The Supreme Court’s recent decision in Chiles v. Salazar looks at first glance to be a lopsided triumph of First Amendment values—a ringing endorsement of the principle that the state can’t prescribe what is orthodox in the marketplace of ideas.
Eight justices agreed that Colorado had violated the First Amendment by taking sides in a social controversy involving so-called “conversion therapy” on minors—efforts to alter a child’s chosen sexual orientation or gender identity.
Colorado in fact did take sides in that debate. But a closer look reveals that while the majority’s decision might be doctrinally sound, it is at the same time shortsighted and harmful.
In 2019, the Colorado legislature banned licensed counselors from practicing conversion therapy on minors. It did so on the basis of overwhelming professional consensus: The American Psychological Association had found no empirical evidence that any therapy can alter a child’s sexual orientation or gender identity; former participants reported lasting psychological harm, including depression, PTSD, suicidal ideation, and family rupture; and the therapy’s core premise—that a gay or transgender identity is something to be fixed—was itself found to stigmatize patients in ways that cause long-term emotional distress.
The plaintiff was a licensed Colorado counselor who identifies as a conservative Christian and who wanted to continue offering exactly the kind of talk therapy the legislature had determined is harmful. She did not argue that conversion therapy is always effective or benign. Rather, she said that because her version of treatment involves only speech—no physical interventions, no medications—the First Amendment shields it from regulation.
The court, 8–1, agreed. Justice Neil Gorsuch wrote the majority. Liberal Justices Elena Kagan and Sonia Sotomayor joined a concurrence. The majority held that Colorado was practicing blatant viewpoint discrimination, which is nearly always impermissible. Colorado’s law permitted counselors to affirm a client’s gender identity while forbidding speech aimed at changing it.
That asymmetry, Gorsuch concluded, is precisely what the First Amendment prohibits: the government picking sides in an ideological debate and licensing only one viewpoint within the treatment relationship. He closed by invoking the foundational premise of First Amendment law: “a faith in the free marketplace of ideas as the best means for finding truth.”
Justice Ketanji Brown Jackson wrote a lone dissent, and she took the extraordinary step of reading her dissent from the bench, a gesture reserved for rare, vehement disagreement with the majority. Her anchor point was that the majority had failed to appreciate the crucial context: The state had good reason for its action, anchored not in viewpoint discrimination but in the regulation of medical practice—a classic proper area of review for the states.
The marketplace model works when the harm from speech is epistemic: A listener hears a harmful idea, and the antidote is exposure to better ideas. More speech corrects bad speech. That is sensible for political debate, for journalism, for the ordinary exchange of views.
It is not a sensible model for the clinical relationship between a licensed therapist and a vulnerable minor.
Consider recovered memory therapy—one of the most catastrophic therapeutic fads in American history, practiced in the 1980s and ’90s by licensed clinicians using nothing but verbal suggestion, guided imagery, and hypnosis. Therapists convinced patients, including children, that they had repressed memories of childhood sexual abuse that had never occurred.
The results: shattered families, wrongful prosecutions of wholly innocent childcare workers, patient suicides. More speech didn’t un-implant those false memories. It didn’t un-traumatize children who spent years being convinced they had been abused by their parents or others. States disciplined practitioners. Malpractice verdicts ran into the millions. The harm was medical, not informational, and the marketplace of ideas had nothing to offer it.
Jackson’s core argument is that talk therapy is a medical treatment, and medical treatments are subject to state regulation. The fact that this particular treatment is delivered through speech rather than a scalpel or a syringe does not exempt it from the rules that govern every other form of medical care.
Chiles remains free to write papers defending conversion therapy, give speeches praising it, tell patients she thinks Colorado’s law is wrong. What she cannot do is practice the therapy. The restriction falls on the treatment, not the speech.
The stakes are not abstract. Jackson’s dissent catalogs what the majority has put at risk. Mandatory reporting laws that compel a therapist to speak when a patient presents a threat involve only speech. Prohibitions on guaranteeing cures—speech. Ethics codes requiring humane treatment—speech. Licensing boards’ authority to discipline an incompetent counselor—speech, if the incompetence consists of saying the wrong things. On the majority’s logic, providers who offer cruel speech-only therapies can assert a First Amendment right to carry on.
The court’s answer is that malpractice handles it. But as recovered memory therapy demonstrated, by the time malpractice is litigated, lives are already destroyed. Prophylactic regulation exists to stop harm before it happens. That is what a license is for. That is what a standard of care means.
It’s facile if accurate to call that discrimination based on viewpoint. But it’s myopic to miss the more central context of regulating medical practice and shielding citizens from quackery. In Chiles, Jackson alone had the clear long-range vision to see the flaws in the court’s brittle approach to the case.