Texas Department of Insurance and Cassie Brown, in Her Official Capacity as Commissioner of the Texas Department of Insurance v. Stonewater Roofing, Ltd. Co.

Supreme Court of Texas
══════════
No. 22-0427
══════════

Texas Department of Insurance and Cassie Brown, in her Official
Capacity as Commissioner of the Texas Department of Insurance,
Petitioners,
v.
Stonewater Roofing, Ltd. Co.,
Respondent

═══════════════════════════════════════
On Petition for Review from the
Court of Appeals for the Seventh District of Texas
═══════════════════════════════════════

JUSTICE YOUNG, concurring.

I join the opinion of the Court but write separately to emphasize
my understanding of two points with significance for future cases.
First, the federal due-process clause (or the Texas due-course
clause) is generally satisfied when the State rationally regulates
professional conduct to ensure competence and safeguard public safety.
Stonewater does not challenge the statutory licensure and conflict-of-
interest requirements on due-process grounds but as violations of the
First Amendment’s free-speech clause. Because of the Court’s narrow
statutory construction, any speech that that the statute touches is only
incidental to conduct, the regulation of which is of even less concern to
the First Amendment than it is to the due-process clause. I agree with
the Court, therefore, that today’s case turns out to be easier than it might
first have appeared. But I emphasize that, to benefit from today’s
holding, it is not enough for the State to call something conduct. The
State wins today despite, not because of, its overweening theory of what
constitutes “conduct” that it may subject to professional regulation.
Second, I am concerned about what comes next. Not every case
will be so easy. It will not be possible or proper to construe every statute
that regulates professions as only incidentally burdening speech and
targeting only non-expressive conduct. What then? As it stands today,
the relevant First Amendment doctrine is a mind-numbing morass of
tangled precedents developed in contexts very different from professional
licensing. There is just enough of a whiff of original meaning to disguise
a stew of ad hoc conclusions—the way that heavy sauces can fool diners
into enjoying meat that sat for hours out in the sun.
The doctrine as we have it seems poorly equipped to address
legitimate public-licensing regulation that does affect speech or expressive
conduct more than “incidentally.” The outcome is basically determined
at the first move: if conduct, the regulation survives; if speech, it is
doomed. Worse, the conduct-speech dichotomy is, to put it mildly, rather
malleable. The First Amendment can surely be obeyed with better
rationales—and recent cases provide some reason for hope that the U.S.
Supreme Court will clarify and rationalize its jurisprudence. If so, the
speech implications for professional licensure will likewise become clearer.
I

The State’s theory of its authority to impose professional licenses
without violating the First Amendment is too vast.

2
A

The normal framework for challenging professional licensure
sounds in due process. “Competence” and “public safety” are the kinds of
neutral criteria that the police power allows the State to invoke to defend
regulations like licensing regimes. “[T]he state may have an interest in
shielding the public against the untrustworthy, the incompetent, or the
irresponsible, or against unauthorized representation of agency. A usual
method of performing this function is through a licensing system.”
Thomas v. Collins, 323 U.S. 516, 545 (1945) (Jackson, J., concurring).
Stonewater does not contend that the statutory provisions it challenges
fail this constitutional requirement. We must assume that the licensing
and conflict-of-interest provisions are important and rational measures
to protect the public, and indeed various amici have explained in detail
why this is so. See, e.g., Brief for American Property Casualty Insurance
Association et al. as Amici Curiae 15–17 (highlighting the risks to the
public of unlicensed and unregulated contractors pocketing money from
settled insurance claims).
Legislation must satisfy not only due process, of course, but also
every other constitutional requirement. And with the expansion of
professional licensing to a greater number of professions, challenges like
this one increasingly sound in free speech. Asserting that a licensure
requirement burdens protected speech does not make it so, but neither
is it inherently implausible. Imagine licensing not just the structural
engineer who will ensure that a new cathedral does not collapse, but also
the bishop who will preach in it. Or not just the truck driver who
transports stacks of hot-off-the-press newspapers, but the journalists who

3
write the articles printed in them. Likewise for poets, painters, political
consultants, and on and on. Would such licenses satisfy the free-speech
clause (and perhaps other clauses)?
The scope of the State’s theory is not entirely clear. But, as I
understand it, that theory encompasses examples like these by converting
speech into conduct, much as nominalizations convert verbs into nouns:
the “act” of doing a job that involves speech, especially when it is a paid
“act.” Under this view, only “conduct” is reached—“journalism” becomes
the “act” of taking money from employers to produce news articles for
those employers, for instance. The State’s theory seems to be that it gets
to decide who is competent to undertake conduct and can impose a
licensure requirement without offending—or even implicating—the free-
speech clause. The State’s argument appears to at least agree that “pure
speech” cannot be nominalized into mere conduct to evade First-
Amendment review—but I am not sure the State truly concedes that any
profession involves “pure speech.” It describes “prototypical professional
conduct” as “taking defined actions on behalf of a client in exchange for
payment.”
The problem, therefore, is not that the State denies that expression
is protected. Rather, it is how broadly the State may seek to define
“conduct.” And the more broadly one defines “conduct”—using the
formula “acting as [fill-in-the-blank]”—the less room there is for speech.
The less speech, the less likely that any regulation is subject to an
exacting judicial inquiry. Thus, even accepting the true rule that the
First Amendment permits only incidental burdens on speech without
heightening the scrutiny, the effect of that rule depends entirely on what

4
we classify as speech, conduct, expressive, non-expressive, and the like.
Said differently, it is understandable—in the context of someone
being paid to do a job, after all—to regard the resulting work as just paid-
for conduct rather than something that implicates the First Amendment.
In many contexts that is true enough. But in others, it is just a way to
subtly erase the role of the First Amendment. Jack Phillips was paid to
make cakes—and so the State of Colorado did not think it was a big deal
to demand that he toe the line, despite the expressive nature of the
custom cakes that he designed to convey deep meaning. See Masterpiece
Cakeshop, Ltd. v. Colo. Civ. Rts. Comm’n, 584 U.S. 617 (2018); see also
303 Creative LLC v. Elenis, 600 U.S. 570 (2023) (protecting a graphic-
design maker from being compelled to create expressive designs when the
designer disagrees with the messages the designs convey).
Our conduct-speech dichotomy lends itself to confusion and abuse
because conduct and speech are not hermetically sealed categories.
Burning a piece of cloth is conduct; banning the burning of cloth in public
spaces regulates that conduct. But banning burning pieces of cloth in
public only if the cloth has alternating red and white stripes and a blue
field studded with fifty white stars is to regulate conduct that is imbued
with speech. See Texas v. Johnson, 491 U.S. 397 (1989). I make such
elementary points precisely because the State’s theory—and I guess I
cannot blame it for trying—seems largely to elide, or at least downplay,
these fundamental principles. Its focus is on “acting” in a given way,
transforming it into “conduct” that can be regulated, period.
I understand the Court to reject the State’s blunt theory, too. The
most important word of the most important sentence in the opinion, to

5
me, is “nonexpressive”: “The gravamen of the defined profession is the
role a person plays in a nonexpressive commercial transaction, not what
anyone may or may not say.” Ante at 14. Without the word
“nonexpressive,” I do not see why the State could not require licenses and
impose restrictions (like those here) for portraitists, political consultants,
journalists, ministers of the Gospel, and so many others, including as to
the parts of their jobs that only convey messages.
The Court also, however, emphasizes that, in this case, “the
profession’s actuating activity and dominant focus is employment in a
representative (or agency) capacity.” Id. Although I join the Court’s
opinion, I do so only on the understanding that this sentiment, which is
sprinkled throughout, is understood to address nonexpressive conduct. To
be honest, I doubt that “representative (or agency) capacity,” by itself, has
anything to do with the First Amendment analysis. Speech and
expressive conduct are no less protected because they are made on behalf
of another or for compensation. “[T]he First Amendment extends to all
persons engaged in expressive conduct, including those who seek profit
(such as speechwriters, artists, and website designers).” 303 Creative,
600 U.S. at 600. “It is well settled that a speaker’s rights are not lost
merely because compensation is received; a speaker is no less a speaker
because he or she is paid to speak.” Riley v. Nat’l Fed’n. of the Blind of
N.C., Inc., 487 U.S. 781, 801 (1988).
Agency, in my view, is therefore all but irrelevant to the First
Amendment analysis—except that it is too easily looked upon to abridge
First-Amendment rights. Regulating expressive conduct taken “in a
representative (or agency) capacity” is subject to no less scrutiny than the

6
same conduct taken in a non-representative or non-agency capacity. It is
not the details of the commercial relationship that matter, but whether
expression is at the core of the undertaking.
Because the Court concludes that the statute, as construed,
regulates nonexpressive conduct, there is no option but to reverse. As I
describe below, however, a harder case will come to us, sooner or later.
It could even come in this context—the legislature could, for example,
amend the law at issue in a way that really does constrict speech.

B

Today’s narrow statutory construction allows the State to bar
Stonewater from undertaking Stonewater’s desired conduct—adjusting
insurance claims while financially benefiting from that work in a
different capacity. But today’s construction allows Stonewater to do quite
a lot that the government may want to forbid—including discussing, in
detail, the damage and costs of repair with the insurance company. See
ante at 24 (opinion of the Court); ante at 2–3 (Blacklock, J., concurring in
the judgment). This narrow construction avoids serious constitutional
problems.
Even if the legislature amends the law to impose greater
restrictions on parties like Stonewater, any burdens on speech may still
be merely incidental to the conduct that the Court describes. The vast
majority of states, as the Court observes, see ante at 2, 6 n.17, do what the
two challenged statutes do here: (1) require that insurance adjusters be
licensed and (2) prevent conflict-of-interest problems by prohibiting dual-
capacity arrangements (preventing the same actor from being both a
contractor and an insurance adjuster).

7
The State’s argument about why its regulation is permissible, even
with a less narrowly construed statute, emphasizes cases like the Eleventh
Circuit’s decision in Del Castillo v. Secretary, Florida Department of
Health, 26 F.4th 1214 (11th Cir. 2022). That case upheld a Florida law
that regulates the practice of nutrition and dietetics, including “advising
and assisting individuals or groups on appropriate nutrition intake by
integrating information from the nutrition assessment.” Fla. Stat.
§ 468.503(10). Del Castillo alleged that the law violated her First
Amendment free-speech rights. The Eleventh Circuit held that
“[a]ssessing a client’s nutrition needs, conducting nutrition research,
developing a nutrition care system, and integrating information from a
nutrition assessment are not speech. They are ‘occupational conduct’;
they’re what a dietician or nutritionist does as part of her professional
services.” Del Castillo, 26 F.4th at 1225–26. The court therefore concluded
that the licensing scheme for dieticians and nutritionists regulated
professional conduct and only incidentally burdened speech. Id. at 1226.
The First Amendment, it held, did not require heightened scrutiny. Id.
If a dietician’s “advising” or “counseling” people about nutrition is
not speech, then public insurance adjusting is not either. Like providing
nutrition and diet counseling, public insurance adjusters must engage
mainly in conduct for which communicating with an insurer is only
incidental. The Court recognizes the important conduct involved in
settling an insured’s claim: “evaluating insurance coverage, assessing
property value, assessing property damage, and calculating repair costs”
and ultimately “payment, satisfaction, or final adjustment” of the damage
claim. Ante at 15–16. Stonewater’s position certainly does not lack force,

8
but especially as the Court construes the statute, communicating the
value of the claim to the insurer is not remotely the defining characteristic
of a public insurance adjuster’s job. It is as incidental to professional
conduct as speech can be—far more incidental than communicating a
diet and nutrition plan formulated according to professional standards,
which is why the Eleventh Circuit reached the result it did in Del Castillo.
In other words, if the dietician’s job is mainly to figure out what the health
and nutrition needs of a client are, the public insurance adjuster’s
primary role is to figure out what a proper insurance claim is in light of a
damage-causing event.
Assessing the constitutionality of the specific provisions
Stonewater challenges—Insurance Code §§ 4102.051(a) and 4102.163(a)—
is therefore straightforward. Section 4102.051(a) bars an individual from
acting or holding himself out as a public insurance adjuster without a
license. The first half of this prohibition is constitutional under the
analysis laid out above. I agree with the Court and with Justice Blacklock
that the latter part of this prohibition is constitutional too—holding
oneself out as an adjuster restricts speech, but it merely bars false
commercial statements, which the Constitution does not protect. Ante at
12–13 (opinion of the Court); ante at 3 (Blacklock, J., concurring in
judgment); see, e.g., Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n
of N.Y., 447 U.S. 557, 563–64, 566 (1980).
In my view, at least as to this case, the Court correctly concludes
that there is no colorable basis for a constitutional challenge. To prevail,
the State does not need such an overreaching theory.

9
II

On the other hand, it is not clear that Del Castillo is exactly right.
Really, no speech interests are involved in a dietician giving nutrition
advice and counseling? That work is nothing but nonexpressive conduct?
Is the conduct–speech divide even the right line to look for—does
it ask the right questions? Current First Amendment doctrine does not,
formally at least, care too much about the context in which speech is
infringed, including if it is a statute imposing professional-licensure
requirements that does so. “[T]he traditional conduct-versus-speech
dichotomy” remains the doctrinally mandated way to determine whether
such a requirement violates the First Amendment. Vizaline, L.L.C. v.
Tracy, 949 F.3d 927, 932 (5th Cir. 2020) (citing Nat’l Inst. of Fam. & Life
Advocs. v. Becerra, 585 U.S. 755, 771–75 (2018) (NIFLA)).
In cases like this one that challenge a professional-licensure
regime, therefore, courts must ask whether the regulation burdens
(1) speech, (2) conduct, or (3) speech incidental to conduct. The answer to
this question may raise additional questions—if the answer is conduct,
for example, whether it is inherently expressive conduct. The Court today
concludes that this case is easy. Even so, drawing the line is often hard.
Del Castillo strikes me as harder than today’s case—but even assuming
that case got it just right, cases pushing the line are coming.

A

When they come, how will courts react? One problem is that
drawing the line is somewhat questionable. The “enterprise of labeling
certain verbal or written communications ‘speech’ and others ‘conduct’ is
unprincipled and susceptible to manipulation.” King v. Governor, N.J.,

10
767 F.3d 216, 228 (3d Cir. 2014), abrogated on other grounds by NIFLA,
585 U.S. 755. The Eleventh Circuit—which decided Del Castillo—has
made the exact same observation. Wollschlaeger v. Governor, Fla., 848
F.3d 1293, 1308 (11th Cir. 2017) (en banc) (quoting King, 767 F.3d
at 228); see also, e.g., Otto v. City of Boca Raton, 981 F.3d 854, 865 (11th
Cir. 2020). If courts must decide these cases by drawing this line, and if
drawing this line is a manipulable exercise, then it may turn out that
First Amendment jurisprudence does care an awful lot about the context
in which a challenge arises, even if the formal doctrine purports to be the
same always and everywhere.
The importance of protecting speech means that courts should be
wary of this conduct-speech dichotomy, and particularly of too quickly or
without rigorous reasons concluding that a regulation touches just
conduct. It is precisely because the doctrine formally does not distinguish
among contexts that a lack of discipline in this area is troubling. If it is
comparatively easy to hold that any professional licensure is “just
conduct,” what will stop regulation of the same “conduct” outside that
regulatory context? Justice Cardozo wrote of “the tendency of a principle
to expand itself to the limit of its logic.” Benjamin N. Cardozo, The Nature
of the Judicial Process 51 (1921). Where is the limit here, exactly, that
does not threaten individual liberties more generally? This concern is one
of the reasons why I observed above that it is a mistake to be led into
thinking that there is any lesser protection for those who are engaged in
remunerative conduct in an “agency” or “representative” capacity.

B

The distinction between speech and conduct is also significant

11
because it dictates the appropriate level of scrutiny that courts must
apply. A regulation invites strict scrutiny, which is typically fatal, when
the regulation “ ‘targets speech based on its communicative content’—that
is, if it ‘applies to particular speech because of the topic discussed or the
idea or message expressed.’ ” City of Austin v. Reagan Nat’l Advert. of
Austin, LLC, 596 U.S. 61, 69 (2022) (alteration omitted) (quoting Reed
v. Town of Gilbert, 576 U.S. 155, 163 (2015)). Regulation of a different
profession—one that involves expressive conduct (perhaps the legal or
medical professions)—might demand O’Brien’s intermediate scrutiny,
which is sometimes fatal but possible to survive. See United States v.
O’Brien, 391 U.S. 367 (1968). Or we may apply no scrutiny under the
First Amendment at all when, as here, the target is legitimately seen as
nonexpressive conduct. See, e.g., Hines v. Alldredge, 783 F.3d 197, 201–
02 (5th Cir. 2015) (citing Lowe v. SEC, 472 U.S. 181, 211 (1985) (White,
J., concurring in the result)), abrogated on other grounds by NIFLA, 585
U.S. 755. In such cases, only rational-basis review under the due-process
clause remains—and it is rare indeed for a governmental action to
succumb to that level of scrutiny.
Combining the conduct-speech dichotomy’s apparent malleability
with the all-but-determinative level of scrutiny yields serious problems
for courts, the other branches of government, and the regulated public.
Public suspicion that courts work backwards—that they categorize laws
so that the desired level of scrutiny applies, not the other way around—
could follow from a perception, whether fair or not, that this endeavor
lends itself both to inconsistent categorization (speech, expressive
conduct, neither) and to wildly divergent results based on the chosen

12
categorization. At the least, the very risk of inconsistency makes the
whole endeavor open to the charge.
I am hardly alone in wondering if the tiers of scrutiny are moored
in the Constitution’s text and original meaning. Jurists from every
perspective have expressed concern. “[T]he label the Court affixes to its
level of scrutiny in assessing whether the government can restrict a given
right—be it ‘rational basis,’ intermediate, strict, or something else—is
increasingly a meaningless formalism. As the Court applies whatever
standard it likes to any given case, nothing but empty words separates
our constitutional decisions from judicial fiat.” Whole Woman’s Health v.
Hellerstedt, 136 S. Ct. 2292, 2326–27 (2016) (Thomas, J., dissenting)
(emphasis added). Compounding the problem, judges treat the tiers of
scrutiny as “guidelines informing [their] approach to the case at hand, not
tests to be mechanically applied.” Williams-Yulee v. Fla. Bar, 575 U.S.
433, 457 (2015) (Breyer, J., concurring). “Such an amorphous inquiry
risks . . . judges upholding or invalidating . . . laws at will—without
respect to the original public meaning of the” relevant constitutional
provision. United States v. Jimenez-Shilon, 34 F.4th 1042, 1051–52 (11th
Cir. 2022) (Newsom, J., concurring).
Scholars have reasonably doubted whether the tiers of scrutiny
“have [any] basis in the text or original meaning of the Constitution. They
emerged as a political solution invented by the justices to navigate
internal factions at the Supreme Court, and they do not withstand critical
analysis even on their own terms.” Joel Alicea & John D. Ohlendorf,
Against the Tiers of Constitutional Scrutiny, 41 Nat’l Affs. 72, 73 (2019).
Using tiers of scrutiny to “displace longstanding national traditions as the

13
primary determinant of what the Constitution means” is troubling at best.
United States v. Virginia, 518 U.S. 515, 570 (1996) (Scalia, J., dissenting).
Today’s Supreme Court—some Justices more than others—
appears to have found new doubts about the proper role of the tiers of
scrutiny. See, e.g., N.Y. State Rifle & Pistol Ass’n v. Bruen 597 U.S. 1
(2022); Students for Fair Admissions, Inc. v. President & Fellows of Harv.
Coll., 600 U.S. 181 (2023) (Thomas, J., concurring). Indeed, the Bruen
majority claimed that its new text, history, and tradition test “comports”
and “accords with” how the Court protects free-speech rights. 597 U.S.
at 24–25. That is at least partially true. The Court has looked at history
to define categories of unprotected or lesser-protected speech. See Tingley
v. Ferguson, 144 S. Ct. 33, 35 (2023) (Thomas, J., dissenting from denial
of certiorari) (“Accordingly, the Court has instructed that states may not
‘impose content-based restrictions on speech without “persuasive
evidence . . . of a long (if heretofore unrecognized) tradition” to that
effect.’ ” (quoting NIFLA, 585 U.S. at 767)). In other contexts too, like the
religion clauses, the Court has construed the scope of the asserted right
“by ‘reference to historical practices and understandings.’ ” Kennedy v.
Bremerton Sch. Dist., 597 U.S. 507, 510 (2022) (quoting Town of Greece v.
Galloway, 572 U.S. 565, 576 (2014)).
But the Supreme Court has continued to embrace balancing tests
to determine whether a regulation imposed on protected speech is
constitutional—and so, therefore, do the lower courts. Decreasing the
role of these balancing tests and increasing consideration of the history of
regulating certain professions, at least as a first step and in this kind of
context, might bring the judicial approach to First Amendment

14
challenges to professional-licensure regulations more in line with how we
assess due-process claims. After all, that doctrine is the one under which
professional-licensure challenges have traditionally been raised. The
U.S. Supreme Court has said that “the Due Process Clause specially
protects those fundamental rights and liberties which are, objectively,
deeply rooted in this Nation’s history and tradition.” Washington v.
Glucksberg, 521 U.S. 702, 720–21 (1997) (internal quotation marks
omitted). Perhaps a similar inquiry as a starting point might be a good
first step in isolating novel licensure regimes that warrant First
Amendment scrutiny from those that have clear historical roots such that
heightened scrutiny would be more problematic (or more needed).
Whether a better and clearer approach to assessing First
Amendment challenges will emerge, even in limited contexts like
professional licensure, remains to be seen. I reserve further thoughts for
future cases, if they come. This Court, like all courts, must of course
follow the Supreme Court’s First Amendment guidance, and I hope that
Court will continue to refine its jurisprudence in this area. But it is for
this Court to determine the proper analytical framework for cases that
arise under the Texas Constitution’s free-speech clause, see Tex. Const.
art. I, § 8—if parties raise, research, preserve, and press contentions
under that provision, which Stonewater did not.

Evan A. Young
Justice
OPINION FILED: June 7, 2024

15