Chiles v. Salazar - Constitutional Challenge to Conversion Therapy Ban

(Slip Opinion) OCTOBER TERM, 2025 1 Syllabus

NOTE: Where it is feasible, a syllabus (headnote) will be released, as is being done in connection with this case, at the time the opinion is issued. The syllabus constitutes no part of the opinion of the Court but has been prepared by the Reporter of Decisions for the convenience of the reader. See United States v. Detroit Timber & Lumber Co., 200 U. S. 321, 337.

Syllabus

CHILES v. SALAZAR, EXECUTIVE DIRECTOR OF THE COLORADO DEPARTMENT OF REGULATORY AGENCIES, ET AL.

CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE TENTH CIRCUIT No. 24–539. Argued October 7, 2025—Decided March 31, 2026 Kaley Chiles holds a master’s degree in clinical mental health and a state counseling license in Colorado. Ms. Chiles does not begin counseling with any predetermined goals; instead, she sits down with clients, dis- cusses their goals, and then formulates methods of counseling that will most benefit them, seeking throughout to respect her clients’ funda- mental right of self-determination. On matters of sexuality and gen- der, Ms. Chiles’s clients, including young people, often have different goals: Some are content with their sexual orientation and gender iden- tity and want help with social issues or family relationships, while oth- ers hope to reduce or eliminate unwanted sexual attractions, change sexual behaviors, or grow in the experience of harmony with their bod- ies. With all those clients, Ms. Chiles seeks to help them reach their stated objectives. And she employs only talk therapy. In 2019, Colorado adopted a law prohibiting licensed counselors from engaging in “conversion therapy” with minors, Colo. Rev. Stat. §12–245–224(1)(t)(V), defining the term to include “any practice or treatment . . . that attempts . . . to change an individual’s sexual ori- entation or gender identity,” as well as any “effor[t] to change behav- iors or gender expressions or to eliminate or reduce sexual or romantic attractions toward individuals of the same sex,” §12–245–202(3.5)(a). Yet the law explicitly allows counselors to provide “[a]cceptance, sup- port, and understanding for . . . identity exploration and development,” §12–245–202(3.5)(b)(I), and to assist persons “undergoing gender tran- sition,” §12–245–202(3.5)(b)(II). Ms. Chiles filed suit in federal court

Syllabus seeking a preliminary injunction, raising a First Amendment chal- lenge to the law as it applies to her talk therapy. Both the district court and the Tenth Circuit determined that Ms. Chiles had Article III standing to pursue her as-applied pre-enforce- ment challenge. On the merits, however, both courts denied Ms. Chiles’s request for a preliminary injunction, reasoning that Colo- rado’s law is best understood as regulating professional conduct and that it regulates speech only incidentally, thus triggering no more than rational-basis review under the First Amendment. This Court granted certiorari to resolve a circuit conflict over how the First Amendment interacts with laws like Colorado’s when those laws are applied to talk therapy. Held: Colorado’s law banning conversion therapy, as applied to Ms. Chiles’s talk therapy, regulates speech based on viewpoint, and the lower courts erred by failing to apply sufficiently rigorous First Amendment scrutiny. Pp. 7–23. (a) The First Amendment protects the inalienable right of every in- dividual to decide for himself “how best to speak,” Riley v. National Federation of Blind of N. C., Inc., 487 U. S. 781, 791, and laws regulat- ing speech based on its subject matter or “communicative content” are “presumptively unconstitutional,” triggering “strict scrutiny” that re- quires the government to prove its restriction is “narrowly tailored to serve compelling state interests,” Reed v. Town of Gilbert, 576 U. S. 155, 163. “Viewpoint discrimination” represents an even more “egre- gious form” of content regulation from which governments must nearly always “abstain.” Rosenberger v. Rector and Visitors of Univ. of Va., 515 U. S. 819, 829. The Court has recognized only a “few historic and traditional cate- gories of expression”—such as fraud, defamation, and “fighting words”—where content-based restrictions do not automatically trigger strict scrutiny. United States v. Alvarez, 567 U. S. 709, 717. These categories are narrowly drawn and share a long and well-recognized historical pedigree. A law regulating the content of speech cannot avoid searching First Amendment review just because it mostly regulates non-expressive conduct. What matters is whether, in fact, the law regulates speech in the case at hand, as illustrated by Cohen v. California, 403 U. S. 15, and Holder v. Humanitarian Law Project, 561 U. S. 1. And the First Amendment’s protections extend to licensed professionals much as they do everyone else. National Institute of Family and Life Advocates v. Becerra, 585 U. S. 755, 766–767. Pp. 8–11. (b) As applied to Ms. Chiles, Colorado’s law regulates the content of her speech and goes further to prescribe what views she may and may not express, discriminating on the basis of viewpoint. The law permits

3 Cite as: 607 U. S. ___ (2026) Syllabus her to express acceptance and support for clients exploring their iden- tity or undergoing gender transition, §12–245–202(3.5)(b), but forbids her from saying anything that attempts to change a client’s “sexual orientation or gender identity,” including efforts to change “behaviors,” “gender expressions,” or “romantic attraction[s],” §12–245–202(3.5)(a). Her speech does not become “conduct” just because a government says so or because it may be described as a “treatment” or “therapeutic mo- dality.” The First Amendment is no word game, and “the exercise of constitutional rights” cannot be circumscribed “by mere labels.” NAACP v. Button, 371 U. S. 415, 429. The fact that the State’s viewpoint regulation falls only on licensed health care professionals does not change the equation. The First Amendment protects the right of all to speak their minds, and NIFLA expressly rejected the notion that professional speech is subject to “ ‘di- minished constitutional protection.’ ” 585 U. S., at 767. History is lit- tered with examples of governments that have sought to manipulate professional speech “ to increase state power, ” “ suppress minorities, ” and censor “ ‘unpopular ideas.’ ” Id., at 771. Colorado’s law does not implicate any recognized exception to the Court’s usual First Amendment rules. It does not require disclosure of “factual, noncontroversial information in . . . ‘commercial speech,’ ” id., at 768, and as applied to Ms. Chiles, it does not regulate conduct in a way that only “incidentally burden[s] speech,” id., at 769. All she does is speak, and speech is all Colorado seeks to regulate. Colorado’s argument that the law regulates speech only incidentally fails because the Court’s speech-incident-to-conduct doctrine asks whether the law restricts speech only because it is integrally related to unlawful conduct, or whether the law restricts expressive conduct only for reasons unrelated to its content. Colorado’s law does neither: Ms. Chiles’s speech does not bear a close causal connection to any sep- arately unlawful conduct, and the State’s law trains directly on the content of her speech, permitting some viewpoints but not others. Pp. 11–17. (c) Colorado cannot establish that applying its law to Ms. Chiles falls within a long tradition of permissible content regulation. Colorado’s arguments proceed at far too high a level of generality, asking the Court to recognize a broad “ ‘First Amendment Free Zone’ ” for speech the State considers “substandard care.” United States v. Stevens, 559 U. S. 460, 469. The Court’s precedents foreclose aggre- gating discrete traditions of content-based regulations to sustain some new and broader category of lesser-protected speech. Even taking each of the traditions Colorado invokes on its own

Syllabus terms, none supports the State’s position. Colorado’s suggestion that the statute represents a traditional law licensing of medicine fails be- cause the State has not presented persuasive evidence of a historic tra- dition—the first state “counselor-licensure bill” was adopted only in 1976—and because licensing laws have traditionally addressed quali- fications, not dictated a professional’s point of view. Colorado’s anal- ogy to informed-consent laws fails because such laws regulate speech only incident to separate physical conduct and usually require disclo- sure of only factual and uncontroversial information, whereas Colo- rado’s law as applied to Ms. Chiles seeks to silence a viewpoint she wishes to express. Finally, Colorado’s invocation of traditional tort claims for malpractice fails because malpractice actions require exact- ing proof of injury caused by breach of duty, “provid[ing] breathing room for protected speech,” Illinois ex rel. Madigan v. Telemarketing Associates, Inc., 538 U. S. 600, 620, whereas Colorado’s law threatens fines, probation, and loss of license simply for expressing a particular view, and does not allow clients to consent to practices that depart from the prevailing standard of care. Pp. 18–22. (d) The First Amendment stands as a bulwark against any effort to prescribe an orthodoxy of views, reflecting a belief that each American enjoys an inalienable right to speak his mind and a faith in the free marketplace of ideas as the best means for finding truth. Laws like Colorado’s, which suppress speech based on viewpoint, represent an egregious assault on both commitments. P. 23. 116 F. 4th 1178, reversed and remanded. GORSUCH, J., delivered the opinion of the Court, in which ROBERTS,

1. J., and T HOMAS, ALITO, SOTOMAYOR, KAGAN, KAVANAUGH, and BARRETT, JJ., joined. KAGAN, J., filed a concurring opinion, in which SOTOMAYOR, J., joined. JACKSON, J., filed a dissenting opinion.

ANOTICE: This opinion is subject to formal revision before publication in the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, pio@supremecourt.gov, of any typographical or other formal errors.

ET AL.

JUSTICE GORSUCH delivered the opinion of the Court. Kaley Chiles is a mental-health counselor in Colorado. In this case, we consider her First Amendment challenge to a state law regulating what she may say when speaking with her clients. I According to Ms. Chiles’s verified complaint, she holds a master’s degree in clinical mental health and a state coun- seling license. App. to Pet. for Cert. 212a (App.). Clients seek her help on a wide variety of mental-health issues, in- cluding trauma, addiction, “eating disorders, gender dys- phoria[,] and sexuality.” Id., at 206a–207a, 215a. Ms. Chiles “does not begin counseling” on any topic “with any predetermined goals.” Id., at 207a. Nor does she seek to “impose her values or beliefs” on clients. Id., at 212a. In- stead, she “sits down . . . and talks to them about their goals.” Id., at 207a; see also id., at 213a. Only after clients have identified their own aspirations does Ms. Chiles begin

“formulat[ing] methods of counseling that will most benefit” them. Id., at 207a. In any counseling that follows, as well, Ms. Chiles seeks to respect her “clients’ fundamental right of self-determination.” Ibid. On matters of sexuality and gender, Ms. Chiles’s clients, including minors, come to her with different goals in mind. Some “are content with” their sexual orientation and gen- der identity and seek assistance only with “social issues, family relationships,” and the like. Id., at 207a, 213a–214a. In cases like those, Ms. Chiles does not try to persuade her clients to “change their attractions, behavior, or identity,” but aims instead to help them address their stated goals.

Id., at 214a; see also id., at 207a. Other clients, however, come to her hoping to “reduce or eliminate unwanted sexual attractions, change sexual behaviors, or grow in the experi- ence of harmony with [their] bod[ies].” Ibid. And in these cases, too, Ms. Chiles seeks to help her clients reach their own stated objectives. Ibid. In doing so, she does not pre- scribe any medicines, perform any physical treatments, or engage in any coercive or aversive practices. Id., at 205a– 207a. All Ms. Chiles offers is talk therapy. Ibid. B In 2019, Colorado adopted a law prohibiting licensed counselors from engaging in “conversion therapy” with mi- nors. Colo. Rev. Stat. §12–245–224(1)(t)(V) (2025). The State reports that it adopted the law “in response to a grow- ing mental health crisis among Colorado teenagers and mounting evidence that conversion therapy is associated with increased depression, anxiety, suicidal thoughts, and suicide attempts.” Brief for Respondents 11–12. Any Col- oradan who thinks a licensed counselor is engaging in con- version therapy may file a complaint with a regulatory board. See §12–245–226(1)(a)(II)(B). A complaint, in turn, triggers a disciplinary review process that can yield a fine,

probation, or the loss of a license. See §12–245–225; Brief for Respondents 9. The term “conversion therapy” may evoke physical tech- niques such as “ ‘electric shoc[k]’ ” therapy aimed at chang-

ing an individual’s sexual orientation or gender identity.

Id., at 4. But Colorado’s ban on conversion therapy reaches

further, forbidding “any practice or treatment . . . that at- tempts . . . to change an individual’s sexual orientation or gender identity.” §12–245–202(3.5)(a) (emphasis added). The law forbids as well any “effor[t] to change behaviors or

gender expressions or to eliminate or reduce sexual or ro-

sex.” Ibid. At the same time, the law explicitly allows coun- selors to engage in “practices” that provide “[a]cceptance, support, and understanding for the facilitation of an indi- vidual’s . . . identity exploration and development.” §12– 245–202(3.5)(b)(I). Likewise, the law allows counselors to provide “[a]ssistance to a person undergoing gender transi- tion.” §12–245–202(3.5)(b)(II). After Colorado adopted its new law, Ms. Chiles filed suit in federal court and sought a preliminary injunction prohib- iting the State from enforcing it against her. She did not dispute that the statute has many valid applications. In- deed, Ms. Chiles did not take issue with Colorado’s effort to ban what she herself calls “long-abandoned, aversive” phys- ical interventions. Brief for Petitioner 10. Instead, Ms. Chiles objected to Colorado’s law only as it applies to her talk therapy, therapy that involves no physical interven- tions or medications, only the spoken word. Motion for Pre- liminary Injunction in No. 22–cv–2287 (D Colo.), ECF Doc. 29, pp. 2–3, 15. Ms. Chiles’s as-applied challenge ran this way. With re- spect to gender identity, she claimed, the law permits her to speak in ways that encourage a client “‘undergoing gen-

der transition,’ ” but the law prohibits her from speaking in ways that help a client “realign [his] identity with [his] sex.”

Brief for Petitioner 11; see ECF Doc. 29, at 17–20. With respect to sexual orientation, Ms. Chiles continued, Colo- rado’s law similarly allows her to affirm a client’s sexual orientation, but prohibits her from speaking in any way that helps a client “change” his sexual attractions or behav- iors. Brief for Petitioner 11; see ECF Doc. 29, at 17–20. Even though Colorado’s law surely has other constitutional applications, she insisted, these constraints strip her of her First Amendment right to speak freely with her clients in ways she believes might help them meet “their own goals.” Brief for Petitioner 12; see ECF Doc. 29, at 2–3. C Both the district court and the Tenth Circuit determined that Ms. Chiles had Article III standing to pursue her as- applied pre-enforcement challenge. 116 F. 4th 1178, 1199, 1201 (2024); Civ. Action No. 1:22–cv–2287 (D Colo., Dec. 19, 2022), App. 139a. In support of their conclusion, both courts read Colorado’s law as prohibiting licensed counselors like Ms. Chiles from engaging in any attempt—including through speech—to help a minor client change his gender identity or sexual ori- entation. 116 F. 4th, at 1197; App. 141a–142a. Both courts understood this prohibition as extending to any attempt— including through speech—to change a client’s “behaviors or gender expressions or to eliminate or reduce sexual or romantic attraction or feelings toward individuals of the same sex.” §12–245–202(3.5)(a); see 116 F. 4th, at 1197; App. 141a–142a. As both courts saw it, too, Ms. Chiles had previously spoken in ways the law now forbids, and she would continue speaking the same way but for Colorado’s new law. 116 F. 4th, at 1196–1198; App. 141a–142a. Fi- nally, both courts concluded that Ms. Chiles had alleged a “credible threat” that the State would enforce its law against her if she continued speaking as she had in the past and wished to do in the future. Indeed, both courts

5 Cite as: 607 U. S. ____ (2026)

observed, Colorado authorities had refused to disavow bringing enforcement actions against her. 116 F. 4th, at 1198–1199; App. 142a–144a. Under these circumstances, the courts held, Ms. Chiles faced an “ongoing injury resulting from the statute’s chilling effect” on her speech. 116 F. 4th, at 1195 (internal quotation marks omitted); see App. 140a–141a. Satisfied as well that Ms. Chiles’s injury was traceable to Colorado’s law and redressable by an as-applied constitutional ruling in her favor, both courts held that she possessed Article III standing to bring suit. 116 F. 4th, at 1194, 1199; App. 145a,

1. 5. In coming to this conclusion, the lower courts joined others that have found standing in similar cases involving similar statutes. See, e.g., Tingley v. Ferguson, 47 F. 4th 1055, 1066 (CA9 2022); Catholic Charities of Jackson, Le-

nawee and Hillsdale Ctys. v. Whitmer, 162 F. 4th 686, 691 (CA6 2025).*

—————— *We, too, agree that Ms. Chiles has standing. Before us, Colorado only halfheartedly contests the point, suggesting Ms. Chiles cannot establish a present “intention” to speak in a way the law forbids. Brief for Re- spondents 23, n. 18. But the State’s argument is based not on some new insight about Ms. Chiles’s state of mind, only a new and narrowed con- struction of the statute that the State advances for the first time in this Court. Id., at 18–19. Colorado’s late-breaking construction of its law, however, would render much of the law’s language superfluous. Reply Brief 2–6; People v. Rodriguez-Morelos, 562 P. 3d 71, 73 (Colo. 2025) (re- citing the presumption against surplusage when construing state stat- utes). The State’s new interpretation also seemingly stands at odds with how the State itself understood its law in proceedings below. See, e.g., Motion to Dismiss Complaint in No. 22–cv–2287 (D Colo.), ECF Doc. 52,

1. 5, n. 3. Separately, Colorado suggests that Ms. Chiles has not shown
   a credible threat that the State will enforce its law against her. Brief for Respondents 23, n. 18. But Colorado has fought this suit through three courts over three years and, at argument here, expressly declined to dis- avow enforcement against Ms. Chiles. Tr. of Oral Arg. 80–81. As the lower courts held, no more is required under this Court’s precedents. 116

2. 4th, at 1199; App. 145a; see Susan B. Anthony List v. Driehaus, 573

3. S. 149, 159 (2014).
   6 CHILES v. SALAZAR

Turning to the merits, both the district court and the Tenth Circuit denied Ms. Chiles’s request for a preliminary injunction. The courts recognized that Ms. Chiles provides only “talk therapy.” 116 F. 4th, at 1193; see App. 152a. And they acknowledged that Colorado’s law regulates the “ver- bal language” she may use. 116 F. 4th, at 1208; see App. 152a. But, the courts held, the main thrust of the State’s law is to delineate which “treatments” and “therapeutic mo- dalit[ies]” are permissible. 116 F. 4th, at 1208, 1210; see App. 151a. Accordingly, the courts reasoned that Colo- rado’s law is best understood as regulating “professional conduct.” 116 F. 4th, at 1206, 1208, 1214; see App. 151a, 155a. At most, they continued, Colorado’s law regulates speech only “incidentally” to professional conduct. 116

1. 4th, at 1204; see App. 155a. As a result, the courts con-
   cluded, Colorado’s law triggers no more than “rational basis review” under the First Amendment, requiring the State to show merely that its law is rationally related to a legitimate governmental interest. 116 F. 4th, at 1215; App. 156a. Be- cause the State satisfied that standard, the courts held that Ms. Chiles was not entitled to the relief she sought. 116

2. 4th, at 1220–1221; App. 157a.
   At the Tenth Circuit, Judge Hartz dissented. As he saw it, the majority committed “several fundamental errors.” 116 F. 4th, at 1227. While Colorado’s law may ban conduct in other applications, he observed, with respect to Ms. Chiles it regulates only what she may and may not say. Id., at 1231. And under this Court’s precedents, he argued, that kind of speech regulation triggers far more demanding scru- tiny than mere rational-basis review. Id., at 1226–1229. Nor, Judge Hartz continued, does the fact that Ms. Chiles happens to hold a professional license change the equation.

Id., at 1229–1230. To the contrary, he said, the majority’s effort to recast Ms. Chiles’s speech as “professional conduct” amounted to little more “than a labeling game.” Id., at 1231 (internal quotation marks omitted).

7 Cite as: 607 U. S. ____ (2026)

After the Tenth Circuit ruled, Ms. Chiles sought certio- rari. In doing so, she argued that the Tenth Circuit’s deci- sion “worsened a circuit conflict” over how the First Amend- ment interacts with laws like Colorado’s. Pet. for Cert. 16; compare 116 F. 4th, at 1221, and Tingley, 47 F. 4th, at 1077, with Catholic Charities of Jackson, 162 F. 4th, at 696;

Otto v. Boca Raton, 981 F. 3d 854, 868–870 (CA11 2020); King v. Governor of New Jersey, 767 F. 3d 216, 229 (CA3 2014), abrogated in part by National Institute of Family

and Life Advocates v. Becerra, 585 U. S. 755, 767–769 (2018) (NIFLA); Tingley v. Ferguson, 57 F. 4th 1072, 1083– 1085 (CA9 2023) (Bumatay, J., dissenting from denial of re- hearing en banc). We agreed to hear the case. 604 U. S. 1193 (2024). II The question before us is a narrow one. Ms. Chiles does not question that Colorado’s law banning conversion ther- apy has some constitutionally sound applications. See Brief for Petitioner 53. She does not take issue with the State’s effort to prohibit what she herself calls “long-abandoned, aversive” physical interventions. Id., at 10. Instead, Ms. Chiles stresses that she provides only talk therapy, employ- ing no physical techniques or medications. Yet, she argues, Colorado’s law still applies to her, prescribing what she may say in “voluntary counseling conversations” with her cli- ents. Id., at 50. And because that application of the law strikes at the heart of the First Amendment’s protections for free speech, she contends, it warrants considerably more searching scrutiny than the rational-basis review the Tenth Circuit applied in this case or the intermediate-scrutiny re- view some other lower courts have employed in cases like hers. Id., at 38; see, e.g., King, 767 F. 3d, at 237. We agree. To explain why, we begin by outlining the relevant First Amendment principles that govern us before discussing how they apply here.

8 CHILES v. SALAZAR

A The First Amendment “envisions the United States as a rich and complex place” where all enjoy the “‘freedom to think as you will and to speak as you think.’” 303 Creative

LLC v. Elenis, 600 U. S. 570, 584, 603 (2023) (quoting Boy Scouts of America v. Dale, 530 U. S. 640, 660–661 (2000)). Often, speech may prove illuminating and inspiring. Some- times, it can be misguided, offensive, or cause “incalculable grief.” Snyder v. Phelps, 562 U. S. 443, 456 (2011). But either way, the First Amendment protects the inalienable right of every individual to decide for himself “how best to speak.” Riley v. National Federation of Blind of N. C., Inc., 487 U. S. 781, 791 (1988). In this Nation, no official—“high or petty”—may command our tongues or silence our voices.

West Virginia Bd. of Ed. v. Barnette, 319 U. S. 624, 642 (1943). Consistent with the First Amendment’s jealous protec- tions for the individual’s right to think and speak freely, this Court has long held that laws regulating speech based on its subject matter or “communicative content” are “pre- sumptively unconstitutional.” Reed v. Town of Gilbert, 576

1. S. 155, 163 (2015). As a general rule, such “content- based” restrictions trigger “strict scrutiny,” a demanding standard that requires the government to prove its re- striction on speech is “narrowly tailored to serve compelling state interests.” Ibid. Under that test, it is “ ‘rare that a

regulation . . . will ever be permissible.’ ” Brown v. Enter- tainment Merchants Assn., 564 U. S. 786, 799 (2011) (quot-

ing United States v. Playboy Entertainment Group, Inc., 529

1. S. 803, 818 (2000)). We have recognized, as well, the even greater dangers as- sociated with regulations that discriminate based on the speaker’s point of view. When the government seeks not just to restrict speech based on its subject matter, but also seeks to dictate what particular “opinion or perspective” in- dividuals may express on that subject, “the violation of the

9 Cite as: 607 U. S. ____ (2026)

First Amendment is all the more blatant.” Rosenberger v.

Rector and Visitors of Univ. of Va., 515 U. S. 819, 829

(1995). “Viewpoint discrimination,” as we have put it, rep- resents “an egregious form” of content regulation, and gov- ernments in this country must nearly always “abstain” from it. Ibid.; see also Iancu v. Brunetti, 588 U. S. 388, 393 (2019) (describing “the bedrock First Amendment principle that the government cannot discriminate” based on view- point (internal quotation marks omitted)); Good News Club

1. Milford Central School, 533 U. S. 98, 112–113 (2001);

Barnette, 319 U. S., at 642.

Of course, with almost any rule comes exceptions. And this Court has recognized a “few historic and traditional categories of expression long familiar to the bar” where content-based restrictions on speech will not automatically trigger strict scrutiny—categories that include fraud, defa- mation, and “fighting words.” United States v. Alvarez, 567

1. S. 709, 717 (2012) (plurality opinion) (alterations and in-
   ternal quotation marks omitted). But, as we have taken pains to emphasize, these exceptional categories are few and narrowly drawn, and all share a long and well- recognized historical pedigree. Ibid.; see NIFLA, 585 U. S., at 767. Indeed, even within these categories we have some- times still applied strict scrutiny when governments have sought to regulate speech based on viewpoint. See, e.g.,

2. A. V. v. St. Paul, 505 U. S. 377, 392, 395–396 (1992) (ad-
   dressing an ordinance that barred certain “fighting words” based on viewpoint). From these general principles, other more specific ones follow. So, for example, a law regulating the content of speech cannot avoid searching First Amendment review just because it mostly regulates non-expressive conduct. Take a classic illustration: Cohen v. California, 403 U. S. 15 (1971). There, the State of California charged Paul Cohen with “maliciously and willfully disturb[ing] the peace.” Id., at 16 (internal quotation marks omitted). Often, of course,

10 CHILES v. SALAZAR

a person disturbs the peace through conduct alone (say, by brawling at a city council meeting). But that is not always true. And in Mr. Cohen’s case, California charged him for disturbing the peace because he wore a jacket bearing the words “ ‘Fuck the Draft’ ” in the corridor of a municipal

courthouse. Ibid. As applied to him, the Court recognized,

the law implicated core First Amendment concerns because the only “ ‘conduct’ ” he engaged in was the speech he dis-

played. Id., at 18. And, we held, California could not con-

stitutionally punish him because of the “content” of his mes- sage. Ibid. We repeated the point in Holder v. Humanitarian Law

Project, 561 U. S. 1 (2010). That case involved a federal law

banning the provision of “ ‘material support’ ” to certain for-

eign terrorist organizations. Id., at 8–9. Much as Califor-

nia had in Cohen, the federal government in Holder argued that the law did not trigger strict scrutiny because it ad- dressed “conduct, not speech.” 561 U. S., at 26. We disa- greed. True, we acknowledged, the law often might regu- late conduct. But, we observed, in the case before us the government threatened to prosecute lawyers, doctors, and others for providing spoken training and expert advice (such as “ ‘how to use humanitarian and international law

to peacefully resolve disputes’ ”) to certain groups. Id., at 16, 21–22. And that application of the law, we held, sought

to “regulat[e] speech on the basis of its content” and thus demanded strict-scrutiny review. Id., at 27–28. As Holder indicates, too, the First Amendment’s protec- tions extend to licensed professionals much as they do to everyone else. It’s a point we have since discussed at length in NIFLA. There, California sought to require crisis preg- nancy clinics to make certain statements to their clients. The State argued that its law did not trigger demanding First Amendment review because it sought to regulate only “ ‘professional speech’ ” by state license holders. 585 U. S., at 766–767. We rejected that move. By compelling clinics

11 Cite as: 607 U. S. ____ (2026)

to speak the State’s message, the law regulated speech based on its content. Id., at 766. And, we held, California had failed to “identif[y] a persuasive reason for treating pro- fessional speech as a unique category . . . exempt from ordi- nary First Amendment principles.” Id., at 773. In reaching that conclusion, to be sure, we acknowledged two kinds of content-based restrictions that can apply to professional speech without triggering strict scrutiny. First, courts generally deploy less searching review when faced with laws that require speakers to disclose only fac- tual, noncontroversial information in “ ‘commercial speech.’ ” Id., at 768; Zauderer v. Office of Disciplinary

Counsel of Supreme Court of Ohio, 471 U. S. 626, 650–653 (1985). Second, laws regulating conduct in ways that inci- dentally sweep in speech may also generally avoid strict scrutiny. NIFLA, 585 U. S., at 769. As with laws address- ing fraud, defamation, and “fighting words,” laws regulat- ing speech along these two lines enjoy a long historical tra- dition. Id., at 767. But, we stressed, neither “turn[s] on the fact” that a licensed professional happens to be speaking.

Id., at 768. Nor, we emphasized, do these narrow categories

of lesser-protected speech warrant a new rule exempting a broader “category called ‘professional speech’ ” from de-

manding First Amendment review. Ibid.

B Applying these principles, we conclude that the courts be- low failed to apply sufficiently rigorous First Amendment scrutiny in this case. Start with the most obvious point. While the First Amendment protects many and varied forms of expression, the spoken word is perhaps the quintessential form of pro- tected speech. And that is exactly the kind of expression in which Ms. Chiles seeks to engage. As a talk therapist, all Ms. Chiles does is speak with clients; she does not prescribe

12 CHILES v. SALAZAR

medication, use medical devices, or employ any physical methods. App. 205a–207a. Next, and nearly as clear to our eyes, Colorado seeks to regulate the content of Ms. Chiles’s speech. When it comes to issues of human sexuality, some of her clients “are con- tent with” their sexual identity and orientation and want help only “with social issues [or] family relationships.” Id., at 207a, 214a. But other clients seek her counsel on how to “reduce or eliminate unwanted sexual attractions, change sexual behaviors, or grow in the experience of harmony with [their] bod[ies].” Id., at 207a. And in those cases, Col- orado regulates how Ms. Chiles may respond. Under its law, she may not speak in any way that attempts to change a client’s “sexual orientation or gender identity”—including a client’s “behaviors or gender expressions”—or in any way that seeks to “eliminate or reduce” a client’s “sexual or ro- sex.” §12–245–202(3.5)(a). Doubtless, Colorado sees things differently. The State in- sists, and the Tenth Circuit agreed, that its law does not “regulate expression” at all, only “conduct,” “treatment,” or a “therapeutic modality.” Brief for Respondents 2, 36–37, 47; 116 F. 4th, at 1208. As a result, Colorado reasons, its law triggers no more than rational-basis or intermediate- scrutiny review. Brief for Respondents 42–44. But the State’s premise is simply mistaken. In many applications, the State’s law banning “conversion therapy” may address conduct—such as aversive physical interventions. But here, Ms. Chiles seeks to engage only in speech, and as ap- plied to her the law regulates what she may say. Her speech does not become conduct just because the State may call it that. Nor does her speech become conduct just be- cause it can also be described as a “treatment,” a “therapeu- tic modality,” or anything else. The First Amendment is no word game. And the rights it protects cannot be renamed

13 Cite as: 607 U. S. ____ (2026)

away or their protections nullified by “mere labels.”

NAACP v. Button, 371 U. S. 415, 429 (1963).

Our precedents have long made that much clear. Califor- nia faulted Mr. Cohen for the “conduct” of wearing an offen- sive jacket. Cohen, 403 U. S., at 16. The federal govern- ment insisted that its law banning support to terrorists regulated “conduct” even as applied to the written and spo- ken advice of professionals like lawyers and doctors.

Holder, 561 U. S., at 10, 26. But the effort to recast speech

as conduct failed in those cases—and it must here too. Un- der the First Amendment, what matters is not how a gov- ernment describes its law or whether the law may regulate conduct in other circumstances. What matters is whether, in fact, the law regulates speech in the case at hand. As applied here, Colorado’s law does not just regulate the content of Ms. Chiles’s speech. It goes a step further, pre- scribing what views she may and may not express. For a gay client, Ms. Chiles may express “[a]cceptance, support, and understanding for the facilitation of . . . identity explo- ration.” §12–245–202(3.5)(b)(I). For a client “undergoing gender transition,” Ms. Chiles may likewise offer words of “[a]ssistance.” §12–245–202(3.5)(b)(II). But if a gay or transgender client seeks her counsel in the hope of chang- ing his sexual orientation or gender identity, Ms. Chiles cannot provide it. The law forbids her from saying anything that “attempts . . . to change” a client’s “sexual orientation or gender identity,” including anything that might repre- sent an “effor[t] to change [her client’s] behaviors or gender expressions or . . . romantic attraction[s].” §12–245– 202(3.5)(a). Colorado disputes none of this; neither does the dissent. See post, at 20–21 (opinion of J ACKSON, J.) (ac- knowledging that Colorado has engaged in viewpoint dis- crimination). Of course, Ms. Chiles remains free to say other things. As Colorado and the dissent emphasize, she may “shar[e] in- formation” about sexual orientation or gender identity.

14 CHILES v. SALAZAR

Brief for Respondents 20. She can “criticiz[e] Colorado’s law.” Ibid. She can “writ[e] papers” espousing her views.

Post, at 20 (opinion of J ACKSON, J.). She may even encour- age a client to seek advice from someone else who doesn’t hold a state license. Brief for Respondents 20. But true as all that may be, it is also true that she cannot voice certain “perspective[s]” the State disfavors when speaking with consenting clients. Rosenberger, 515 U. S., at 829. And, under our precedents, viewpoint restrictions like that are not subject to mere rational-basis review or intermediate scrutiny. Rather, they represent “an egregious form of con- tent discrimination” where First Amendment concerns are at their most “blatant.” Ibid.; see also Reed, 576 U. S., at 168–169; contra, post, at 8, 20–21 (J ACKSON, J., dissenting) (contending that “heightened scrutiny . . . is not warranted” even while admitting that Colorado’s law discriminates based on viewpoint). The fact that the State’s viewpoint regulation targets only licensed healthcare professionals like Ms. Chiles changes nothing. Colorado and the dissent may believe that the First Amendment should carry “far less salience” for the Nation’s millions of “medical professionals” than for everyone else. Post, at 2 (opinion of J ACKSON, J.); post, at 17–26; Brief for Respondents 23. They may believe that state-imposed orthodoxies in speech pose few dangers and many benefits in this field (and who knows what others). But their policy is not the First Amendment’s. The Consti- tution does not protect the right of some to speak freely; it protects the right of all. It safeguards not only popular ideas; it secures, even and especially, the right to voice dis- senting views. Consistent with these principles, our prece- dents have expressly rejected the State and dissent’s notion that “professional speech” represents some “separate cate- gory of speech” subject to “diminished constitutional protec- tion.” NIFLA, 585 U. S., at 767 (internal quotation marks omitted). History is littered with examples of official efforts

15 Cite as: 607 U. S. ____ (2026)

to manipulate and control professional speech—including “the content of doctor-patient discourse”—in ways designed “to increase state power,” “suppress minorities,” and muz- zle “unpopular ideas.” Id., at 771 (internal quotation marks omitted) (recounting examples). And the “dangers associ- ated with” censorship, we have recognized, are no less acute “in the fields of medicine and public health” than they are anywhere else. Ibid. (internal quotation marks omitted). Nor does Colorado’s law implicate any recognized excep- tion to our usual First Amendment rules. As we have seen, some laws regulating speech based on its content—like ones addressing fraud, defamation, and “fighting words”—do not generally trigger heightened scrutiny because of their long historical pedigree. See Part II–A, supra. As we have seen, too, NIFLA recognized that two kinds of such laws some- times apply to professionals. See 585 U. S., at 768. But Colorado’s law fits in neither category. The State does not require professionals to disclose “factual, noncontroversial information in their commercial speech.” Ibid. (internal quotation marks omitted). Instead, Colorado seeks to sup- press views Ms. Chiles wishes to express. Nor, with respect to Ms. Chiles, does Colorado’s law regulate conduct in a way that only “incidentally burden[s] speech.” Id., at 769. All Ms. Chiles does is speak—and, as far as she is concerned, speech is all Colorado seeks to regulate. Resisting this conclusion, Colorado and the dissent try to shoehorn the State’s statute into the latter category. See,

e.g., Brief for Respondents 37. By defining prohibited “con-

version therapy” broadly, Colorado observes, its law pro- scribes a “wide range” of “treatments.” Id., at 18. Any ac- tivity that seeks to change a client’s gender identity or sexual orientation—whether it involves “electric shocks” or consists of “words only”—is banned. Ibid. And considering the full scope of the law’s many applications, the State sub- mits, it seeks to regulate mostly conduct and reaches speech only incidentally. Ibid. The dissent pursues a similar

16 CHILES v. SALAZAR

theme, insisting that Colorado has only incidentally prohib- ited Ms. Chiles’s speech because the law’s “primary objec- tive” is to regulate medical treatments. Post, at 8–26 (Opin- ion of J ACKSON, J.). This argument echoes Colorado’s claim that it seeks to regulate only conduct, and they falter for similar reasons. If a government could reclassify talk therapy as speech in- cident to conduct, it might just as easily do the same for speech incident to “teaching or protesting.” Otto, 981 F. 3d, at 865. “[B]oth are activities, after all.” Ibid. Were that the rule, “[w]hat an opportunity for [the] suppression of dis- sent this would offer.” 116 F. 4th, at 1228 (Hartz, J., dis- senting). Governments could easily wield all manner of laws regulating some conduct to silence speech they disfa- vor. It is a result that would not “compor[t] with the First Amendment’s animating principles” so much as betray

ACKSON, J., dissenting).them. Contra, post, at 17 (J Recognizing as much, our precedents in Cohen and

Holder already foreclose exactly this move. California pros-

ecuted Mr. Cohen under a law banning disturbances of the peace. Cohen, 403 U. S., at 16. The federal government threatened lawyers and doctors with prosecution under a law prohibiting support for terrorists. Holder, 561 U. S., at 10, 14–15, 25. In both cases, the government defended its actions on the ground that the law in question was gener- ally aimed at certain types of conduct. But in both cases, we emphatically rejected that argument. Just because a law may “generally functio[n] as a regulation of conduct,” we held, does not exempt it from demanding First Amend- ment review when a government seeks to apply that law to speech alone. Id., at 27–28. At bottom, Colorado and the dissent fundamentally mis- conceive this Court’s speech-incident-to-conduct prece- dents. In these cases, the question is not whether a law mostly addresses conduct and only sometimes sweeps in speech. Instead, the focus lies on two entirely different

17 Cite as: 607 U. S. ____ (2026)

questions: whether the law in question restricts speech only because it is integrally related to unlawful conduct—or whether the law restricts expressive conduct only for rea- sons unrelated to its content. Illustrative of the first cate- gory, this Court has held that strict scrutiny does not apply to regulations aimed at speech promoting the sale of con- traband because such speech is often bound up with tradi- tional criminal conduct. United States v. Williams, 553

1. S. 285, 297–298 (2008); see also Giboney v. Empire Stor- age & Ice Co., 336 U. S. 490, 502 (1949); Brief for Eugene

Volokh as Amicus Curiae 3–11. Illustrative of the second category, “an ordinance against outdoor fires” would not re- quire a court to apply strict scrutiny even if it prohibited burning a flag in protest, because the law forbids conduct without regard to the message it may convey. Sorrell v.

IMS Health Inc., 564 U. S. 552, 567 (2011).

Colorado’s law does not regulate speech incident to con- duct under either test. The State does not dictate what Ms. Chiles may say because her speech bears a close causal con- nection to some separately unlawful conduct like a tradi- tional crime. Rather, Ms. Chiles seeks to speak with inter- ested clients about steps they might take to change unwanted behaviors, expressions, or attractions related to sexual orientation or gender identity—conduct Colorado it- self does not dispute those clients (or anyone else) may law- fully undertake. Cf. Tr. of Oral Arg. 58 (conceding that “life coaches” and other non-licensees can lawfully convey the advice Ms. Chiles seeks to offer). Nor does Colorado seek to regulate Ms. Chiles’s speech for reasons unrelated to its content, like a ban on outdoor fires that happens to sweep in flag burning. Instead, the State’s law trains directly on the content of her speech and permits her to express some viewpoints but not others. Colorado does not regulate speech incident to conduct; it regulates “speech as speech.”

NIFLA, 585 U. S., at 770.

18 CHILES v. SALAZAR

C Perhaps sensing that this Court’s existing precedents of- fer it little support, Colorado ultimately pivots, urging us to reshape our doctrine. In doing so, the State begins by point- ing again to NIFLA. There, recall, the Court observed that our cases have identified “two circumstances” in which speech by professionals may warrant “diminished” First Amendment protection: when a law seeks to compel disclo- sure of factual, noncontroversial information in commercial speech; and when a law regulates speech only incidentally to conduct. Id., at 767–768 (internal quotation marks omit- ted). Still, Colorado emphasizes, NIFLA left open the pos- sibility that a future party might present “persuasive evi- dence . . . of a long (if heretofore unrecognized) tradition” of content regulation regarding additional categories of pro- fessional (or other) speech that might likewise warrant only “diminished” First Amendment protection. Id., at 767 (in- ternal quotation marks omitted); see also United States v.

Stevens, 559 U. S. 460, 472 (2010).

Seeking to take up this challenge, Colorado contends that its law falls within a long tradition of permissible content regulation. It’s a line of argument that comes with a daunt- ing burden. Under our precedents, the State must present “persuasive” historical evidence in order to overcome our “especia[l]” “reluctan[ce] to mark off new categories of speech for diminished constitutional protection.” NIFLA, 585 U. S., at 767 (internal quotation marks omitted). Still, Colorado insists, it can carry that burden in this case be- cause States have traditionally enjoyed wide latitude to proscribe “substandard care” even when that involves reg- ulating the content of speech. Brief for Respondents 16. As evidence, Colorado points to the history of state laws licens- ing the practice of medicine, regulating informed consent, and permitting tort suits for medical malpractice. Id., at 23–36. The dissent pursues the same point citing the same authorities. See post, at 9–11, 26–31, 35 (opinion of

19 Cite as: 607 U. S. ____ (2026)

JACKSON, J.) (suggesting that States may ban “substandard care” provided “via speech” in light of licensing, informed consent, and malpractice laws). This argument stumbles out of the gate, for it proceeds at far too high a level of generality. From three specific sets of laws, Colorado and the dissent ask us to recognize a cav- ernous “First Amendment Free Zone,” Stevens, 559 U. S., at 469 (internal quotation marks omitted), one in which States may censor almost any speech they consider “substandard care.” It is, once more, an approach our precedents already foreclose. Consider a couple examples. In Alvarez, the federal gov- ernment pointed to specific, historical laws proscribing fraud and defamation and asked us to extrapolate from them a much broader rule allowing it to enforce content- based restrictions on any “false statements.” In NIFLA, California suggested that informed-consent laws and our speech-incident-to-conduct doctrine might be cobbled to- gether to sustain the recognition of a field of “professional speech” subject only to rational-basis review. We rejected both efforts, stressing instead that discrete traditions of content-based regulations cannot be aggregated together to sustain some new and broader category of lesser-protected speech. Alvarez, 567 U. S., at 718–722 (plurality opinion);

id., at 736 (Breyer, J., concurring in judgment); NIFLA, 585

1. S., at 767–768; see also Button, 371 U. S., at 439–440 (distinguishing State’s law from laws proscribing accepted “common-law offenses”); Florida Star v. B. J. F., 491 U. S. 524, 539 (1989) (similar). And Colorado’s similar effort can succeed no more than others like it have in the past. Beyond that problem lies another. Taking each of the three traditions Colorado and the dissent invoke on its own terms—as we must—none delivers the support they sup- pose. Start with Colorado’s suggestion, endorsed by the dis- sent, that the State’s statute represents nothing more than

20 CHILES v. SALAZAR

a traditional law licensing the practice of medicine. See Brief for Respondents 27–28; post, at 26–27 (opinion of

JACKSON, J.). We cannot agree for at least two reasons.

First, the State has not presented persuasive evidence that its law is part of a historical tradition. When assessed at the level of generality our precedents demand, what Colo- rado describes turns out to be a relatively recent innova- tion. Indeed, the briefing before us suggests that the very first state “counselor-licensure bill” was adopted only in

1. National Academies, Institute of Medicine, Provision
   of Mental Health Counseling Services Under TRICARE 94 (2010); D. Bergman, The Role of Government and Lobbying in the Creation of a Health Profession: The Legal Founda- tions of Counseling, 91 J. Counseling & Development 61 (2013) (Bergman); see also Brief for Petitioner 42. And that is far from the sort of “persuasive evidence” of a historically grounded practice our precedents require. NIFLA, 585

2. S., at 767 (internal quotation marks omitted).
   Second, licensing laws have traditionally addressed what qualifications an individual must possess before practicing a particular profession. See Bergman 62 (licensing laws generally “establis[h] minimum standards of preparation and ensur[e] the professional is qualified”); R. Kry, The “Watchman for Truth”: Professional Licensing and the First Amendment, 23 Seattle U. L. Rev. 885, 887 (2000) (a “licen- sure scheme typically sets out certain requirements that the professional must fulfill before practicing”). And what- ever traditional interest a State may have in ensuring a professional possesses a particular set of qualifications, that interest does not automatically entail a right to dictate a professional’s point of view. NIFLA, 585 U. S., at 773. Turn now to Colorado’s effort, again echoed by the dis- sent, to compare its statute to informed-consent laws. See Brief for Respondents 34–36; post, at 9–11 (opinion of JACKSON, J.). Those laws generally require a doctor to in- form a patient about “the nature of [a proposed] procedure”

21 Cite as: 607 U. S. ____ (2026)

and its attendant “risks.” NIFLA, 585 U. S., at 769–770 (in- ternal quotation marks omitted). As such, informed- consent laws do not usually trigger strict scrutiny because they regulate speech only incident to separate physical con- duct that would, “without [the] patient’s consent,” amount to “an assault.” Id., at 770 (internal quotation marks omit- ted). Usually, too, informed-consent laws require practi- tioners to disclose only factual and uncontroversial infor- mation. See id., at 769–770. But all that is a far cry from how Colorado’s law operates here. As applied to Ms. Chiles, the State seeks neither to regulate her speech incident to any conduct, nor does it seek to compel disclosure of factual and uncontroversial information. Instead, it seeks to si- lence a viewpoint she wishes to express. Colorado and the dissent close by asking us to analogize the State’s law to traditional tort malpractice claims. See Brief for Respondents 25–26; post, at 27–29 (opinion of JACKSON, J.). But here again the differences are impossible to ignore. In a traditional malpractice action, liability at- taches only if the plaintiff shows, among other things, that he has suffered an injury caused by the defendant’s breach of the applicable duty of care. See, e.g., Day v. Johnson, 255

1. 3d 1064, 1068–1069 (Colo. 2011). Those kinds of “[e]xact- ing proof requirements,” we have observed, may “provide sufficient breathing room for protected speech.” Illinois ex

rel. Madigan v. Telemarketing Associates, Inc., 538 U. S. 600, 620 (2003). Yet Colorado’s law contains nothing like them, instead threatening individuals with fines, proba- tion, and the loss of their licenses simply for expressing a particular view. See §§12–245–202(3.5), 12–245–224(1)(t), 12–245–225. Nor does Colorado’s law allow clients to con- sent to practices that depart from the prevailing standard of care, while malpractice law sometimes does. See Re- statement (Third) of Torts: Medical Malpractice §11 (2024);

Schneider v. Revici, 817 F. 2d 987, 995 (CA2 1987).

22 CHILES v. SALAZAR

Consider, too, where the State and dissent’s logic leads. Not long ago, many medical experts and organizations, in- cluding the American Psychiatric Association, considered homosexuality a mental disorder. See American Psychiat- ric Association, Diagnostic and Statistical Manual of Men- tal Disorders 38–39 (1952); cf. American Psychiatric Asso- ciation, Position Statement on Homosexuality and Civil Rights, 131 Am. J. Psychiatry 497 (1974). On the view Col- orado and the dissent advance, a law adopted during that era prohibiting counselors from engaging in the “substandard care” of affirming their clients’ homosexuality would have been subject to only rational-basis or interme- diate-scrutiny review—and likely upheld. See 116 F. 4th, at 1237–1238 (Hartz, J., dissenting). Today, tomorrow, and forever, too, any professional speech that deviates from “current beliefs about the safety and efficacy of various medical treatments” could be silenced with relative ease.

ACKSON, J.). It is a consequence Post, at 24 (opinion of J

Colorado freely acknowledges. Tr. of Oral Arg. 61–63. And one the dissent embraces. Post, at 23–26, and n. 9 (opinion of JACKSON, J.). So what if that kind of reflexive deference to currently prevailing professional views may not always end well? Cf. Buck v. Bell, 274 U. S. 200, 205–207 (1927). Fortunately, that is not the world the First Amendment envisions for us. Licensed professionals “have a host of good-faith disagreements” about the “prudence” and “eth- ics” of various practices in their fields. NIFLA, 585 U. S., at 772. Medical consensus, too, is not static; it evolves and always has. A prevailing standard of care may reflect what most practitioners believe today, but it cannot mark the outer boundary of what they may say tomorrow. Far from a test of professional consensus, the First Amendment rests instead on a simple truth: “[T]he people lose” whenever the government transforms prevailing opinion into enforced conformity. Ibid.

23 Cite as: 607 U. S. ____ (2026)

III We do not doubt that the question “how best to help mi- nors” struggling with issues of gender identity or sexual ori- entation is presently a subject of “fierce public debate.” Tin-

gley v. Ferguson, 601 U. S. __, __ (2023) (T HOMAS, J., dissenting from denial of certiorari) (slip op., at 1). But Col- orado’s law addressing conversion therapy does not just ban physical interventions. In cases like this, it censors speech based on viewpoint. Colorado may regard its policy as es- sential to public health and safety. Certainly, censorious governments throughout history have believed the same. But the First Amendment stands as a shield against any effort to enforce orthodoxy in thought or speech in this coun- try. It reflects instead a judgment that every American pos- sesses an inalienable right to think and speak freely, and a faith in the free marketplace of ideas as the best means for discovering truth. However well-intentioned, any law that suppresses speech based on viewpoint represents an “egre- gious” assault on both of those commitments. Rosenberger, 515 U. S., at 829. The judgment of the Tenth Circuit is reversed, and the case remanded for further proceedings consistent with this opinion.

It is so ordered.

ET AL.

JUSTICE KAGAN, with whom JUSTICE SOTOMAYOR joins, concurring. The Court today decides that the Colorado law challenged here, as applied to talk therapy, conflicts with core First Amendment principles because it regulates speech based on viewpoint. See ante, at 23. I agree. I write only to note that if Colorado had instead enacted a content-based but viewpoint-neutral law, it would raise a different and more difficult question. As the Court states, governments must “nearly always” abstain from adopting viewpoint-based restrictions. Ante, at 9. Those laws represent a particularly “egregious form” of content-based regulation, implicating First Amendment concerns to the highest possible degree. Ibid.; see Iancu v. Brunetti, 588 U. S. 388, 393 (2019). A law drawing a line based on the “ideology” of the speaker—disadvantaging one view and advantaging another—skews the marketplace of ideas our society depends on to discover truth. Rosenberger

1. Rector and Visitors of Univ. of Va., 515 U. S. 819, 829 (1995). And such a law suggests an impermissible motive— that the government is regulating speech because of its own “hostility” toward the targeted messages. R. A. V. v. St. Paul, 505 U. S. 377, 386 (1992). If the First Amendment

prohibits anything, it is the “official suppression of ideas.” Id., at 390; see Reed v. Town of Gilbert, 576 U. S. 155, 181– 183 (2015) (K AGAN, J., concurring in judgment). Because viewpoint-based laws always raise that specter, they are the most suspect of all speech regulations. So much so that this Court has refused to permit viewpoint discrimination even within unprotected categories of speech, like fighting words or obscenity. See R. A. V., 505 U. S., at 384–390. Colorado’s law, as applied to talk therapy, regulates based on viewpoint, for the reasons the Court gives. See ante, at 13–14. The law forbids a counselor to provide ther- apy designed to “change [a minor’s] sexual orientation or gender identity.” Colo. Rev. Stat. §12–245–202(3.5)(a) (2025); see §12–245–224(1)(t)(V). At the same time, the law specifically allows a counselor to offer therapy expressing “[a]cceptance, support,” and other affirmation of the minor’s “identity exploration.” §12–245–202(3.5)(b)(I)–(II). So, for example, the law prevents a therapist from saying she can help a minor change his same-sex orientation, but permits her to say that such a goal is impossible and so she will help him accept his gay identity. Colorado does not dispute that point. See Tr. of Oral Arg. 78. Nor does it dispute that un- der normal First Amendment principles, that difference constitutes viewpoint discrimination. See ibid. Indeed, the case is textbook. The law “distinguishes between two op- posed sets of ideas”—the one resisting, the other reflecting, the State’s own view of how to speak with minors about sex- ual orientation and gender identity. Iancu, 588 U. S., at

1. Or said just a bit differently, the law draws a line
   based on the speaker’s “opinion or perspective,” and thus enables “speech on only one side”—the State’s preferred side—of an ideologically charged issue. Rosenberger, 515

2. S., at 829; McCullen v. Coakley, 573 U. S. 464, 485
   (2014). Of course, it does not matter what the State’s preferred side is. Consider a hypothetical law that is the mirror

image of Colorado’s. Instead of barring talk therapy de- signed to change a minor’s sexual orientation or gender identity, this law bars therapy affirming those things. As Ms. Chiles readily acknowledges, the First Amendment would apply in the identical way. See Tr. of Oral Arg. 25– 26; see also id., at 37–38 (United States as amicus curiae agreeing). Once again, because the State has suppressed one side of a debate, while aiding the other, the constitu- tional issue is straightforward. It would, however, be less so if the law under review was content based but viewpoint neutral. Such content-based laws, as the Court explains, trigger strict scrutiny “[a]s a general rule.” Ante, at 8. But our precedents respecting those laws recognize complexity and nuance. We apply our most demanding standard when there is any “realistic pos- sibility that official suppression of ideas is afoot”—when, that is, a (merely) content-based law may reasonably be thought to pose the dangers that viewpoint-based laws al- ways do. Davenport v. Washington Ed. Assn., 551 U. S. 177, 189 (2007); see supra, at 1–2. But when that is not the case—when a law, though based on content, raises no real concern that the government is censoring disfavored ideas—then we have not infrequently “relax[ed] our guard.”

AGAN, J.); see Daven-Reed, 576 U. S., at 183 (opinion of K

port, 551 U. S., at 188 (noting the “numerous situations in which [the] risk” of a content-based law “driv[ing] certain ideas or viewpoints from the marketplace” is “attenuated” or “inconsequential, so that strict scrutiny is unwar- ranted”). Just two Terms ago, for example, the Court de- clined to apply strict scrutiny to a content-based but view- point-neutral trademark restriction. See Vidal v. Elster, 602 U. S. 286, 295 (2024); id., at 312 (B ARRETT, J., concur- ring in part); id., at 329–330 (S OTOMAYOR, J., concurring in judgment). In the trademark context, as in some others, experience and reason alike showed “no significant danger of idea or viewpoint” bias. R. A. V., 505 U. S., at 388.

The same may well be true of content-based but view- point-neutral laws regulating speech in doctors’ and coun- selors’ offices.* Medical care typically involves speech, so the regulation of medical care (which is, of course, perva- sive) may involve speech restrictions. And those re- strictions will generally refer to the speech’s content. Cf. Reed, 576 U. S., at 177 (Breyer, J., concurring in judgment) (noting that “[r]egulatory programs” addressing speech “in- evitably involve content discrimination”). But laws of that kind may not pose the risk of censorship—of “official sup- pression of ideas”—that appropriately triggers our most rigorous review. R. A. V., 505 U. S., at 390. And that means the “difference between viewpoint-based and view- point-neutral content discrimination” in the health-care context could prove “decisive.” Vidal, 602 U. S., at 330 (opinion of S OTOMAYOR, J.). Fuller consideration of that question, though, can wait for another day. We need not here decide how to assess viewpoint-neutral laws regulat- ing health providers’ expression because, as the Court holds, Colorado’s is not one.

—————— *JUSTICE JACKSON’s dissenting opinion claims that this is a small, or even nonexistent, category. See post, at 21–22, n. 8. But even her own opinion, when listing laws supposedly put at risk today, offers quite a few examples. See post, at 32–33, and n. 13. Her view to the contrary rests on reimagining—and in that way collapsing—the well-settled dis- tinction between viewpoint-based and other content-based speech re- strictions. See, e.g., Vidal v. Elster, 602 U. S. 286, 292–293 (2024) (ex- plaining the difference).

ET AL.

JUSTICE JACKSON, dissenting. “[T]here is no right to practice medicine which is not sub- ordinate to the police power of the States.” Lambert v. Yel-

lowley, 272 U. S. 581, 596 (1926). This was true 100 years ago, and it should be true today. Many States have now chosen to exercise their police powers to ban “conversion therapy” based on the medical profession’s broad consensus that this medical treatment (which seeks to change a gay or transgender person’s sexual orientation or gender identity) is ineffective and harmful. This case involves the Colorado Legislature’s policy deci- sion to prohibit licensed medical professionals from offering or providing conversion therapy to minors in that State. Petitioner Kaley Chiles is a licensed counselor who works in the State of Colorado. She does not dispute that conver- sion therapy can be harmful to minors in certain circum- stances. Nor does she contest that Colorado has a signifi- cant interest in protecting minors from harm. Chiles complains nevertheless that, because the particular form of conversion therapy she wants to offer clients utilizes only speech, the First Amendment prevents Colorado from pro- hibiting that treatment. But “[t]he power of government to regulate the professions is not lost whenever the practice of

a profession entails speech.” Lowe v. SEC, 472 U. S. 181, 228 (1985) (White, J., concurring in result). And “[m]edical professionals do not, generally speaking, have a right to use the Constitution as a weapon allowing them rigorously to control the content of . . . reasonable conditions” that a State imposes on licensed healthcare providers for the pro- tection of its residents. National Institute of Family and

Life Advocates v. Becerra, 585 U. S. 755, 785 (2018) (Breyer, J., dissenting) (NIFLA). So, I respectfully dissent. Stated simply, the majority has failed to appreciate the crucial context in which Chiles’s constitutional claims have arisen. Chiles is not speaking in the ether; she is providing therapy to minors as a licensed healthcare professional. The Tenth Circuit was correct to observe that “[t]here is a long-established history of states regulating the healthcare professions.” 116 F. 4th 1178, 1206 (2024). And, until today, the First Amendment has not blocked their way. For good reason: Under our prece- dents, bedrock First Amendment principles have far less salience when the speakers are medical professionals and their treatment-related speech is being restricted inci- dentally to the State’s regulation of the provision of medical care. No one directly disputes that Colorado has the power to regulate the medical treatments that state-licensed profes- sionals provide to patients. Nor is it asserted that, when doing so, a State always runs afoul of the Constitution. So, in my view, it cannot also be the case that Colorado’s deci- sion to restrict a dangerous therapy modality that, inci- dentally, involves provider speech is presumptively uncon- stitutional. In concluding otherwise, the Court’s opinion misreads our precedents, is unprincipled and unworkable, and will eventually prove untenable for those who rely upon the long-recognized responsibility of States to regulate the medical profession for the protection of public health.

I To properly evaluate the First Amendment claim at issue in this case, one must first understand the impetus for Col- orado’s regulation, what that law requires, and the nature of the speech it implicates. A Conversion therapy is designed to “convert” a person’s sexual orientation or gender identity, so that the person will become heterosexual or cisgender. Generally speaking, conversion therapy began as an attempt to “cure” gay and transgender people of their “nonconforming” orientations or identities. Brief for American Psychological Association et al. as Amici Curiae 13. Conversion-therapy efforts have historically included aversive therapeutic modalities. Those ranged from induc- ing nausea, vomiting, or paralysis in patients or subjecting them to severe electric shocks to telling patients to snap an elastic band on their wrists in response to nonconforming thoughts. Aversive therapies have now fallen out of fash- ion; nonaversive treatments—primarily, talk therapy—are currently the predominant form of conversion therapy. All such therapies seek to encourage patients to change their behavior in an attempt to “change” their identity. Over the past few decades, however, the premise of con- version therapy (in whatever form) has been widely discred- ited within the medical and scientific community. Conver- sion therapy is, at bottom, “based on a view of gender diversity that runs counter to scientific consensus.” Sub- stance Abuse and Mental Health Services Administration (SAMHSA) Report, 2 App. 570. That is, contrary to the core beliefs that undergird conversion therapy, a robust profes- sional consensus now acknowledges that sexual orienta- tions and gender identities range widely. And it no longer regards nonheterosexual orientations or noncisgender iden- tities as “nonconforming.”

Because people’s identities are simply “a part of the nor- mal spectrum of human diversity,” id., at 535, the medical community has determined that efforts to change a pa- tient’s sexual orientation or gender identity will necessarily be ineffective. The American Psychological Association (APA), for example, has found “no empirical evidence that providing any type of therapy in childhood can alter adult same-sex sexual orientation.” APA Report, 1 App. 360. And “[n]o research has been published in the peer-reviewed lit- erature that demonstrates the efficacy of conversion ther- apy efforts with gender minority youth, nor any benefits of such interventions to children and their families.” SAMHSA Report, 2 App. 569. Not only is conversion therapy ineffective, former partic- ipants of conversion therapy report that it causes lasting psychological harm. Gay and transgender children who un- derwent nonaversive conversion therapy say they were taught to feel shame and self-hatred. See Brief for Conver- sion Therapy Survivor Network et al. as Amici Curiae 11–

1. And survivors of conversion therapy continue to suffer from PTSD, anxiety, and suicidal ideation. Id., at 19–22. As one survivor put it, conversion therapy “‘came close to killing me.’ ” Id., at 17. 1

—————— Consider a specific example: An amicus who received conversion ther-1 apy as a child, Mathew Shurka, was told that his sexual orientation was a disorder either rooted in childhood trauma or stemming from an over- bearing mother or absent father. Brief for Mathew Shurka as Amicus

Curiae 6. In the course of this therapy, Mathew’s therapist said that being gay was a mental illness that could be cured and that, unless he was cured, he could never live a happy and fulfilled life. Id., at 3. Mathew’s conversion therapists eventually hypothesized that his mother was the source of his sexual orientation and instructed him to avoid speaking with her. Id., at 8. For three years, he had barely any contact with his mother. Ibid. The years of conversion therapy brought Mathew nothing but increased isolation from his family, worsening depression, and suicidal ideation. Id., at 11. Thus, as Mathew’s experience illus- trates, the harm from conversion therapy extends to the patient’s family. See also Brief for Parents of Conversion-Therapy Participants et al. as

5 Cite as: 607 U. S. ____ (2026) J

The scientific literature confirms what anecdotal experi- ences suggest: Conversion therapy has harmed patients, particularly minors. The APA found that “the reported neg- ative social and emotional consequences [of conversion therapy] include self-reports of anger, anxiety, confusion, depression, grief, guilt, hopelessness, deteriorated relation- ships with family, loss of social support, loss of faith, poor self-image, social isolation, intimacy difficulties, intrusive imagery, suicidal ideation, self-hatred, and sexual dysfunc- tion.” APA Report, 1 App. 253–254. Even for those study participants who reported positive effects initially, many described experiencing the negative effects later. Id., at

1. Moreover, studies show that children often feel the ef- fects of the shame and stigma from conversion therapy even more vividly than adults due to their “increased emotional vulnerability and less developed capacity to cope effectively with the harm of discrimination.” Glassgold Decl., 1 App. 53–54, ¶ 50. Ultimately, scientific evidence supports the conclusion that the anticipated harms from conversion therapy are twofold. First, conversion therapy stigmatizes the patient, telling them that their gender identity or sexual orientation is something to be fixed, rather than accepted. This rejec- tion can lead to shame and guilt, which in turn can cause long-term emotional distress. Second, conversion therapy sets patients up to fail by giving them an unattainable goal. Some patients have described that experience of failure “as a significant cause of emotional and spiritual distress and negative self-image.” Id., at 63, ¶ 66. B In 2019, Colorado joined 25 other States in banning the practice of conversion therapy for minors. Colorado’s law— titled the Minor Conversion Therapy Law (MCTL)—

——————

Amici Curiae 9–19 (describing the impact of conversion therapy on

amici’s families, including the loss of loved ones to suicide).

6 CHILES v. SALAZAR

prohibits licensed healthcare professionals from practicing conversion therapy with children. It defines conversion therapy as “any practice or treatment” that “attempts or purports to change an individual’s sexual orientation or gender identity, including efforts to change behaviors or gender expressions or to eliminate or reduce sexual or ro- sex.” Colo. Rev. Stat. §12–245–202(3.5)(a) (2025). 2 Not all therapeutic discussions of sexuality and gender identity are prohibited by the MCTL. The law allows “practices or treatments that provide: “(I) Acceptance, support, and understanding for the facilitation of an individual’s coping, social support, and identity exploration and development, including sexual-orientation-neutral interventions to prevent or address unlawful conduct or unsafe sexual practices, as long as the counseling does not seek to change sexual orientation or gender identity; or “(II) Assistance to a person undergoing gender tran- sition.” §12–245–202(3.5)(b). The Colorado Legislature made these allowances after crediting witness testimony and the professional consensus about conversion therapy—namely, that it is harmful, and that the preferred treatment for minors relating to their sexual orientation and gender identity is affirming care (i.e., medical care that helps minors focus on acceptance, support, coping, and identity exploration and development). Prohibit Conversion Therapy for a Minor: Hearing on H. B.

——————

The MCTL primarily applies to licensed healthcare professionals; 2

such professionals are already subject to a number of other restrictions on their professional practice. Most relevant here, under Colorado law, licensed therapists must provide therapy that is consistent with the standard of care, defined as “the standards of practice generally recog- nized by state and national associations of practitioners in the field of the person’s professional discipline.” Colo. Rev. Stat. §12–245– 224(1)(g)(I).

7 Cite as: 607 U. S. ____ (2026) J

19–1129 before the House Committee on Public Health Care & Hum. Servs., 2019 Leg., 72d Gen. Sess. (Colo., Feb. 13, 2019). C Chiles insists that, although she is a counselor licensed by Colorado, she has a constitutional right to flout Colo- rado’s statute and the standard of care it incorporates if a client asks her to do so. Never mind that medical profes- sionals—including counselors like Chiles—are generally bound to follow medical standards and state licensing re- quirements when they provide medical care to patients. Chiles wants to offer patients conversion therapy despite the MCTL and the medical consensus it reflects. So she has invoked the First Amendment, arguing that, because talk therapy is speech, no State can impose treatment standards like the MCTL on licensed talk therapists without first sat- isfying heightened scrutiny. As applied to Chiles, the MCTL treats the talk-therapy form of conversion therapy as a prohibited medical treat- ment. But Chiles is free to express her opinion about the efficacy of conversion therapy or her disagreement with Colorado’s conclusion that such therapy is harmful to mi- nors. Colorado’s law does not target or prohibit the expres- sion of such views by anyone in any form—including by li- censed healthcare providers in discussions with patients and their families. All that Colorado’s law proscribes is the provision of such therapy to minors. This means that, while Chiles can freely promote conversion therapy and vocifer- ously decry the State’s prohibition, she cannot practice that therapy without being subject to professional discipline un- der Colorado law. II I begin my analysis with a simple observation: Our First Amendment jurisprudence does not treat speech as existing

8 CHILES v. SALAZAR J

in a vacuum. Instead, how the First Amendment applies to a State’s power to regulate speech depends upon the context in which the regulation of speech occurs. See, e.g., Vidal v.

Elster, 602 U. S. 286 (2024) (trademark context); Tinker v.

Des Moines Independent Community School Dist., 393 U. S.

503 (1969) (school context). We have not mechanically held that the First Amendment protects all communicative con- tent; rather, we have evaluated First Amendment claims in a nuanced way, sensitive to both core principles and the specific circumstances under which the claim arises. See,

e.g., Virginia Bd. of Pharmacy v. Virginia Citizens Con- sumer Council, Inc., 425 U. S. 748, 758, 762–770 (1976)

(considering First Amendment principles in the commercial speech context). In my view, then, it matters for First Amendment pur- poses that the MCTL restricts treatment-related speech ut- tered by medical professionals only as part of a larger reg- ulatory scheme aimed at ensuring that providers tender high-quality medical care to patients. In Part II–A, I explain that this way of conceptualizing the question before us is not novel—we have long under- stood that States have the power to regulate medical pro- fessionals. And our precedents demonstrate that, when a healthcare provider’s speech is incidentally restricted as part of a state-law scheme regulating the provision of med- ical treatments, the heightened scrutiny we reflexively ap- ply in other situations is not warranted. In Part II–B, I show that First Amendment principles are not offended when lesser scrutiny is applied to a state law regulating medical treatments in a manner that incidentally restricts a provider’s professional medical speech. A 1 A case that we decided in 2018—NIFLA—shows us the way to determine the appropriate level of constitutional

9 Cite as: 607 U. S. ____ (2026) J

scrutiny here. 585 U. S. 755. In that case, we began by explaining that “[s]peech is not unprotected merely because it is uttered by ‘professionals.’ ” Id., at 767; see also ante, at

10–11, 14. But, critically, we also acknowledged that the

Court has afforded less protection for professional speech in two circumstances—one of which occurs when a State “reg- ulate[s] professional conduct, even though that conduct in- cidentally involves speech.” NIFLA, 585 U. S., at 768; see

ante, at 11. This acknowledgement was grounded in an al-

ready well-established principle: “[T]he First Amendment does not prevent restrictions directed at . . . conduct from imposing incidental burdens on speech.” Sorrell v. IMS

Health Inc., 564 U. S. 552, 567 (2011).

To further explain the contours of this presumptively per- missible speech restriction, NIFLA cited Planned

Parenthood of Southeastern Pa. v. Casey, 505 U. S. 833 (1992). In Casey, the Court assessed “an asserted First Amendment right of a physician not to provide information about the risks of abortion, and childbirth, in a manner mandated by the State.” Id., at 884 (joint opinion of O’Con- nor, Kennedy, and Souter, JJ.). More specifically, Casey in- volved a challenge to a Pennsylvania law requiring that “a doctor give a woman certain information as part of obtain- We held that Penn-ing her consent to an abortion.” Ibid. 3 sylvania’s informed-consent mandate did not violate the First Amendment. “To be sure, the physician’s First Amendment rights not to speak [were] implicated” by Penn- sylvania’s law. Ibid. But we emphasized that those rights were implicated “only as part of the practice of medicine, subject to reasonable licensing and regulation by the State.”

Ibid. (emphasis added). Thus, there was “no constitutional

——————

The First Amendment challenge to Pennsylvania’s law in Casey was 3

effectively the flipside of the one Chiles brings here: While the doctors in

Casey complained that Pennsylvania was forcing them to speak when they did not want to, Chiles asserts that Colorado is preventing her from saying what she wants to say.

10 CHILES v. SALAZAR J

infirmity in the requirement that the physician provide the information mandated by the State.” Ibid. In NIFLA, we reaffirmed the principle from Casey that the First Amendment inquiry requires consideration of whether the regulated speech was made during the provi- sion of medical care. NIFLA involved a challenge to a Cal- ifornia law that required certain crisis pregnancy centers to post notices in their waiting rooms informing low-income patients that California paid for qualifying abortions. 585

1. S., at 762–763. We asked whether, on the one hand, this law was regulating the clinics’ speech qua speech, or whether, on the other, the notice requirement was actually regulating the clinics’ professional conduct and only inci- dentally restricting speech. If the latter, the NIFLA Court explained, California’s notice requirement would fit into the category of cases that Casey illustrated; namely, those in which “this Court has upheld regulations of professional conduct that incidentally burden speech.” 585 U. S., at 769. Relying in part on Casey’s analytical framework, the

NIFLA Court held that California’s law regulated “speech

as speech.” 585 U. S., at 770. We explained this conclusion by contrasting the Pennsylvania regulation at issue in Ca-

sey: While the notice requirement in Casey restricted doc-

tors’ speech, it did so “only ‘as part of the practice of medi- cine.’ ” 585 U. S., at 770. The notice requirement at issue

in NIFLA, by contrast, was “not an informed-consent re-

quirement or any other regulation of professional conduct.”

Ibid. “In fact,” California’s notice was “not tied to a proce- dure at all. It applie[d] to all interactions between a covered facility and its clients, regardless of whether a medical pro- cedure [was] ever sought, offered, or performed.” Ibid. (em- phasis added). So, we reasoned, unlike Pennsylvania’s in- formed-consent requirement in Casey, California’s notice mandate warranted heightened scrutiny because it “regu- late[d] speech as speech.” 585 U. S., at 770.

11 Cite as: 607 U. S. ____ (2026) J

The takeaway from NIFLA is that Casey applied a lower level of scrutiny because the law in Casey restricted speech uttered in the course of—and as a part of—providing pro- fessional medical care. By contrast, the notice requirement in NIFLA was not “tied to a procedure at all” and was there- fore meaningfully different: That law restricted “speech as speech.” 585 U. S., at 770. Thus, the key distinction, as the

NIFLA Court saw it, was whether the challenged law was

a regulation of speech as such or a regulation of “profes- sional conduct that incidentally burden[ed] speech.” Id., at

769.

2 Given all this, one might think today’s majority would make more of an effort to explain why the MCTL does not likewise qualify as a regulation of “professional conduct that incidentally burden[s] speech.” Ibid.; see ante, at 14–

1. Such an inquiry would entail evaluating whether the MCTL’s restriction on Chiles’s therapy only “incidentally” restricts Chiles’s speech by virtue of the fact that the medi- cal care she provides is delivered orally. It would also re- quire acknowledging that the MCTL’s restriction on Chiles

is plainly “tied to [the provision of] a [medical] procedure,” NIFLA, 585 U. S., at 770—one that, but for Colorado’s law,

a licensed counselor like Chiles might offer to minors. In my view, it is obvious that the MCTL is regulating pro- fessional conduct insofar as it prohibits providing a partic- ular therapy; the aim of the statute is not suppressing speech. Indeed, Chiles’s claim that her (otherwise pro- tected) speech is being swept up by Colorado’s (otherwise valid) treatment prohibition proves that very point. This set of circumstances seems to fit NIFLA’s idea of permissi- ble state “regulation of professional conduct” that “inci- dentally burdens speech” to a “T.” Yet, the majority strangely suggests otherwise with the opinion it hands down today. Ante, at 14–17. The majority

12 CHILES v. SALAZAR J

does this primarily by eschewing serious engagement with the interaction between NIFLA and Casey. Its workaround seems to be: The First Amendment applies full bore here because Chiles’s speech is being impacted; after all, she is a

talk therapy provider. Ante, at 12–17. But when NIFLA’s teachings are properly understood, this comeback is no an- swer. Yes, Chiles happens to be talking when she’s provid- ing therapy to patients, but the MCTL regulates the provi- sion of medical treatments by licensed medical professionals, which States are fully empowered to do. That Chiles’s kind of medical care involves talk therapy is, in

NIFLA’s words, merely “incidenta[l].” 585 U. S., at 769.

I am the first to admit that, as applied to talk therapists like Chiles, the MCTL restricts speech—I do not argue that this law really just limits Chiles’s professional conduct. See

ante, at 12–13, 16 (characterizing Colorado’s argument). Similarly, I do not maintain that, because this law primar- ily regulates talk-therapists’ professional conduct, it should not be conceived of as a speech restriction. See ante, at 16. I agree with the majority that, in cases like Cohen v. Cali-

fornia, 403 U. S. 15 (1971), we firmly rejected a State’s at-

tempt to suppress free speech by calling the restricted ex- pression “conduct.” Ante, at 16. But, here, the observation that the MCTL indeed restricts Chiles’s “speech” (not refor- mulated as conduct) just raises the question that this case presents: Whether the MCTL is restricting Chiles’s speech “incidentally” to its regulation of medical professionals’ treatment-related conduct, such that the law warrants less scrutiny under the First Amendment than a law that re- stricts her speech “as speech.” NIFLA, 585 U. S., at 769–

770.

The majority’s failure to acknowledge that this is the ac- tual issue here—not just whether Chiles’s “speech” or “con- duct” is being restricted, but what the State is doing—ig- nores what has always been true under our precedents. The real lesson of NIFLA’s discussion of Casey is this: When

13 Cite as: 607 U. S. ____ (2026)

a healthcare professional’s speech is not being targeted “as speech” (because it conveys an idea) but is instead “inci- dentally” restricted due to a State’s otherwise legitimate regulation of the medical treatments being offered to pa- tients, heightened scrutiny is not warranted. 585 U. S., at 769–770; Casey, 505 U. S., at 884 (joint opinion of O’Connor, Kennedy, and Souter, JJ.). 3 NIFLA’s focus on a State’s regulation of medical treat- ments also undermines the position the United States has taken on today’s First Amendment question. The United States, joining this case as amicus curiae, insists, in es- sence, that when the NIFLA Court acknowledged that States can regulate professional conduct while “inci- dentally” restricting speech, it was blessing only two cir- cumstances: (1) where the speech that is being restricted is inextricably tied to an act of the speaker (so, speech-only therapists are never included), or (2) where speech is being regulated in a manner entirely unrelated to its content. Brief for United States as Amicus Curiae 17–21; see ante, at 11, 15–16. But, under NIFLA’s reasoning, that cannot be so—at least not logically. With respect to the first category, the United States mis- takenly swaps “integral” (i.e., whether the restricted speech is bound up with the regulated conduct) for “incidental”— but these are two completely different concepts. Given 4

—————— Under the reading the United States advocates, the line between per-4 missible and impermissible state regulation seems to turn on the rela- tionship between the restricted speech and a physical act of the speaker. See Tr. of Oral Arg. 35–36. So, the argument goes, the First Amendment allows state regulation if the speech restriction is “incidental” to a phys- ical act of the healthcare provider, but prohibits it if there is no physical act for the speech to be “incidental” to because the provider treats pa- tients solely with speech. See id., at 35 (“[T]his Court hasn’t drawn a particularly clear line about when speech is close enough to conduct to be viewed as incidental, but, here, again, this is an easy case because

14 CHILES v. SALAZAR J

NIFLA’s focus on the “incidental” nature of the challenged speech restriction, the State’s objective is the actual ful- crum: We ask whether, on the one hand, the State’s law is restricting the provider’s speech “incidentally” (meaning in the course of the State’s pursuit of its primary objective of regulating the provision of medical care), or whether, on the other, the State’s law is restricting her speech “as speech” (primarily to suppress its message or expressive content). 585 U. S., at 769–770. The appropriate scrutiny level logi- cally follows: Incidentally restricting speech needs less scrutiny because we view the State as generally regulating the provision of medical care, while restricting “speech as speech” receives heightened scrutiny because the State is aiming at professional speech qua speech. The second “speech incident to conduct” category the United States advances—whether the restriction on speech is unrelated to its content—is also fatally flawed. It ignores what NIFLA plainly recognizes: that States can regulate the medical treatments healthcare professionals provide to patients without running afoul of the First Amendment, even if the regulation applies to and restricts speech based on its content. 585 U. S., at 769–770. In other words, what mattered to NIFLA’s analysis was not that the regulation was content-based, but instead that the speech was being restricted incidentally.

—————— there is no conduct”); id., at 33 (emphasizing that, because Chiles is a talk therapist, “[t]here’s no separate non-speech conduct being regulated here”); see also Brief for United States as Amicus Curiae 24, 26 (noting that “regulations of the mental-health profession are less likely to qualify as ‘incidental’ burdens on speech . . . because much mental-health treat- ment is conducted using only speech”). This logic rests upon the integral nature of the speech to the conduct that is being regulated. But that is different from—and says nothing about—whether the speech is being regulated “incidentally,” which is the line that NIFLA draws. See Na-

tional Institute of Family and Life Advocates v. Becerra, 585 U. S. 755, 769–770 (2018).

Cite as: 607 U. S. ____ (2026) 15

So it is here. Talk therapy is a medical treatment. 116

1. 4th, at 1206, 1208–1210. So, why wouldn’t such speech- based medical treatments be subject to reasonable state regulation like any other kind of medical care? The United States and the majority just insist that a law that under- takes to regulate speech-based medical treatments is pre- sumptively unconstitutional because the treatment is being administered solely through speech. But that reasoning is maddeningly circular, and it is based on happenstance, not logic. Even more important, it is not the rationale upon which NIFLA’s analysis relies. To the contrary, with its de- scription of Casey, NIFLA recognizes precisely the oppo- site—a State can regulate professionals’ treatment-related conduct even if doing so impacts treatment-related speech. Again, what distinguished NIFLA from Casey was the fact that Pennsylvania’s speech-related mandate was aimed at regulating the provision of medical treatments to patients; the fact that the particular medical treatment at issue in Casey involved a physical (instead of a verbal) act was of no moment. NIFLA, 585 U. S., at 769–770. The reading of NIFLA the United States favors—which the ma- jority appears to endorse in part, see ante, at 16–17—is ir- rational because, for purposes of the State’s regulation of harmful professional conduct, treatments administered through words versus treatments administered through acts are not meaningfully different. 5

——————

JThe majority appears to adopt a theory of the “speech incident to con-5

duct” doctrine that is analytically similar to, but ultimately narrower than, the recitation adopted by the United States. Ante, at 16–17. But, just like the United States, the majority fails to account for NIFLA’s con- ceptualization of the doctrine as turning on the State’s objectives. In- deed, the majority’s analysis offers no cohesive narrative to explain ei- ther why this exception to heightened scrutiny exists or how—like all exceptions—it operates to consistently effect a balance of the public’s in- terests (here, free speech and the personal safety that medical standards secure).

16 CHILES v. SALAZAR J

By contrast, NIFLA’s actual line marks a real, constitu- tionally relevant distinction: Restrictions of speech that oc- cur when a State undertakes to regulate the treatments that professionals provide to patients are merely “inci- dental”; they are materially different from speech re- strictions that are not “tied to a [medical] procedure at all.” 585 U. S., at 769–770; cf. Black’s Law Dictionary 686 (5th ed. 1979) (defining “incidental” as “[d]epending upon or ap- pertaining to something else as primary,” such as “some- thing incidental to the main purpose”). The latter warrants strict scrutiny since the State is regulating “speech as speech,” while in the former case—where the State is merely restricting speech due to its regulation of medical treatments—heightened scrutiny is not needed. 585 U. S., at 770. The “speech incident to conduct” doctrine thus ably bal- ances the interests at stake by accommodating a State’s tra- ditional police power to regulate the practice of medicine for the protection of its residents while also ensuring that speech is not being targeted. As we explained in NIFLA, that doctrine recognizes that the treatments provided by li- censed medical professionals can be prohibited even if, by doing so, the State incidentally restricts those providers’ speech. Id., at 769–770. This is constitutionally permissi- ble precisely because the restricted speech is not being reg- ulated “as speech”—i.e., based on or due to its message or expressive content. Rather, the speech restriction is a mere byproduct of the State’s healthcare regulation. So, at the end of the day, I think what we have here is what Casey involved and NIFLA did not: a State restricting a medical provider’s speech only as part of its regulation of the provision of medical treatments to individual patients. See NIFLA, 585 U. S., at 769–770; Casey, 505 U. S., at 884 (joint opinion of O’Connor, Kennedy, and Souter, JJ.). And it is precisely because the MCTL is restricting Chiles’s speech “only as part of [her] practice of medicine” that the

17 Cite as: 607 U. S. ____ (2026) J

First Amendment is not particularly bothered despite the impact on her speech. Casey, 505 U. S., at 884 (same). Ac- cordingly, talk therapists like Chiles—just like any other healthcare provider seeking to treat patients—can pre- sumptively be “subject[ed] to reasonable licensing and reg- ulation by the State.” Ibid. B The conclusion that a State can regulate the provision of medical care even if, in so doing, it incidentally restricts the speech of some providers, fully comports with the First Amendment’s animating principles. These principles in- clude the well-settled notion that context matters when evaluating First Amendment challenges to state regula- tion. See Virginia Bd. of Pharmacy, 425 U. S., at 762–768 (analyzing the First Amendment protections due speech in the commercial context); cf. Central Hudson Gas & Elec.

Corp. v. Public Serv. Comm’n of N. Y., 447 U. S. 557, 564,

1. 6 (1980) (explaining how the context and characteristics of commercial speech justify less scrutiny). The context that frames today’s debate is the kind of speech that is at issue here—what I am calling (as short- hand) “professional medical speech.” This is the only type of speech the MCTL restricts. 1 Properly defined, “professional medical speech” is a nar- row category. It is not all speech “uttered by ‘profession- als.’ ” NIFLA, 585 U. S., at 767. Rather, it is speech by

healthcare professionals made as part of their provision of

medical care to patients. To be even more specific, profes- sional medical speech occurs when a medical professional speaks to a client (1) in the context of the professional-pa- tient relationship; (2) on matters within the provider’s pro- fessional expertise as defined by the medical community; (3) for the purpose of providing medical care. See C. Haupt,

18 CHILES v. SALAZAR J

Professional Speech, 125 Yale L. J. 1238, 1247–1248 (2016);

1. Post, Informed Consent to Abortion: A First Amendment
   Analysis of Compelled Physician Speech, 2007 U. Ill.

2. Rev. 939, 947; D. Halberstam, Commercial Speech, Pro-
   fessional Speech, and the Constitutional Status of Social In- stitutions, 147 U. Pa. L. Rev. 771, 834 (1999). First, professional medical speech is speech uttered within the bounds of the professional-patient relationship. See, e.g., Haupt, 125 Yale L. J., at 1254–1255. That rela- tionship imposes certain duties and restrictions on the med- ical professional. For example, medical providers are bound by the twin duties of beneficence (the obligation to act for the benefit of the patient) and nonmaleficence (the obliga- tion not to harm the patient). B. Varkey, Principles of Clin- ical Ethics and Their Application to Practice, 2020 Med. Principles and Prac. 17, 18. 6 Second, professional medical speech is speech within the healthcare provider’s area of expertise as a member of the medical community. Haupt, 125 Yale L. J., at 1248–1251. Within the professional-patient relationship, the profes- sional has knowledge that the patient does not have, includ- ing knowledge of which medical treatments are appropriate and how to administer them. The patient comes to the pro- vider to access that expertise, which is informed by—and constrained by—what the medical community knows. See

id., at 1243.

Finally, and most importantly, professional medical speech is made for the purpose of providing the patient with

—————— This means, of course, that a provider is not employing professional6 medical speech when speaking outside of the professional-patient rela- tionship. The provider who gives a speech touting the benefits of conver- sion therapy, or writes a paper criticizing those who do not practice con- version therapy, or even expresses to a patient her general (non- treatment-related) views about conversion therapy does not have the du- ties that arise in the context of the professional-patient relationship and, accordingly, is not engaging in professional medical speech.

19 Cite as: 607 U. S. ____ (2026) J

medical care. See id., at 1255. This speech is a tool em- ployed to treat patients. In this sense, professional medical speech facilitates the professional’s goal of providing the pa- tient with the treatment, procedure, or healthcare that is within her expertise and that forms the basis of the profes- sional-patient relationship. 2 Keeping in mind these characteristics of professional medical speech, consider the First Amendment principles that serve as guideposts for determining the level of scru- tiny that a government restriction of such speech deserves. First, and most fundamentally, is preservation of the marketplace of ideas. See Abrams v. United States, 250

1. S. 616, 630 (1919) (Holmes, J., dissenting); Meyer v.

Grant, 486 U. S. 414, 421 (1988). Indeed, the “whole project

of the First Amendment” stemmed from the Founders’ de- sire to protect the “critically important” goal of having “a well-functioning sphere of expression, in which citizens have access to information from many sources.” Moody v.

NetChoice, LLC, 603 U. S. 707, 732 (2024). Within the mar- ketplace of ideas, speech that is expressive of the speaker’s thoughts and views is, generally speaking, highly valued. See Leathers v. Medlock, 499 U. S. 439, 447 (1991); Ashcroft

1. American Civil Liberties Union, 535 U. S. 564, 573–574 (2002). But professional medical speech does not intersect with the marketplace of ideas: “[I]n the context of medical prac- tice we insist upon competence, not debate.” Post, 2007 U. Ill. L. Rev., at 950. The degree to which medical providers speaking within the boundaries of providing patient care can express themselves is limited because their interactions with patients are constrained by their well-established du- ties to those patients and the requirement that they meet the standard of care. Moreover, given these limits, profes- sional medical speech does not necessarily involve the

20 CHILES v. SALAZAR J

expression of ideas or messages, so it does not provide sig- nificant value to the general marketplace. See Dun &

Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U. S. 749,

758–759 (1985) (plurality opinion). That’s not to say that there isn’t a robust marketplace of ideas within the medical community. Medical professionals contribute to that particular marketplace by writing pa- pers, giving speeches, and pushing the bounds of the com- munity’s knowledge through experimentation. And, in- deed, the standard of care for a medical treatment can be greatly influenced and changed by virtue of such speech. It is there that truth competes for “accept[ance] in the . . . market.” Abrams, 250 U. S., at 630 (Holmes, J., dissent- ing). But that marketplace exists outside the confines of the professional-patient relationship. See Haupt, 125 Yale L. J., at 1243–1244 (discussing the epistemic marketplace among medical professionals). Within the confines of the professional-patient relation- ship, treatment-related “truths” are a given—they are set by licensing and malpractice standards, and it is not un- common that such regulation incidentally restricts provider speech. Moreover, regulation of the practice of medicine is

pervasively and unavoidably viewpoint based. The majority

and the concurrence both resist this: They relentlessly de- ride Colorado for engaging in “viewpoint discrimination” by banning conversion therapy but permitting affirming care. AGAN, J., Ante, at 13–14 (majority opinion); ante, at 1–2 (K concurring). But context makes that point ring hollow. When a State establishes a standard of care, or punishes a doctor for providing care outside of that standard, it nec- essarily limits what medical professionals can say and do on the basis of viewpoint. A State can permissibly “pro- hibi[t] the administration of specific drugs for particular medical uses” but not for others. United States v. Skrmetti,

21 Cite as: 607 U. S. ____ (2026)

605 U. S. 495, 516 (2025). So, too, may it prohibit a doctor 7 from encouraging a patient to commit suicide, see Tr. of Oral Arg. 43–45, or a dietician from telling an anorexic pa- tient to eat less, see id., at 22–23. Likewise, no one would bat an eye if a State required its doctors to discourage, but not encourage, smoking tobacco. Even though these kinds of regulations are inherently viewpoint based, in the context of medical care, a State can certainly require the medical professionals it licenses to stand on one side of an issue. See Collins v. Texas, 223 U. S. 288, 297–298 (1912) (recognizing the “right of the State to adopt a policy even upon medical matters concerning which there is difference of opinion and dispute”). Though these proscriptions certainly promote a viewpoint, in this context, that alone does not suffice to establish a presumptive First Amendment violation. Instead, under the “speech incident to conduct” doctrine, the challenged laws must also operate as speech-suppression tools, designed to vanquish free ex- pression. But, here, Colorado’s clear aim is enforcement of a stand- ard of care that is indisputably applicable to the State’s li- censed healthcare professionals. Taking a position as to how those providers should handle a medical issue is the very essence of standard-setting—once again, this kind of viewpoint-based regulation ensures “competence, not de- bate.” Post, 2007 U. Ill. L. Rev., at 950. My colleagues’ con- trary conclusions are puzzling, for a standards-based healthcare scheme cannot function unless its regulators are permitted to choose sides. 8

—————— Of course, when the State discriminates “on the basis of sex and 7 transgender status” with respect to the administration of specific drugs, that discrimination implicates the Equal Protection Clause and requires heightened scrutiny for purposes of the Fourteenth Amendment. See

Skrmetti, 605 U. S., at 579 (S JOTOMAYOR, J., dissenting). Faulting Colorado for legislating based on its view that conversion 8 therapy is harmful for minors and that affirming care is the better

22 CHILES v. SALAZAR

A second and corollary First Amendment principle is the listener’s interest in receiving information. See Murthy v.

Missouri, 603 U. S. 43, 75 (2024); accord, Kleindienst v.

Mandel, 408 U. S. 753, 762 (1972). In the professional med- ical context, however, informational asymmetry shapes the listener’s interest. To be sure, “[r]espect for patients’ au- tonomy is a cornerstone of medical ethics.” American Med- ical Association Code of Medical Ethics, Opinion 11.2.4: Transparency in Health Care (2026). But that interest is not served by receiving all existing opinions—only infor- mation about treatments that are within the standard of care advances patients’ interests. Ibid. (“[P]hysicians have

—————— treatment, the concurrence purports to save “for another day” the ques- tion whether “content-based but viewpoint-neutral laws regulating speech in doctors’ and counselors’ offices” comport with the First Amend- ment. Ante, at 4 (opinion of K JAGAN, J.). But that magnanimity is a mi- rage. Standards-based regulations exist in the medical context precisely

because the State has a view about safety or efficacy; regulation is a State’s police-power prerogative to promote those views (as the standard of care) while simultaneously rejecting all others. The laws I reference in Part IV, infra, are not examples of content- based, viewpoint-neutral laws, as the concurrence maintains. Contra,

ante, at 4, n.*. Rather, when properly analyzed, those laws are either

facially viewpoint based—see, e.g., the requirement that the medical pro- fessional must not provide care “in a cruel manner,” Kan. Admin. Regs. 102–3–12a(b)(11) (2022)—or unavoidably viewpoint based in applica- tion. Consider, for example, a therapist disciplined for failing to provide care that promotes the “best interests” of her client. See Ga. Comp. Rules & Regs., Rule 135–7–.01(1) (2026). Punishment for a violation of that standard requires the State to impose its view of what a therapist should have said or done, and would necessarily “reflect the [State’s] disap- proval” of the speech the therapist actually employed. See Matal v. Tam, 582 U. S. 218, 249 (2017) (Kennedy, J., concurring in part and concurring in judgment). But, of course, imposing the State’s view of what is appro- priate is the entire point of standards-based regulation. The First Amendment allows this because the State is regulating professional con- duct and this professional’s speech is only being incidentally restricted; the analysis does not turn on whether the State’s regulation is viewpoint neutral. Neutrality is not—and cannot be—the touchstone of the laws that govern the quality of care that professionals provide to patients.

23 Cite as: 607 U. S. ____ (2026) J

an obligation to inform patients about all appropriate treat- ment options” (emphasis added)). Patients are not in a po- sition to wade through medical discourse and inde- pendently evaluate the best treatment for their circumstances. Their interests as listeners are thus limited by the nature and purpose of the professional-patient rela- tionship. Third, and finally, the First Amendment protects a speaker’s autonomy. “ ‘[T]he fundamental rule of protection under the First Amendment [is] that a speaker has the au- tonomy to choose the content of his own message.’”

Rumsfeld v. Forum for Academic and Institutional Rights,

Inc., 547 U. S. 47, 63–64 (2006) (quoting Hurley v. Irish-

American Gay, Lesbian and Bisexual Group of Boston, Inc.,

515 U. S. 557, 573 (1995)). But, here again, with respect to professional medical speech, healthcare providers do not have autonomy; when it comes to providing treatments for their patients, they are bound by the standard of care and are not generally free to “choose the content” of their mes- sage. See Haupt, 125 Yale L. J., at 1272; Halberstam, 147

1. Pa. L. Rev., at 867. Put differently, although medical professionals do have an autonomy interest in communi- cating their ideas to the patients they are treating, that in- terest only extends to treatment-related advice and infor- mation that is consistent with the standard of care. In my view, the majority is mistaken to equate treat- ment-related speech rendered in the context of providing medical care with any spoken words uttered by any other speaker. See, e.g., ante, at 11 (“While the First Amendment protects many and varied forms of expression, the spoken word is perhaps the quintessential form of protected speech. And that is exactly the kind of expression in which Ms. Chiles seeks to engage”). The majority is also wrong to insist that it is antithetical to the First Amendment for a State to incidentally restrict a healthcare provider’s treat- ment-related speech based on a “prevailing ‘standard of

24 CHILES v. SALAZAR J

care’ ” because “[m]edical consensus . . . is not static; it evolves and always has.” Ante, at 22. The mutability of medical standards tells us little about the First Amend- ment’s scope in a country where medical standards are en- forceable by law and govern the treatment-related conduct of professional healthcare providers. Like it or not, treatment standards exist in America. And those standards necessarily reflect the expert medical com- munity’s current beliefs about the safety and efficacy of var- ious medical treatments, whatever those beliefs might be. Medical standards are driven by science (objective facts and data), but, naturally, they are not viewpoint neutral. Con- sequently, the people win—not lose—when a State incorpo- rates the medical profession’s viewpoint into laws that re- quire licensed treatment providers to conform to prevailing standards of care. Contra, ante, at 22 (suggesting other- wise). For this reason, the Court has long recognized a State’s power to regulate to protect its residents even in the face of uncertainty. Cf. Gonzales v. Carhart, 550 U. S. 124, 163 (2007) (collecting cases and noting the “wide discretion” afforded state legislatures to “pass legislation in areas where there is medical and scientific uncertainty”). 9

—————— The majority laments that, because medical consensus is “not static,” 9 a law like the MCTL might operate to “silenc[e]” professional speech go- ing forward even if medical consensus swings the other way. Ante, at 22. Illustrating this problem, the majority points to shameful parts of this country’s past to show the dangers that can come from regulation that relies on outdated medical practices. Ibid. (citing Buck v. Bell, 274 U. S. 200, 205–207 (1927)). But the majority does not mention that, if the standard of care does change, the state legislature has the power to change the law in response to that evidence. The majority’s point seems to be that States should not be permitted to enact (rigid) laws based on current scientific thought because expert opinions might shift over time. But those uncertainties—which have always existed—are no reason to abandon medical standards or to alter how the law has traditionally ac- commodated scientific discoveries. The potential that medical consensus may change in the future does not mean that the Constitution prevents

25 Cite as: 607 U. S. ____ (2026) J

Put differently, States impose treatment standards incor- porating the current consensus of medical experts to protect state residents from harm. And they do this to ensure that professionals provide patients with high-quality care. A State that, alternatively, pursues an agenda of purposefully silencing critics, muzzling opponents, or targeting views it considers threatening would, of course, violate the First Amendment. But it behooves us all (and especially courts) to see and know the difference. Ultimately, then, no traditional First Amendment princi- ple justifies preventing a State from regulating medical care simply and solely because its law happens to restrict treatment-related speech. And in this case, there is zero evidence that Colorado has engaged in the corrosive and il- licit suppression of ideas that the First Amendment val- iantly repels. The record here does not show that Chiles is being “target[ed]” or “muzzle[d]” or “silenced” or “cen- sor[ed],” as the majority suggests. Ante, at 14, 15, 22, 23. Instead, as a healthcare provider licensed by the State of Colorado, she is simply being held to the same standard of care that all other licensed medical professionals in that The MCTL’s conversion-therapy ban State must follow. 10

—————— a State from acting today to protect its residents from what medical ex- perts currently believe is a harmful medical treatment.

Under my analysis, evidence of speech targeting or suppression10

could include the fact that the challenged state regulation does not, in fact, reflect current medical consensus. See ante, at 22 (noting the mu- tability of the medical consensus). If a State enacts a treatment prohibi- tion that substantially diverges from the medical community’s present beliefs, the law might well be a pretext for illicit speech-targeting objec- tives. Far from requiring “reflexive deference,” ibid., proof of such moti- vation would be unearthed, and carefully examined, as part and parcel of a court’s proper “speech incident to conduct” inquiry, since the doctrine is only applicable to reasonable State regulations. See Planned

Parenthood of Southeastern Pa. v. Casey, 505 U. S. 833, 884 (1992) (joint opinion of O’Connor, Kennedy, and Souter, JJ.) (noting that medical pro- fessionals are “subject to reasonable licensing and regulation by the State”); NIFLA, 585 U. S., at 785 (Breyer, J., dissenting) (stating that

26 CHILES v. SALAZAR J

only incidentally restricts professional medical speech as a result of Colorado’s regulation of a harmful medical treat- ment; nothing compels the conclusion that a state regula- tion that operates to restrict this kind of communication in this way is targeting speech qua speech. III The centuries-long tradition of States using their police powers to establish and enforce the standards of care that bind medical professionals—including those who use speech to administer treatments—is another indication that heightened scrutiny does not and need not apply here. The majority’s opinion largely omits this broader historical record. But, when consulted, that history demonstrates un- equivocally that the MCTL is neither unusual nor inher- ently suspect. States have always had “broad power to establish stand- ards for licensing practitioners and regulating the practice of professions.” Goldfarb v. Virginia State Bar, 421 U. S. 773, 792 (1975). With respect to the medical profession in particular, States have used that power to control how med- icine is practiced “from time immemorial.” Dent v. West Vir-

ginia, 129 U. S. 114, 122 (1889).

States have historically regulated the medical profession in two complementary ways: licensing schemes and medi- cal-malpractice liability. Both necessarily encompass re- strictions on professional medical speech through the regu- lation of the provision of medical care. Medical licensing began as early as 1639, before this country was founded. R. Horowitz, In the Public Interest: Medical Licensing and the Disciplinary Process 39 (2013). Many States deregulated in the mid-1800s by abolishing their licensing schemes. Id., at 40. But regulation through licensing was not abandoned for long: By the turn of the

—————— the First Amendment yields to “reasonable conditions” that States im- pose on medical providers).

27 Cite as: 607 U. S. ____ (2026) J

20th century, 12 States had adopted licensing laws. D. Johnson & H. Chaudhry, Medical Licensing and Discipline in America: A History of the Federation of State Medical Boards 23 (2012). We held that such laws were a permissi- ble exercise of a State’s traditional police powers, declaring that “[t]he power of the state to provide for the general wel- fare of its people authorizes it to prescribe all such regula- tions as, in its judgment, will secure or tend to secure them against the consequences of ignorance and incapacity as well as of deception and fraud.” Dent, 129 U. S., at 122; see also Watson v. Maryland, 218 U. S. 173, 176 (1910) (reject- ing challenge to medical-licensing law). Today, every State has a medical-licensing scheme. See App. to Brief for Health Law Scholars as Amici Curiae. In practice, medical licensing serves two functions, both essen- tial to patient safety. First, as the majority emphasizes, li- censing sets limits on who may practice medicine. See P. Larkin, M. Fishpaw, & L. McCarthy, Telemedicine and Oc- cupational Licensing, 73 Admin. L. Rev. 747, 774 (2021);

ante, at 20. But the second function is more important for

this case: State licensing laws also regulate how those pro- fessionals may practice, by requiring them to adhere to a standard of care. See Brief for Respondents 27; see also App. to id., at 1a–8a (listing state laws that require medical and mental health professionals to be licensed and to com- ply with professional standards). While licensing regulates medical professionals ex ante, medical-malpractice lawsuits enforce those standards

ex post. And just like medical licensing, the tort of medical malpractice has a long pedigree. English common law held doctors liable for harm caused by their negligent medical treatment. 3 W. Blackstone, Commentaries on the Laws of England *122 (W. Lewis ed.

1. (1768). The English tradition carried forward; this legal claim has been available in our country since the founding. See K. De Ville, Medical Malpractice in

28 CHILES v. SALAZAR J

Nineteenth-Century America: Origins and Legacy 3, 5 (1990) (De Ville). In the mid-1800s, the number of malprac- tice cases skyrocketed, filling the void left by the States’ temporary deregulation of medical licensing. See id., at 25–

34; see also id., at 115–137; Graham v. Gautier, 21 Tex. 111, 117–118 (1858) (observing that the lack of a licensing scheme demanded enforcement of a corresponding stand- ard of care). Historically, the medical-malpractice tort generally re- quired the plaintiff to prove three things: (1) that there was a professional-patient relationship, (2) that the medical professional had caused him harm, and (3) that the pro- vider had departed from a standard of care. De Ville 46–

1. The majority focuses on the requirement to show harm.

Ante, at 21. But the entire point of the third element was

to reinforce the standards that govern medical practition- ers. For that reason, the harmed plaintiff was required to establish that the accused physician had not practiced with “ordinary care, diligence, and skill.” See De Ville 49 (em- phasis deleted). State enforcement of the standard of care—i.e., “the care, skill, and knowledge regarded as competent among similar medical providers in the same or similar circumstances,” Restatement (Third) of Torts: Medical Malpractice §5 (Tent. Draft No. 2, Mar. 2024)—has continued over time and still serves as the touchstone for both licensing schemes and medical-malpractice lawsuits. Brief for Respondents 25– 26, and App. to id., at 1a–8a. That is, States have consist- ently regulated medical professionals’ conduct to ensure that modern healthcare practices conform to the standard of care through both medical-malpractice law and profes- sional licensing. Note, too, that such state regulation has not been limited to medical procedures: Physicians have historically been held liable for what they said when administering those procedures as well. See Graham, 21 Tex., at 119–120

29 Cite as: 607 U. S. ____ (2026) J

(imposing liability for medical advice); Edwards v. Lamb, 69 N. H. 599, 45 A. 480 (1899); Skillings v. Allen, 143 Minn. 323, 173 N. W. 663 (1919). Moreover, in the past, States have—as part of their licensing regulations—defined the “practice of medicine” to cover practitioner speech. See, e.g.,

Smith v. People, 51 Colo. 270, 272, 117 P. 612, 613 (1911)

(noting that the “practice of medicine” as regulated by Col- orado’s licensing law included “suggestion[s or] recommen- dation[s] . . . of treatment”). In short, States have regulated professional conduct re- lated to the provision of all kinds of medical care—and inci- dentally restricted speech—without constitutional affront for eons. Though the majority averts its gaze, even a cur- sory glance at the broader historical record is illuminating, for it reveals that States have traditionally played a signif- icant role in setting the standards that govern the medical profession. See Washington v. Glucksberg, 521 U. S. 702, 731 (1997) (emphasizing the state interest “in protecting the integrity and ethics of the medical profession”); Barsky

1. Board of Regents of Univ. of N. Y., 347 U. S. 442, 451 (1954) (same). With the MCTL, Colorado has merely taken up that same mantle. That law operates by prohibiting a particular med- ical treatment the State considers harmful, and nothing about it implicates Chiles’s First Amendment rights in a markedly different fashion than other States’ traditional ef- forts to regulate and enforce the standard of care. 11

—————— The majority’s observation that “counselor-licensure bill[s]” are a rel-11 atively recent innovation, ante, at 20, is an interesting diversion. But that is all. The relevant historical question is whether States have his- torically regulated the medical care that licensed professionals provide to patients, including treatments that are delivered via speech. The an- swer is yes; the fact that counselors have only recently been included in the regulated category of “licensed medical professionals” is beside the point.

30 CHILES v. SALAZAR

One more thought on this: The majority rigidly imposes a history-and-tradition test that treats the plethora of his- torical examples as insufficient. See ante, at 18–22. But it should instead find the long tradition of state laws setting standards of care by regulating the professional conduct of medical providers—including those who treat with speech—doubly reassuring. For one thing, this history helps us to be confident that what Colorado is doing here is actually regulating medical care, not suppressing messages. The record shows that States have routinely enacted laws that establish and en- force the standard of care, and that serves as a backdrop for an understanding of how States have acted historically to protect their residents from harm. The majority is right about one thing, however: A State will always say that its law just regulates the provision of medical treatments, while the challenger will inevitably argue that the State’s law nefariously targets speech qua speech. See ante, at 16. A lengthy tradition of similar regulatory efforts by States— or the absence of one—helps courts to ferret out who has the better of that argument. The history also helpfully demonstrates that a lower level of scrutiny is appropriate here, despite the impact of the MCTL on Chiles’s speech. We can rest easy, comforted by the fact that this law is not actually operating to suppress the expression of thoughts, messages, or ideas about con- version therapy; instead, the MCTL restricts talk thera- pists in the same way and to the same extent as other healthcare professionals have historically been limited when treating patients. Like other valid licensing re- strictions, the MCTL does not prevent Chiles from speaking out in favor of conversion therapy, promoting conversion therapy, or otherwise lending credence to efforts to validate that therapy. All this law does is prohibit Chiles from

providing this treatment to minor patients—no different

than what Colorado and other States have been doing in the

31 Cite as: 607 U. S. ____ (2026) J

indisputably valid exercise of their police powers for centu- ries. 12 All things considered, then, I reach a different conclusion in this case than the majority does because precedent, prin- ciples, and history point in the same direction: No height- ened scrutiny is warranted here. The First Amendment cares about government efforts to suppress “speech as speech” (based on its expressive content), not laws that, like the MCTL, restrict speech “incidentally,” due to the govern- ment’s traditional, garden-variety regulation of such speak- ers’ professional conduct.

—————— Suggesting otherwise, the majority places great stock in our decision 12 in Holder v. Humanitarian Law Project, 561 U. S. 1 (2010). See ante, at 10, 16. Holder involved a law that prevented lawyers and doctors from providing “ ‘material support’ ” for others’ terrorist activities by word or deed. 561 U. S., at 8–9. We subjected the law to strict scrutiny because, as applied to the plaintiffs, the law was aimed at preventing profession- als from “communicating a message.” Id., at 28. Such a regulation plainly raised the specter of suppression—i.e., that what the United States was really aiming to do was prevent those professionals from ex- pressing support for something the United States found distasteful. In other words, the challenged law sought to punish the plaintiffs based on the expressive content of their speech. That is not what we have here. The MCTL—which follows in a long line of state regulation of healthcare providers’ treatment-related conduct—does not restrict or punish medi- cal professionals because of the expressive content of their communica- tions. Rather, the speech restriction happens only incidentally; the MCTL’s indisputable objective is prohibiting a harmful medical treat- ment. To put a finer point on this: The equivalent of Chiles’s First Amendment claim, transported to the Holder context, would be as if the United States in Holder had said: “you professionals are prohibited from committing acts of terrorism,” and the lawyers among them responded, “it is unconstitutional to apply your ‘no terrorist acts’ prohibition to us because we want to commit the prohibited terrorist acts with our speech.” But, of course, the First Amendment would not prevent the United States from prohibiting all terrorist acts even if, by doing so, it inci- dentally restricts the speech that some actors might otherwise have used to behave in the manner the law prohibits. In my view, that is how Chiles’s constitutional claim works (and also why it fails).

32 CHILES v. SALAZAR J

IV Ultimately, because the majority plays with fire in this case, I fear that the people of this country will get burned. Before now, licensed medical professionals had to adhere to standards when treating patients: They could neither do nor say whatever they want. Largely due to such State reg- ulation, Americans have been privileged to enjoy a long and successful tradition of high-quality medical care. Today, the Court turns its back on that tradition. And, to be completely frank, no one knows what will happen now. This decision might make speech-only therapies and other medical treatments involving practitioner speech effec- tively unregulatable—not to be reached via licensing stand- ards, medical-malpractice liability, or any other means of state control. Who knows? Certainly not the majority. It appears to have made this momentous decision without ad- equately grappling with the potential long-term and disas- trous implications of this ruling. The fallout could be catastrophic. Many regulations im- pact the speech of medical professionals in the context of their provision of healthcare to patients; the possibilities go far beyond talk therapy and informed consent. For exam- ple, many States require that medical professionals “make every reasonable effort to promote the welfare, autonomy and best interests of ” the client. Ga. Comp. Rules & Regs., Rule 135–7–.01(1) (2026); see Ind. Admin. Code, tit. 839, §1–§5–5(1) (2026); Conn. Gen. Stat. §17a–542 (requiring “[h]umane and dignified treatment”) (2025); Ala. Admin. Code Rule 255–X–11–.01 (Supp. 2016) (requiring that the professional “assure client welfare and protection” during medical care). Some States further prohibit medical profes- sionals from treating a patient “in a cruel manner.” Kan. Admin. Regs. 102–3–12a(b)(11) (2022). Similarly, some

33 Cite as: 607 U. S. ____ (2026)

licensing boards can discipline a provider who “is incompe- tent.” Alaska Stat. §08.29.400(4) (2025). 13 On the majority’s view, these kinds of regulations become unenforceable if the healthcare provider risks harming pa- tients with their speech rather than an operation. Provid- ers who offer “cruel” speech-only therapies or who use speech to (intentionally or incompetently) harm the welfare of patients, for example, can now assert a First Amendment right to carry on, regardless of these standards. So, to put it bluntly, the Court could be ushering in an era of unprofessional and unsafe medical care administered by effectively unsupervised healthcare providers. A state license used to mean something to the patients who entrust their care to licensed professionals—i.e., that the person is certified to be one who provides treatments that are con- sistent with the standard of care. That stops today. Indeed, it is not at all clear how, or to what extent, state regulation of medical care involving practitioner speech can survive this holding. We are on a slippery slope now: For the first time, the Supreme Court has interpreted the First Amendment to bless a risk of ther- apeutic harm to children by limiting the State’s ability to regulate medical providers who treat patients with speech. What’s next? In the worst-case scenario, our medical sys- tem unravels as various licensed healthcare professionals— talk therapists, psychiatrists, and presumably anyone else who claims to utilize speech when administering

—————— Those are not the only restrictions that constrain the speech of med-13 ical professionals. Some States require that a counselor make a treat- ment plan specifying goals and methods, and ensure that such plan is “viab[le] and effectiv[e].” Ariz. Admin. Code, Rule R4–6–1102 (Sept. 2025). Others prohibit “making claims of professional superiority that one cannot substantiate” or “guaranteeing that satisfaction or a cure will result from the performance of professional services.” Kan. Admin. Regs. 102–3–12a(b)(31), (32). Still others prohibit the professional from “exer- cis[ing] undue influence on the client.” Colo. Rev. Stat. §12–245–224(j).

34 CHILES v. SALAZAR J

treatments to patients—start broadly wielding their new- found constitutional right to provide substandard medical care. It is baffling that we could now be standing on the edge of a precipitous drop in the quality of healthcare services in America. But the Court sees fit to bring us one step closer to that fate today. Stranger still is the fact that this possi- bility looms in the 21st century—given what science now en- ables us to know about medical conditions and treatments, what our cases say, and what we all should have learned by now from history. Somehow, Justices from eras past have always understood that (as I stated at the outset) “there is no right to practice medicine which is not subordinate to the police power of the States.” Lambert, 272 U. S., at 596. They correctly applied that simple but powerful under- standing of our Constitution across the board—to all healthcare professionals, including those with practices that happen to involve treatment-related speech. We do harm to both the Nation’s medical system and our First Amendment jurisprudence by ignoring that wisdom today.

• * * The First Amendment requires heightened scrutiny when States regulate “speech as speech” but not when speech is restricted “incidentally.” NIFLA, 585 U. S., at 769–770. The latter occurs where, as here, a State seeks to prohibit healthcare professionals from providing a danger- ous medical treatment in all of its forms, including the speech-related variety. States have traditionally regulated the provision of medical care through licensing schemes and malpractice regimes without constitutional incident. And no core principle of our First Amendment jurisprudence leads inexorably to the conclusion that it violates the Con- stitution for a State to prevent its licensed talk therapists from using speech to harm the minors in their care. Hold- ing otherwise, as the majority does now, flouts centuries of

35 Cite as: 607 U. S. ____ (2026) J

state-standardized regulation of medical care and is, ulti- mately, nonsensical. The Constitution does not pose a bar- rier to reasonable regulation of harmful medical treatments just because substandard care comes via speech instead of scalpel. Accordingly, I cannot agree with the majority’s analysis or its conclusions in this case. The majority finds, at bot- tom, that Colorado likely cannot legislate to protect the chil- dren of its State if, by doing so, it happens to keep state- licensed healthcare providers from saying what they want to say to minors. And the majority’s holding means, in ef- fect, that just because Chiles is a talk therapist—and not, say, a surgeon—a State can be prevented from incidentally imposing reasonable restrictions on the treatments she pro- vides. Our precedents do not compel this conclusion. In fact, NIFLA draws a different line, and the correct course of action here is to hold it: Speech uttered for purposes of providing medical treatment may be restricted incidentally when the State reasonably regulates the speaker’s provi- sion of medical treatments to patients. Id., at 769–770. To do anything else opens a dangerous can of worms. It threatens to impair States’ ability to regulate the provision of medical care in any respect. It extends the Constitution into uncharted territory in an utterly irrational fashion. And it ultimately risks grave harm to Americans’ health and wellbeing.