Photographer: KIRILL KUDRJAVTSEV/AFP/Getty Images

Closing the Safe Harbor for Libelous Fake News

There's a way to enhance the truth while preserving the First Amendment -- and still prevent the next Pizzagate.

As things stand, Comet Ping Pong, the Washington restaurant falsely smeared as a hub for child sex trafficking in the Pizzagate mess, doesn’t have much in the way of a legal remedy. It could sue the anonymous conspiracy theory purveyors for libel, but even if it can find them, they probably don’t have any money to recover, and the damage to the restaurant’s reputation is already done.

In Europe, however, things might be different. The European Union recognizes a “right to be forgotten” on the internet, which under some conditions allows posts to be removed or blocked from search engines. Extending that kind of right to the American victims of demonstrably false news stories might actually help victims like Comet Ping Pong, who’ve been tagged with a falsehood that otherwise just won’t go away.

In the spirit of end-of-year creativity, it’s worth asking: Could Congress craft a legal remedy for victims of fake news that would be consistent with the First Amendment? The answer is complicated, but I would suggest that it is a qualified yes. U.S. law could authorize a judicial order to bar the publication -- and maybe searchability -- of specific, libelous statements that have been shown to be false in court. The First Amendment would still protect writing about those stories. But it would not protect the statements themselves.

To be sure, the U.S. Supreme Court has gone very far -- too far, probably -- in protecting false statements. In the leading case, U.S. v. Alvarez, decided in 2012, Justice Anthony Kennedy wrote a plurality opinion that extended full free-speech protection to Alvarez’s knowing and intentional lie about having received the Medal of Honor. Under that precedent, Congress can’t outlaw falsehood as a general matter.

But Kennedy’s opinion left an important loophole: the traditional First Amendment exception for libel. Although false statements outside the libel context are protected speech since the Alvarez decision, the same is not true of libelous statements. That means the government can't punish you for saying 1+1=3, because while that is false, it doesn't defame anybody. But it can impose damages for making up the story that a specified pizzeria is the hub of an underage sex ring run by specific people, because the statement is knowingly false and defames the owners and employees of the pizza shop as well as the specific politicians said to be involved.

When the libel is against a private party, the fact that the statement is false and defamatory is enough to remove it from free-speech protection. When it’s against a public figure, the New York Times v. Sullivan precedent requires that the defamatory statement be not only false but also maliciously so, meaning that its author knowingly lied or spoke with reckless disregard for the truth.

The remedy I am considering would build on this libel exception. Congress might say, for example, that there exists a private right for libelous statements to be removed from the internet. It might add that there is a general legal obligation not to publish such libelous statements. Further, and perhaps more doubtfully, it could prohibit enabling publication via search functions.

To protect against rampant overuse of the provision that would chill free speech, Congress could require a trial-like hearing in federal court before the remedy kicked in. Thus, Comet Ping Pong could go to court and sue the people who libeled it. If the court found the Pizzagate allegations to be false and defamatory, it would then have statutory authority to order the statements to be taken down.

If Comet Ping Pong counts as a public figure, the stories would have to be shown to be maliciously false. It might even be desirable to extend this greater protection to everyone libeled under the law, just to protect free speech that much more.

There are some possible First Amendment objections to this idea, but they all have adequate answers. One might be the view that the traditional libel exception doesn’t ban speech -- it just makes the person who published the libel pay damages for publishing it. It could be argued that an outright ban goes too far.

But that objection doesn’t seem quite right. Someone who violated the court order not to publish the libel in my proposed system would after all be hit with damages, the same deterrent used in ordinary libel.

Another possible objection would be that this law reaches third-party speakers, not only the publisher of the libel in the first place. But this, too, misses the mark. It’s libel for me to repeat an existing libel, and I can be found liable for doing so. The exact form of words shouldn’t matter.

And if there is a dispute about whether a given statement is the same as the outlawed libel, the court that issued the initial order can resolve it.

Regulating search engines by legal order, the way the European right-to-forget system does, is the probably the trickiest part, legally speaking. It’s also the most practically important. U.S. takedown orders won’t reach all who publish fake news. But they might well cover the major search engines that people use to spread the articles, as well as social networking sites that spread fake news.

Google or Facebook could try to say that they shouldn’t be covered because they aren’t publishing the libel, just letting others access the published information. Currently, the Digital Millennium Copyright Act safe harbor provision protects internet providers and similar services against being sued for content posted by users.

The law I’m imagining would have to lift that safe harbor for libelous fake news. If it did, the Constitution wouldn’t be violated. Traditionally, to repeat a libel is to publish it. If Google or Facebook reproduces a libelous statement on its proprietary page, the First Amendment allows that to be treated as a libel.

In short, a law against fake news wouldn’t have to create a Ministry of Truth. It’s enough for it to build on the tradition of judicial verdicts against false, defamatory statements. The First Amendment would not have to be infringed. And truth -- the traditional objective of free speech -- would be enhanced, not harmed.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.