Outlaw social media censorship?

Samuel E. Miller writes,

Owners of real property know that both contractual covenants and statutes can restrict freedom of use. . .

Why should limitations to the otherwise free use of private property not extend to private property owned by tech giants? Congress could, for example, enact a statute prohibiting social media companies from tampering with non-criminal speech on their platforms.

An interesting take. Regulate the platforms by obligating them not to censor. Pretty much the opposite of what most people want, which is for the companies to apply censorship the “right way,” whatever that is. From the best magazine on the Web, Quillette.

35 thoughts on “Outlaw social media censorship?

  1. I’ve written this before, but I don’t think in its pure form this kind of policy would be acceptable to voters. Force social media companies to publish Nazi or NAMBLA commentary? I’m not sure most people are even that absolutist on the legality of speech, let alone on social media moderation. I think some allowance for ‘reasonable’ censorship would be incorporated into any such regulation, and from there I can easily see it creeping toward legally sanctioned selective censorship. I also don’t think there’s a clear definition of what exactly is or isn’t a platform, which could lead to issues.

  2. There are ‘common carrier’ situations for other networked industries; but overall, as you aptly point out, at least half the country doesn’t want that solution ported to social media. First, law enforcement is involved if someone contracts with a freight carrier to do something illegal – but law enforcement and speech codes are a fraught combination, especially when the speech codes are partisan by nature. Second, the freight companies are not in the business of making everybody’s goods discoverable by everyone else. It’s in the implementation of these algorithms that most of the trouble has come. If facebook had stayed true to its initial privacy and exclusivity, it wouldn’t have had viral possibilities. The things that made it popular are the things that made it problematic; because the real problem is that some people disapprove of (and are scared by, offended by) things that are popular – like populism.

    • “at least half the country doesn’t want that solution ported to social media”.

      Can you cite some authority for this proposition? Link please.

      Seems to me most people are very much in favor of common carrier rules in lots of other industries. Heck, increasingly, we don’t even let companies or single-property landlords discriminate against customers on the basis of whether or not they pay their bills: the core property right.

      “law enforcement and speech codes are a fraught combination.”

      Laws against prohibited communications are enforced all the time, as has been the tradition for centuries, and it works fine. The issue is whether one discriminates or is neutral as regards viewpoint, what you call “partisan”. If partisan codes are the problem, then what’s the matter with prohibiting partisan codes at social media companies? Many states already do this for regular private companies, for example, in the case of employment law and prohibited basis for termination: See: This recent post from Volokh and also read his empspeech paper, it’s excellent. Here’s from a footnote:

      The South Carolina law stated that any company or corporation that had a legislative charter could not “discharge, or threaten to discharge, from employment . . .any operative or employee, . . . for or on account of his political opinion, or for voting or attempting to vote as he or they may desire,” and provided both for civil liability and for cancelation of the corporate charter.

      I guess the majority of voters in all those states are somehow ok with prohibiting private companies from discriminating on the basis of political opinion and viewpoint. Weird!

      • I think you have made an attempt to delineate between what is acceptable and what is unacceptable censorship, but the lines you have drawn don’t really work either.

        If a group of commenters attempt what is commonly referred to here as “cancellation”, i.e. an opinion such as “this person’s viewpoint is vile and we don’t want him/her in our organization”, is it censorship to block that type of swarming opinion?

        Is it an acceptable viewpoint to find a viewpoint unacceptable?

        If an organization has expressed a set of core operational objectives which they deem important to their brand identity, and an employee wants to publicly express the opinion that they disagree with those objectives, does the company have to accept that public dissent from an employee under the principle of free political speech?

        We haven’t even reached a point where we demand that speech has a verifiable origination on social media platforms.
        Traditional limits on free speech such as defamation, copyright infringement, etc… are currently not possible, and hundreds of billions of dollars hinge on how we decide to deal with these issues in the next decade. We have a collision of values and we need to work a lot harder at figuring how to move forward.

        • Why does this argument not apply to discrimination in general?

          “Traditional limits on free speech such as defamation, copyright infringement, etc… are currently not possible”

          Not possible? What on earth are you talking about? Are you even familiar with this topic?

          The fact that it’s perfectly possible and being done all the time is *precisely* the issue, it is the very root of the problem, because now those capabilities are being applied to viewpoint discrimination censorship.

          Companies with intellectual property have become increasingly bold, aggressive, and successful in asserting their rights against various internet platforms and social media companies with billions of users, something that newer machine-learning tech is making easier, cheaper, faster, and harder to circumvent all the time.

          Look at what’s going on with Twitch right now – suddenly there is a lot of extremely successful copyright protection in a place that was recently rampant with infringement violations.

          Various Anglosphere countries are lobbying the EU to amend their ‘ePrivacy’ law so as to make exemptions so that social media companies can track people enough to help eradicate child pornography and prosecute those handling and especially making the stuff. The problem is not at all the technical capability to do this at the scale of literally *the entire human population*, it is mostly a matter of legal constraints, liabilities, and incentives.

          To understand this, one has to grasp that the bureaucratic capacity powers behind James C. Scott’s “Seeing Like A State” have been growing exponentially too along with Moore’s Law. Everyone worried what it would be like one day in the far future to live in a world where large, rich, powerful entities could literally watch every single thing every human does, but then keep imagining that it all still far away when in fact it was already feasible and practicable years ago, hence, the reality of China’s Social Credit system today.

          Such capabilities mean that literally none of this stuff is “impossible”. There are now more than enough affordable servers and processing units and advanced algorithms and “AI” tools to scrutinize every single thing every human says or does for any kind of definable transgression whatsoever.

          Which is *precisely why* it’s so critical at this particular moment to establish hard lines and prohibitions so that such capabilities don’t land us in a permanent and irreversible tyranny of fashionable elite opinion.

          Speech is Special. Once you lose the ability to express viewpoints, you won’t be allowed to argue to get that liberty back, either.

          • Fair enough. I could have been more careful with my language. Certainly some activities here are possible. Content gets taken down. In some cases, accounts can be suspended.

            But currently, accounts are predominantly unverified identities and actual users face virtually no consequences for violations other than the need to abandon a login.

            Holding users accountable for defamation and copyright infringement is not a thing online, at least not yet. Content can be injected by a non discernible identity, and that isn’t changing any time soon.

            —–

            “Which is *precisely why* it’s so critical at this particular moment to establish hard lines and prohibitions so that such capabilities don’t land us in a permanent and irreversible tyranny of fashionable elite opinion.”

            The point of my post was to show that there are no such “hard lines”, and that free speech works both ways. You haven’t expressed much support for “viewpoints” when the free speech is incoming criticism.

            Your use of the term “irreversible tyranny of fashionable elite opinion” clearly shows you refuse to accept the full consequences of free speech.

            There is no such thing as free speech if you can’t be subject to the heat of fashionable elite opinion. Criticism is free speech.

  3. “Pretty much the opposite of what most people want, which is for the companies to apply censorship the “right way,” whatever that is.”

    How do you know this? Was there a poll or study, “Are you cool with social media censorship?”

    I question this, because I often hear things like this in a “culture war has been lost” sense when 95% of HEE’s or establishment figures support something, but then it turns out like 60% of those polled are against it and it loses elections when actually put to the test.

    Also, I think “right way” is a bit slippery and unfair.

    Let’s say one is in favor of a First Amendment standard for social media companies, in other words, the same rules that apply to the state. The First Amendment prohibits viewpoint discrimination, but allows restrictions for time, place, and manner, in addition to liability for traditional exceptions like obscenity, defamation, copyright violation, etc. Does that mean one is in favor of “censorship applied the right way”, or would it be more fair to distinguish this from “viewpoint-based discrimination”? How about just for liability for copyright violation? Is that also “favoring right-way censorship?”

    My claim is that if you put the question more precisely and to a poll or a vote, “Are you in favor of prohibiting social media companies from discriminating between users based on their viewpoint” or “Are you in favor of making social media companies follow the same First Amendment rules as the government” that “most” hypothesis would not be verified.

    • Yeah, if you make your poll vague, about whether you favor “free speech”, you might find most people in agreement, but if you get specific (e.g., ask about Nazi speech, like a previous commenter said) I think you would find a lot of sentiment in favor of censorship. So the whole question is, where is the line? There is great disagreement over that, and I think it’s true that most people are not really in favor of free speech.

      • The ACLU doesn’t even favor free speech anymore, and they used to be the most free-speech organization around (Nazis in Skokie, etc.).

      • How about any poll result at all? If one is going to say “most”, one believes there is some poll, somewhere, in some language, vague or not, where the number comes out above 50%, and usually we use “most” to have some cushion above a bare majority, so maybe 55% or 60%. I asked Kling how he knows this and I asked BenK for a link, so I’ll ask you for the same.

        • How about just looking at pundits, politicians, and the news over the past several years and seeing many many people against free speech and no one of any importance favoring it?

          • How about when you can’t substantiate “most people” you just admit, “Well, I tried googling around for this, and it turns out I couldn’t find any evidence to back up the claim, so how about we move the goal posts from ‘people’ to ‘progressive elites’?”

            The whole point of these proposals is because a lot of people – I say a lot because I can’t prove ‘most’, see how that works? – don’t want to be ruled by the latest whims of progressive elites, and lately, the political situation in our society means that unfortunately only the state has the capacity to serve as a countervailing force to mitigate the kinds of social pressure and soft coercion those elites can impose on everyone by credibly threatening to cancel their lives.

            As it happens, there have been politicians of importance like Senators Hawley and Cruz who have indeed been proposing ways to reign in Social Media viewpoint discrimination, and there was enough of a backlash to Twitter spiking the (accurate) story about Hunter Biden that even Jack Dorsey had to back down and fake-apologize that it was wrong for him to order the deletion of all tweets with links to the story, and to suspend the New York Post’s account for as long as they did (message sent though, right?)

          • Handle — how about we turn this around, and you tell us what evidence you have that a majority of people support free speech that they don’t like, or from people that they don’t like.

  4. Arnold, today you say “pretty much the opposite of what most people want…”. Yesterday you argue about loyalty but your definition acknowledges that it amounts to some preferential treatment, that is discrimination, and you didn’t say “pretty much the opposite of what most people want…”. You were right yesterday, but wrong today.

    Recognizing people a legal right not to be censored by social media (or if you prefer imposing the legal obligation of social media not to exclude potential users) aims at preventing “arbitrary” discrimination (“arbitrary” as in “you cannot use it because I don’t like what you will say or write”). Recognizing loyalty as a social bond doesn’t involve rights or obligations, but instances amounting to some “extreme” form of disloyalty like betrayal or treason may be legal wrongs and punished. We accept that the type of discrimination involved in loyalty is based on relationships protected by the law. Recognizing people a legal right not to be censored by social media is based on freedom of speech protected by the law. But the goods protected by law are never absolute.

    Yes, today some bad people want to constrain freedom of speech. They are not a majority, however. They are the ones that control most of the press and social media and make a lot of noise. Yes, the rotten and corrupt democrats are relying on them to censor the opposition.

    I strongly recommend anyone willing to give an opinion on these issues to read at least Cooter & Ulen’s “Law&Economics”, and Cooter’s “The Strategic Constitution”.

  5. >—“Congress could, for example, enact a statute prohibiting social media companies from tampering with non-criminal speech on their platforms.”

    Of course this would immediately switch the debate to which speech was criminal. So then, to cite a “hypothetical,” if a President decided that the top leaders of the opposing political party were guilty of criminally treasonous speech and should be arrested, would that justify “tampering with” their “treasonous” opposition speech and proceeding to “lock them up”?

    It seems like we are rapidly heading toward the situation we had in the early days of the country when media outlets were simply expected to be partisan and no one found that shocking. Of course that led to the Alien and Sedition Acts which were probably the worst censorship this country has ever seen.

    New social media platforms spring up all the time especially when many people are dissatisfied with the old ones. I like Quillette a lot but I don’t go there because I am under the illusion they are neutral platform letting through much stuff they disagree with. I go there because they have a definite point of view I want to hear.

    • This was good enough for the readers of the Washington Post: “We must treat the Hunter Biden leaks as if they were a foreign intelligence operation–even if they probably aren’t.”

      Instead of cancelling their subscriptions or spitting out their coffee they just nodded in agreement. Perfectly content. Acting as if it was a vast foreign plot, but at the same time knowing it wasn’t. Because these are readers who expect their paper to be partisan.

      The Alien and Sedition Acts were repealed but the Logan Act was left on the books. So when Joe Biden was Vice-President he could talk about locking people up for treason under this vaguest of laws.

  6. Seems to me the social media companies want it both ways, to have the protection of common carriers, or open platforms, but then limit speech.

    Perhaps we need not “force” social media companies to do anything, but simply give a choice:

    1) You can say you are common carrier, like a phone company, and not limit speech, anymore than phone companies stop political actors or even criminal cabals from calling friends and spreading the most vile lies, or plotting crimes.

    2) You can limit speech, but then you are like a newspaper, and liable for what is on your site.

    So….

    Presently, we are seeing a melding of establishment, government and business censorship, and news manipulation, at work.

    • I think Ben Cole’s construct is great! The best path forward.

      Voters would understand that.

      “Newspapers and TV stations like the NY Times and CNN and FOX have to follow certain laws. Internet providers like Verizon have to follow other laws…

      …Facebook, Youtube, and Twitter claim they don’t have to follow EITHER set of laws. That’s not okay. They have to follow one. We are offering to let them CHOOSE but they have to follow one or the other.”

  7. Pick whatever analogy you want that tech companies are just like [thing they aren’t really like] and whatever outcome you want is going to run face first into the First Amendment.

    There are scores of other problems if you could simply wish your way past the 1st. No one wants all legal activity on every platform.

    1) Trolls, porn, grifters, just-shy-of-fraud would make user generated content on any platform essentially unusable.
    2) Any service that relies on reviews (AirBNB, Yelp, etc) couldn’t withstand the avalanche of defamation suits.
    3) Moderation at scale is *hard* if not impossible.

    • This exactly. You can’t write a rule that prevents the kind of political discrimination you dislike without also preventing the kind of banal moderation that keeps UGC heavy social spaces from being totally unusable cesspools.

      • Yup. Consider how without any moderation most internet comment sections degenerate quickly to being filled with troll posts. The relatively light hand that Arnold exercises here makes all the difference in keeping this comment section interesting and worthwhile. The quality would drop quickly with no moderation because it would signal that anything goes.

        Somebody is always going to scream censorship when you remove things and occasionally they will be right. As always there are tradeoffs.

  8. This has a slippery slope feel to it. If we allow the government to remove content moderation in the name of free speech on private platforms, what else might they force on the private sector in the name of freedom or rights? Or how far might they extend that regulation? What if it extended to beyond big tech? Even to blogs…

    • I don’t see why it has to be a slippery slope. The big social media companies are natural monopolies due to network effects, so there’s a clear distinction as to why you’d choose to regulate them more than (say) individual blogs.

      If you feel that Twitter is censoring you, sure you can go to Parler, but it only has 2.3 million active daily users, vs. 187 million on Twitter. It’s always going to be a subpar experience, with less content being provided from a far smaller pool of individuals and organizations.

      • I am coming to agree that, because of network effects, it probably would make sense to regulate Facebook and Twitter as public utilities.

  9. –“An interesting take. Regulate the platforms by obligating them not to censor. Pretty much the opposite of what most people want, which is for the companies to apply censorship the “right way,” whatever that is.”–

    I think the least bad approach is for accurate censorship controls to be given to the user. Consider VidAngel, for TV and movies, as a model. It lets you filter out violent or otherwise explicit content, bad language, even (delightfully) Jar Jar Binks, but people who are comfortable with those things do get to see them.

    Granted, people who disagree with you get to view content that you don’t happen to like, but there’s probably no way around that in a diverse, pluralistic society. And it’s probably for the best. Our culture often associates censorship with powerful people attempting to hide the truth, and unless your censoring is so perfect that no one knows its being done, it probably just adds to confirmation bias.

    • The New York Times has invented something similar, called the New York Times.

      By filtering out embarrassing news about Eric Swalwell, for example, the New York Times reduces the stress levels of its readers and makes them feel happier, less anxious.

      What gets published is designed to activate the pleasure points within the readers.

      • Very true.

        When it comes to social media, I think the least bad thing we could do is to allow users to self-censor.

        People aren’t going to be happy if they are censored by Zuckerberg & Co, but they will probably not even notice and hardly care if Bob Smith who they went to Middle School with blocks all of their political content.

        Granted, neither conservatives or liberals will be fully satisfied with this arrangement, as each wants the other side exposed to their ideas, but it’s an improvement for conservatives who are finding the opinions of conservatives they agree with censored.

        Ideally, news would either convey facts in a nonpartisan fashion or allow a trusted person from each significant ideological view to present a point/counterpoint view of the topic, but I’m not sure how you’re going to be able to enforce such a thing, as many people now get much of their news in the form of commentary from tweets and blog posts.

  10. Article 19 of the Universal Declaration of Human Rights states that “Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers.”

    Furthermore, Article 2 states “Everyone is entitled to all the rights and freedoms set forth in this Declaration, without distinction of any kind, such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status. Furthermore, no distinction shall be made on the basis of the political, jurisdictional or international status of the country or territory to which a person belongs, whether it be independent, trust, non-self-governing or under any other limitation of sovereignty.” And Articles 7 and 8 provide that “All are equal before the law and are entitled without any discrimination to equal protection of the law. All are entitled to equal protection against any discrimination in violation of this Declaration and against any incitement to such discrimination,” and “Everyone has the right to an effective remedy by the competent national tribunals for acts violating the fundamental rights granted him by the constitution or by law.”

    https://www.un.org/en/universal-declaration-human-rights/index.html

    The campaign by Twitter, the other tech giants, payment processors, web hosting services, and progressive political to suppress the expression of dissident political expression constitutes a violation of human rights and the USA is obligated to provide a remedy under the terms of the Universal Declaration. The sorts of statutes that Samuel E. Miller advocates for here are therefore mandated by the Universal Declaration of Human Rights. Moreover, the USA’s failure to provide such protection of dissident expression is a violation of human rights.

  11. Just for argument’s sake — okay, trolling — imagine a slightly different history where we had had something like Twitter and Facebook back in the good old 1930s and ’40s, and the comment streams were furious during the battles with discussions of Allied and and Axis strategy and philosophical inquiries about whether Jews and Gypsies and Commies and Fascists were making the world a better or worse place. Imagine, just to be really hypothetical, that those comment makers were not all good old truth- and freedom-loving Americans but included some Russians and Germans and Italians , not all forthright about their identities, and various individuals who sort of thought that maybe those Nazi guys had some decent ideas even if you couldn’t quite say so in public …. Can any of you imagine such a thing?

    How would all you folks deal with free speech and section 230-like issues in this ever-so-much-more exciting and technologically advanced version of things? Feel free to be idealistic and absolutist in your prescriptions, as you’re trying to be here.

    • Well, The War Powers Act of 1941 provided that:

      “Whenever, during the present war, the President shall deem . . . public safety demands it, he may cause to be censored under such rules and regulations as he may, from time to time establish, communications by mail, cable, radio, or other means of transmission passing between the United States and any foreign country.”

      So that would just about cover everything. Maybe you are right and that is what is needed today. It seems to be more and more like a war time situation.

  12. The First Amendment “legal speech” should be more relied on and used more.

    Mostly we need user ability to avoid seeing what they don’t want to see, as well as better labeling of messages.

    I like the (1) common carrier idea, but social media is too far away from phone company.

    We need a Digital Utilities Commission – regulators. For free speech, within reason. Not so easy. Will be “captured” in the not too distant future. And always a political issue.
    Like reality.

    • Yes, it’s always a political issue. That is why in line with what reader Benjamin Cole said, I support the idea that by definition “social media” should be common carriers to be regulated and supervised by some DUC as you mention. If you don’t want to provide a platform under a new law of common carriers, then you have the choice of running a “club” under your rules but you are co-responsible for your members’ crimes.

      To confirm that it’s always a political issue, the problem has not been created by a new technology but by the old temptation of absolute power. The rotten and corrupt democrats have colluded with radical leftists whose position has always been to destroy their enemies’ culture. Democrats are relying on these radicals to do the dirty work but without their sponsorship and protection the radicals cannot cancel their enemies’ culture.

Comments are closed.