The Lawfare Podcast: Jawboning at the Supreme Court

[Audio Excerpt]

Alex Abdo

It is unconstitutional, the plaintiffs argue, for the federal government to try to persuade platforms to take down constitutionally protected speech. And that's why a lot of the questions at argument yesterday were focused on all of the circumstances in which the government, historically and up to the present day, does that routinely, as a matter of course. And a lot of justices were trying to grapple with what the implications of the plaintiff's very broad legal theory would be. But if they accept that legal theory, then the plaintiffs probably have a strong case. But to accept that legal theory would be to essentially undo 60 years of legal precedent in the lower court, focusing on the line between persuasion and coercion.

[Main Podcast]

Quinta Jurecic

I'm Quinta Jurecic, a Senior Editor at Lawfare, and this is the Lawfare Podcast, March 21^st, 2024. Today we're bringing you an episode of Arbiters of Truth, our series on the information ecosystem. On March 18th, the Supreme Court heard oral arguments in Murthy v. Missouri concerning the potential First Amendment implications of government outreach to social media platforms, what's sometimes known as jawboning. The case arrived at the Supreme Court with a somewhat shaky evidentiary record, but the legal questions raised by government requests or demands to remove online content are real. To make sense of it all, Matt Perault and I called up Alex Abdo, the Litigation Director of the Knight First Amendment Institute at Columbia University. While the law in this area is unsettled, the Supreme Court seemed skeptical of the plaintiff's claims of government censorship. But what's the best way to determine what contacts and government requests are and aren't permissible?

It's the Lawfare Podcast, March 21^st: Jawboning at the Supreme Court.

It's been a long and winding road to get here. So to start off, can you just give us an overview of what exactly this case is and how on earth we ended up here?

Alex Abdo

Yeah, the case is about whether both the Biden and Trump administrations violated the First Amendment by pressuring the major social media platforms to take down what the administrations claimed was disinformation. The cases were filed by a couple of states and a couple of individual plaintiffs arguing that their posts relating to COVID-19 and the vaccines and the 2020 elections had been, in their words, censored by Facebook and Twitter at the behest of the government. And so the question that came before the Supreme Court was whether the plaintiffs had raised plausible constitutional claims under the First Amendment, that the administrations had coerced the platforms into doing its dirty work for them.

Quinta Jurecic

Before we get more into the details, I do want to make sure we touch on the procedural posture of the case, because that's part of the complication here. So first, there was this case before Judge Terry Doughty who issued an extremely broad injunction. Then it went up to the Fifth Circuit, which narrowed it somewhat, but it was still pretty broad, and then all that was stayed, well, this was before the Supreme Court. Can you just talk a little bit about that procedural route that it took and the scope of the injunction? Because I do think that strangeness in the underlying courts, which I do want to make sure we talk about a little bit more, is important to understanding some of the subtext of the arguments.

Alex Abdo

Yeah, of course. So the case went up on a preliminary injunction issued by the district court, like you said, and the district court took in a voluminous record submitted by the parties and made findings of fact as the basis for preliminary injunction that it entered. It issued this extremely broad preliminary injunction that essentially forbade the federal government from interacting with the social media platforms in a variety of ways, including a lot of ways that would have been perfectly constitutional for the federal government to do. And it even reached private researchers who wanted to engage with the government about their research in ways where the researchers were welcoming the interactions.

The Fifth Circuit broadly endorsed the findings of the district court in most respects, not in all, but then significantly narrowed the scope of the injunction in an effort to make sure it didn't reach a lot of perfectly permissible interactions between the government and the social media platforms and the government and researchers. It still reached, I think, quite a bit more broadly than the First Amendment actually requires the government to--than the First Amendment line that limits government interactions with the platforms. But that's what went up to the Supreme Court. And the reason why that's relevant is because the preliminary injunction was based on findings of fact by the district court, and usually findings of fact are reviewed very deferentially by appellate courts, including the Supreme Court.

And that question came up at argument yesterday in a couple different cases where various justices were basically asking, how much deference do we owe the district court's findings of fact? And by the way, a lot of people have pointed out, Mike Masnick did a great run through of all the factual findings or a lot of the factual findings by the district court, some of the amicus briefs submitted did this, pointing out that the district court just got a lot of things wrong, which was very strange, and suggested that there was motivated reasoning going on and you saw some justices grappling with that. Justice Sotomayor at one point basically accused the plaintiff's lawyer of misrepresenting facts or omitting crucial context. And it led to this very bizarre exchange where he apologized to the Court for not having been in his words, I think, “as forthright as the brief ought to have been,” which is a very, you don't want to be on your back foot like that in the Supreme Court. You want to make sure your brief is as accurate as possible. So it came up at the argument yesterday and it may very well appear in the Supreme Court's decision, this question of the deference that the Supreme Court owes to the factual findings.

Matt Perault

So could we, could we actually just try to cut to the end for a second? We want to obviously unpack a lot of the specifics of the argument and you're alluding to some of them here, but we're curious about your view of where this ends up. Where do you think the Court's going to land based on what you heard on Monday?

Alex Abdo

Yeah, so I don't have a great prediction for how precisely the Court will rule, but I think in one form or another it will be largely, if not entirely, a victory for the Biden administration. There was not a majority of the Court, at least an oral argument, that seemed convinced of the core allegations being made by the plaintiffs that there was a conspiracy between the federal government and the social media platforms to censor speech that the government doesn't like. Justice Kagan at one point said that she didn't think any of the allegations made by the plaintiffs suggested that the platforms were enforcing their policies at the behest of the government, as opposed to on the basis of their own editorial decisions to enforce their own content moderation policies. The only justice who seemed potentially convinced of the plaintiff's theory was Justice Alito. Oftentimes, Justice Thomas is in the same boat as Justice Alito, but I don't recall him even really saying that he was convinced or expressing any agreement with the underlying theory. He seemed really focused on a different question.

So I think the Biden administration will win. I think there's some question as to how, whether it'll be on the merits, in other words, whether the plaintiffs could show that the government had coerced the platforms, or maybe on some procedural question, the federal government focused a lot on the question of standing. So I don't know which route the Court will take, but I think it will very likely be a victory for the Biden administration.

Quinta Jurecic

So I definitely want to make sure we talk about the doctrine here, which is kind of a mess, as the justices seem to be struggling with. But before we do, let's jump back briefly to these questions about the factual record. So as you say, there was this incredible moment where Benjamin Aguiñaga, the Solicitor General of Louisiana, seemed chastened by Justice Sotomayor saying that she had a problem with his brief. There, there was also a, I thought, really striking moment where Aguiñaga at one point made a reference to material posted by Jim Hoft, who runs the right-wing Gateway Pundit website as being part of a spreadsheet that was maintained by CISA, the Cybersecurity and Information Security Agency, implying that CISA was using that as a tracking mechanism to, I think the phrase he used was “go after Hoft” or take Hoft's post down. And during the rebuttal, Fletcher, who's arguing for the government, actually says “This wasn't maintained. This wasn't a a CISA database. This was CISA being aware of notifications to platforms sent by the Stanford Election Integrity Partnership, which was a private organization, not an arm of the government or anything like that,” which I think is representative of some of the factual weirdnesses here. There's also some strangenesses, which Justice Kagan seemed really frustrated by in terms of whose posts were taken down and to what extent that was traceable to government action. Traceability was an issue that came up again and again.

So how should we understand the factual weirdnesses here? Not only in and of themselves, but also to what extent should they shape our understandings of the legal issues raised? Like, is this just a weird case, I guess is one way to phrase what I'm asking.

Alex Abdo

Yeah, it is a weird case. Most jawboning cases, cases where a plaintiff says the government violated the First Amendment by pressuring a speech intermediary, like the platforms or, in the days of old, the book distributors, most of them involve pretty discreet conduct, challenged conduct by the government. For example, like the case that was heard right after Murthy, the NRA v. Vullo case, is also a jawboning case, and it centered on a handful of very specific interactions between New York's Department of Financial Services and the NRA's insurance providers. And that's how most jawboning cases arise in the courts.

This case is much more atmospheric in nature. It posits a wide conspiracy among the federal agencies to engage in all sorts of disparate conduct. Any individual instance of which doesn't really look like jawboning, but collectively make you scratch your heads and wonder why was the government so invested in responding to all of these different alleged pieces of disinformation on the platforms. I will acknowledge that as a First Amendment advocate, it is a little bit bizarre how invested the federal government was in responding to all this disinformation. And we pointed out in the amicus brief that we filed that there was one exchange in particular that seemed to cross the constitutional line where you had a couple of White House officials berate Facebook for failing to take down, I think it was vaccine hesitancy information quickly enough. And then when Facebook didn't respond the way they wanted, they said, “We're considering our other options.” That to me seemed like crediting the plaintiff's allegations, giving them fair inferences, seemed like an implied threat and that maybe they should be allowed to go forward on that allegation. But the case as a whole is not just about that in one instance. It's about this broader pattern of conduct. And again, as Justice Kagan said, it's very hard to connect any of it to any actual instance of suppression that was taken on the basis of government coercion as opposed to the platform's decisions to enforce content moderation policies that they seemingly independently adopted.

And that's why so much in the case turns on what you think the right legal standard is. The standard that the lower courts have been operating under since 1963, when the Supreme Court decided the Bantam Books case, is the line between coercion and persuasion. All the lower courts have understood that the First Amendment prohibits the government from coercing intermediaries into suppressing speech, but allows them to persuade. But the plaintiffs in this case have actually offered a really radically different vision for what the First Amendment prohibits. Their argument really is that the government cannot set about to accomplish the suppression of speech, even if all that it uses is persuasion. It is unconstitutional, the plaintiffs argue, for the federal government to try to persuade platforms to take down constitutionally protected speech. And that's why a lot of the questions at argument yesterday were focused on all of the circumstances in which the government historically and up to the present day does that routinely as a matter of course. And a lot of justices were trying to grapple with what the implications of the plaintiff's very broad legal theory would be. But if they accept that legal theory, then the plaintiffs probably have a strong case. But to accept that legal theory would be to essentially undo 60 years of legal precedent in the lower court focusing on the line between persuasion and coercion.

Matt Perault

So I think Quinta and I have a bunch of questions about this aspect of the doctrine. Quinta's framing a minute ago of it being a weird, strange case I think it's spot on. I also think it's just a weird, hard issue, actually. I think people have a sense here that there is a line where you cross it where the conduct would feel problematic. And then there's also all this benign conduct that I think the justices were referring to repeatedly about educational-style communication, letting platforms know about health consequences or emergency situations, that I think most people would think is clearly on the right side of the line. And the complexity, which you got into in your excellent brief, is how we draw that line. And I take your point about the value of the Bantam test of coercion versus persuasion, but do you think that's a clear line? Does that feel to you like you have a sense just based on those terms of what's on one side and what's on the other?

Alex Abdo

No, absolutely not. It’s a spectrum and that's the challenge with it. These cases are all decided individually on a case-by-case basis, which is the most frustrating kind of legal test that the law has to offer. But it's hard to imagine a better alternative. If administrability is your biggest concern, a clear rule, if that's your primary concern, then you either have one that is prophylactically protective of free speech rights, which I think the plaintiff's legal test would be, or you have one that is prophylactically unprotective of free speech rights. And nobody was really offering one of those. I think the only benefit that I can see to the plaintiff's theory that the government can't even persuade the suppression of constitutionally protected speech is that it is a clear test. Right? That's the advantage of it.

But I think it would be a disaster for democracy if that is the test that the Supreme Court endorses. Because one reason why people elect governments, elect officials, is to represent their views in political discourse. To try to persuade the public to care about particular issues, to galvanize public opinion, to steer public attention to issues that the government thinks are important. But all of that could be seen depending on the context as efforts by the government to persuade private actors to hold or host different views. And on the plaintiff's theory of the case, all of that would be unconstitutional. And that's a hard pill to swallow in a democracy that relies on having a government that can govern effectively because speech is part of governing effectively.

Matt Perault

So I thought they were--I didn't hear their theory as quite so broad. I thought they were pretty clear that real informational-type communication, like, “this is what we're seeing about how COVID misinformation is spreading” would be permissible. I thought they were conceding that. It seemed like on the other end of the spectrum, there was clear acknowledgement of coercion. Like if you said, we're going to prosecute you unless you remove this content, that seemed like clearly coercive. The thing that is unsettling to me about the test is it just seems like it depends on a level of nuance that's almost impossible to imagine being clear, at least from a platform perspective, I assume from a government perspective as well in any exchange. Some of the hypotheticals we're pushing on things like if you said, “This is really a problematic for health, and therefore, we want you to take it down and take it down now. We will be really concerned if you don't take it down now,” that seems to tilt toward some level of threat that might be problematic. It feels like a lot is riding on a couple of additional words that might come at the end of a government communication. And my sense from a platform's perspective, platforms don't need an explicit or implicit threat in a statement to know that there's the possibility that action is going to be taken against them if they don't agree to do what the government does. I don't know. Do you feel like that's an overblown concern?

Alex Abdo

I don't think that's an overblown concern. You need a test that is sensitive enough to recognize threats that are implied. Because if the doctrine prohibits only explicit threats, then it's basically a roadmap for the government to engage in coercion by another name. And I think that that test would not serve First Amendment values well. But as soon as you recognize the constitutionally problematic nature of implied threats, you have to figure out how to detect them, how to determine whether something is an implied threat. And that can be especially challenging in the context of a social media platform where they are repeat players when it comes to interactions with the government. Any one interaction between a platform and the government is part of a broader context. And also the platforms are behemoths. There's not like there's a single person that the government is trying to pressure or persuade, depending on your view, into taking down particular speech. They might be in one instance talking to somebody who feels pretty exposed, who feels as though they don't have the authority to push back in the way that maybe Mark Zuckerberg might have if he were the one in the meeting, and how is the First Amendment meant to deal with those potential asymmetries?

And I think the answer is unsatisfying. I think the answer is you have to take all of those contextual clues into account. But I also think there's the possibility of structural protections that might be helpful. So for example, one thing I find attractive is a proposal made by the Cato Institute, which we pointed to in our brief, which would require the government to be more transparent about its interactions with the platforms. I think that would do a couple of things. I think it would make it harder for the government to pressure the platforms free from any kind of accountability, whether it's political or judicial, because we would have a better record about what's going on. And I think it would make it easier for the platforms to push back publicly. I imagine that one consideration, and maybe I'm curious, Matt, if this was something that occurred to you when you were on the other side of these exchanges, and one thing that is always available to a platform is just to publicize these efforts by the government to pressure it to take down speech. But I imagine the decision about whether to do so is weighted down by the consideration of how that would affect the platform's relationship if they published all these emails. And a transparency law would basically take that decision out of the hands of the platforms. Those exchanges would just become public as a matter of course and the platforms wouldn't have to decide whether publicizing those efforts would itself come at a cost for the relationship with the government, which for understandable reasons they want to preserve.

Matt Perault

Yeah, I think those are exactly the kinds of “resolutions,” quote unquote, that are helpful to think about. You guys, at Knight, organized really just an extraordinary convening, I think it was back in October, about this issue, and a bunch of people wrote pieces that, I think, fleshed out various different ways to think about this. I saw in your brief that you cited several of them. Katie Harbath, who used to be at Facebook, and I wrote a piece where we explored a little bit of the transparency-type options. We were exploring the idea of some option that allowed platforms to report ex parte communications from government to them. So there are procedures in place for the inverse, for government to report ex parte communications that they get from private entities to try to influence government decision-making. And we were exploring the idea of the inverse, of platforms reporting government interactions outside of formal democratic channels. And I think that's a good direction of travel to explore, but the skepticism that you're expressing, I think is the right one. And I talked about it with some platforms after the fact, and I think they would be hesitant, for the reasons that you described, to emphasize the government action because they would be concerned, I think, about what the implications of that would be for their ongoing relationships with the government.

But I do think overall, one of the feelings that I had coming out of yesterday's argument was just that there's, I feel, some dissatisfaction of a judicial process rooted in the First Amendment being the way to resolve entirely this issue. And it seems like there should be there should be a First Amendment bar on certain kinds of conduct and courts should approach that bar with rigor and close analysis of issues like standing and causation and fealty of the factual record and the things that the justices were pushing on yesterday. And when they do that it seems to me like it may be that only a small percentage of the things that we might want to be concerned about here will be addressed. And so transparency maybe gets at some of that, but my guess is it wouldn't fully address it either.

Alex Abdo

I think that's right. And this is why we had the last section of our brief, which actually came up during oral argument. In the last section of our amicus brief, we cautioned the Court against trying to use jawboning doctrine, the First Amendment, to solve all of the problems of social media, many of which are attributable to the concentration of private power over public discourse. And that concentration absolutely exacerbates the problem of jawboning because it makes it easier for the government to have a handful of targets to go after to coerce the suppression of protected speech.

But concentration is a much broader problem than just the First Amendment problem of jawboning. And there are solutions, I think, to concentration that would benefit that would alleviate some of the concerns that you have, Matt, as well as some of the other concerns that people have with social media. And that would make it unnecessary to try to rely on jawboning doctrine as the prophylactic solution for all of these other problems. Because I think it would be problematic to distort or stretch jawboning doctrine to meet all of those challenges, because the effect would be to disable democratically elected governments from trying to steer public attention or have important conversations with speech intermediaries.

And just to make clear why I think that matters and be a little bit more concrete, I suppose. There are a lot of private conversations that take place between the government and speakers or the government and hosts for other speech that I think most people would recognize are socially beneficial and don't really raise any significant First Amendment problem. The example that we gave in our brief was an exchange that took place between the Bush administration and the New York Times. Two New York times reporters had learned of the Bush administration's warrantless wiretapping program and they were going to publish. And they went to the Bush administration for a reaction and the Bush administration spent a lot of time trying to convince the Times that publishing that story would endanger intelligence-gathering by the Bush administration. It would make it harder for them to spy on al-Qaeda terrorists. And the New York Times held on to the story for a year. I think that was too long. I don't think they should have held on to it for that long. But they held on to it for a year because they were convinced and the Bush administration had something to offer in that conversation that the Times couldn't have gotten on its own, which was a perspective on how revealing the warrantless wiretapping program might implicate national security.

Now, the response that the plaintiff's lawyer gave during oral argument to not that specific hypo, but hypos like that, was to basically say, “Well, terrorist speech is not protected or national security can always be enforced. It always satisfies strict scrutiny.” But I don't think that's right. I think in the New York Times example, it was clearly constitutional or protected against criminal law enforcement for the New York Times to publish that story. There was nothing the government could have done to prevent the New York Times from publishing that story. We have the Pentagon Papers case from the ‘70s that said a prior restraint in that context would have been unconstitutional, and there's no broad category of terrorist speech that is unprotected by the First Amendment. And so I don't think some broad national security interest solves the plaintiff's problem in all of those hypotheticals. I think their theory really would have made it very difficult for the government to engage in all these kinds of, again, what most people probably would agree are socially beneficial exchanges between the government and private speech intermediaries.

Quinta Jurecic

Yeah, I will say I was really struck by that exchange over terrorist speech, because actually, pretty closely analogous conversation happened during the NetChoice arguments over these restrictive Texas and Florida laws that restrict platforms ability to moderate content, where I believe it was the Texas Solicitor General made a similar argument at one point. “Oh, well, it would be fine if platforms took down terrorist speech under our law because we have a carve out for illegal content.” And one of the justices, it may have been, I think it was Justice Kagan or Kavanaugh, it was Kagan in this instance, in the Murthy argument who said, “There's plenty of terrorist speech that's protected by the First Amendment.” And I do think that eliding that complication, the fact that the states, Texas in that instance, Louisiana and Missouri in this instance, the fact that they keep sliding past it is really representative of this effort to ignore how complicated this can be.

To your point about the potentially really important role of government interactions, I think one thing that was touched on briefly in this argument by the Solicitor General's office, but deserves more attention, is how this case has really led the federal government to pull back from partnerships with tech companies before the 2024 election in ways that I find really concerning.

And this is even after the injunction had been stayed by the Supreme Court. I did a piece for Lawfare with Eugenia Lostri on how many efforts from CISA, from the FBI have basically just stopped in terms of communications with tech companies. Facebook had a report out in November that said we have heard absolutely nothing from the federal government in terms of foreign election interference communications since July. And the question of First Amendment implications, of coordination in private are complicated as we're talking about here. But I do think, it seems pretty clear to me that existing in this situation where nobody knows what the rules are and the federal government is therefore scared to do anything is really bad, especially heading into an election where it seems, I think, fair to say we're expecting a certain amount of foreign interference and also the potential for real political violence. That seems not great to me.

Alex Abdo

I think that these cases have also cast a pall over independent research that is really important for understanding all of those potential threats to our elections, to our public discourse and independent research, private research done by academic researchers, done by journalists is absolutely constitutionally protected. Even if it results in the suppression of speech on the platforms, it's constitutionally protected. And it also helps us understand our digital environment. But often those researchers rely on partnerships with the federal government, or they rely on funding from the federal government, at least in part, or they rely on the willingness of the platforms to talk to them. And the Murthy litigation, and then a whole bunch of other lawsuits filed against the researchers, have chilled some of that research. They've made it harder for the researchers to do their work, and I'm worried, echoing what you just said, Quinta, that we're heading into this election season with our heads in the sand because we don't have the sensors, S-E-N, not C-E-N, we don't have the researchers to help us understand what's going to take place during these elections and whether Russian interference is going to meaningfully impact political ads or political conversations on the platforms, whether misinformation about the time and place of elections is going to affect people's ability to exercise their right to vote. I really worry about that a lot too.

Matt Perault

Yoel Roth wrote a great piece as part of the symposium about some of the nature of their interactions between Twitter and the FBI specifically, and how benign and helpful they are in most cases, and that's consistent with my experience at Facebook. My sense is a lot of those interactions are low intensity, high value, very informal and collaborative in a way that I think is very productive. So I share your concern.

So going forward, we should have an opinion here at some point before the end of June. So maybe part of the election season, we won't have guidance on this, but then presumably for the summer and fall, we will. Do you anticipate, given the direction that you think the Court's going in, do you anticipate a ruling that is favorable for the Biden administration? How will that alter practice going into the election? Do you think we will be where we are now with government agencies being hesitant to communicate with social media platforms? Or do you think we'd see some of that communication come back?

Alex Abdo

That's a really great question. And I think it might depend on the way the Court rules. If the Court gets rid of the case on some procedural ground without clarifying what the constitutional limits are on government conversations with the platforms, then maybe not. Maybe we won't see the government resume these conversations because they'll continue to be worried about how the Supreme Court would view their constitutionality. And maybe that's an especially good argument for the court to dive into the doctrine and clarify what the test is. And even if that's as simple as reaffirming the line from Bantam Books, which again, was that 1963 case. Even if it's just reaffirming that Bantam Books provides the right line, I think that would be valuable.

But I think it'd be valuable for the Court to go a little bit further than that, just because as Matt, as you've pointed out, Bantam is not the clearest of tests. It has not been applied in the clearest fashion in the lower courts. And there's actually a lot of confusion in the lower courts. I said that most courts have understood that to be the line or courts have understood that to be the line, but there’s a competing line that a handful of other courts have relied on that is different from Bantam. And to my mind, it's really perplexing why they rely on this other line of cases where you have a direct one from the Supreme Court in Bantam Books that sets out what I think is the constitutional line. So I think that clarity would be helpful. And if the Court provides it, then maybe that would be a basis on which the federal government could, probably the SG’s office or the AG’s office could tell the agencies, could give them guidance on how to resume those conversations without running into the risk of violating the First Amendment.

Quinta Jurecic

Yeah. So let's talk about the doctrine a little bit more and Bantam Books in this competing way of viewing these issues. So you wrote in your amicus brief that the relevant First Amendment doctrine is in disarray, which I think is an accurate and depressing way of characterizing just the total confusion that everyone seemed to be grappling with here. So maybe, can you spell it a little, in a little more detail, what exactly the Bantam Books line of cases is here? And then, what this competing argument about state action is in this other case, Blum v. Yaretsky?

Alex Abdo

Yeah. So Bantam was a classic jawboning case. It was a First Amendment case where book distributors sued the Rhode Island Commission on Morality and Youth, arguing that this commission had violated the First Amendment by threatening book distributors for liability for carrying books that the commission said were objectionable. Not because they were necessarily obscene, which is what the law actually forbade them from selling, but objectionable in some way. And the result was what the Supreme Court called this informal system of censorship, in which the commission was able to carry out its censorial aims without all the constitutional protections that would normally apply to the sale of books. And the Court said that what was problematic about this system was that it was coercive. And following that ruling, lower courts drew a distinction between coercion on the one hand, which the court said was unconstitutional, and persuasion, where the government is offering either good faith advice to private actors on what their legal obligations are or just trying to persuade them to share the government's views. So that's the test that most courts have applied.

But there are a handful of courts that apply the test from Blum v. Yaretsky, which is not a First Amendment case. It's a state action case. It's a case about when either the government is responsible for the conduct of private actors, or when you can actually sue the private actors themselves as though they were the government. And the facts of the case are not all that relevant, but just to prove the point that it wasn't a First Amendment case, it had to do with private nursing homes that had discharged patients without what the plaintiffs claimed were the constitutionally required procedures for discharging patients. The plaintiffs basically claimed that the private nursing homes were entangled sufficiently with the government that they that the government could be held liable to the Fifth Amendment's requirement of due process for the private nursing home’s decision. So again, it was not a First Amendment case. It was a state action case.

And the test that the court set out in that case was a really stringent one. It said you can hold the government liable for the conduct of private actors when the government has so significantly coerced or encouraged the private conduct that, and this is a rough quote, I'm not going to get it exactly right, but “so that the choice in law made by the private actor must be deemed to be that of the state.” And it's a very high standard, deliberately so, because a consequence of a finding of state action is that private actors can be held to constitutional standards, which is a significant thing. It's a significant thing to override a private individual's right to behave the way they want and to say that they are actually bound by the Constitution. And it's especially significant in the First Amendment context because one consequence of a finding, for example, that Facebook was a state actor in its exchanges with the government and in its decisions to take down certain speech is that Facebook can be held to First Amendment standards, which implicates Facebook's own First Amendment right to adopt content moderation policies of its own choosing. And it implicates the First Amendment right of Facebook users to have a community that is policed according to those private editorial decisions, rather than the First Amendment standards, which would require Facebook to allow all sorts of speech that Facebook's users might not want to interact with on the platform.

So those are the two competing lines of precedent that the courts have relied on. And I think I've foregrounded why I think the Blum test is not the right one. I think the Bantam test is the right one. So let me just explain why I think there is confusion. It's because both cases use the language of coercion and Bantam says coercion interpreted by lower courts to be coercion versus persuasion. Blum says, such significant coercion that the choice in law must be deemed to be out of the state. I think it's basically coercion plus. It's extremely significant coercion. And I think that matters because the state action test is trying to answer a very different question than the First Amendment test is trying to answer. And there's a risk in applying that test in jawboning cases.

Now, it may be that there is a subset of jawboning cases where the plaintiff could demonstrate not just the lower level of coercion required to make out a First Amendment violation, but the higher level of coercion so that you can hold the government accountable for the private actor's decisions and also even hold the private actor to constitutional standards. It may be there's a subset of those cases that reach that level and maybe even a subset of those cases where it would be constitutionally tolerable for a court to impose a remedy against the private actor that would require them to do something that the First Amendment normally forbids courts from requiring private actors to do, which is to host speech they don't want to host. Those are going to be, I think, extremely rare, that category of cases. If and when a court finds that that's appropriate, it will need to grapple with the First Amendment implications of issuing a remedy against a private actor like that, which I think would be significant. I don't think I can imagine circumstances where it would be appropriate, but they are not the mind run of jawboning cases, which I think again should be resolved under the First Amendment case, Bantam Books. I hope that was not too confusing.

Quinta Jurecic

If it was confusing, I think it's confusing because the doctrine is confusing, right? So the Fifth Circuit seemed persuaded by this argument about Blum and state action. The Court, it seemed to me, was really not so interested in this line of argument, and it seemed relatively clear that they were, to the extent that what's going on in this case is a choice between Bantam Books and Blum, that they were going to go with Bantam Books. Does that seem fair?

Alex Abdo

I think so. I didn't really try to count the votes on that particular question, but it was interesting to me that Brian Fletcher, who was representing the SG's office, I think, maybe I'm getting this wrong, but I think he said that the tests are really doing very similar work in this context. And maybe that's right if you understand Blum just to work very differently in the First Amendment context. But I do worry a little bit because I think what motivated the Fifth Circuit was that it didn't think Bantam was broad enough to encompass unconstitutional efforts by the government not to coerce, but to induce platforms into censoring constitutionally protected speech. And so instead, what the Court did was to basically dilute, to water down, the test from Blum to reach encouragement, which is the word the Fifth Circuit used. And that word was used in Blum. But it was also surrounded by a bunch of other limiting words in Blum. Again, such “substantial encouragement that the choice in law must be deemed to be that of the state,” which is a very demanding test to satisfy.

And I think there's a risk in watering down the test for state action, because normally, again, speech intermediaries like Facebook should not be subject to First Amendment limitations in the way that they moderate speech on their platform. To the contrary, they're protected by the First Amendment in their content moderation decision. So I think there's a danger in conceptually thinking of state action as being the right test in all but a very, very narrow set of cases that is probably vanishingly small.

Matt Perault

So I'm curious about your take on the politics here. It seems like this has a particular valence because the issue was the Biden administration putting pressure on tech companies related to information that I think many people on the left think tech companies should do a better job of moderating. So I was struck by some of the headlines yesterday, like in anticipation of the argument, NPR's headline was “Supreme Court examines whether government can combat disinformation online.” And when I was listening to the PBS live stream of it, the PBS heading for it was “Supreme Court hears case on collaboration between government and social media companies.” And I think the framing, if the issue were coming from, for example, the Trump administration on content moderation approaches that are more supported by the right and disfavored by the left, those headlines would be very different. And one question I had here is whether people who might have concern about the kinds of pressures that a Trump administration would impose on social media companies would actually want a strong rule in this case? Not to constrain the Biden administration, but to constrain the potential Trump administration or future administrations who might have a broad view of how aggressive they can be with respect to social media companies. I'm curious how you see that playing out. Let's say that Trump is elected in November. Is this issue going to take on a different resonance with some of the communities that have been supportive of one point of view to date?

Alex Abdo

It’s a great question. I think it might, which is why I think we really need a rule that everybody would be satisfied with, no matter which political party happens to be in power. And even in this case, some of the plaintiff's allegations date back to the Trump administration. They're about efforts by the Trump administration to police disinformation about COVID-19. I think the 2020 election stuff, I think for the most part took place after the end of the Trump administration. But you want a rule that satisfies everybody. And that means it needs to apply in a way that draws a line that is defensible no matter the politics. And I will say there's a lot of concern in the free speech community that given the partisan valence of these cases, and not just these cases but the NetChoice cases and also the Gonzalez and Taamneh cases last term, that given the partisan valence of these cases, they were worried about rulings from the Supreme Court that reflected that tenor, and for that reason, parted way with the First Amendment principles, their free speech principles.

And I have been encouraged by the way, at least by the oral arguments in the NetChoice cases, and in Murthy and Vullo the NRA case, and also by the decision in the Gonzalez case, where the Court basically punted on the Section 230 question. I’ve been encouraged by the care with which the Supreme Court seems to be going about the business of figuring out what the right rules are to determine whether the government has crossed the First Amendment line. They seem really attentive to the interests on both on both sides of the First Amendment equation.

Matt Perault

So I would love to think there's some chance here of a satisfying rule, but I thought, maybe I'm hearing it wrong, but I thought the theme of this conversation was basically, the Supreme Court might reach the right outcome, and it will probably be dissatisfying in some ways to everyone, as opposed to the inverse of they'll find their way to a rule that will like govern every case effectively. Do you think it's possible that there's going to be a rule that would like work in a range of circumstances, not just governing the Biden administration, but governing the Trump administration as well? Because my feeling coming out of this in hearing the argument and the necessary and important discipline of the justices was that the law is designed to resolve cases narrowly. And that's a good thing. We should want that, we should invite it.

Alex Abdo

Yeah.

Matt Perault

This is an issue that's broader than any one particular case and there might be a lot of things that we want the government not to do, independent of whether there's actually a First Amendment violation that could be held up in court.

Alex Abdo

Well, I think there are really two questions bound up there. I think what we want is a rule that is apolitical. We want a rule that either equally satisfies or equally dissatisfies everybody. But then there's a separate question of can we actually fashion a rule that protects the First Amendment values we want to protect? And this goes back to where I think we started, which is that it's very hard to come up--even accepting that the line between persuasion and coercion is the right one, which is what I think, even accepting that, it's very hard to draw that line. It's going to be case-by-case. And I just don't think there's a reasonable alternative on the table to case-by-case adjudication of these kinds of claims.

And I would love it if we had a clearer rule. There are a lot of contexts in the law where I would love it if we had a clearer rule because it would be more predictable for everybody involved. I just worry that the only way to clarify this line is, again, either to be prophylactically protective of the speech rights of platforms and users at the expense of the ability of governments to govern. Which some people frame that as being about government speech, which I think is the wrong way to frame it because I don't really think the government has a speech right. I think the public has an interest in having a government that can be effective, which is something that I think actually Brian Fletcher did a really nice job of explaining this in the argument yesterday, that that is a structural necessity that emerges from the Constitution. You need a government that can govern that involves speaking and it involves taking advantage of the bully pulpit to try to convince people. Or you have a rule again, that is prophylactically unprotective of speech rights, which I think would be problematic for the other reason, which is that it would give the government way too much power to shape public discourse, to distort it to political or partisan ends and allow the government to set orthodoxy in our political discourse. And I think it's worth giving up the administrability that would come with either one of those prophylactic rules for the more intellectually coherent, but fuzzier rule that is somewhere in the middle. And I share your concerns, Matt, that it's a fuzzy rule. I just don't know what the alternative is.

Quinta Jurecic

So we've mentioned the NetChoice cases a few times, and I did want to circle back to them at the close here and ask, now that we've heard the Supreme Court have oral arguments in both the NetChoice cases and Murthy, if there's any connections you see between this group of First Amendment internet law-adjacent litigation that's been before the court. Is there anything that we can say about how the justices seem to be thinking about these issues or anything that's revealed by looking at these cases alongside one another?

Alex Abdo

Yeah, I think thematically, to my mind, the most important connection is that both cases are about the extent to which the government can regulate social media. In NetChoice, it's about direct regulation and in the jawboning cases, it's about indirect regulation through either threats or pressure or cajoling, what have you, browbeating, whatever you want to call it.

There's another connection that some people pointed to after the argument yesterday, which is that some of the questions yesterday seem to take for granted that the platforms had First Amendment rights, which is, of course, one of the very questions at issue in the NetChoice case, because Texas and Florida, in those cases, argue that they can regulate the content moderation decisions of the platforms or require them to carry speech they don't want to carry or, to take another framing, forbid them from discriminating against the speech of their users because the platforms themselves don't have any free speech rights because they are just either common carriers that can be regulated or neutral conduits for the speech of others. And I'm a little bit less convinced that the Court in ruling in the Murthy case has to accept that premise. But I think it's a good sign if they do, because I think the platforms do have free speech rights and their free speech rights are intimately connected with the free speech rights of their users, who I think benefit from having platforms that can construct communities that are responsive to the needs and desires and wants of the platform's users.

Quinta Jurecic

Let's leave it there. Alex, thank you so much for coming on.

Alex Abdo

Thanks for having me.

Quinta Jurecic

You've been listening to Arbiters of Truth, a Lawfare Podcast series on the information ecosystem. You can find past episodes in the Lawfare Podcast feed.

The Lawfare Podcast is produced in cooperation with the Brookings Institution. You can get ad-free versions of this and other Lawfare podcasts by becoming a Lawfare material supporter through our website, lawfaremedia.org/support. You'll also get access to special events and other content available only to our supporters.

The podcast is edited by Jen Patja and your audio engineer this episode was Noam Osband of Goat Rodeo. Our music is performed by Sophia Yan. As always, thanks for listening.