Opinion Breakdown: Why Kagan and Sotomayor Joined Chiles—and Why Jackson Didn’t

In one of the most fervent battles of the Supreme Court Term -- the unexpected happened -- two of the Court's more liberal justices joined the conservative majority for an 8-1 vote.

At first glance, Chiles v. Salazar looks like one more example of the Supreme Court’s conservative supermajority expanding the First Amendment in a politically charged setting. Justice Gorsuch wrote for the Court, joined by Chief Justice Roberts and Justices Thomas, Alito, Kavanaugh, and Barrett. But that is not the full story. Justice Kagan joined that opinion too, and Justice Sotomayor joined both the Court’s opinion and Kagan’s concurrence. Justice Jackson dissented alone. The lineup is real, the split is notable, and neither is well explained by a simple left-right label.

The better explanation is narrower and more revealing. Kagan and Sotomayor appear to have thought that Colorado’s law, as applied to licensed talk therapy, crossed a clear First Amendment line because it regulated speech based on viewpoint. But Kagan’s concurrence also worked to preserve room for future healthcare regulation that is content based yet viewpoint neutral. Jackson rejected both the majority’s classification of the case and the concurrence’s middle ground. In her view, this was not an ordinary speech case at all. It was a case about the state’s power to regulate licensed treatment and enforce a standard of care.

That distinction matters because it helps explain both the vote and the case’s likely afterlife.

The Coalition Is More Unusual in First Amendment Cases Than on the Docket as a Whole

The first graph clears away the least useful explanation. Kagan and Sotomayor joining a conservative-direction result (per the Supreme Court Database) is not especially rare across the Court’s docket. The first graph shows that the overall bar is plainly higher than the First Amendment bar, and that the same pattern persists in the Jackson-era panel. In other words, the surprise of Chiles is not that Kagan and Sotomayor ever join conservative-direction majorities. They do. The surprise is that they did so here, in a First Amendment case, where the gap between full-docket voting and speech-case voting is more meaningful.

That is the first piece of context. The second is in the pairwise agreement data.

The heatmap shows that Kagan, Sotomayor, and Jackson have generally traveled together in recent First Amendment cases. That is why Chiles deserves more than a passing note about an unusual coalition. The more telling fact is not just that Kagan and Sotomayor crossed ideological lines. It is that Jackson broke from justices with whom she has otherwise aligned frequently in this area.

The third graph sharpens the point from Jackson’s side.

Jackson’s dissent profile does not show a justice who is constantly isolated. It shows a justice who is willing, in selected areas, to mark out a sharper position than the Court’s other left-leaning members. That matters in Chiles. Her dissent is not just another vote on the losing side. It is a sustained argument that the Court has misclassified the case at a foundational level.

So, the empirical baseline narrows the puzzle. Kagan and Sotomayor sometimes join conservative-direction majorities. Jackson sometimes dissents sharply. But Chiles is especially interesting because the break happened in a First Amendment case and because the opinions reveal a three-part split: majority, concurrence, and lone dissent.

The Majority Treated Chiles as an Ordinary Speech Case

The majority’s theory is blunt. Colorado’s law, as applied to Chiles’s talk therapy, did not merely regulate treatment. It regulated speech, and worse, it regulated speech based on viewpoint. The law allowed Chiles to provide counseling that affirmed a minor’s existing sexual orientation or gender identity, but barred counseling that sought to help the minor move in the other direction. For the Court, that was enough to make this a straightforward First Amendment problem. Colorado could not save the law by relabeling speech as “treatment” or “therapeutic modality.” “The First Amendment is no word game,” Gorsuch wrote, and what Colorado was regulating was “speech as speech.”

That framing carries broader implications. The majority leaned heavily on NIFLA v. Becerra, where the Court said it had never recognized “professional speech” as a separate category of speech subject to different rules and stressed that speech is not unprotected merely because it is uttered by professionals. In Chiles, Gorsuch used NIFLA to reject the idea that licensed counselors should receive diminished First Amendment protection and to insist that Colorado had failed to identify a persuasive reason to carve out a special rule for state-regulated therapeutic speech.

The rhetoric reflects that choice. The Court said the First Amendment stands against “orthodoxy,” warned against “enforced conformity,” and described the law as an “egregious assault” on free speech and the marketplace of ideas. That is the language of public debate and anti-censorship and not the language of ordinary professional regulation.

Kagan and Sotomayor Joined the Result, But Not Necessarily the Full Sweep

That is where the concurrence matters and as Epstein, Landes, and Posner point out, concurrences are not merely pure agreements with the majority.

The easiest mistake is to treat Kagan and Sotomayor as either reluctant passengers in a conservative opinion or full converts to the majority’s broad anti-regulatory stance. The concurrence says neither. Kagan agreed that the case before the Court was easy because, as applied to talk therapy, Colorado’s law was “textbook” viewpoint discrimination. The statute, she wrote, enabled “speech on only one side” of an ideologically charged issue. On that point, the concurrence is with the majority on the core classification of this law.

But Kagan immediately narrowed the path. If Colorado had enacted a content-based but viewpoint-neutral law, she wrote, that would raise “a different and more difficult question.” She then invoked the distinction the Court had recently been drawing in Vidal v. Elster, where the Court refused to equate every content-based rule with the kind of viewpoint discrimination that most directly threatens “official suppression of ideas.” In Vidal, the Court emphasized that its decision was “narrow” and did not set out a comprehensive framework for all content-based but viewpoint-neutral trademark restrictions. Kagan’s concurrence in Chiles performs the same function in a different doctrinal neighborhood. It says, in effect: this law falls because it is viewpoint based; that does not mean every content-based regulation of professional speech in healthcare should fall too.

That is the nuance the coalition data needs. Kagan and Sotomayor joined because they thought this law crossed a bright First Amendment line. They did not say the state lacks broad authority to regulate all speech in doctors’ and counselors’ offices. On future cases, the concurrence is narrower than the majority and more careful about preserving regulatory space.

Jackson Rejected Both the Majority’s Framing and the Concurrence’s Middle Ground

Jackson’s dissent is not just a vote against the result. It is a different description of the case.

For Jackson, Chiles belongs in the line of cases that treat speech by licensed professionals differently when that speech occurs “only as part of [the] practice of medicine.” She builds that position directly from Casey, where the joint opinion said a physician’s First Amendment interests were implicated “only as part of the practice of medicine, subject to reasonable licensing and regulation by the State,” and from NIFLA, which she reads as preserving the distinction between regulations of speech “as speech” and regulations of professional conduct that only incidentally burden speech. In her telling, talk therapy is still medical treatment, and the fact that it is delivered through words does not magically move it into the ordinary marketplace of ideas.

That is why Jackson goes further than the concurrence in resisting the Court’s framework. She does not just disagree with invalidating Colorado’s law. She disputes the possibility of a stable middle category between viewpoint discrimination and professional standard-setting. Her dissent defines “professional medical speech” as speech by healthcare professionals in the professional-patient relationship, on matters within their expertise, for the purpose of providing medical care. In that setting, she argues, providers “do not have autonomy” in the same way ordinary speakers do, because their treatment-related advice is bounded by a standard of care. And she makes the point even more directly: “the people win—not lose” when a state incorporates the medical profession’s current views into laws that require licensed providers to conform to prevailing standards.

That is a sharp answer to both Gorsuch and Kagan. To the majority, Jackson says the Court has confused treatment regulation with censorship. To the concurrence, she says the supposed distinction between viewpoint-based and viewpoint-neutral regulation is unstable in medicine, because standards of care inevitably require the state to choose among competing views about what is safe, effective, or appropriate treatment. Her view is that context matters, and that healthcare regulation cannot be analyzed as though it were a fight over school flags or parade banners.

The Comparison Cases Make the Split Easier to See

The small comparison set underscores the point.

NIFLA helps explain the majority. There the Court rejected a broad “professional speech” carveout and said content-based regulations of professional speech still require ordinary First Amendment justification absent a recognized exception. That case supplied the anti-exceptionalism logic Gorsuch used in Chiles.

Casey helps explain Jackson. Whatever else it is famous for, Casey also contains the line Jackson needs most: physicians’ speech rights are implicated only as part of medical practice, “subject to reasonable licensing and regulation by the State.” That is the doctrinal foundation for her insistence that Colorado was regulating treatment, not policing dissenting ideas.

Vidal helps explain Kagan. The Court there said the challenged trademark rule was content based but not viewpoint based, stressed the narrowness of the decision, and reserved the broader question for another day. Kagan imported that logic into Chiles to preserve a lane for future healthcare regulation that is content based yet viewpoint neutral.

And 303 Creative helps explain what Chiles is not. There, the Court’s compelled-speech analysis produced a more familiar ideological split, with Sotomayor, Kagan, and Jackson together in dissent. That contrast matters since it suggests that Chiles was not simply a case in which the Court’s left flank embraced the majority’s broader anti-coercion rhetoric. Sotomayor’s 303 Creative dissent insisted that public accommodations law regulates conduct, not speech, and warned against constitutionalizing a right to deny service. In Chiles, by contrast, Sotomayor joined Kagan in treating the law before the Court as viewpoint discrimination.

What the Case Really Shows

The cleanest way to put it is this: the concurrence is closer to the majority on the immediate case, and closer to Jackson on the scope of future cases.

Kagan and Sotomayor did not join merely to minimize damage. They agreed that Colorado’s law, as applied, was unconstitutional because it discriminated on the basis of viewpoint. But they also refused to let that conclusion settle every future dispute over professional speech and healthcare regulation. Jackson rejected that narrower compromise. In her view, medicine is too context-dependent and too saturated with professional judgments for the concurrence’s middle category to do much real work.

That makes Chiles more than a curious coalition case. It is a case about how the Court should classify professional speech itself. The majority’s answer is formal and speech protective. The concurrence’s answer is formal for now, but cabined. Jackson’s answer is contextual and regulatory. The next cases in this area will likely test whether Kagan’s middle position is workable—or whether Jackson is right that the line between viewpoint discrimination and ordinary standard-of-care regulation is far harder to maintain once the state is regulating treatment rather than public debate.