David Inserra
Americans generally don’t have to worry about online platforms removing speech because one person complains. We get into Facebook arguments, engage in debates on Reddit, criticize a podcaster or journalist on YouTube, laugh or wince at gamer hot takes on Twitch, or complain about the fairness of a Wikipedia entry all the time—but we aren’t concerned that some offended party can shut down an entire discussion and scrub it entirely from the internet.
But a recent case in Portugal shows how, without strong liability protections for online platforms, the future of online speech is under threat not just in Europe but around the world as well.
The case involved Cesar DePaco, a Portuguese businessman and former honorary consul of Portugal and Cape Verde to Florida, who resigned due to criminal allegations from 1989 and more recent allegations of donations to a right-wing populist party. While DePaco sued various news organizations and journalists for making false claims about him, he also sued Wikipedia for allowing such allegations on his Wikipedia page.
But Wikipedia said that it didn’t write or control the content—that was left to the community of Wikipedia editors, private individuals who create, debate, and update Wikipedia entries. Wikipedia said it was not liable for such content, but the Portuguese courts disagreed, ordering Wikipedia to remove significant parts of DePaco’s Wikipedia page and to identify the anonymous editors. Wikipedia is appealing to the European Court of Human Rights, but the Court’s precedents do not inspire confidence.
Setting aside whether or not the claims were libelous or not, this case should be deeply concerning because a European court just ordered speech removed from the entirety of the internet. Taken together with the European “right to be forgotten” and other judicial precedents, this court case threatens the global internet. If platforms are to be held liable for third-party content and compelled to delete anything off the global internet, then we may increasingly have to wonder whether what we are reading online has been censored by a foreign government. The rules set in Brussels are directly impacting Americans’ ability to speak and access information in this case.
And such censorship will often not be so explicit. The mere threat of liability and punishment now creates an incentive for platforms and websites like Wikipedia—whenever anyone threatens to sue for supposedly libelous or harmful content, it would be legally safer to quickly remove the content. The English Wikipedia has around 7 million articles, and across all languages, it’s more than 65 million. Many of those articles contain descriptions of people that some may find misleading, libelous, or just unwanted. If everyone who doesn’t like a Wikipedia entry is allowed to sue Wikipedia as European laws allow, then every piece of content is a potential legal and financial risk.
But how can Wikipedia possibly adjudicate what is truth and what is libel for millions of articles containing billions of factual claims? The answer is that it simply can’t. And so it is left in a hard place: continue to provide users with a vast and free source of information and face significant legal threats, or remove speech that could pose a legal risk. Such a system empowers those with more money, more interest in litigation, and more desire to suppress speech.
In the early days of the internet, a similar regime began to emerge in the United States. The first big court cases in the US were Cubby v. CompuServe and Stratton Oakmont v. Prodigy, and together they created a set of perverse incentives not so different than what we see in Wikipedia’s case. Companies could take a completely hands-off approach to reduce liability but would still be liable for failure to take action once notified that some content could be defamatory. On the other hand, if they tried to set any content policies or engage in any moderation, then they would be responsible for correctly policing all the third-party content posted on their platform.
Section 230 of the Communications Act, however, solved these perverse incentives. It provides online service providers like Wikipedia with liability protection so that they can run their platforms as they see fit. The creators of libelous content can be sued, but not Wikipedia for hosting it. Since they don’t need to worry about being held liable for the speech they carry, online platforms are free to allow a great deal more speech, as well as moderate and curate content as they see fit. Section 230 has allowed online speech and innovation to flourish in the US.
It is worth emphasizing just how different our online experience would be if we didn’t have Section 230. Speech that would be illegal or pose a legal risk to online platforms in Europe and elsewhere is just a normal part of our online discourse. Saying mildly humorous and personal things about people in your community has resulted in criminal punishment in Sweden. Creating a platform that allows users to share files resulted in a 5‑year court battle in Japan. Europe’s Digital Service Act creates massive fines—as large as 6 percent of global revenue—for companies that fail to sufficiently moderate content and follow a whole host of other rules.
Brazil had—at least in principle—protections for online speech under its Macro Civil “Internet Constitution.” That was until the Brazilian Supreme Court decided to progressively shred the law, issuing all sorts of mass, secretive censorship orders to online platforms, before recently declaring the Marco Civil unconstitutional and rewriting the liability standard on its own.
The liability protections offered by Section 230, however, have allowed the truly unique growth of online speech and platforms in the US. It is not a coincidence that so many successful technology companies are based in the US. This is especially true of platforms that allow social third-party content, like social media.
But over the past few congresses and administrations, calls to weaken or even repeal Section 230 have grown. While few US policymakers are likely paying attention to what’s going on in the Portuguese legal system, it is a striking reminder of how great we have it under Section 230. The rest of the world is a cautionary tale of censorship, overmoderation, and a less vibrant market for users if we weaken our commitment to robust online liability protections.