The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Free Speech

The Return to Criminal Law as a Remedy: Libel


I'm continue to serialize my forthcoming UC Davis Law Review article What Cheap Speech Has Done: (Greater) Equality and Its Discontents; you can read the Introduction, but in this post I'm talking about how "cheap speech" has brought back criminal remedies for libel. Recall that the article is mostly descriptive, focusing on what's happening, for better or worse.

[A.] The Traditional Civil Damages Model

For decades, protecting people's reputation from defamatory falsehood had been left to libel damages liability. Damages liability is supposed to compensate the injured target of the speech. It is supposed to deter libelers. And it is supposed to encourage libelers to promptly retract their false charges once threatened with a lawsuit.

This mechanism worked to some degree, however imperfectly, for the pre-Internet mass media. Because such media organizations had money, they tended to worry about libel liability. And because they had money, plaintiffs (or plaintiffs' lawyers) had some prospect of recovering their fees, if they had very strong libel claims. Libel law also worked to some degree for libel lawsuits against employers, business rivals, and similar commercial actors.

This is, of course, an oversimplification. Libel cases were often hard to win, because of the Supreme Court's decisions reining in libel law. The availability of libel insurance also likely made the deterrent effect of libel law more complex. And even in the past, there were judgment-proof libel defendants: "[M]ost libellers are penniless," an 1881 treatise author wrote, though perhaps exaggerating, "and a civil action has no terrors for them." Still, on balance, tort law tended to serve its compensatory and deterrent function here, at least to some extent.

But the risk of civil liability doesn't much affect speakers who have no money. Suing such a speaker is a sure money pit: you have to pay your lawyer, and you know you'll never recover any of that expense, much less get compensated for your damaged reputation.

Knowing this, judgment-proof speakers aren't much deterred by the risk of a libel lawsuit up front, before they make their statements. And even if they get a letter demanding that they take down the statements from a blog or a Facebook page, they can feel relatively safe playing chicken. True, even poor speakers can have some assets that could be seized, so they risk some pain from a libel lawsuit. But such speakers can usually be fairly confident that their target won't invest the money in getting and enforcing a judgment.[1]

[B.] 47 U.S.C. § 230

Of course, Internet speech, even from judgment-proof speakers, comes through platforms owned by businesses that have ample assets. Blogs are hosted on some company's computer systems. Consumer reviews are posted on some company's site, such as Yelp or RipoffReport. Revenge porn is often posted on sites devoted to pornography. And this material is usually found by users using search engines.

But all those non-judgment-proof intermediaries are, with few exceptions, not liable for what users post, and generally aren't even subject to injunctions to remove or block such user posts.[2] Title 47 U.S.C. § 230, enacted in 1996, expressly provides that such Internet content and service providers can't be treated as publishers or speakers of material posted by others. Courts have read this immunity broadly, to bar nearly every theory of civil liability that plaintiffs have tried to impose on such companies.

And the immunity applies whether or not service providers decide to control what is posted on their sites. Service providers are thus free to choose whether to take down some material that they conclude is defamatory or otherwise offensive, or whether to keep it up. In either case, they are immune from liability (except as to material that infringes federal copyright or trademark law).

Thus, for much online material, there is no potential institutional defendant who might be held accountable. Plaintiffs can sue the individual authors — but if such a lawsuit doesn't give the plaintiffs the relief they seek, no other defendants are available.

Some of the problems discussed in this Article could be ameliorated by repealing or limiting § 230, and thus by giving organizations that are vulnerable to civil liability an incentive to police speech. Of course, this would exacerbate other problems, chiefly by giving the organizations too much of an incentive to police even protected speech. For our purposes, I will assume that § 230 endures, though the concerns discussed in this Article may lead some readers to reflect on whether § 230 ought to be modified — say, by instituting a limited notice-and-takedown provision, such as the one provided for copyright infringement under the Digital Millennium Copyright Act — or whether such calls should be resisted.

[C.] Anti-Libel Injunctions and Criminal Contempt

As compensatory damages have become practically unavailable to more and more libel victims, courts have shifted to a remedy that had long been seen as categorically forbidden — injunctions against libel.[3] And this trend seems to have accelerated as the Internet has democratized access to the media.

As Judge Posner noted in one recent Internet libel case, the traditional ban on libel injunctions "would make an impecunious defamer undeterrable. He would continue defaming the plaintiff, who after discovering that the defamer was judgment proof would cease suing, as he would have nothing to gain from the suit, even if he won a judgment." In traditional equity terms, the assumption that libel plaintiffs have an "adequate remedy at law" in the form of a damages claim is especially inapt when it comes to the judgment-proof defendant. And the Internet makes it easier than ever for judgment-proof speakers to cause damage that is substantial, yet financially irremediable.

Anti-libel injunctions can avoid this problem by adding a potent enforcement tool: the threat of jail. Continuing to libel someone in violation of an anti-libel injunction is criminal contempt, punishable by jail time. Failing to take down libelous material in violation of a takedown order may also be civil contempt, which could lead to the threat of jailing until the defendant complies with the order. Even if we're judgment-proof, we aren't jail-proof (unless we're safely anonymous or outside the jurisdiction).

I discuss the mechanics of anti-libel injunctions in much more detail elsewhere, and talk there about what First Amendment protections those injunctions have to contain. For now, though, the point is simple: greater equality of access to speech has meant more speech that is widely distributed, libelous, and said by poor speakers; and that in turn has led to more calls for a remedy that, at bottom, rests on the threat of criminal enforcement.

[D.] Criminal Libel: Survival and Revival

Libels by the judgment-proof have also led to the use of a mechanism that is all about criminal enforcement: criminal libel. Even if criminal libel prosecutions are rare enough that they won't provide general deterrence of Internet speakers at large, they seem likely to yield a prompt takedown of the allegedly libelous speech, and a prompt suspension of such speech during the prosecution. Once an indictment is filed, only rare speakers would boldly continue the same behavior that got them prosecuted.

And criminal libel prosecution can also benefit poor victims of libel, because the state pays the legal costs. The victims may get little financial compensation: restitution appears not to be a common remedy in criminal libel cases — and even if restitution were made available, and were easier to get through the criminal process than through the civil process, you can't get blood from a stone even through a criminal prosecution. But you can get some sense of vindication, and likely removal of the reputation-damaging material.

Criminal libel law is often described as essentially dead. But it is constitutionally permissible if it's properly limited to knowing falsehoods.

Many criminal libel statutes did not survive the Court's libel revolution as well as civil liability has, partly because statutes are less supple than common-law tort rules. Because libel was a common-law tort, state courts could easily preserve a constitutionally narrowed form of civil libel action just by adapting state tort law rules to fit the Court's emerging libel caselaw, and doing so with each new Court decision. But by 1964, the criminal law, including the law of criminal libel, had been codified in most states. The Supreme Court's cases rendered those statutes unconstitutionally overbroad.

And when the statutes were challenged, courts were often inclined to just strike them down rather than to narrow them by essentially adding new limiting language to them. Since 1964, courts in several states have struck down the old statutes, and in most of those states the legislatures did not reenact narrower, constitutionally valid versions. Indeed, in some states, legislatures just repealed the criminal libel statutes altogether. Criminal libel laws are thus indeed less popular now with legislatures than in the past.

But in about a dozen states, the laws remain on the books. And recent years have begun to see something of a revival in criminal libel enforcement, at least in some states. As best I can tell at this point (I'm writing a broader article called Criminal Libel: Survival and Revival in which I hope to canvass this in more detail), there are likely about twenty criminal libel prosecutions per year throughout the country.

Twenty cases a year is not a vast amount. (Libel injunctions, which I mentioned above, are more significant.) But the prosecutions show that some prosecutors do see criminal libel as a valuable tool; and what some prosecutors do now, others can do in the future. Indeed, there is some evidence from Wisconsin that criminal libel prosecutions rose from 1991–99 to 2000–07, the era during which Internet use surged. And most of the prosecutions involve ordinary people lying about each other online — impersonating each other in reputation-damaging ways, accusing each other of child molestation, and more.

Sometimes, the prosecutions or threatened prosecutions do appear to be political abuses. Consider, for instance, the case of the Louisiana sheriff who went after an anonymous online critic who had claimed that the sheriff had improperly given a local businessman a contract.

The sheriff got a search warrant based on the theory that the criticism was criminal libel of the businessman, and managed to identify the critic as a result. But the businessman was himself a local government official, and Louisiana courts had already held the state criminal libel statute unconstitutional as to public officials, or for that matter as to anyone involved in a matter of public concern. The Louisiana Court of Appeals therefore set aside the warrant as "lack[ing] probable cause because the conduct complained of is not a criminally actionable offense" — but only after the critic was identified as a police officer from a neighboring jurisdiction.

It's possible, then, that criminal libel law is unduly chilling, and subject to potential political abuse. Maybe it should be categorically barred as to speech on matters of public concern: punitive damages are barred in public-concern cases, unless "actual malice" is shown — perhaps criminal libel law should be even more severely limited.

Or perhaps criminal libel laws should be invalidated or repealed altogether, for instance because the line between speech on matters of public concern and private concern is too hard to draw, or because we think the legal system already criminalizes too much, and adding even misdemeanor punishments will only exacerbate the problem. But if we do set criminal libel law aside, we have to acknowledge that we're setting aside what might often be the only effective tool for punishing and deterring intentional libels.

[E.] Criminal Libel by Another Name

Indeed, one state — my own California — appears to be reinventing criminal libel law after a decades-long break. In 1976, a California appellate court struck down the California criminal libel statute, in a case involving a publication about the famous actress Angie Dickinson. Ten years later, the California Legislature repealed the statute.

But two recent California Court of Appeal decisions have read an identity theft statute as essentially recriminalizing libel (though with no evidence that the Legislature contemplated this). The statute criminalizes "willfully obtain[ing] personal identifying information . . . of another person" and using it "for any unlawful purpose, including to obtain, or attempt to obtain, credit, goods, services, real property, or medical information without the consent of that person." "Personal identifying information" includes "any name, address, telephone number" alongside other identifying items (such as Social Security number, bank account number, and the like).

Though the statute is colloquially called the "identity theft statute," California courts have held that the statute isn't limited to behavior generally viewed as "identity theft," such as impersonation, or to financial fraud. The statute, they have held, "contains no requirement that the defendant hold himself out as someone else," nor does it "require an intent to defraud or to cause harm or loss to another."

And courts have held that the "unlawful purpose" could be a purpose to commit a tort, such as libel, and not necessarily a crime.[4] Since nearly all libels would involve the use of at least one piece of "personal identifying information" — the subject's name — nearly all knowing libels are thus criminal under this interpretation. Indeed, two California appellate decisions have expressly upheld criminal convictions for posting libels.

I think the California Court of Appeal has erred in reading the statute so broadly. A step as significant as recriminalizing libel, following express legislative repeal, ought not be lightly taken, especially since such a reading may well not be what the Legislature intended — the requirement of "unlawful purpose" in the criminal statute may easily have been intended to refer to criminal purpose.

Nonetheless, the impulse behind these decisions — the impulse of prosecutors who argued for this theory, and the judges who adopted it — shows the appeal of criminal libel prosecutions, even when a statute has to be stretched for that purpose. The law ought to do something about knowing lies about people, the impulse suggests, and the civil law of libel alone does virtually nothing when the libelers are judgment-proof. That is part of the reaction that I'm aiming to describe.

Modern "criminal harassment" and "cyberstalking" laws are also being adapted to revive aspects of criminal libel law. Traditionally, such laws have banned unwanted speech to a person (such as telephone harassment or in-person approaches). But increasingly they also ban unwanted speech about a person, if it's intended to "harass," "annoy," "alarm," or "embarrass"; and much libelous speech can be said to qualify. Indeed, several state and federal cases have allowed such statutes to be used to criminally punish speech that in earlier decades might have been punished as criminal libel.[5]


[1] Occasionally, plaintiffs will litigate such cases, if they think that they (1) have a great deal at stake, and (2) can persuade even judgment-proof defendants that having a judgment against them is such a hassle that it's better to agree to take down and stop the libels. See, e.g., Obsidian Fin. Grp., LLC v. Cox, 740 F.3d 1284 (9th Cir. 2014). But my sense is that this is relatively rare.

[2] Ben Ezra, Weinstein, & Co. v. Am. Online Inc., 206 F.3d 980, 983-84 (10th Cir. 2000); Hassell v. Bird, 420 P.3d 776, 785 (Cal. 2018); Kathleen R. v. City of Liverpool, 104 Cal. Rptr. 2d 772, 780 (Ct. App. 2001); Giordano v. Romeo, 76 So. 3d 1100, 1102 (Fla. Ct. App. 2011); Reit v. Yelp!, Inc., 907 N.Y.S.2d 411, 414 (Sup. Ct. 2010).

[3] See Eugene Volokh, Anti-Libel Injunctions, 168 U. Pa. L. Rev. 73 (2019).

[4] See People v. Bollaert, 203 Cal. Rptr. 3d 814, 825 (Ct. App. 2016); People v. Casco, No. G049375, 2015 WL 2455083, at *6 (Cal. Ct. App. May 22, 2015) (nonprecedential); In re Rolando S., 129 Cal. Rptr. 3d 49, 58 (Ct. App. 2011). A nonprecedential case, Clear v. Superior Court, No. E050414, 2010 WL 2029016, at *1 (Cal. Ct. App. May 24, 2010), had concluded that "There is no authority that the commission of civil tort, such as defamation, constitutes an unlawful purpose." But there is such authority now, and in precedential opinions (In re Rolando S. and Bollaert).

[5] Criminal Complaint, United States v. Thompson, No. 17 MAG 1532 (S.D.N.Y. Mar. 1, 2017) (cyberstalking complaint based on defendant's impersonating the victim and sending threats in her name); Burroughs v. Corey, 92 F. Supp. 3d 1201, 1208-09 (M.D. Fla. 2015) (upholding Florida criminal harassment statute in part because it "prohibits unprotected conduct," such as "defamation"); United States v. Matusiewicz, 84 F. Supp. 3d 363, 371-72 (D. Del. 2015) (allowing criminal cyberstalking prosecution on the grounds that the distressing speech in that case was "defamation"); United States v. Sergentakis, 2015 WL 3763988, at *7-8 (S.D.N.Y. June 15, 2015); United States v. Sayer, 2012 WL 1714746, at *4 (D. Me. May 15, 2012) (likewise); Commonwealth v. Cox, 72 A.3d 719, 721-22 (Pa. Super. Ct. 2013) (upholding harassment conviction for publicly accusing someone of having a sexually transmitted disease, a classic example of libel); Order Imposing Sentence, Commonwealth v. Abrams, No. MJ-3810+NT-0000217-2014, at 1 (Pa. Mag. Ct. Sept. 15, 2014) (sentencing for criminal harassment based on allegedly "slanderous" statements against a business, see Private Criminal Complaint, id. at para. 2 (July 7, 2014)).