As every first year law student learns, the spending clause is a significant source of federal power. Because Congress can spend money on programs that would otherwise fall outside the enumerated list of powers in Article I, Congress can use grants and subsidies to significantly increase its regulatory reach. As every first year law student also learns, however, this broad power cannot be used to obliterate all other constitutional constraints on the federal government. Most relevant to this series, the federal government cannot use its spending power to do an end-run around the First Amendment.

The First Amendment constrains the federal government in this context in two important, and importantly separate, ways. As others in this series have explored, the First Amendment constrains the government’s formal use of its spending power through the unconstitutional conditions doctrine, which holds that the federal government cannot attach conditions to governmental benefits that require the recipient to give up their right to engage in protected expression. This simple rule can be complicated in practice. The line between the government’s (important, constitutional) power to define the uses to which its money is spent, and (impermissible, unconstitutional) attempts to leverage that power to suppress protected speech is, as Chief Justice Roberts once put it, “not always self-evident.” But the unconstitutional conditions doctrine is nonetheless an important constraint that the First Amendment imposes on government actors when they fund private persons or institutions. It isn’t the only constraint that the First Amendment imposes, however.

Importantly, the First Amendment also constrains the government when it attempts to informally leverage its spending power to suppress protected expression. The government cannot throw its considerable economic weight around by, say, threatening to withdraw or cut off funding in order to coerce private actors to stop speaking or suppress the speech of others. As Genevieve Lakier has explained, this rule against what is known as “jawboning” seeks to ensure the government does not evade constitutional limits on its powers, and prevents governmental actors from exploiting the incentives of private actors who are vulnerable to all sorts of governmental pressure, well beyond formal legal commands.

This bar on the informal use of the funding power is, in a crucial way, broader than the unconstitutional conditions doctrine. It constrains the government from using informal threats as leverage even if those threats, if formalized in law or binding contracts, would be otherwise perfectly lawful, under the unconstitutional conditions doctrine or any other legal rule for that matter.

The rest of this post first explains how and why this is so, and its import right now. It then turns to the less sunny side of the story: despite its formidable breadth, the protection the anti-jawboning rule provides will often be unavailable to those who most need it due to the willing acquiescence of the very private actors it is intended to protect. This makes the anti-jawboning rule in some respects an all-too-weak safeguard against governmental abuse of its spending power.

The Breadth of the Rule Against Jawboning

As the Supreme Court unanimously affirmed just last term, “a government official cannot do indirectly what she is barred from doing directly.” That is, “[a] government official cannot coerce a private party to punish or suppress disfavored speech on her behalf.” Stated this way, the rule against jawboning seems perhaps logical and unremarkable. Constitutional rights would be meaningless if the government could evade them simply by enlisting private actors to do its dirty work.

But less appreciated is that the rule, first announced over 60 years ago in Bantam Books and reaffirmed in Vullo last year, does not merely mean that the government cannot use informal power to do things it could not formally do. Sometimes, the government can’t do something informally even if it could do it formally.

The facts of Vullo make this clear. In that case, a New York state official, Vullo, allegedly used her regulatory oversight over banks to pressure them into terminating their business relationships with the National Rifle Association (NRA) in order to suppress the NRA’s gun advocacy. There was no doubt that Vullo was lawfully entitled to investigate and pursue violations of New York state insurance law. There was also no doubt that those laws ordinarily raised no First Amendment issues. And indeed, the banks in question had conceded that some of their activities were illegal. But the Court held Vullo’s actions violated the First Amendment nonetheless because of her censorious motivation. “[A]lthough Vullo can pursue violations of state insurance law,” Justice Sotomayor wrote for the Court, “she cannot do so in order to punish or suppress the NRA’s protected expression.” Therefore, the fact that “the NRA and the insurers violated New York law does not excuse Vullo from allegedly employing coercive threats to stifle gun-promotion advocacy.” The unconstitutional motivation to coerce private parties into censoring the speech of others made Vullo’s otherwise lawful use of power unlawful.

In holding as much, the Vullo Court was not breaking new ground. It was simply following the Court’s approach in Bantam Books. In that case, the Rhode Island Commission to Encourage Morality in Youth had sent letters to book distributors in order to intimidate them into ceasing to sell certain books the Commission found “objectionable.” Although it was conceded that some of the books were not obscene (and therefore protected), the Court did not actually stop to determine how many of the books were protected or unprotected. It did not matter. Because the letters were intended to, and in fact did, informally suppress expression, they were unconstitutional, even if the government might have been able to prosecute the booksellers for distribution of obscenity using ordinary formal mechanisms. What made its actions unlawful was the intentional use of informal threats to suppress speech that may (or may not!) have been protected.

Current events should make clear why the distinction the doctrine draws between informal acts of speech suppression (never allowed) and formal acts of speech suppression (sometimes allowed) has significant practical implications. The government is currently terminating, withdrawing and threatening to withhold, or attempting to impose new conditions on, many kinds of federal funding, particularly across higher education. Perhaps the government will argue that these terminations are done in accordance with contractual terms. Perhaps the government will say any new funding conditions simply define the purposes to which that federal funding should be put, and do not violate the rule against unconstitutional conditions. I am not taking any position here on the strength of these arguments—and I don’t need to. My point is that even if they are true, the government cannot take these actions—and indeed cannot even threaten to take those actions in informal communications—if its intention is to suppress protected expression.

So, to be concrete, when the Trump administration terminates billions of dollars of federal funding to Harvard in order to get the university to change its own expressive choices, and to suppress the speech of the university’s community members, the rule against jawboning means that courts do not need to inquire whether the funding terminations would be otherwise lawful, or whether the speech sought to be suppressed is protected or not. The unconstitutional motivation of attempting to make Harvard self-censor and suppress the speech of others without using formal speech regulation makes the government’s informal threats and other formal actions, well, unconstitutional.

At first glance, this may seem like a remarkable rule. Why should the Constitution constrain the government from informal actions that would be constitutional if undertaken formally? Courts do not usually let any form of speech-suppressive intent doom government action. The government gets to suppress speech all the time, formally—particularly when it acts in a content-neutral manner. And as mentioned earlier, it can even attach certain viewpoint-discriminatory conditions to government funding in certain circumstances.

The reason is the Court’s acknowledgment in these cases of the unique evils of informal government pressure to suppress speech. The “vice of [such a] system” is that it obviates the need to employ formal law and therefore “at the same time eliminate[s] the safeguards” of the legal process. Therefore, “[i]t is a form of regulation that creates hazards to protected freedoms markedly greater than those that attend reliance upon the criminal law.” That is: the use of informal governmental power can be more dangerous than the use of formal power.

Jawboning can be an extremely powerful tool of speech suppression not only because it takes place without legal safeguards, but also because it seeks to exploit the incentives of actors in the private sphere. The private sphere is made up of intermediary institutions—universities, social media platforms, law firms—that host or support the speech of others. But these intermediaries might be especially susceptible to governmental pressure to suppress that third-party speech, when push comes to shove. As Justice Sotomayor explained in Vullo, the “intermediary strategy” of targeting such institutions can be “more effective” because “intermediaries will often be less invested in the speaker’s message [than the speaker] and thus less likely to risk the regulator’s ire.” This is why, as the Trump administration is proving, the power of the purse is an especially effective jawbone. The loss of funding is such a direct and tangible harm to an institution, while the loss to that intermediary of third-party expression is intangible and perhaps immaterial.

It is for this reason that, ironically enough, despite the breadth of the rule against jawboning, there are limits to the extent to which courts will be able to provide protection against the effective use of the intermediary strategy.

The Limits of Anti-Jawboning Doctrine

The core weakness of the rule against jawboning is the very same reason the rule is designed so broadly: intermediary institutions are often not adequately incentivized to resist informal demands to suppress the speech of others.

This risk is perhaps especially acute in the context of federal funding, because while there is a First Amendment right to be free of current threats, there is no such right to receive future funds. Therefore, a First Amendment jawboning lawsuit could succeed in preventing the government from rescinding or cancelling current funding but may do little to protect the billions of dollars of future funding that universities no doubt hope and expect to receive in coming years. This is why the funding power is an especially effective jawbone—because once the threat is made, even once, the intermediary institution understands that its long-term financial health is at risk. It is reasonable to assume, for example, that this is why Columbia University has not yet challenged the Trump administration’s threats to revoke current funds, even though they would have a strong First Amendment claim to do so. It does not want to win the First Amendment battle but lose the funding war.

This problem of intermediary incentives is exacerbated by the obstacles that the Court has suggested will exist for third-party plaintiffs that seek to challenge governmental jawboning of intermediaries who have suppressed their speech. In last term’s other jawboning decision, Murthy v. Missouri, the Court suggested that such plaintiffs may struggle to establish standing if the intermediary might have “independent incentives” to suppress speech and is exercising “their own judgment.” Thus, when Columbia suggests (somewhat improbably) that acceding to the Trump administration’s specific demands just happened to align with its “key priorities” and is part of “advanc[ing] [Columbia’s] mission,” it may be making it harder for those whose speech is suppressed as a result to challenge the administration’s (successful) jawboning.

What is the answer to the problem of intermediary incentives? At the very least, as Genevieve Lakier and I wrote last year, courts should follow Judge Richard Posner’s approach in Backpage.com v. Dart, and not take such university statements that they were not acting as a result of government pressure at face value. Targets of government pressure “will be reluctant to acknowledge that [they are] submitting to threats” and will instead insist that they are acting for some independent reason like “having discovered that [the activity] offends [their] moral principles.”

But the more important response is not a legal one. As the above examples illustrate, a broad and strong First Amendment rule on paper means little without institutional and societal commitments to uphold free speech values. The only real answer, therefore, is for leaders to refuse to capitulate to these kinds of informal demands. The government cannot be permitted to make an end-run around the First Amendment, and acquiescence only gives it greater incentive to do so again in the future.

This means it is important to recognize that what the Trump administration is doing right now is unconstitutional full stop, no matter what Title VI says, or particular contractual clauses say, or what a court later finds to be an incident of Executive Power. The informal threats are themselves a violation of the First Amendment because they are intended to, and are succeeding in, suppressing speech. It is up to every institution to stand up for the free speech values the First Amendment seeks to protect—the First Amendment cannot do it alone.