Last Monday, following over a month in exile after being dropped from Amazon Web Services, Google’s Play Store, and Apple’s App Store, the social media app Parler returned to life with a new hosting service and new content guidelines. Parler has become popular among the pro-Donald Trump right, as tech companies have increasingly cracked down on far-right content over the last year, and it now boasts 20 million users on what it calls “the world’s premier free speech platform.” Parler is just one part of a mass conservative reaction to increasing social media scrutiny of Republican commentators and politicians, culminating in Twitter’s permanent ban of President Trump. But although legal experts agree the social media bans that have led to the migration to Parler do not violate free speech, the fact that so few companies control the dissemination of speech in America should concern those across the political spectrum. 

It is generally accepted by legal scholars that freedom of speech does not mean that social media platforms cannot ban or restrict the speech of whoever they like, just as any business can deny service based on non-demographic categories. This is explicitly encoded in the Section 230 provision of the 1996 Communications Decency Act, which allows platform hosts to restrict user-submitted content as long as they are doing so in a good-faith attempt to remove objectionable content. But free speech, in common parlance, is more than just a legal principle. It is an important principle of an open society. The idea that everyone should be able to have a say no matter their opinion is central to the functioning of a liberal democracy, and most Americans agree.

When it comes to hateful speech in particular, though, opinions are more split. Philosopher Karl Popper popularized the “paradox of tolerance,” the idea that a society that is limitlessly tolerant would be eventually destroyed by the intolerant, and therefore that intolerance should not be protected. But this idea is far from universally accepted. John Rawls wrote in “A Theory of Justice” that any intolerance, even towards intolerance itself, is unjust, except when it materially limits the liberty of others; for Rawls, the right to equal liberty of speech superseded the harms of intolerance.

Of course, as we’ve already established, none of this takes away the right of corporations like Facebook or Twitter to restrict speech as much as they like on their own platforms. But in the 21st century, when fewer and fewer companies have oligopolies over avenues of user-submitted speech, these restrictions have shifted from a free-speech issue to one of corporate control. It isn’t a new phenomenon for speech to be controlled by corporations — the average person has a far greater likelihood of getting a message out to people today than they did before the Internet — but now the same handful of companies control speech everywhere in the country. This includes platforms like Facebook, Twitter, Reddit, and YouTube, but we’ve also seen drastic actions taken by web hosting companies like Amazon (as with Parler), or payment service companies like Mastercard and Visa. 

In a pre-Internet world, it was clearly reasonable that the local newspaper or a journal of amateur fiction could restrict what users could submit to their pages. But what if you’re a long-form video creator, and there’s only one platform that most Americans view videos on? Social media platforms, which now serve as leading forms of entertainment, news, interpersonal interaction, and activism, are increasingly becoming a public square for our collective expression. It should then be concerning that we could count on two hands the number of companies it would take to effectively wipe content off the Internet.

Unfortunately, the defense of social media speech is a hard line to walk in Washington, since it requires defending the people you despise most across political lines. (Even as I write this article, it’s difficult to defend controversial social media posts when the ones that precipitated post-Jan. 6 social media takedowns were mostly conspiratorial, anti-democratic nonsense.) There’s a reason that social media CEOs keep getting hauled into congressional hearings, and it’s because they’re the few unlucky people in America that both Democrats and Republicans agree to dislike. Both Trump and President Joe Biden have expressed interest in repealing Section 230 protections, which would make social media companies legally liable for content posted on their platforms. 

Yet while many Democrats see Section 230 as protecting social media companies from having to remove objectionable content, and Republicans see it as protecting social media bans of conservatives, a repeal of Section 230 is a misguided move that would get neither side what they want. To the GOP’s point, repealing the bill would make platforms legally liable for user-submitted statements on their platforms, which is patently ridiculous when there are billions of posts made every day and no feasible way to screen them effectively. Platforms would effectively be forced to prevent any speech that could legally damage them, leading to speech restrictions far outstripping what we see today. (It’s also ironic that the “pro-business” Republicans would endorse a Section 230 repeal that would be far more threatening to corporate independence than any law the Democrats could even dream of passing.)

To the Democrats’ objections, though Section 230 does keep social media companies from being legally responsible for hateful or violent user posts, a repeal of Section 230 would remove the law that gives those platforms the right to safely remove that content without expecting a costly First Amendment suit. In summary, no matter how bad the Internet might seem now, a regime that makes Internet companies legally responsible for the content they do not create and cannot feasibly review is not just unjust, but also unwise. 

One of the reasons that this issue is so difficult to solve is that our interests in freedom of speech usually do not extend to speech by the other side. While liberals and leftists have had a tendency to rejoice when social media platforms take down the accounts of prominent conservatives, they have reason for concern as well. Not only do bans of conservatives help fuel the sorts of alt-right platforms that have radicalized so many, but left-wing groups have also been the target of mass social media bans. It’s not hard to see that escalating, especially as Amazon battles against efforts to unionize. Even if many on the left might agree with social media companies’ bans of conservative figures, they would be wise to doubt whether international megacorporations are the best actors to be handing them out.

Of course, that selective attention to speech rights isn’t restricted to liberals. Take Parler, for example, which touts itself as a bastion of free speech and intellectual openness. Meanwhile, in its actual policy, Parler is actually a far more restrictive social media network than its competitors: it has forbidden content such as fecal matter, obscene usernames, “unrelated comments like ‘f— you,’” and the promotion of marijuana, all of which are allowed on Twitter. Furthermore, Parler has banned parody accounts and accounts critical of Parler, and the platform has conducted mass purges of left-wing accounts and of those supporting Antifa. For all the cries of social media censorship from the right, the network they promote to counter this is far more restrictive, except for the sorts of content that Parler has actually become the new home for far-right conspiracy theories, conservative echo chambers, and a collective dissociation from reality.

If not Section 230 or alternative platforms, what are the actual answers to free-speech issues on social media? Unfortunately, there aren’t any easy ones. It’s not legally reasonable to mandate platforms to host content they do not want to, nor is it reasonable to repeal Section 230 and hold them responsible for the content they do host. The former would be a massive overreach into companies’ rights, and the latter would, as discussed, end up changing the Internet for the worse. At the end of the day, this is a constitutional dilemma that requires us to weigh our society’s value for free speech and expression with the right to deny service. Both these rights are stretched to their limits, as mainstream political speech has increasingly radicalized, and that denial of service has ever-greater consequences due to corporate centralization. 

So even if they cannot solve our legal issues, it would help us all if we returned to the aforementioned dialogue on the paradox of tolerance. Is it indeed the case that we as a society cannot tolerate intolerance, lest that very intolerance destroy us? Or should we only restrict speech when it violates others’ liberties — and did Trump violate our liberties by attacking democracy? The answers to these questions are not legal, or even rational, but moral. But it is only with these answers that we can come up with consistent guidelines around speech and, one way or another, save the Internet.

Art by Andrew Diep for the UC San Diego Guardian.