Why Public Officials Have a Right to Block People on Social Media

August 21 Update: I made some important mistakes in the article below, and I have since drafted a new article dealing with the same issues. Please see the new article for my developed views. I am leaving up the text below, despite its problems, as an archive. Please do not quote from it as though it reflected my developed view. My basic mistake was to assume that, because social media companies block people, therefore government may not use social media for official forums of public commentary. But government may do so, I now conclude, so long as they also provide a means to comment outside of social media. I apologize for the confusion caused by the release of the draft below. However, I wouldn’t have made the advances in my thinking that I did without publishing the initial draft, so I have a hard time regretting it. —Ari Armstrong

The right to freedom of speech entails the right not to speak; the right to freedom of association entails the right not to associate. If government may force you to speak when you do not wish to speak (aside from cases involving obstruction of justice or the like), or associate with people with whom you do not wish to associate (aside from lawful imprisonment or the like), then rights of speech and association are not rights at all, but merely government-granted privileges. (I speak here of rights in the moral sense, not to refer to whatever politicians and judges happen to say rights are.)

If we take seriously rights of speech and association—rights enshrined in (not merely granted by!) the First Amendment—then we must take seriously (among other things) a couple of derivative principles. First, private organizations (with “private” understood here broadly to include all non-governmental corporations) have a right to establish terms of association with respect to the organization. Private organizations do not lose their rights by virtue of operating via the Internet. Second, individuals have a right to say what they want to say and to whom they want to say it (leaving aside speech that is inherently rights-violating, such as incitement of violence). Individuals do not lose their rights by virtue of being elected to public office.

Social media sites such as Facebook and Twitter are private organizations in the relevant sense. They are not arms or branches of the government. They are not guilty of censorship when they remove content or ban users that they deem in violation of their terms of service. So, for example, the fact that GoDaddy and Google refused to host a white supremacist site, and the fact that Twitter suspended the account of the white supremacist group, does not constitute censorship. Private organizations are not government, they are not aspects of government, and they properly may set their terms of use however they like (however vague, stupid, or irregularly applied those terms may be).

Private organizations may not properly be compelled by government to conduct or facilitate official government business—such use of force would violate rights of speech and association. (If there are exceptions to this they would apply in very narrow and unusual circumstances.)

Yet, if we say that public officials may not block users or users’ comments on social media, we in effect say that government may force private organizations to conduct or facilitate official government business. Why? Bear in mind that social media sites already block certain people from using their services—so obviously those people cannot interact with public officials on those sites. If we are to be consistent in saying that public officials may not block users, then we must also declare that the social media sites themselves cannot block users with respect to their access to public officials. If we declare that public officials violate people’s First Amendment rights by blocking them, then we must also declare that the social media sites themselves violate people’s First Amendment rights merely by enforcing their terms of service. Obviously that is an absurd conclusion, the effect of which is to declare that government must violate the First Amendment rights of social media companies to protect the First Amendment rights of social media users.

The obvious alternative—and the one consistent with individual rights—is to recognize that a public official may not unilaterally impose upon a private company the conduction of official government business. And insofar as a public official is not conducting official government business, the official has the same rights as everyone else to speak and associate (or not speak and not associate) freely. This includes the right to block users and users’ comments on social media as consistent with the relevant terms of service.

Let’s take a more tangible example to clarify the relevant issues. If a public official rented a private hall for official government business—say, to hold a vote on a town ordinance—the public official would need explicit buy-in from the hall, and the hall would not be able to ban participants. For example, if the hall’s owners had banned some white supremacists from the hall for giving a Nazi salute during a meeting, those people could not be banned from the official government meeting (assuming the meeting is open to the public, as is usual). Government may not exclude people who in the past gave a Nazi salute; private parties may. Insofar as a hall agrees to host an official government function, the hall in effect is not “private” for the duration of that event. It operates in the service of government and must play by government rules.

If Facebook explicitly agreed to host an official government function—again, say a vote by city officials on an ordinance—Facebook would not be able to block certain people simply because they violated Facebook’s terms of service then or in the past. For purposes of that meeting, Facebook effectively would be acting as an agent of the government, again subject to government rules.

We can take an extreme example from the The Circle, a film about a Facebook-like social media site. One proposal that comes up in the film is to run official public voting for government officials through the social media site. The implication is that literally everyone would have to be allowed onto the site. Obviously government could not contract with a private party to operate official voting and allow exclusions based on the company’s terms of service.

Anyone who claims that a public official may not block users on social media on First Amendment grounds, but that the social media service in question may itself block people from participating (because they are blocked generally), in effect calls for a two-tiered system in which some people have certain rights that others do not have. Such a position is untenable, legally and morally.

The above analysis has obvious application for several current news stories, the most prominent involving a case of people suing Donald Trump for blocking them on Twitter. One of the arguments for the suit, the New York Times summarizes, is that “Trump was imposing an unconstitutional restriction on the plaintiffs’ ability to participate in a designated public forum.” But Trump cannot unilaterally declare a privately owned forum a “public forum” in the sense relevant to official government action. If he could, then obviously Twitter would not be able to block anyone, possibly except in very narrow circumstances, from interacting with Trump’s account, regardless of what Twitter’s terms of service might say.

Now someone might say that Trump, as a public official, has a positive obligation to release comments related to government business only in a forum open to the public, where only the First Amendment, and not a private company’s terms of service, applies. The implication is that Trump would simply have to stop using Twitter. But such a policy would violate Trump’s rights, as Trump has a right to freedom of speech just like the rest of us do. Government officials are not above the law, but they are not beneath the law, either.

I therefore agree with analysis of Michael H. Baer of the Department of Justice:

It would send the First Amendment deep into uncharted waters to hold that a president’s choices about whom to follow, and whom to block, on Twitter—a privately run website that, as a central feature of its social-media platform, enables all users to block particular individuals from viewing posts—violate the Constitution.

Oddly (to me), Jacob Sullum writes on this matter for Reason magazine—an overtly libertarian publication—without addressing the problems that I bring up. Sullum writes, “The crucial question is whether Trump has created the constitutional equivalent of a town hall on Twitter.” That’s wrong. The central question is whether a government official may unilaterally impose on a private party the demands of a government-regulated town hall. The ultimate implication of the sort of legal reasoning that Sullum reviews is that government should treat social media companies as public utilities, at least in important ways. I take it that the partial (or total) nationalization of Twitter and other social media companies is not now a libertarian position.

A second case involves a county official in Virginia who blocked people from her Facebook page. A federal court found in favor of the people suing the official, on grounds that the official operated the page in question in an official capacity. But the ruling ignores the fact that a government official may not impose on a private party—in this case, Facebook—the obligations of hosting an official government function. What the court should have said is that, regardless of how the public official handled the account, it simply wasn’t a public forum in the sense relevant to First Amendment law.

A third case involves Colorado State Senator Ray Scott. As the Daily Sentinel relates, Scott blocked several people from posting to his Facebook page. An editorial by the Sentinel argues that Scott was legally wrong to do so. But none of the complaints counter my main two claims: First, that government may not unilaterally impose on private parties the requirements of official government purposes, and, second, that public officials have First Amendment rights of speech and association, too.

One thorny issue is what constitutes official government action. I think that only a narrow interpretation of official government business is legally justified. Obviously a meeting to vote on policy, to solicit comments explicitly from the entire public, or the like is an official government function. But we cannot sensibly extend officialdom to anything a public official says that pertains to their office, or public officials would have severely curtailed First Amendment rights. So, for example, government cannot rightly bar public officials from holding exclusionary rallies, private fundraisers, private dinner parties where participants discuss politics, and so on.

If we say that public officials, whenever they discuss policies relevant to their office, may not block participants, then the implication is that politicians must not use private social media accounts to discuss politics—because some potential participants already are blocked by social media sites. The proper conclusion, however, is that public officials should be free to use social media to discuss politics, and that such use does not constitute official government business in the sense relevant to First Amendment law.

This brings me to a peculiar area of law pertaining to open meetings. According to a summary from Colorado’s Office of Legislative Services:

The Open Meetings Law (OML), which is part of the Colorado Sunshine Law, generally requires any state or local governmental body to discuss public business or to take formal action in meetings that are open to the public. A “meeting” refers to any kind of gathering, convened to discuss public business, whether in person, by telephone, electronically, or by other means of communication. The Colorado Supreme Court has held that “a meeting must be part of the policy-making process to be subject to the requirements of the OML.” (endnotes omitted)

I’m certainly no expert in this area of law, but the measure seems highly ambiguous to me. If the law means that a few legislators cannot get together over dinner to discuss a legislative strategy, then that would constitute a violation of the legislators’ rights of speech and association. This seems to be the meaning of the law; the summary says of a “state body,” “Two or more members of the body conducting business are subject to this law.”

The dividing line should be actions that anyone could do versus actions that only public officials could do. Can private citizens get together to discuss legislative strategy? Obviously. So public officials have a right to do the same thing, and state laws to the contrary should be tossed on First Amendment grounds. The Bill of Rights applies to public officials too! Can private citizens get together to alter city ordinances or state statutes? Obviously not. So when public officials are involved in such actions, they properly are covered by open-meeting laws. So I do see the propriety of open-meetings laws, so long as they are narrowly tailored to actual government actions.

We should bear in mind here that the First Amendment is part of a Constitution that was drafted in private. The people who drafted and adopted the First Amendment certainly did not think that public officials meeting in private thereby violated anyone’s rights. I do think that narrowly tailored open-meeting laws are compatible with the Bill of Rights, but that broader restrictions on the speech of public officials are not.

No doubt critics of my narrow interpretation of public business would complain about the loss of “openness.” But openness is not the point of the Bill of Rights; the protection of individual rights, is. One theme of the Bill of Rights is to protect individuals from unwanted “openness” to government and the public. Government actions per se should be public (with exceptions for national defense and the like); the actions of public officials, except insofar as they directly pertain to an official government action, should not be public by legal mandate.

Of course, none of this pertains to how we, as private citizens, should respond to public officials. We have every right to complain when public officials block people from their social media pages for arbitrary reasons. We have every right to complain about “back-room deals” that take place in smoke-filled bars and the like—and we have every right to vote with such actions in mind. What we do not have is a right to violate the right to freedom of speech of public officials.

Let me close with a comment regarding my degree of certainly about all of this. First-Amendment law is complicated. Sunshine laws are complicated. I’m “pretty sure” that my analysis and conclusions here are solid. But I recognize that there are very smart people who specialize in these areas who might be able to bring facts or arguments to my attention that I have so far neglected. At this time I cannot imagine anything that would shake my confidence in these matters, but I have to recognize the limits of my knowledge here. Obviously I’ll post an update if I find that my analysis is in any important way wrong or incomplete.

I wanted to post these comments, even if somewhat tentative, because, so far as I’ve seen, several of the views that I’m expressing here have not been part of the discussion. I think they should be because they’re right and important. Getting these issues wrong means that government will violate someone’s rights.