It is almost impossible to believe, as the nation stands on the precipice of one of the most important elections in our lifetime, if not the nation’s history, that Democrats would want to stick it to young people, whose enthusiasm and activism they desperately need to prevail in November. But in trying to ram through Congress the so-called “Kids Online Safety Act” (the title is a misnomer, especially given that the law would actually put vulnerable youth in jeopardy by cutting them off from resources and support) this sort of political foot-shooting would seem to be exactly what leading Senate Democrats were determined to make their fellow Democrats do, no matter how much this exhibition of self-defeating political ineptitude would make them all resented in November.

Fortunately, House Republicans inexplicably decided to help Democrats dodge their own bullet by scuttling the bill, at least for now. Hopefully that will afford a chance for clearer minds to prevail and recognize the political insanity of it should it come back to the floor, or any other similar bill percolating through legislatures around the country, many of which inexplicably have Democratic support. As courts are already starting to find, all of them stand to be a First Amendment disaster. But they also stand to be a political one too.

No matter how they are framed, just like book bans, each of these bills ultimately seeks to use government power to unduly infantilize young people, ignore their First Amendment rights, and remove their practical ability to become educated, informed, and socially connected citizens. While some of the worst of these bills may have been dulled a bit in the Senate’s version of KOSA, it still remained a blunt and potent weapon able to cause a lot of harm to young people’s ability to access information online.

Just as book bans are intended to do, laws like these are intentionally designed to cut young people off from a universe of ideas and information. Sure, maybe up until they turn 18 they could still passively watch TV (almost entirely corporately owned, by very few companies), or listen to the radio (do young people even still do that?), or read newspapers and magazines (but only the dead tree versions, if they still exist, because everything digital stands to fall under these laws’ age-restricting requirements, which means young people can forget about reading things like Teen Vogue, or plenty of other digital-only publications). But the richness of the Internet will largely be out of reach.

Furthermore, these bills manage to be even worse than book bans in how they are also inherently designed to harm young people’s ability to communicate online as well, cutting them off from a needed sense of community and opportunities for their own self-expression. None of these outcomes would arrive accidentally; they are indeed the very point of laws like these, to disconnect young people from others’ expression, and others from theirs.

Yet despite this deliberate deprivation somehow Democrats nevertheless seem to expect young people to emerge from the mandatory isolation these laws would impose on their youth as eager supporters of the very same politicians who had sought to keep them in the dark up to then. Such an expectation is beyond naïve. Like with the TikTok ban, laws like KOSA would represent yet another Democrat own-goal taking aim at the very means that political excitement is engendered, including that which has tended to benefit Democrats in particular.

Laws like KOSA seem to reflect the unreasonable belief that at the stroke of midnight on their 18th birthdays young people somehow magically suddenly become qualified to avail themselves of the medium. At least that’s the belief for now, because any law championed to unilaterally ignore First Amendment rights at one age it could similarly be championed to unilaterally ignore them at other ages too. So even if laws like KOSA only dismiss the rights of young people under 18, maybe the next bill proposed by its proponents will strike at the rights of young people up to the age of 21. Or 24. Or 30, etc. Or maybe the next bill after that will work the other way and take First Amendment rights away from people lawmakers now think are too old.

As they stand now laws like these already harm the First Amendment rights of everyone, including all adults, which is yet another reason that bills like KOSA are so toxic to Democrats’ political prospects. They explicitly teach everyone that Democrats don’t care any more about the Constitution than their opponents. For proponents of these bills the First Amendment is a mere triviality that can be ignored whenever politically expedient. Bills like these run roughshod over the First Amendment rights of so many, including everyone who runs a website, who will now be burdened with onerous if not impossible compliance obligations, which is why young people will end up getting cut off from nearly all digital information if they go into force. And also every adult Internet user as well, who would now have to divulge sensitive personal information to be allowed to read or say anything online, irrespective of how the First Amendment is supposed to protect their rights to read and speak anonymously.

And of course it offends the rights of young people themselves, just as they are when books get excised from curricula and removed from libraries, which is why the book bans being pushed by Republican officials are so odious: they are always about cutting people off from ideas and information, Constitution be damned. But there is no reason why it would be bad only when Republicans engage in such censorship and not when Democrats do it too (and with Republican help, of course, as KOSA was only get passed with bipartisan effort).

In fact, if anything laws like KOSA are even worse than book bans, because people are not just getting cut off from the ideas and information that certain government officials don’t like, but potentially ALL ideas and information, including plenty that young minds very much need to grow up into educated adults, well-equipped and ready to vote (including for Democrats!).

The problem, of course, as per usual, is that this rush to “think of the children!” is not thinking about how much what a law is trying to do will instead actually harm them, and what the results of that harm will inevitably be. If the alleged goal of a law like KOSA is to help ensure the minds of the next generation can develop into healthy adults it is indeed an odd policy to pursue, to purposefully starve those minds of informational resources, social connections, and their own expressive outlets. And politically obtuse to think that, once grown, these same young minds won’t remember who left them hungry.

This week all nine Supreme Court justices found in favor of the NRA. Not because they all like what the NRA is selling (although some of them probably do) but because the behavior of New York State, to try to silence the NRA by threatening third parties, was so constitutionally alarming. If New York could get away with doing what it had done, and threaten a speaker’s business relationships as a means of punishing the speaker, then so could any other state against any other speaker, including those who might be trying to speak out against the NRA. Like with the 303 Creative decision, the merit of this decision does not hinge on the merit of the prevailing party, because it is one that serves to protect every speaker of any merit (including those at odds with, say, the preferred policies of states like Texas and Florida, which would cover those conveying pretty much every liberal viewpoint).

The decision was written by Justice Sotomayor, which was something of a welcome surprise given how she’s gotten the First Amendment badly wrong in some of her more recent jurisprudence, including her dissent in 303 Creative and her decision in the Warhol case, where its expressive protection was conspicuously, and alarmingly, absent from her analysis entirely. But in this case she produced a good and important decision that contemporizes earlier First Amendment precedent, and, importantly, in a way entirely consistent with it. In doing so the Court has strengthened the hand of advocates seeking to protect speakers from a certain type of injury that state actors have been trying to use to silence them.

The Court does not break new ground in deciding this case. It only reaffirms the general principle from Bantam Books that where, as here, the complaint plausibly alleges coercive threats aimed at punishing or suppressing disfavored speech, the plaintiff states a First Amendment claim. [p.18]

In these cases it’s not a direct injury, because the First Amendment pretty clearly says that state actors cannot directly silence expression they do not like (although, true, we still see cases where the government has nevertheless tried to go that route). What this decision says is that state actors also cannot try to silence speakers indirectly by threatening anyone they need to interact with to no longer interact with them.

[A] government official cannot do indirectly what she is barred from doing directly: A government official cannot coerce a private party to punish or suppress disfavored speech on her behalf. [p.11]

Here, the New York official, Vullo, pressured insurance companies she regulated to not do business with the NRA.

As superintendent of the New York Department of Financial Services, Vullo allegedly pressured regulated entities to help her stifle the NRA’s pro-gun advocacy by threatening enforcement actions against those entities that refused to disassociate from the NRA and other gun-promotion advocacy groups. Those allegations, if true, state a First Amendment claim. [p. 1]

As alleged Vullo did more than argue that the companies not do business with the NRA, which might be a legitimate exercise of a government official’s ability to try to persuade.

A government official can share her views freely and criticize particular beliefs, and she can do so forcefully in the hopes of persuading others to follow her lead. In doing so, she can rely on the merits and force of her ideas, the strength of her convictions, and her ability to inspire others. What she cannot do, however, is use the power of the State to punish or suppress disfavored expression. See Rosenberger, 515 U. S., at 830 (explaining that governmental actions seeking to suppress a speaker’s particular views are presumptively unconstitutional). In such cases, it is “the application of state power which we are asked to scrutinize.” NAACP v. Alabama ex rel. Patterson, 357 U. S. 449, 463 (1958). [p.8-9]

What she did also went beyond a legitimate exercise of regulatory authority.

In sum, the complaint, assessed as a whole, plausibly alleges that Vullo threatened to wield her power against those refusing to aid her campaign to punish the NRA’s gun-promotion advocacy. If true, that violates the First Amendment. [p.15]

[A]lthough Vullo can pursue violations of state insurance law, she cannot do so in order to punish or suppress the NRA’s protected expression. So, the contention that the NRA and the insurers violated New York law does not excuse Vullo from allegedly employing coercive threats to stifle gun-promotion advocacy. [p.17]

It was using that regulatory authority against a third party as a means of punishing a speaker for its views that violated the First Amendment.

As discussed below, Vullo was free to criticize the NRA and pursue the conceded violations of New York insurance law. She could not wield her power, however, to threaten enforcement actions against DFS-regulated entities in order to punish or suppress the NRA’s gun-promotion advocacy. Because the complaint plausibly alleges that Vullo did just that, the Court holds that the NRA stated a First Amendment violation. [p.8]

Nothing in this case gives advocacy groups like the NRA a “right to absolute immunity from [government] investigation,” or a “right to disregard [state or federal] laws.” Patterson, 357 U. S., at 463. Similarly, nothing here prevents government officials from forcefully condemning views with which they disagree. For those permissible actions, the Constitution “relies first and foremost on the ballot box, not on rules against viewpoint discrimination, to check the government when it speaks.” Shurtleff v. Boston, 596 U. S. 243, 252 (2022). Yet where, as here, a government official makes coercive threats in a private meeting behind closed doors, the “ballot box” is an especially poor check on that official’s authority. Ultimately, the critical takeaway is that the First Amendment prohibits government officials from wielding their power selectively to punish or suppress speech, directly or (as alleged here) through private intermediaries. [p.19]

This decision is not the first time that courts have said no to this sort of siege warfare state officials have tried to wage against speakers they don’t like, to cut them off from relationships the speakers depend on when they can’t attack the speakers directly.

The NRA’s allegations, if true, highlight the constitutional concerns with the kind of intermediary strategy that Vullo purportedly adopted to target the NRA’s advocacy. Such a strategy allows government officials to “expand their regulatory jurisdiction to suppress the speech of organizations that they have no direct control over.” Brief for First Amendment Scholars as Amici Curiae Supporting Petitioner 8. It also allows government officials to be more effective in their speech-suppression efforts “[b]ecause intermediaries will often be less invested in the speaker’s message and thus less likely to risk the regulator’s ire.” [p.19]

One such earlier decision that we’ve discussed here is Backpage v. Dart, where the Seventh Circuit said no to government actors flexing their enforcement muscles against third parties in a way calculated to hurt the speaker they are really trying to target. But instead of there being just a few such decisions binding on just a few courts, suddenly there is a Supreme Court decision saying no to this practice now binding on all courts.

The big question for the moment is what happens next. There are still several cases pending before the Supreme Court – the two NetChoice/CCIA cases and Murthy v. Missouri – which all involve questions of whether the government has acted in a way designed to silence a speaker. The NetChoice/CCIA cases are framed a bit differently than this case, with the central question being whether state regulation of a platform directly implicates the platform’s own First Amendment rights, but for the Court to rule in NetChoice and CCIA’s favor and find that platforms do have such rights it would need to recognize that what Texas and Florida are trying to do in regulating Internet platforms is punish viewpoints they don’t favor. But if the Court could recognize that sort of viewpoint punishment is what the state of New York was trying to do indirectly here, perhaps it can also recognize that these other states are trying to do it directly there.

Meanwhile, in Murthy v. Missouri, the legal question is closer to the one raised here, and indeed the case was even heard on the same day. In that case the federal government is alleged to have unconstitutionally pressured platforms to cut certain speakers off from their services. It would be the same unconstitutional mechanics, to punish a speaker by coming after a third party the speaker depends on, but as even this decision suggests, only if the conduct of the government was in fact coercive and not simply an expression of preference the platforms were free to take or leave.

Which is why the concurrences from Justices Gorsuch and Jackson may be meaningful, if not for this NRA case but for others. With the latter concurrence Jackson appears to want to ensure that government actors are not chilled from exercising legitimate enforcement authority if they also disfavor the speaker who is in their regulatory sights.

The lesson of Bantam Books is that “a government official cannot do indirectly what she is barred from doing directly.” Ante, at 11. That case does not hold that government coercion alone violates the First Amendment. And recognizing the distinction between government coercion and a First Amendment violation is important because our democracy can function only if the government can effectively enforce the rules embodied in legislation; by its nature, such enforcement often involves coercion in the form of legal sanctions. The existence of an allegation of government coercion of a third party thus merely invites, rather than answers, the question whether that coercion indirectly worked a violation of the plaintiff’s First Amendment rights. [p.2 Jackson concurrence]

In her view, the earlier Bantam Books case the decision is rooted in is not the correct precedent; Jackson would instead look at cases challenging retaliatory actions by the government as a First Amendment violation, and here she thinks that analytical shoe better fits.

[It] does suggest that our First Amendment retaliation cases might provide a better framework for analyzing these kinds of allegations—i.e., coercion claims that are not directly related to the publication or distribution of speech. And, fortunately for the NRA, the complaint in this case alleges both censorship and retaliation theories for how Vullo violated the First Amendment—theories that, in my opinion, deserve separate analyses. [p.4 Jackson concurrence]

As for the Gorsuch concurrence, it is quite brief, and follows here in its entirety:

I write separately to explain my understanding of the Court’s opinion, which I join in full. Today we reaffirm a well-settled principle: “A government official cannot coerce a private party to punish or suppress disfavored speech on her behalf.” Ante, at 11. As the Court mentions, many lower courts have taken to analyzing this kind of coercion claim under a four-pronged “multifactor test.” Ibid. These tests, the Court explains, might serve “as a useful, though nonexhaustive, guide.” Ante, at 12. But sometimes they might not. Cf. Axon Enterprise, Inc. v. FTC, 598 U. S. 175, 205–207 (2023) (G ORSUCH , J., concurring in judgment). Indeed, the Second Circuit’s decision to break up its analysis into discrete parts and “tak[e] the [complaint’s] allegations in isolation” appears only to have contributed to its mistaken conclusion that the National Rifle Association failed to state a claim. Ante, at 15. Lower courts would therefore do well to heed this Court’s directive: Whatever value these “guideposts” serve, they remain “just” that and nothing more. Ante, at 12. “Ultimately, the critical” question is whether the plaintiff has “plausibly allege[d] conduct that, viewed in context, could be reasonably understood to convey a threat of adverse government action in order to punish or suppress the plaintiff ’s speech.” Ante, at 12, 19.

What seems key to him is the last line, and reads like a canary of an issue potentially splitting the Court in Murthy, where there the government clearly engaged in communications with intermediary platforms but the question is whether those communications amounted to attempts at persuasion, which is lawful, or coercion, which is not.

Meanwhile, this case itself will now be remanded. The Court ruled based on the facts as the NRA pled them – as was procedurally proper to do at this stage of the litigation – but it’s conceivable that when put to a standard of proof there won’t be enough to maintain its First Amendment claim. And even if the claim survives, the state for its part can still litigate whether it has an immunity defense to this alleged constitutional injury. So the matter has not yet been put to rest, but presumably the underlying First Amendment question it raised now has.