Columbia law professor Philip Hamburger published a detailed post in response to my earlier argument that courts should focus on coercion in Murthy v. Missouri, the case in which two state governments and other plaintiffs allege that various federal agencies violated the First Amendment by pressuring social media platforms to ban various posts from their sites. I appreciate Prof. Hamburger’s thoughtful post. But I remain unconvinced.
Prof. Hamburger relies heavily on the use of “abridging” in the free speech clause, as opposed to the use of “prohibit” in the free exercise clause:
The First Amendment, however, has something to say about coercion. Prof. Somin recognizes the argument I make in Courting Censorship that the First Amendment prevents the government from “restricting” free speech, and therefore prevents it from curtailing that freedom. But he fails to note that the amendment also prohibits the government from “prohibiting” the free exercise of religion. The contrasting uses of the amendment of abbreviating AND prohibiting they are significant. IDat 254.
The contrast reveals that Prof. Somin’s argument about coercion incorrectly places free speech under the standard used by the Free Exercise Amendment. The word prohibiting appears to refer to various forms of coercion. So when he talks about the First Amendment instead abbreviating freedom of speech, would seem to take a different measure of government action for that right. Free speech is violated simply by curtailing it, whether through coercion or not.
Therefore, Prof. Hamburger concludes that “even when the government acts through entirely voluntary cooperation, without even a hint of coercion, it can still limit free speech.”
There may actually be a meaningful distinction between “reduce” and “prohibit.” The former is a lower standard than the latter, suggesting that even relatively minor restrictions are unconstitutional. But the fact still remains that, for there to be a violation of the Constitution, what the government must limit (or “shorten”) is “freedom of speech” (italics mine). In the absence of some kind of constraint, there is no loss of freedom.
Indeed, voluntarily choosing not to engage in some kind of discourse is itself a mistake exercise of that freedom. Both common sense and long-standing Supreme Court precedent indicate that the right to refrain from speaking is itself a right to free speech. If a private company voluntarily chooses not to engage in free speech activities, there is no violation of free speech, only the exercise of it. And this is true even if the decision was made in response to non-coercive government persuasion.
If a government official convinces Reason to prevent me from posting on their website because, it claims, my posts are unfair to the president and misrepresent the effects of his wonderful policies, there is no violation of free speech. In reverse, Reasons the decision would constitute an exercise of that freedom.
Prof. Hamburger tries to distinguish this type of situation from the case of social media because, in the latter, the consensus of the speakers was not obtained:
One could protest, as the professor does. Somin, that when the government convinces Reason do not allow him to post on this site, the government is not thereby violating his First Amendment rights. In his opinion, it is “no different” from when the government convinces platforms to suppress Dr Jayanta Bhattacharya’s posts. Truly, no difference? Prof. Somin publishes here as a member of the blog, which he was personally invited to publish under his masthead. While newspapers publish their choice of editorial submissions, blogs publish anything a member of the blog posts, but in any case, nothing is published or posted except what has been chosen by the newspaper or blog through the selection of an editorial or a blogger….
On the contrary, any individual can post on the platforms, at least until their posts are removed, and this is the issue of individuals, not that of the platforms. Not even the platforms claim otherwise…. Therefore, when the government consensually obtains a platform to remove posts, it is suppressing the speech of third parties without their consent.
In reality, it is not true that “any individual can post on the platforms, at least until their posts are removed”. They can only do this if they first agree to the platforms’ “terms of service,” which include restrictions on the types of speech allowed, and generally reserve the right to unilaterally change the terms of service. This last point refutes Prof. Hamburger’s thesis that users have not consented to the removal of content that does not violate the terms of service. If platforms impose restrictions that somehow go beyond the terms of service, users may have the right to sue for breach of contract. But that still wouldn’t be a violation of the First Amendment. In summary, while Facebook or Twitter/X impose fewer editorial constraints on content than New York Times OR Reasonthey still have them.
Prof. Hamburger also relies on contract law’s recognition of various forms of information asymmetry and “pressure” that do not involve coercion, but could still vitiate a contract. As the libertarian that I am, I condemn many of these changes in contract law. But whether they are defensible or not, they still do not demonstrate that there can be a violation of “free speech” without coercion. Indeed, these restrictions on contractual rights tend to be paternalistic in nature. They do not promote freedom, but rather limit it in the name of promoting welfare, out of fear that people might make bad choices in response to information asymmetries or social pressure.
Finally, it is ironic and internally contradictory that Prof. Hamburger has an expansive view of what is prohibited by the First Amendment when it comes to noncoercive government pressure to block social media posts, but a very narrow view when deals with Texas and Florida. attempts to force social media companies to host speech they disapprove of. In the latter situation there is clear and obvious coercion.
My position, by contrast, is entirely consistent: the government can use persuasion, but not coercion, in both situations. It cannot force social media platforms to publish material they disapprove of or ban what private companies would prefer to allow. But, in both situations, he can use non-coercive persuasion. As noted in my previous post, various veiled threats can also be coercive. Identifying them can, in some cases, create evidentiary problems.
Perhaps there is a disagreement here that goes beyond legal considerations. Professor Hamburger is, I believe, deeply concerned that social media companies that delete posts severely limit the marketplace of ideas. I’m not. The types of speech that federal agencies have tried to get social media companies to remove (e.g. anti-vaxxer speech, claims that the 2020 election was stolen from Trump, etc.) nevertheless remain widely available online and elsewhere . You can find plenty of them on social media platforms too! If there is a market demand for this type of speech, media companies will have strong incentives to provide it, as long as the government does not suppress them.
Professor Hamburger is right that speech can be a “public good” (though it can often also be a public bad, as when misinformation leads to the adoption of harmful government policies). But the willful decisions of social media companies pose no significant threat to the public good. Only systematic government coercion can truly endanger it.