Evelyn sits down with Genevieve Lakier, a Professor at University of Chicago Law School, to discuss the 5th Circuit Court of Appeals ruling in Missouri v. Biden, narrowing but affirming a district court injunction prohibiting large parts of the federal government from communicating with platforms about content moderation.
Genevieve Lakier:
I'll just say, if any moderated content listener has a spare weekend and wants to provide an annotated analysis of the factual record in the Fifth Circuit opinion where they could just go and look at the record and help us understand where this is coming from and what are the countervailing facts that the court's ignoring, that would be helpful.
Evelyn Douek:
Hello and welcome to Moderated Content, podcast content about content moderation moderated by me, Evelyn Douek. Just after promising you, our listeners, that we were back for regularly scheduled programming, we are interrupting your regularly scheduled programming because in a Friday night news dump, the Fifth Circuit Court of Appeals weighed in in Missouri v. Biden, the case that's been playing out in the Fifth Circuit about government platform cooperation, communication, and what is colloquially known as jawboning.
This is a decision that we have talked about, the district court decision we talked about at length with one Genevieve Lakier, professor of law and the Herbert and Marjorie Fried Teaching Scholar at the University of Chicago on this podcast before. She's kindly agreed to join me again to discuss the Fifth Circuit Court of Appeals decision. Welcome, Genevieve, and thank you.
Genevieve Lakier:
Thanks for having me.
Evelyn Douek:
Sure. I want to start with previously on Moderated Content to recap where we left off, which is we spoke last time shortly after Judge Doughty's July 4th ruling in this case. That was a 150-page decision and I think we spent some hour plus talking about it on this podcast. I'm not going to be able to go into all of it here, but I think the general broad brush of where we left off is that the judge set out this extensive factual record of communications between the platforms and many different kinds of government officials, including the White House, the CDC, the FBI, lots of different kinds of communications.
The content moderation that those government officials were encouraging platforms to do around a wide variety of topics, but in particular things like the pandemic, COVID misinformation, and Hunter Biden's laptop, and concluded that the platforms had violated the First Amendment in being unduly coercive or significantly encouraging them to take down legal speech and issued an injunction, an extremely broad injunction, that basically enjoined huge swathes of the federal government from having any and all contact with platforms about their content moderation practices.
Here we are a couple of months later now and the Fifth Circuit has issued its decision on appeal, which is very different in tone. The Judge Doughty's decision for those who remember was very alarmist about the Orwellian nature of what has been going on in this country, whereas the Fifth Circuit Court of Appeals adopts a much more measured approach and it does narrow the injunction. But the question that we're going to be talking about is how different is it in substance. Let's start at the top. Tell us what did the Fifth Circuit Court of Appeals say?
Genevieve Lakier:
Okay, well, I'll just say that the Fifth Circuit Court of Appeals did what I predicted on this podcast it would do, which is essentially affirm the district court opinion, but narrow it somewhat and narrow the injunction. I'm validated in my prediction.
Evelyn Douek:
One of the best things about having a podcast is the ability to be able to say, "I told you so." And when you get things wrong, you just don't say anything at all and no one ever calls you out on it. It's great.
Genevieve Lakier:
It's fantastic, but I will say that it narrowed it less significantly than I anticipated. One of the striking features about the district court opinion was that it takes this complaint, which is against all different kinds of government actors. On the one hand, there's the White House aids who are sending angry emails to officials at Facebook and Twitter demanding action on particular posts or strongly urging them to take action on particular posts, to the CDC, which is providing information about vaccines and what constitutes good information about the quickly changing medical information about the pandemic, to the FBI, to the Stanford Internet Observatory.
I mean, it's just a motley array of defendants who are named in the district court decision and finds that there's unconstitutional coercion, there's a First Amendment violation with respect to all of them, and writes with this very broad brush. I was hoping, predicting, anticipating that when the Fifth Circuit Court of Appeals took the case, they would reach that conclusion only with respect to some of the defendants, not others, and that is true. That happened.
The private groups that were named in the complaint and then the district court enjoined, like the Stanford Internet Observatory and some of the state department defendants, they're excluded. The circuit says there's not enough evidence that there's an unconstitutional violation, but it finds that there is First Amendment violation with respect to the White House officials, with respect to the CDC, with respect to the Surgeon General, and with respect to the FBI. It narrows the opinion, but only slightly.
Evelyn Douek:
One of the things that I have been surprised by in the few days of commentary in the aftermath of this decision is the level of praise or the way that it's been positively received by a number of people, a number of commentators. To me, that speaks to a bit of an anchoring effect. You've got a good cop/bad cop dynamic playing out here in the Fifth Circuit, where the district court comes in, issues this wild opinion that just smooshes a whole bunch of things together and enjoins half the federal government, why not, from doing basically anything and ever speaking.
And then the Court of Appeals comes in and gives something of a little bit more measured of approach and says, "Hold on here. Maybe some of these things aren't like the others." And suddenly people are like, "Oh wow. Thank goodness the adults have entered the room and this is a much, much better decision." But I still think if we had started here, I think this decision still leaves a lot to be desired.
It's absolutely true that the Court of Appeals is painting with a less broad brush. But in terms of the way that it handles the facts and then the handles the law, and I think we should step through each of these in turn, there's still a lot of, let's say, innovation going on here. But before we get really specific, I'm just curious for your high level thoughts on how you read this decision.
Genevieve Lakier:
I guess I agree that there's a lot to be desired. My general takeaway about the decision is that it is slippery. It's slippery both on the facts and on the law. I would really prefer opinions on an incredibly important and contested area of law not be so slippery, not be so hard to pass, and not do what you suggested podcast host can do, which is just not mention things you don't want to have to mention. I would prefer that our Courts of Appeal, they take account of all the contrary facts and they help us work through them.
And that's not at all what the Fifth Circuit did here. I agree, a lot to be desired. I guess I have a more sympathetic read of why there has been some amount of praise for the decision, which is up until now or up until the district court opinion in this case, the jawboning case law in the First Amendment had been so... Except for several important, but I think pretty unusual counter examples like the back page case that we talked about last time.
For the most part, the courts have been really hostile to these kinds of claims, and the result has been that the First Amendment has been interpreted so rigidly and so narrowly that even when there is a lot of reason to think that there was government coercion, that platforms took action as a result of government pressure or wanting to make the government happy. Courts have said, "Oh no. No First Amendment problem." And then the district court opinion, in this case, reaches the opposite conclusion, but as you say, really extraordinarily broad crazy way.
I don't think it's so surprising that here we have a seemingly more nuanced, more thoughtful approach to the jawboning analysis in First Amendment cases that doesn't get us to such a rigidly narrow result. For First Amendment libertarians, speech protective people, this feels like, I can imagine, it's such a better path to go down because at least it's opening up the possibility that the First Amendment is available. I'm sympathetic with that. I also think that the case law up until now has been too narrow, but I agree with you that this opinion leaves a lot to be desired.
Evelyn Douek:
Okay, so let's get more specific then. Let's start with the factual record and talk about that. One of the big and really important critiques of district court opinion, I think, was it's extremely poor handling of the facts. For example, I could point listeners to the amicus brief submitted by Stanford on behalf of my colleagues here at Stanford and the Stanford Internet Observatory, which we have mentioned, outlining all of this plain factual errors that the district court made, including just making up quotes or quoting so selectively as to just completely distort the meaning of what certain researchers said and things like that.
Of course, this is the amicus brief submitted by Stanford, and so is focusing on the issues with Stanford, but it makes you nervous about what else is in the opinion. There've been many others who have also pointed out other similar instances of where the court was just saying things about what had happened, about the level of pressure or what the government was achieving in terms of its communication from platforms and omitting things. Very often the platforms pushed back and didn't take down this content.
That was a really important critique of the district court opinion, I think, and it makes it really difficult, I think, for some of us who think transparency is super important when we're talking about jawboning. One of the problems in this area is we don't know a lot of the time what kind of communication is occurring between government actors and social media platforms. You might have thought that getting judicial discovery of the kinds of communication that's taking place could be a saluatory development in terms of thinking about this more sophisticated and more accurate.
But the district court opinion really gives you pause as to whether judicial fact finding is going to be a good answer to that problem in the current state of the judiciary and the current state of the polarized, politicized nature of these issues, where it was pretty clear that the district court was trying to paint a certain picture of what was going on to reach certain conclusions. Let's turn then to the Court of Appeals. How does it handle the facts in this case? Do you think that there's progress?
Genevieve Lakier:
No, maybe it's even worse.
Evelyn Douek:
Big call. Wow, okay.
Genevieve Lakier:
I mean, I think so. The district court errs in all these ways makes stuff up, attributes quotes to people that they didn't actually say, so this seems like a problem.
Evelyn Douek:
Yeah, okay, but that's the good one.
Genevieve Lakier:
But also has citations to the record, speaks specifically names, who is the author of the alleged unconstitutional threat and who is the recipient. If anything, the problem with the district court is that it's so detailed that it makes it really hard to synthesize for the reader and the analyst. It's hard to deal with this overwhelming, massive information. What the Court of Appeals does to respond to that problem is it just simplifies everything by getting rid of any citations to the record.
It'll be quoting people. I have to say, I teach law and citation to authority is really important when you're writing legal briefs and papers, and I make my students always cite for factual support. They can't just tell me things happen in the world, they have to show me that things happen in the world. This is the Fifth Circuit Court of Appeals and it is quoting people all the time without any citation to any record and not naming their name or who they are talking to. It was an incredibly disheartening reading experience because I had no way of checking them.
At least we know that the district court made stuff up and messed the facts around. But the Fifth Circuit opinion makes it really hard to understand that. This is one of the ways in which it's slippery. It paints a story, it tells us a story that seems as if it's relatively simple, it's really relatively clear, but that's just because it's gotten rid of all the pesky actually proving things and basing it in the dense factual record that was developed through discovery. It leaves us in the situation where it's just really hard in some cases to figure out what it's talking about. That's not good.
Evelyn Douek:
Yes. And in particular, it just talks about the platforms as a monolith and says, "The platforms did this, the platforms did that, or this official said this to the platforms," and it doesn't break out how different platforms acted differently and responded differently and had completely different relationships with many of these officials. Again, I think slippery is a very, very good word for it and misrepresenting. In some ways, it is still blatantly misrepresenting.
There's this quote in here that's saying, "In response to this pressure, the platforms responded with total compliance," which is just an incorrect quote. We can talk about whether these relationships were appropriate, and I think you and I are on the record and say that in many cases, in some ways they weren't, but the idea that they acted with total compliance with the government's wishes is just not established in the record at all.
Genevieve Lakier:
Agreed. What's so interesting, this gets us more to the law, and I want to stick with the fact for one more second, but what's so interesting, and this is another way in which the opinion is slippery, is that at the end of the analysis, the court says, "In fact, for pressure coercion or encouragement to violate the First Amendment, it doesn't have to succeed." The court didn't have to say total compliance. It could just have given us a clear analysis of what kind of compliance, what degree of compliance is necessary to establish a unconstitutional violation.
But it just wanted to tell us a simple story. There was unrelenting pressure and then there was total compliance, even though in both those cases, the story is much more complicated. I think that is incredibly frustrating, not only because it misrepresents the facts, but it doesn't help us get to clarity about what the law should be because it might be the case. I mean, I think it is the case that every instance of unconstitutional pressure doesn't have to succeed. The government can try and do something unconstitutional.
The platform may resist it for whatever reason. It doesn't mean that we don't have a First Amendment problem, but courts need to acknowledge the fact that in many of these cases... The reason why there is this unrelenting and ramping up pressure campaign as it describes is because the platforms are resisting and not doing what the government wants. And that the story told here doesn't take account of that and doesn't integrate that into the legal analysis is very frustrating.
I'll just say, if any moderated content listener has a spare weekend and wants to provide an annotated analysis of the factual record in the Fifth Circuit opinion where they could just go and look at the record and help us understand where this is coming from and what are the countervailing facts that the court's ignoring, that would be helpful because there's so much that is omitted and that's hard to figure out what it's talking about. And then the countervailing facts that it's just not even discussing, it makes it hard to analyze.
Evelyn Douek:
Right. Again, I think here is where an anchoring effect is present. Because I mean, it didn't succumb to the rhetoric of the district court and it didn't invoke Orwell, which the district court did, but it couldn't help itself that the court does say at some point, "The Supreme Court has rarely been faced with a coordinated campaign of this magnitude orchestrated by federal officials that jeopardized a fundamental aspect of American life." There is clearly a point of view here about what's going on. I think though the last point about the...
Or not the last point, just on the facts though, I mean, one of the things that makes this really difficult, I think, is how to deal with something that is so big and complex, how to deal with facts that are like this. One of the points that the court is trying to do here, which I have some sympathy with, is it's saying, look, you can't take isolated statements in isolation out of their context and say, "This one wasn't sufficiently threatening. This one wasn't sufficiently threatening," because you need to understand the whole relationship between government officials and platforms and understand them in the course of that relationship.
I think that's right, that's obviously true that you can't just say, "Well, this email wasn't threatening. This email wasn't threatening," and then the fact that you have, I don't know, thousands of emails becomes irrelevant because you look at each one in isolation. At the same time, it then becomes like if you go too far in the opposite direction, it feels like you're kicking up dust. You're just pointing to facts all over the place and not being really specific about what exactly was the constitutional violation that occurred here.
I don't really know how to think about that. I mean, beyond the fact that clearly this lack of citation and the lack of specificity is problematic especially in terms of checking, it does create this problem where I think if we try and get too specific about what exactly is the problem, you are missing the forest for the trees.
Genevieve Lakier:
Oh, I agree. There is this tendency in First Amendment cases, not this one, but in others that that's precisely what courts do. They focus on an individual speech act and assess its constitutionality without thinking about the pattern of behavior or the larger context. I guess here is where I'll say something positive about this opinion, which is that it is trying to think about how repetition or longstanding campaigns can change the character of communication. If you say one time, "Hey, can you take that down," it's just a different situation altogether than if you say it 12 times.
Both the district court and the Fifth Circuit are really cognizant of this fact, and so trying to pay attention to, I guess, the social meaning, how it affects the platforms when government officials continue to ask and demand. There's lots of people making demands on them and they're fending off this fire hose of pressure. I like that. But the way to deal with that is not to say, "Oh, it's just too difficult then when we've got so many individual defendants and so many instances of alleged pressure to make sense of it in any reasonable length of page numbers, length of opinion, and so we're just going to simplify everything down."
I don't think that's the response. I was wondering, I mean, I don't know, but I was wondering whether this case suggests that there should be limits on the variety of defendants you bring into any one suit. It seems to me that the claims against the White House aids are pretty separate from the claims against the CDC are pretty separate from the claims against the FBI. At the very least, there should be an extensive discussion of what happened in those cases, the pushback and the analysis, the constitutional analysis that goes with that.
But if that's all in one opinion, it's going to be, it's true, an incredibly long opinion. I think that this is just too big a case and that there's nothing to be done about it, but break it up. But I'm not sure.
Evelyn Douek:
Yeah, that's a really helpful point. It is kind of like 10 cases combined into one. How to deal with that? One last thing on the facts before we move on to the law, the question about what exactly were the threats or what was the level of coercion or the encouragement that was happening here?
One of the things that I think is notable is even in this decision, there is no single point where the court points to a statement that says, "You need to take down this piece of content or we will do this," such that the relationship's really clear. It's much more general than that. What exactly is the unproblematic, sorry, the unconstitutional, the problematic threat in this case?
Genevieve Lakier:
I mean, again, so many defendants, so many allegations, so it's threats in the plural. Again, this is an area where I think the opinion has something going for it. In the past, the cases have suggested that in the absence of some kind of explicit or strongly suggested implicit threat, it's going to be really hard to find a constitutional violation. The Fifth Circuit here says that's bananas. It's often the case that when a government act or speaks this kind of implicit threat, so a lot of the threats are implicit, that if you don't take action, we are going to do something bad to you.
Something bad is implicit, but we look at the position of authority that the speaker has over the defendant, over the platform, sorry, and read that as itself enough to establish a threat. I don't think that that's crazy. When a platform is talking to an important government official, it wants to carry favor such that the threat of the withdrawal of that favor is a meaningful threat. The Fifth Circuit takes it pretty far though.
It suggests at times in the opinion, I think, that any strongly worded statement from a law enforcement officer about the badness of a particular piece of content or policy is an implicit threat. That's incredibly broad. That's a real departure from how courts have read it before. Here is also an area where I think the court's doing interesting stuff, but it's being so slippery about it, it's not entirely clear to me how it's limiting this category of implicit threats, how we know that there's an implicit threat, what's the kind of implicit threats.
And then in other areas of the opinion, it talks about the pretty standard kind of threats that are often invoked in jawboning cases like the threat of regulatory reform. And here too, there is a lot of questions about whether that's a kind of unconstitutional threat. If what you're saying to a platform is if you don't... For example, this threat of Section 230 reform, we're going to remove Section 230 immunity for particular kinds of misinformation if you don't change your ways.
Is that an unconstitutional threat or is that a government official saying, "If you show that you are causing harm in a way that we didn't anticipate when we wrote this law, we're going to change the law to take account of that harm?" I'm not sure that that's the kind of threat that we should recognize as unconstitutional in First Amendment jawboning cases, but the court is not worried. It's lumping in these implicit threats, these kinds of vague warnings of regulatory change, and it's treating them all as threatening.
Evelyn Douek:
Let's turn to the doctrine then. I strongly believe that there's no one in the world that knows First Amendment jawboning cases as well as you, so you're very well-placed to walk us through this. Really reading this opinion, the court really seems to be suggesting that the law is pretty clear and settled and here's what the law says, and we're just following the law, we're just doing what we're told, ma'am. There's nothing funny happening here. No innovation. It's just how it's got to be.
And then it sets out these tests as if they are handed down from on high and analyzes these facts that we've been talking about under two separate tests of whether there's been coercion or significant encouragement by the government and comes to its conclusion that in many cases there has. First of all, is that how you read it too? And then second, is that accurate? Is what the court is doing here doctrinally just... I guess on a scale of one to 10, how much innovation is going on here in terms of what the court is doing doctrinally?
Genevieve Lakier:
Oh, there's a lot of innovation, I think, which also maybe why people have been somewhat positive about it. I'm not sure. But it's interesting. This is a super interesting opinion, but it is also, I agree, annoying/interesting that the court is working so hard to portray what it is doing as, this is another way in which it's slippery, to portray what it is doing as simply restating the settled law.
It's interesting in part because you could imagine a Fifth Circuit opinion that plays it completely differently that says, and I think this is more accurate, "We are going to strongly disagree with how the Second and Ninth Circuit treat jawboning cases and do this other thing."
But instead it says, "No, no, no. We're completely in line with what the Second Circuit and the Ninth Circuit are doing. We're even going to borrow their tests for how you establish coercion. But in applying those tests to the facts before us, we're going to do it in a very different way. Rather than reading a test extremely restrictively as the Second Circuit did in this case called Vullo and as the Ninth Circuit did in this case called O'Handley, we're going to read them very expansively."
In some ways, they're applying the same formal rules, but I think they're showing how underspecified these rules are, how much the meaning of the rule depends upon the application. There are other ways in which they innovative, but I think that's the big way.
Evelyn Douek:
I want to come back to the circuit split question, but talk to me more about what's going on here in terms of the doctrine and the innovation in terms of how the court's thinking about significant encouragement and jawboning more generally.
Genevieve Lakier:
To keep things relatively simple, it's been established for a while that government actors can violate the First Amendment if they coerce private actors into taking down speech. And that's been clear since this case, Bantam Books v. Sullivan, which is one of the most important of the jawboning cases and the only clear jawboning case the Supreme Court has ever decided, and it's clearly a coercion case.
There's a later case that is not a First Amendment case and not a jawboning case in which the court also talks about significant encouragement as a way in which the pressure on private actors to do things can violate the Constitution, including the First Amendment. And that makes sense. There's a long history of state action case recognizing that certain kinds of encouragement can violate the constitutional rights guarantees, although most of those are not First Amendment cases.
I think that's all conceptually coherent. I think encouragement is just the flip side of coercion. Coercion is saying, if you don't do this or if you do do this, we're going to do something bad to you, and encouragement is saying, if you don't do this or do this, we're going to do something good to you, or encouragement could also be, we're not going to do something bad to you. I don't actually think they're conceptually distinct. I think they should be read as part of the... Is a government official putting so much pressure on a private actor that the Constitution is offended?
But the court here is distinguishing the two. It splits them apart, and it says there's a separate test for encouragement and there's a separate test for coercion. The test for encouragement is different than the test for joint entanglement. This is all new. It's playing with the cases and it's putting them in a slightly different configuration. The result is to expand the range of actions that might be considered unconstitutional because there are some sets of actors, the White House defendants and the FBI, I think, where it says both the coercion and the encouragement test are satisfied.
The fact that it splits them apart really does mean anything. But the CDC, and I think this is a really interesting part of the opinion, the court says, "Okay, there's no coercion. There's no evidence that there was any implicit threats or explicit threats when the CDC was telling the platforms about their latest view of what was correct information about the COVID pandemic and about what vaccine misinformation looked like and what causes public health harms. This is all expert information about public health. Not always with hindsight correct, but based on the latest scientific knowledge. That's not coercive."
But the court says it's still significant encouragement because the platforms relied upon that to make decisions. That is an incredibly broad view of what significant encouragement is under the First Amendment, and it's a significant innovation.
Evelyn Douek:
Yeah, and I completely agree. It is fascinating how much the effort that the court goes to to suggest that there's no circuit split in these cases. To our non-lawyer listeners or non-American lawyer listeners, a circuit split is important because that's when the Supreme Court is most likely to grant certain here a case, and we'll talk more about that in a minute, but this court, instead of saying, "We completely depart from what the Ninth Circuit does," it's going to a lot of effort to suggest that it's a meeting of the minds.
In particular, it draws on some Ninth Circuit cases. You mentioned O'Handley. There's also the Warren case, and it says we're just adopting those tests, but it comes to the completely opposite conclusion that the Ninth Circuit came to in those tests. O'Handley was a case about Twitter's partner support portal where the state actors were flagging certain tweets to Twitter as election misinformation and Twitter would sometimes take them down, and Warren was a case about Senator Elizabeth Warren sending a letter to Amazon about certain misinformation on its platform.
In both cases, the Ninth Circuit found that there was not a violation of the First Amendment, and the court, in this case, the Fifth Circuit seems to suggest that it doesn't quibble with those answers. The facts here are just so different, that the facts here are just a whole nother level, I find that hard to believe. I find it hard to believe based on the rhetoric and the way that the court is framing it in this case. That if O'Handley and Twitter support portal came before it, it would find that that was not a violation of the First Amendment.
I mean, first of all, do you agree with that? And then second, why is the court doing that? Why is the court not saying, "Look, the Ninth Circuit is just wrong. Those crazy people over on the West coast, they're just completely out on a limb. That's not the correct analysis," instead they're trying to say, "No, no, no. There's nothing really to see here?"
Genevieve Lakier:
I mean, I agree. I think the O'Handley case, the facts that the opinion cites... I don't know if this is actually correct, but again, we're in a world where district courts and federal courts can cite facts and maybe they're not right, but let's just assume that the facts are actually correct. I think in the O'Handley case, the opinion says that 98% of the tweets that were flagged by the government defendants were taken down.
I think given those facts, it's hard to see how the Fifth Circuit given this analysis where it says if there's strong evidence that platforms are being affected by what government actors are doing, then that's enough to violate the First Amendment. It seems like it would come out the same way. I'm hesitating a little with the Warren case though, because the thing that's so interesting about the Warren case is that the Ninth Circuit in that case said it really matters that Warren...
This was the case where Elizabeth Warren sends a letter to Amazon about books that it's promoting on the site that are advancing COVID misinformation, according to Warren's view, I think, vaccine disinfo. The language is pretty strong and Warren says, and it's potentially unlawful behavior. It looks like it's somewhere between an implicit and explicit threat. The Ninth Circuit says, "Nah, this is just a onetime event. There's no sustained campaign."
Evelyn Douek:
Elizabeth Warren doesn't have a sustained campaign against Amazon whatsoever.
Genevieve Lakier:
Exactly. If you just look at the opinion in isolation, the formal rule seems pretty consistent, but there is just no way that the Fifth Circuit would look at Warren in isolation. Here you see how much digital storytelling matters. We would have a 10-year recitation of Elizabeth Warren's sustained unprecedented attack on Amazon and the result would be the opposite. The formal rule is the same, but the application is really, really different. I agree. Also, Second Circuit, when thinking about the FBI defendants, the Fifth Circuit here says there is this implicit threat because you're a law enforcement officer.
It's the FBI speaking to the platforms. There is this implicit threat that they're going to do bad things, even though we have to acknowledge the language is polite and there's no demanding. The language that the FBI uses looks pretty different than the White House official's language. I think there's meaningful distinction there, but the court doesn't really care because it says, "But they're FBI officials. I mean, isn't that enough?" There is direct explicit language in both a DC circuit case called I think it's Miss V Penthouse, and then a Second Circuit case, Zieper v. Metzinger, which the Fifth Circuit cites to the contrary.
Here too, it cites Zieper, it says, "Oh yeah, we're completely consistent here," but reaches the opposite conclusion. Why it is trying so hard to obscure the fact that it's really, really disagreeing with a whole bunch of other circuits might be to shore up what it's doing to say, "We're just applying just the facts, ma'am. We're not doing anything so different." But I don't know. I found it also like you really interesting, noticeable and slippery.
Evelyn Douek:
And slippery. Part of it surely has got to be that it knew that this is a decision that's going to be reaching. It's going to have a broader audience than just First Amendment experts. This is going to be widely read and widely reported on as it has been. I think knowing that potentially the decision was to try and present this as uncontroversial a decision as possible rather than to be inflammatory, because surely the Supreme Court isn't... I mean, I don't know how to say it. Surely the Supreme Court isn't going to be fooled by this presentation, but I guess we'll have to wait and see.
Genevieve Lakier:
There's another possibility about why it's doing it this way, which is that if you're disagreeing with these other circuits, then you have to have a damn good reason to do so. You have to have some kind of policy justification or theoretical justification based in First Amendment's principles and values. I actually would appreciate that. I think ultimately the constitutional law of jawboning does rest on a whole set of policy considerations.
Because on the one hand, we want platforms and government actors to be able to communicate when those government actors have really important information to convey. The Bantam Books case that I talked about earlier, the one Supreme Court jawboning case we have so far on the books, makes that clear. It says there are some kinds of informal context between government and private actors that are totally permissible and we want to allow, but there are others that we don't.
There's always going to be line drawing. That line drawing hopefully is going to be based on some idea of how pressure actually works in the real world and also the cost to the broader system. We want to have a diverse and vigorous public sphere, but we don't want to give the government actors too much power. But on the other hand, we don't want to chill too much useful collaboration. There are all these really difficult interests to take account of, and so you should have a reason why you're adopting the rule you're adopting.
If the Fifth Circuit were to say, "Well, okay, we disagree. We want a much broader First Amendment rule," I think that that is defensible and justifiable, but you have to justify it. Whereas if you say, "Oh yeah, we're just applying settle law," then you don't. There is really, in this opinion, no justification for the conclusions it reaches. It's just doctrinal application of the rules in quotation marks because the court's doing a lot, but it's not explaining why and what it's doing exactly.
Evelyn Douek:
Yeah, and I just want to second that and expand it to what we were saying earlier about the facts as well. For the people that agree with this outcome and think it's a good decision substantively, that might be true and it might be that this is a defensible position substantively, but the way that you get there matters. What I would've liked to see is the court facing head on the facts that are inconvenient for the conclusions that it arrives to. For example, it notes at some point that the platforms only took down 50% of the things that the FBI flagged to them.
It notes that in passing and never returns to that. It's not necessarily the fact that, oh, if it's not 100%, it's not coercive, and so 50% means there's no problem here. It could entirely be the fact that read in context, that 50% is still very problematic. But it would've been nice to see the court wrestle with that head on rather than just willful blindness to these inconvenient facts. I think people should be demanding more of that from the decisions in these cases.
Genevieve Lakier:
Agreed 100%. I will just say for those who haven't read the opinion, by the time I got to the one line in which it says it only took down 50% of what the FBI flagged, I was so pleased to see any acknowledgement whatsoever that it was only only 50%. Because up until this point, there'd be no acknowledgement of anything other than total compliance by the platforms. I thought, okay, there's this one fact that they're going to acknowledge and then deal with and then it disappears
Evelyn Douek:
Never returned to. Never mentioned ever again.
Genevieve Lakier:
Nope. I think that one fact about the opinion, for those who haven't read it, gives you the best flavor of what it's like reading this opinion.
Evelyn Douek:
Right. Let's talk about the injunction then. Of course, this is what a lot of people are reacting to when they are giving their headline takes in reaction to the opinion because it's the thing you jump to first to see what is the bottom line of this case and what happened. This is where definitely the Court of Appeals is adopting a much more narrow approach. It's much closer to a scalpel than a sledgehammer. Talk us through what the Court of Appeals did here.
Genevieve Lakier:
Well, it significantly narrows, I think, what the district court did. I'm confused. I just say, at the end of a slippery opinion, I find the injunction quite slippery as well. But just describing I guess what happened, the district court had this complicated injunction with lots of different parts and it had very specific prohibitions. For example, it said you may not notify social media companies to be on the lookout for postings containing protective free speech.
This is a reference to these Be On The Lookout meetings that the CDC had held with the platforms, or you may not collaborate with the Stanford Internet Observatory, or you may not follow up with social media companies to determine whether they've done what you've said. These are very specific. The Fifth Circuit gets rid of all of those. In fact, in the discussion around the injunction suggests that some of these, like the Be On The Lookout meetings, are fine without more.
It's giving express permission for there to be some kinds of government platform communication that can continue. It modifies the injunction to be only one paragraph, and that paragraph says that basically the government officials who it keeps in the suit shall take no action to coerce or significantly encourage social media companies to change what they're doing. Basically it says the First Amendment is violated if you coerce or significantly encourage private actors to change what they're doing with respect to speech. Hey, government officials, follow the law.
Evelyn Douek:
Right. Clear as mud. Don't do anything illegal. Do only the legal stuff. Thank you.
Genevieve Lakier:
Just follow the First Amendment as we have explicated it. I think in specific instances that helps. I think the government officials are allowed to share information with the platforms when it's not encouragement or coercive. Don't make threats implicit or explicit. Don't tell them that they have to do stuff. This is why I'm confused. Where it gets really complicated is... It's clear that White House officials can no longer write emails to platforms saying, "Hey, you really should take this down,"
And then five days later follow up, "Have you taken this down? We're very unhappy that you're not taking this down." That seems clearly enjoined under the injunction. Where I get confused is when it comes to the FBI and the CDC, the other two big clauses of defendants. The FBI, we are told by virtue of their authority as law enforcement officers at this very powerful scary agency, whenever they communicate with platforms, it seems like the court is suggesting that there is some kind of implicit threat.
How could they comply with this injunction? What kind of information sharing between the FBI and the platforms is okay? If the FBI has a nice prologue where it says, "You may do whatever you like. There is no thread whatsoever," is that enough?
Evelyn Douek:
Lots of smiley faces in the emails.
Genevieve Lakier:
Goodie bags. We love you so much. Positive affirmations. You're doing a great job. Here's how you could do even better. Is that encouragement? And then the CDC, the CDC seems to be told you can have these Be On The Lookout meetings. But if the platforms then take your advice and run with it, now you've established the policy that they're operating under and that's significant encouragement in the courts for you.
Don't do that. Have meetings, but make them very unpersuasive. I am genuinely very pragmatically confused. If I was at the FBI or the CDC, I would really have no idea how to comply with this injunction.
Evelyn Douek:
Yeah, I agree. I skipped to the injunction when I first got it and it totally seemed to me like the command from the court was that the government officials should... You can keep doing the fine stuff, but don't do the bad stuff. Thank you. Have a nice day. I think once you read the opinion in full, there is more specific guidance from the court in terms of how to understand this and saying, "Yes, these Be On The Lookout meetings are okay," and saying, "We wouldn't want to get to a situation where the surgeon general couldn't suggest to platforms to make sure that kids don't see ads for cigarettes."
That's a kind of communication, a kind of government speech which seems important and useful, and so that's clearly on the right side of the line. There is more guidance than just follow the law, but it's still I think very slippery. Now, to be fair to the court, this is a really slippery area and there's not been a lot of good precedent on how to think about significant encouragement in particular. When things aren't just explicitly threatening, how do we think about that?
It is a lot to ask the court to do to give a very clear injunction, good for all rides that it makes it clear, but I think that this still remains particularly unclear. Of course, part of the result of that is to chill government action. If you are in the CDC or the FBI or any of these other agencies now and you're faced with this very unclear injunction, the result you would think would be that you're going to err on the side of caution and chill things that otherwise might have been acceptable and totally okay.
We're going to see, I think, and there has been reporting on this already from the district court opinion. I think that's probably going to continue that many meetings between platforms and officials are going to be halted, pending whatever happens next. That means in the lead up to the 2024 election, things like foreign interference or certain kinds of blatant election misinformation about polling times and things like that might be the kinds of things that government officials get nervous about.
Because of course, this injunction applies to the federal government and these named agencies. But if you're a state government official, you're also thinking about what can and can I not do in terms of talking to platforms. You're also looking for guidance and I don't think that this helps you very much so. Thoughts.
Genevieve Lakier:
Yeah. I mean, I'm sympathetic to the difficulty of crafting a good injunction, but one of the perils, one of the problems with having a very unclear analysis in your opinion is that if you're then just relying on that analysis to craft your injunction, the injunction is just going to be very hard to comply with. They say basically just follow up the analysis we provided about what encouragement means and what coercion means. But then analysis is hard to pass because the court's saying one thing in theory and then applying it in different ways.
To be sure, this is a difficult area of law. I think scholars find it really difficult. Injunctions are really difficult. There's this interesting question that we don't need to go into, but I think it's an interesting and important one, which is that you're enjoining future communication that's not really directed to the plaintiffs in the suit and not in fact involving the same tweets, but you're still enjoining it because it's part of the parcel of the same thing. I think that's right, but it's not exactly how things work in other kinds of cases.
There's all these complicated questions, but there are easy questions that the court did not answer that would've helped a lot. For example, can we just focus Be On The Lookout question because I think just getting a little concrete would be helpful? In the doctrinal analysis, the court says, okay, so the Be On The Lookout meetings where the officers educate the platform about misinformation, hot topics, they're not coercive, but they still violate the First Amendment because they lead the platforms to change their policies.
The court cites as evidence of this that one platform noted that as soon as the CDC updates it, it would change information on its website to comply with the official's view of what was COVID disinformation, et cetera, et cetera, and that violates the First Amendment. That's the analysis we get. It's not coercive, but it is significant encouragement. And then in the injunction portion, the court says, "But you can still have be Be On The Lookout alerts. Just don't violate the First Amendment."
What does that mean? It hasn't explained how you could have a Be On The Lookout meeting that doesn't violate the First Amendment because it hasn't said... These are the kind of considerations, and this is the context in which it's okay, and this is the context in which it's not okay. It's just said this is bad. I don't want to give the Fifth Circuit too much of a pass because I think the injunction is very hard to make sense of because the opinion as a whole is quite hard to make sense of. This is really problematic if we're thinking about speech because of the chilling effect, as you said.
Evelyn Douek:
Great. The question is what happens next? To cut to the chase, what are the chances that you think this ends up in a Supreme Court case soon? It's a bit of a leading question, but I'm just going to throw it over to you. What do you think is going to happen?
Genevieve Lakier:
I mean, I'm not really an expert court watcher, but I don't understand how the court could not take it. This is such a significant...
Evelyn Douek:
Presuming, of course, that the administration appeals, but that seems very likely.
Genevieve Lakier:
It seems very likely. I mean, another explanation for why the court's working so hard to cover up the circuit split is to avoid the court taking cert. But I don't think it's a very good coverup job, so I assume it's likely. I have no idea really. But this injunction, even though it's narrower, is still going to have a very significant impact, I imagine, on the federal government.
It seems surely that the court's going to take cert. And if it does, and then if it also takes cert in the NetChoice cases and it's already agreed to take the social media blocking cases, all of a sudden, next term is going to be a very, very significant social media and free speech term.
Evelyn Douek:
Yeah. I mean, I completely agree with that. It seems like the court should grant cert. You mentioned earlier that there's only really been one proper jawboning case taken by the Supreme Court ever. I mean, it should be implicit in the fact that we've now spent some 50 minutes discussing this, but this is a really fundamental issue in the construction of the modern public sphere, how to think about government speech and government interaction with these intermediaries that are playing this really important governance role in our public sphere.
It's just there's so little guidance and so much conflicting guidance and so much uncertainty around what government actors can do at all levels of government, that it seems like a really important issue for the court to weigh in on. Of course, whether the court will weigh in in a way that totally clarifies everything and gives us a very healthy pro-democracy rules is another question, but that's a question for future podcasts, I guess.
I just want to pick up on the thing you mentioned about the NetChoice cases, because, of course, those are the cases where we have these laws out of Texas and Florida where they're commanding platforms. These are laws commanding platforms to carry certain speech. That is in a sense coercion. It's a legal coercion for the platforms to do certain things about their content moderation. I've seen some people suggesting on its face this seems to conflict, right?
Because of course, the Fifth Circuit upheld the Texas law that says platforms, "You can't viewpoint discriminate. You have to moderate in a certain way." And then on the other hand has said that the government can't have this informal communication with the platforms. Do you agree that there's any conflict, I guess, between those two decisions?
Genevieve Lakier:
Oh, I don't. I think they're completely consistent, and I think both are pillars of a certain kind of conservative judicial viewpoint that is likely to win a lot of favor at the Supreme Court, although I think the politics at the court are going to be complicated. But the harm of jawboning, the distinctive specific harm is the evasion of legal process. In Bantam Books v. Sullivan, the court was very clear that the worry here is that it's not so much that the government is coercing the removal of protected speech.
In fact, in jawboning cases, courts don't always look at whether the speech is protected at all, because the whole point is you're using informal pressure or encouragement, threats or promises to get private actors to take it down without it ever getting court to figure out whether it is protected or not. The harm to the system of free expression in general is the same whether or not that particular speech act happens to be constitution protected or unprotected.
That's very different than when you have a democratically enacted law that is then subject to judicial review about whether or not it's okay or not okay. It's just a very different kind of coercion. It doesn't pose the same kind of harm. Now, we may disagree with the conclusion that Fifth Circuit reached in NetChoice v. Paxton in the NetChoice case saying Texas can require the platforms to carry this kind of speech. But it's just a completely different kind of government action that is in question.
I don't think it's either politically inconsistent or theoretically inconsistent or doctrinally inconsistent for the same court to say it's fine to pass a law requiring platforms to do X, but it's not okay to use informal pressure to do exactly the same thing. I guess if I were a betting person, if, as I think is likely, the court takes cert on both of these cases, it could affirm without inconsistency, it could, it's possible that it could affirm the Fifth Circuit on both counts.
Evelyn Douek:
Well, we will have to have you back on when that happens in order for you to be able to say, "I told you so." If that doesn't pan out, we just will never talk ever again and never mention that this interview ever occurred, and it'll all be totally fine. All right, with that, thanks very much, Genevieve. That was great.
Genevieve Lakier:
Thanks for having me.
Evelyn Douek:
This has been Moderated Content. The show will be available in all the usual places, including Apple Podcasts and Spotify, and transcripts come up at law.stanford.edu/moderatedcontent. The show is produced by Brian Pelletier. Special thanks also to Justin Fu and Rob Huffman. See you next week.