Florida's Social Media Anticensorship Law and the Court's Tortured Legal Logic
Readers may recall Florida Senate Bill 7202, regulating social media platforms, which was passed by the legislature and signed by Governor Ron DeSantis about a year ago. The stated intention of the legislation is to combat social media censorship. This bill had a slate of provisions, including:
- Prohibiting deplatforming of political candidates
- Prohibiting censorship of posts about political candidates
- Prohibiting the removal of posts by “journalistic enterprises” based on content
- Requiring consistent application of censorship, deplatforming, and shadow banning
- Only allowing changes to terms of service once every thirty days
- Users must be able to opt out of curated content feeds, and platforms must allow users who opt out to receive material in sequential order
- Disclosure obligations: social media platforms must provide view counts, publish deplatforming standards, issue detailed explanations for deplatformed users, inform political candidates receiving free advertising of in-kind contributions
- Users who are deplatformed must be given at least sixty days to access their data
Tech companies challenged the law, and the US Court of Appeals for the Eleventh Circuit recently issued an opinion that most of the law’s provisions are likely unconstitutional.
Upon reading the opinion they issued, I have a number of observations.
Property Rights Play Almost No Role in the Court’s Decision.
For those who think it’s none of the government’s business what private companies do with their private property, it should be emphasized that this is not how the court reasons. Rather, the legal arguments have to do with the First Amendment as a positive right, rather than owners’ property rights.
Let’s look at the cases cited in the state of Florida’s argument about why SB 7202 is consistent with precedent and First Amendment jurisprudence. The first is PruneYard Shopping Center v. Robins (1980), in which the owner of a shopping mall challenged the state’s forcing him to allow the public to circulate pamphlets and petitions on his property.
The US Supreme Court affirmed the state court’s decision that the mall owner’s speech rights were not threatened by the petitioners because they didn’t prevent him from speaking. Also apparently relevant is the fact that the owner did not object to the content of the pamphlets. This is relevant because the social media companies’ argument regarding SB 7202 is that they object to the content of the users they censor and thus being forced to allow them to speak would violate the companies’ First Amendment protections (which the court of appeals accepts).
Apparently the state of the law is that it’s perfectly fine for the state to force you to host speech on your (finite) physical property as long as you don’t specifically object to its content, because your own speech is not inhibited, but that the state cannot force you to allow the speech of users on your social media platform if you object to their speech, because that undermines your speech.
Courts Can Work backward from Any Conclusion They Choose.
The second case the state of Florida cited was Rumsfeld v. Forum for Academic and Institutional Rights, Inc. (2006), in which law schools challenged the requirement that they allow military recruiters on campus, arguing that this violated their free speech rights (they wanted to bar recruiters from campus to protest “Don’t Ask, Don’t Tell” policies). In this case, the US Supreme Court decided that being forced to host recruiters didn’t violate law schools’ free speech rights because it “neither limit[ed] what law schools may say nor require[d] them to say anything.” (I am unsure how being forced to allow Alex Jones on Facebook limits what Facebook may say or that it requires Facebook to say anything.)
Hosting military recruiters did “not affect the law schools’ speech,” the court said, “because the schools [were] not speaking when they host[ed] interviews and recruiting receptions.” In fact, recruiting activities aren’t “inherently expressive”—they aren’t speech! (I’m sure the court would feel the same way about a group like the Proud Boys using Twitter to recruit new members.)
So what’s different between Rumsfeld and the SB 7202? The court of appeals argues that a social media platform that “exercises editorial discretion in the selection and presentation of” the content that it disseminates to its users “engages in speech activity,” whereas a law school wanting to express its disagreement with the military isn’t “in the business of disseminating curated collections of speech.” I guess charging tens of thousands of dollars to attend lectures about the law isn’t a business of disseminating curated collections of speech?
Bizarrely, the court of appeals also argues that social media platforms’ clear targeting of certain people and ideas is itself “expressive,” thereby implying that if platforms instead applied their policies evenhandedly, their banning people wouldn’t be so clearly expressive. Thus, the court rewards discrimination based on political ideology. Things would have probably gone better for the PruneYard mall owner had he explicitly disagreed with the content of the pamphlets and petitions forced upon him.
Invoking “Governmental Interests” Means Judges Decide Whatever They Want.
SB 7202 attempts to keep social media companies from unduly influencing elections by censoring or shadow banning political candidates and journalists. Since it’s a First Amendment case, which SCOTUS in its wisdom decided is a “fundamental right,” strict scrutiny is triggered, meaning that the state must prove a compelling, substantial interest in order to abridge that right. The court states, “Put simply, there’s no legitimate—let alone substantial—governmental interest in leveling the expressive playing field.”
This is quite interesting if one considers a famous case involving election finance and political speech, Citizens United v. Federal Election Commission (2010). Justice John Paul Stevens argued that corporations are not members of society and that there are compelling governmental interests to curb corporations’ ability to spend money during elections. So, basically, there’s a compelling government interest when courts want one, and there’s not when they don’t.
The Court Did Not Consider the Entire Law Unconstitutional.
Notably, the court considered most of the disclosure requirements of the law to likely be constitutional. Below is a chart summarizing the court’s opinion on individual provisions within the law.
The disclosure provisions can be important, as they create explicit expectations about a platform’s responsibility to a user. While most social media platforms are “free” to use, the actual exchange is a user’s data for use of the platform. I think the law would treat censorship and shadow banning quite differently if users paid cash to use the platform, as this would create consideration and therefore a contract between the user and the social media company. Censoring users in such an arbitrary fashion would, as such, be considered a possible breach of contract.
For another approach, see Jeff Deist on a tort law approach to social media regulation.
Texas passed an anticensorship bill that is being litigated in the US Court of Appeals for the Fifth Circuit. Given courts’ inconsistencies, strange body of decisions, and ability to make things up, who knows what will happen?