.³⁷³ The Court therefore rejected the proposition that defamatory statements should be protected as long as it is clear that they reflect the author's point of view, or as long as they state accurately the facts on which they are based.³⁷⁴ Such statements are nonetheless actionable if they imply factual assertions that are defamatory and untrue.

Despite the Court's rejection of the "artificial dichotomy"³⁷⁵ between fact and opinion, the Supreme Court did not reject the notion that the First Amendment requires a measure of protection for opinion on matters of public concern.³⁷⁶ Instead, the Court found adequate [*pg 926] protection for opinion -- that is, statements that do not imply an assertion of objective fact -- in existing First Amendment doctrine.³⁷⁷

Of the doctrines identified by the Court,³⁷⁸ two stand out as being relevant to whether a defamatory statement implies an assertion of objective fact. First is the doctrine that requires both public figure and private figure plaintiffs to bear the burden of proving falsity in all defamation actions involving matters of public concern,³⁷⁹ a requirement imposed by the Court in Philadelphia Newspapers, Inc. v. Hepps.³⁸⁰ This requirement protects opinion, as the Court defined it, because statements not capable of being proved false are not actionable as implied assertions of objective fact. Put another way, a statement on a matter of public concern that is not provably false is one type of constitutionally protected opinion.³⁸¹

A second type of statement that does not imply an assertion of objective fact -- and is therefore constitutionally protected -- is a statement that "cannot 'reasonably [be] interpreted as stating actual facts' about an individual."³⁸² The Court gleaned this principle from a line of prior cases protecting rhetorical hyperbole, satire, and parody.³⁸³ The first of these cases, Greenbelt Cooperative Publishing Ass'n v. Bresler,³⁸⁴ was the earliest Supreme Court case to suggest that opinion deserves some constitutional protection. In Bresler, a real es- [*pg 927] tate developer involved in negotiations with the city council sued a newspaper for publishing that some citizens "had characterized [his] negotiating position as 'blackmail.'"³⁸⁵ The Supreme Court held, however, that the First Amendment barred imposing defamation liability because "even the most careless reader" would have interpreted the term "blackmail" as "no more than rhetorical hyperbole,"³⁸⁶ at least where the term "blackmail" arose in the context of a heated public debate over zoning.³⁸⁷ Although the Bresler Court did not frame its analysis in terms of protecting "opinion," the decision nonetheless constitutes strong authority for protecting statements that, while seemingly factual on their face, cannot be construed as stating actual facts when read in context.

This reading of Bresler was confirmed by the Court in No. 496, National Ass'n of Letter Carriers v. Austin,³⁸⁸ a case in which the defendant union members labeled the plaintiffs as "scabs" and then invoked Jack London's famous definition of a "scab" as a "traitor to his God, his country, his family and his class."³⁸⁹ Emphasizing that the statement occurred in the context of a labor dispute, the Court found that the defendants had not accused plaintiffs of treason, but had merely employed the words "in a loose, figurative sense to demonstrate the union's disagreement with the views of those workers [like plaintiffs] who oppose unionization."³⁹⁰

Hustler Magazine, Inc. v. Falwell³⁹¹ presents a more recent statement of this principle. Falwell involved an action of intentional infliction for emotional distress stemming from a highly offensive advertisement parody suggesting that the Reverend Jerry Falwell had sex with his mother in an outhouse.³⁹² The Court viewed the parody at is- [*pg 928] sue as being related, albeit distantly, to political cartoons, satire, and caricature³⁹³ -- all of which contribute color and "'emotional impact'"³⁹⁴ to public discourse. The Court thus held that, at least in cases involving public officials and public figures, the First Amendment bars imposition of liability for "nonfactual" statements like the one at issue in Falwell.³⁹⁵

What these cases establish, and Milkovich confirms, is that the First Amendment protects statements that cannot be interpreted as stating actual facts in order "that public debate will not suffer for lack of 'imaginative expression.'"³⁹⁶ Taken together, therefore, these cases might form the basis for extending broad First Amendment protection to "loose talk" of the sort that is common in cyberspace.³⁹⁷ Bresler, in particular, emphasized the role of social context in transforming an apparently factual accusation of crime into "no more than rhetorical hyperbole,"³⁹⁸ and therefore it might be used as authority for courts to create an opinion privilege responsive to the culture of cyberspace.

In Milkovich itself, however, the Supreme Court failed to specify what role, if any, context plays in interpreting allegedly defamatory statements. The majority's analysis of Diadiun's statements stressed their verifiability rather than their surrounding context; as the Court noted, whether Milkovich actually committed perjury could be objectively verified "by comparing, inter alia, [Milkovich's] testimony before the [association] with his subsequent testimony before the trial court."³⁹⁹ In addition, the Supreme Court observed that neither the [*pg 929] "language" of Diadiun's column nor its "general tenor" negated the impression that he was accusing Coach Milkovich of perjury.⁴⁰⁰ The Court's analysis took little notice of the fact that the alleged accusation of perjury had appeared in a signed editorial sports column, the tone of which, as described by the dissent, was "pointed, exaggerated, and heavily laden with emotional rhetoric and moral outrage."⁴⁰¹ Indeed, the broader social context in which Diadiun's accusation was made appeared to play no role in the Court's analysis.⁴⁰²

The Court's failure to specify what role context plays in determining whether a statement implies an assertion of objective fact is a critical flaw in its analysis. Whether a given statement constitutes rhetorical hyperbole or satire requires close contextual analysis of the sort in which the Court engaged in the Bresler case, but the Milkovich opinion makes it clear that the Court had no intention of immunizing speech just because it appears in "[c]ertain formats -- editorials, reviews, political cartoons, letters to the editor -- [that] signal the reader to anticipate a departure from what is actually known by the author as fact."⁴⁰³ Thus, of the many questions raised by Milkovich,⁴⁰⁴ one of the most vexing is whether, and to what degree, the context in which a defamatory statement appears should be considered by courts when determining whether a statement implies an assertion of objective fact.⁴⁰⁵

[*pg 930]

C. The Opinion Privilege in the Lower Courts, Post-Milkovich

Not surprisingly, the Milkovich decision created confusion in the lower courts on this issue. Some courts continue to rely heavily on context in assessing whether an allegedly defamatory statement implies an assertion of fact.⁴⁰⁶ Other courts downplay context in favor of verifiability.⁴⁰⁷ The epitome of this confusion is the D.C. Circuit's stunning "self-reversal" in Moldea v. New York Times Co. ("Moldea II").⁴⁰⁸ Dan Moldea, an investigative reporter and author, sued the New York Times over a negative book review that stated, among other things, that Moldea's book contained "too much sloppy journalism" and provided several examples to support this conclusion.⁴⁰⁹ The district court, however, granted summary judgment in favor of the Times on the grounds that the review "exemplifies a description of a literary work, from one's personal perspective."⁴¹⁰

In what came to be known as Moldea I, a panel of the District of Columbia Circuit Court of Appeals reversed.⁴¹¹ Applying Milkovich, two of the judges concluded that the allegation that Moldea's book contained "sloppy journalism" implied an assertion of "provable, albeit unstated, defamatory facts" -- including the facts "that Moldea [*pg 931] plays fast and loose with his sources" and "that his allegations are not to be believed."⁴¹² The court thought it irrelevant to the Milkovich analysis that "the challenged statements appeared in a 'book review' rather than in a hard news story,"⁴¹³ and it noted that the appearance of Diadiun's allegations in Milkovich in "a forum well known for spirited expressions of personal opinion" had not affected the outcome of the case. Although Moldea I was a defensible, if highly literal, application of Milkovich, it prompted immediate criticism from journalists and media lawyers alike,⁴¹⁴ and the Times promptly petitioned the panel for rehearing.

Surprisingly, the panel granted this petition, and, without hearing additional argument in the case, "confess[ed] error" and "amend[ed] its earlier decision."⁴¹⁶ Judge Harry T. Edwards, author of the court's opinion in Moldea I, apologized for giving insufficient consideration to the fact that the allegedly defamatory statements had "appeared in the context of a book review, a genre in which readers expect to find spirited critiques of literary works."⁴¹⁷ Judge Edwards conceded that, "[w]hile Milkovich could be interpreted as . . . [the court had in its] initial decision," a better interpretation is that the Supreme Court did not "sweep away so much settled law"⁴¹⁹ but instead retained a focus on "the context in which speech appears."⁴²⁰ The court then held that commentary of the sort at issue in Moldea should be actionable "'only when the [reviewer's] interpretations are unsupportable by reference to the written work.'"⁴²¹ Applying this new "supportable interpretation" standard, the court concluded that the statement that Moldea's book contained "too much sloppy journalism" was "supported by revealed premises that [the court could not] hold to be false in the context of a book review."⁴²² By adopting a strained interpretation of [*pg 932] Milkovich, therefore, the Moldea II court managed to reach a pragmatic result: its decision extended broad protection to an important genre of speech by acknowledging the important role context plays in shaping the meaning of allegedly defamatory statements.⁴²³

As the strange saga of Moldea illustrates, whether a court accepts or rejects contextual analysis plays an important role in the outcome of defamation cases, and lower courts can (and many do) plausibly claim to be following Milkovich no matter which route they choose. Some courts have therefore resolved the difficulties created by Milkovich by interpreting it expansively to allow consideration of social context.⁴²⁴ Others have simply rejected Milkovich entirely, holding that their own state constitutions or the common law requires an Ollman-type "totality of circumstances" approach to separating opinion from fact.⁴²⁵ Nonetheless, the uncertainty surrounding Milkovich forces lower courts to address the fact/opinion determination with little meaningful guidance from the Supreme Court, making it difficult to predict in advance what role context will play in the outcome of any particular defamation action.

D. Adapting the Opinion Privilege to Cyberspace

Given this uncertainty, it is worth asking what role context should play in adapting the opinion privilege to new Internet libel cases. After Milkovich, the opinion privilege applies only to statements that do not imply an assertion of objective facts.⁴²⁶ As this section illustrates, when the new Internet libel cases are analyzed with due consideration for both the inherently speculative nature of "investment advice" generally⁴²⁷ and the specific conventions that govern [*pg 933] discourse on financial message boards, it becomes apparent that many of the allegedly defamatory statements made on financial message boards will not imply assertions of objective fact. In other words, courts should extend the benefit of the constitutional opinion privilege to many (though by no means all) of the defendants in the new Internet libel cases.⁴²⁸

1. An Illustration: Hitsgalore.com, Inc. v. Shell. Consider once more the facts of Hitsgalore.com v. Shell.⁴²⁹ There, numerous John Does had posted a variety of serious allegations about Hitsgalore.com, Inc. and its officers.⁴³⁰ The anonymous posters called the corporation a "scam,"⁴³¹ a "fraud,"⁴³² a "flying turd";⁴³³ they labeled its officers as "criminals,"⁴³⁴ "crooks,"⁴³⁵ "con men,"⁴³⁶ "slime,"⁴³⁷ and "scum"⁴³⁸ who had lied to the SEC⁴³⁹ and duped investors in a "classic pyramid scheme."⁴⁴⁰ Two posters specifically stated that their posts were stating facts rather than opinion,⁴⁴¹ and one even dared the corporation to sue him.⁴⁴²

Most of these accusations are, in Milkovich's terms, "provable" as true or false. Although it may not be verifiable that the corporation [*pg 934] is a "flying turd" or even a "scam,"⁴⁴³ it is certainly verifiable whether its management engaged in fraudulent or crimin