Free speech doctrine thrives on categories and tests. Content-based regulations target speech based on their communicative content; a court analyzing a content-based regulation will apply strict scrutiny, which the government can rarely satisfy. Courts apply intermediate scrutiny to evaluate content-neutral regulations. Conduct regulations that only incidentally burden speech also implicate a test roughly akin to intermediate scrutiny. Laws either regulate conduct or speech (though some conduct can be expressive; there’s a test for that too).

All this has given rise to a formally complicated, almost flow chart-like model for evaluating First Amendment challenges. Does the law regulate speech, or merely conduct? Is it content-based? If so, can the government satisfy strict scrutiny? And if it’s content-neutral, will the parties merely just fight over tailoring? Because of the judicial skepticism of laws that regulate speech, particularly content-based regulations, challengers to government action fight strenuously to persuade the court to identify the action as speech. The government, for its part, will almost always begin with a claim that there’s nothing to see here: this law merely regulates conduct or is government speech, and the court should let it stand.

Formalism may have its virtues in free speech cases, particularly if a rules-based framework minimizes chilling effects and allows speakers to express themselves with less fear of reprisal. But the current overreliance on rules, in my view, has created a predictable result—parties hostile to government regulation try to jam their claims into the First Amendment, hopeful that they can pull off a maneuver that invalidates the entire scheme as content-based, or perhaps not even properly tailored under intermediate scrutiny.

This is where we have arrived: a system in which any legislator or regulator thinking of drafting a a law that regulates information or data has to think “Is this going to survive the Supreme Court’s view of the First Amendment?” And absent a radical narrowing of what falls into First Amendment-protected speech—a move the Court could make, but seems unlikely—I see only one manageable, principled path forward to allow for government regulation of the large swaths of the economy that supposedly implicate speech. The Court must embrace free speech balancing tests.

Calling for balancing inquiries in speech cases may seem politically impossible, absurd, or ill-advised. The need for rules to provide clarity and predictability for speakers, and to limit the possibility of judges upholding the censorship of speech they dislike, has become a near orthodoxy in free speech cases. In my view, though, the Court has itself abandoned this commitment to rules even as First Amendment doctrine has become more convoluted, contradictory, and opaque. Whatever virtues clear rules provided, they have faded.

Consider two decisions from the Court’s last term. Free Speech Coalition v. Paxton concerned a Texas law mandating age verification for access to explicit content online. Justice Thomas’ majority distinguishes the case from ACLU v. Ashcroft, a 2004 case striking down a similar federal law, by arguing that a burden on adults to access explicit content differs from a ban; no matter that the Court had already said that burdens do not qualify the level of scrutiny in U.S. v. Playboy Entertainment Group. The rule of Ashcroft, Playboy Entertainment Group, and other cases became something that the court circumvented because it could.

In TikTok v. Garland, the Court evaluated a federal law that effectively forced TikTok to either sever ties with its Chinese holding company or face a shutdown within the United States. The social media company argued that the law violated the First Amendment; the federal government countered that it had vital interests in preventing foreign governments from manipulating Americans via social media feeds and in collecting data on American users. Justice Kagan noted at oral argument that the former sounded like a content-based restriction; the latter, however, carried the day as the justification for upholding the law. But by smuggling a mixed-motives justification rule into the First Amendment—in which a law with two motives, one that violates the First Amendment and one that does not, can still survive—the Court created a world in which courts can pick and choose which motives to investigate and which to ignore.

Beyond these two cases, other recent decisions show how the Court has allowed its supposedly firm doctrinal rules to permit discretionary choices by judges. NIFLA v. Becerra weakened the standard from Zauderer v. Office of Disciplinary Counsel by allowing challengers to seek invalidation of transparency requirements on “controversial” topics, ignoring how challengers can actually manufacture controversy to begin with. Americans for Prosperity v. Bonta ignored the longstanding ban on “subjective ‘chill’” as a basis for standing in First Amendment cases. 303 Creative v. Elenis created an exemption for antidiscrimination laws applying to businesses engaged in “expressive activity,” conceding that determining what falls in that category might “raise difficult questions.” If the Court once feared allowing standards and judicial discretion in First Amendment cases, it seems to recently have gained some courage—though it hasn’t actually acknowledged it.

Rather than stop this move into standards-based reasoning, I in fact think the Court should more explicitly adopt it. Justice Breyer, on multiple occasions, called for means-ends balancing in First Amendment cases to allow for consideration of governmental motives and speech interests, eschewing “a mechanical use of categories.” Breyer even managed to get the Court to adopt balancing tests in two of his last First Amendment majority opinions, Mahanoy Area School District v. B.L. and Shurtleff v. City of Boston. In Mahanoy, Breyer set forth a number of factors to determine whether a K-12 school could regulate off-campus speech of students. And in Shurtleff, Breyer fashioned a contextual inquiry to determine whether the government speech doctrine applied. While these balancing tests don’t change the core of free speech doctrine, they show a perhaps surprising potential to incorporate standards and balancing into other First Amendment areas.

Beyond the need to better reflect some of the Court’s actual recent free speech decisions, incorporating standards furthers political goals that a reimagined First Amendment must take into account. A brittle, harsh First Amendment system of categories and stark rules makes it difficult, if not impossible to regulate in the areas most essential to promote a contemporary democratic society, including campaign finance, anti-discrimination, and information governance. The Court has recently, belatedly, and partially adopted a Breyerian fondness for standards and balancing in free speech cases. Directly acknowledging and continuing that shift would allow for a healthier First Amendment environment.

Beyond the courts, entities that use First Amendment principles to inform their own speech regulations (such as social media companies, private universities, and some private employers) should more explicitly and transparently adopt balancing frameworks, which could help socialize common practices in the private sphere. Legislators and regulators at all levels of government should contemplate and enact legislation that might force the Supreme Court to reconsider its excessively formalized doctrines. As many social movements have taught in other areas of rights and liberties, while the Court will probably not, on its own, develop a pro-democracy First Amendment, we can attempt to guide it to that necessary end.