The oral argument before the U.S. Supreme Court in Murthy v. Missouri—the case that will determine whether the Biden administration engaged in unlawful “jawboning” of Internet platforms—was strangely anticlimactic. While the justices quizzed the attorneys on either side about standing, standards of review, and hypotheticals, they never engaged deeply with the facts of the case. Similarly, neither side offered a new angle on how courts should divide permissible government speech from impermissible coercion. The Court may well punt by finding that the plaintiffs lack standing, which would be a lost opportunity: Jawboning is a standard feature of the information ecosystem in 2024, and setting out some guidelines or guardrails would be invaluable.

At oral argument, the government focused—correctly, in our view—on the argument that a “significant encouragement” standard for unconstitutional jawboning would tear down the executive branch’s bully pulpit. It would prevent the government from discussing national security issues, antisemitism or harm to children online, with social media platforms as well as other speech distributors like newspapers or broadcasters. And it focused on the fact that the communications about user posts at issue in the case— some of which were undoubtedly heated—took place in the context of a historic pandemic, in which the government was engaged in an effort to get as many Americans as vaccinated as possible. The government also pointed out the clear errors in fact-finding made by the lower courts. Some users’ content was moderated by the platforms years after any allegedly coercive communications with the government or in the absence of any government communications whatsoever. The Fifth Circuit’s injunction also applied to platforms, including LinkedIn, that the government had never communicated with at all.

On the other hand, the parties that won the injunction below argued that government encouragement of the platforms to take down—or even criticism of the platforms’ carriage of—user speech because of its viewpoint “has to” violate the First Amendment rights of its users. Justice Kagan, with support from Justices Barrett and Jackson, characterized that position as an “extremely expansive argument,” one that would cripple the government’s ability to express its points of view on topics like terrorist speech or other emergencies where the government had information that it found necessary in the public interest to share.

The interaction between the justices and the two attorneys, Brian Fletcher (Deputy Solicitor General at the U.S. Department of Justice for Petitioners) and J. Benjamin Aguinaga (Solicitor General of Louisiana for Respondents) proved both the value of oral advocacy skills and the truth of the old adage that while one cannot win a case at oral argument, one can lose it. Fletcher was fast on his feet, precise, and—most important—ready to concede points that did not weaken his position. He came across as authoritative but reasonable. Aguinaga, by contrast, was far less focused, and doggedly refused to retreat from the Fifth Circuit’s extraordinarily expansive definition of jawboning. His position—that any pressure, even rhetorical, from the government about the content moderation practices of platforms violates the First Amendment—was met with frequent disbelief. Justices Kagan and Kavanaugh both noted that, in their experience as government officials, they were constantly in contact with media sources to push the government’s position. Aguinaga thus had to make the claim that this ubiquitous practice contravenes the First Amendment. He quickly ran aground, claiming that “if [a] newspaper declines to run an op-ed because the government asked, that op-ed author can go to any number of other publications,” but an author denied access to Twitter is without recourse. [oral argument transcript pages 73-74] That would come as quite the surprise to the authors and advertisers who have fled Twitter for platforms such as Mastodon, Blue Sky, Reddit, Substack, Truth Social, Threads, and so on.

Aguinaga’s full-throated defense of the Fifth Circuit’s holding was both a tactical and strategic mistake. Tactically, it forced him to claim that any governmental threat—even just to move a reporter’s cubicle down the hall—is barred by the First Amendment. That position runs contrary to the government’s long-established ability to engage in its own speech, while also making the Respondents’ position indifferent to the potential consequences of governmental action. Strategically, it cost both the Court and the rest of us a chance to adopt a less fulsome but more coherent framework for what constitutes impermissible jawboning and what is protected governmental expression. (Pro tip: “significant encouragement” isn’t it, and “mild encouragement” very definitely is not it.)

That is a lost opportunity, because the facts in the Murthy case (despite the hash that the district court and then the Fifth Circuit made of them) do include troubling allegations, such as that the White House pressured social media platforms to remove postings about Hunter Biden’s laptop. Aguinaga’s vastly overexpansive formulation of jawboning (which was certainly not the result of his personal analysis of the case, but rather a decision by the Respondents collectively) cost him credibility with the Court and observers alike. Unfortunately, it also decreased the probability that the eventual Murthy decision will provide meaningful guidance on jawboning.

As for oral advocacy, both Fletcher and Aguinaga faced a key hypothetical from the Court. Justice Barrett asked Fletcher a hypothetical about Facebook deciding to outsource content moderation to the federal government, becoming a quasi-TASS for the 21st century. While noting that the current case didn’t involve that setup, Fletcher grasped the nettle: He conceded that such an arrangement “may…well be state action,” and thus would likely violate the First Amendment. [page 57] Doing so enabled him to distinguish Justice Barrett’s hypothetical joint action arrangement from the respondents’ “encouragement” theory. He said: “the rubric would be I think more sound in the joint action cases than under significant encouragement, which has never been just [the federal government] trying to persuade you to do something.” [page 57] Fletcher conceded a point he didn’t need and built credibility in the process. Aguinaga didn’t fare as well. Justice Jackson asked him about a future teen challenge that involved jumping out of windows: Could the government “call the platforms and say: This information that you are putting up on your platform is creating a serious public health emergency, we are encouraging you to take it down?” [pages 95-96] Aguinaga proceeded to jump out a metaphorical window himself, responding “the moment that the government tries to use its ability as the government and its stature as the government to pressure them to take it down, that is when you're interfering with the third party's speech rights.” Chief Justice Roberts tossed Aguinaga a parachute, asking whether mere governmental encouragement for platforms to remove instructions for the window challenge would violate the First Amendment. (Another tip: If the Chief Justice gives you the opportunity to reconsider a prior answer, take it.) Aguinaga threw the parachute back: “that is a First Amendment problem.” Someone should share with the Louisiana Solicitor General the helpful reminder from the late Justice Robert Jackson that “the Constitution is not a suicide pact.”

Based on the breadth of the Fifth Circuit’s injunction and other errors in factfinding below, Murthy v. Missouri, or at least the version of it argued this week,may not be the right case to decide the bounds of what the government can do to combat disinformation during public emergencies. It may also be so that the case is so embedded with politics that it can’t serve as the basis for an administrable rule. Given that fact, a majority of the Court will remain deeply uncomfortable with the breadth of the relief being sought by the users and states, as well as with the ramifications of the rule they seek to have the Court adopt and its potential effects on the government’s ability to function.

Fortunately, in our view, it seems based on Justice Kavanaugh’s questions that if the Court does decide to provide a bright-line rule, it will focus more on coercion than encouragement. This would take into adequate account the government’s ability to exercise its own speech rights to pursue what it believes to be the public good, including its telling private speakers what that is. But the problem of disinformation online, and what if anything the government can do about it, remains present. Solving it is going to require more careful thinking about what disinformation is, and the connection between false information, our collective information ecosystem, and the First Amendment.

The Murthy oral arguments, and the framing of the case by both sides to date, are disappointing. The federal government’s position is that nothing alleged by the respondents violates the First Amendment. The respondents contend that every communication described does so. The correct answer, normatively and descriptively, almost certainly lies between. But none of the discourse has yielded much insight into four critical questions about jawboning.

First, what sort of express or implied threat of government action counts as coercive? Hyperbolic presidential rhetoric (“They’re killing people”) shouldn’t, but a presidential threat to direct the Federal Trade Commission (FTC) to investigate a platform that doesn’t comply should. It’s not clear what yardstick to use to measure coercion, as Justice Gorsuch observed. [pages 19-20]

Second, how do the power or gravity of the threat and its probability interact? A menacing statement about seeking Section 230 reform is a long way from actionable government action, although complete abolition of the statutory safe harbor might well create significant liability risk for platforms.

Third, how should the courts evaluate allegations of coercion when the purported targets of it—in Murthy, social media platforms—are not complaining of mistreatment and, at least publicly, are pursuing the same content moderation goals that the government espouses?

Lastly, how should the First Amendment treat jawboning when the target is already in the government’s crosshairs? Threats to initiate an FTC investigation may be less frightening if the Commission is already enthusiastically mounting an antitrust action against the platform, but a suggestion that the administration might discourage settlement could create compulsion. The discourse and decisions in Murthy thus far generate little optimism that the Supreme Court will resolve, or even meaningfully address, these issues at the core of jawboning.

Additionally, deciding the extent of the government’s ability to jawbone is only one part of a larger problem. Social media platforms, in the wake of the Murthy litigation below or based on their owners’ views of social media’s role in the marketplace of ideas, are currently engaged in what one of us has called “epistemological retreat”—a refusal to moderate user content based on the platforms’ own determinations of whether or not that content is factually accurate. The truth has been politicized to such a degree that it might be understandable, if not sensible, for platforms to decline entering the fray with respect to true/false calls, and instead let individual users decide for themselves whether what they see online is accurate.

And who is to say that people go online for the truth anyway? Much of the commentary and concerns about disinformation online rest on an implicit presumption: that people are persuaded by what they read. Academic research on that question is at best inconclusive; most studies have shown that, when making up our mind as to whether we believe a statement is true, as between accuracy-motivated and goal-motivated reasoning, the latter usually wins.

But the fact many people are largely unpersuadable doesn’t mean we should accept the death of objective truth, or decide that social media platforms should refrain from moderating false content. Also, the government, just like any of us, has its own views with respect to what is true—and as several members of the Court recognized during the Murthy oral argument, in many areas it has better resources to decide what is true than does the public.

There may well be a business case for disagreement online, but the operation of democracy requires a shared base of truth upon which disagreements of opinion can take place. Moving our current information environment closer to the truth will be difficult, but the First Amendment does not have to stand squarely in the way.