Gonzalez, Jose D

In the Court of Criminal
Appeals of Texas
══════════
No. WR-94,041-01
══════════
EX PARTE JOSE D. GONZALEZ,
Applicant
═══════════════════════════════════════
On Application for a Writ of Habeas Corpus
Cause No. 1191091-A in the 208th District Court
From Harris County
═══════════════════════════════════════
YEARY, J., filed a dissenting opinion.
The Court today grants relief. Regarding Applicant’s actual
innocence claim, I agree with the majority. For the reasons stated in Ex
parte Fournier, Applicant is not actually innocent. See Ex parte
Fournier, 473 S.W.3d 789, 790–96 (Tex. Crim. App. 2015) (explaining
that an actual innocence claim is a factually based claim alleging that
newly discovered evidence shows that the applicant did not commit the
GONZALEZ – 2
charged conduct as a matter of historical fact). In this case, Applicant’s
purely legal basis for his claim that the convicting statue is
unconstitutionally overbroad does not amount to an actual innocence
claim. 1
I nevertheless dissent to the Court’s ultimate disposition of this
1 This case is distinguishable from Ex parte Warfield, in which I agreed
with the Court that the Applicant therein was “actually innocent.” Ex parte
Warfield, 618 S.W.3d 69, 74 (Tex. Crim. App. 2021). In Warfield, the Applicant
pled guilty to fraudulent possession of identifying information and to
possession of more than ten “items . . . possessed,” which made the offense a
second-degree felony. After Warfield’s plea, this Court, in Ex parte Cortez,
construed the phrase “item of . . . identifying information,” in an opinion, for
the very first time. 469 S.W.3d 593 (Tex. Crim. App. 2015). That construction
of the unit of prosecution language revealed that Warfield had actually
possessed only six “items,” meaning that he was, in fact, never guilty of the
second-degree felony, as a matter of historical fact, pursuant to a proper
understanding of the law involved. The Court’s decision in Warfield explained
that, in order to be guilty of the greater offense, Warfield must have possessed
a certain number of “items,” and he in fact never possessed that number of
“items.” It was a factually and historically based actual innocence claim. In
this case, however, Gonzales does not claim that he did not, in fact, commit the
exact conduct for which he was convicted. He merely brings the purely legal
claim that he is entitled to retroactive relief since he was convicted under a
statute that this Court has subsequently declared to be overbroad.
If the Applicant in this case had argued that he was convicted under a
facially unconstitutional statute that is unconstitutional in every possible
application, rather than simply overbroad, I would not distinguish this
Applicant’s claim from that in Warfield. If Applicant had brought such a claim,
he would be arguing that he in fact never committed any constitutionally
proscribable conduct, making him “actually innocent” in the same absolute
sense as the applicant in Warfield. Just as the statute in Warfield meant what
the Court found it to mean since the moment of its enactment, a facially
unconstitutional statute, invalid in every application, is void from the moment
of enactment. Here, with an Applicant seeking retroactive relief from
conviction under a statute with at least some plainly legitimate applications,
it is not necessarily the case that Applicant has demonstrated that he never in
fact committed any constitutionally proscribable conduct. Applicant is bringing
a legal claim without alleging anything about the conduct he in fact committed.
GONZALEZ – 3
case. In 2009, Applicant pled guilty to the state jail felony of improper
photography under Texas Penal Code Section 21.15(b)(1). TEX. PENAL
CODE § 21.15(b)(1). After that plea, this Court held that Section
21.15(b)(1) was facially unconstitutional, as overbroad. Ex parte
Thompson, 442 S.W.3d 325, 351 (Tex. Crim. App. 2014). Subsequently,
Applicant brought this 11.07 collateral attack against his final
conviction, seeking relief under Thompson. See TEX. CODE CRIM. PROC.
art. 11.07. Today, the Court sets aside Applicant’s final conviction
because he was sentenced under a statute later held to be
unconstitutional.
I dissent for the reasons I have stated in many similar dissenting
opinions. Namely, I question the Court’s willingness to give decisions
such as Ex parte Lo and Ex parte Thompson, in which the Court decides
that a statute is facially unconstitutional as overbroad, the power of
retroactive application—even in post-conviction habeas corpus
proceedings. See Ex parte Shay, 507 S.W.3d 731, 738 (Tex. Crim. App.
2016) (Yeary, J., dissenting) (dissenting to retroactive application of
Thompson); Ex parte Chang, 485 S.W.3d 918, 918–19 (Tex. Crim. App.
2016) (Yeary, J., dissenting) (dissenting to retroactive application of
Thompson); Fournier, 473 S.W.3d at 800 (Yeary, J., dissenting)
(dissenting to retroactive application of Ex parte Lo, 424 S.W.3d 10 (Tex.
Crim. app. 2013)).
The Court should more carefully consider whether retroactive
application through post-conviction 11.07 collateral attack is necessary,
given the underlying purpose of the First Amendment overbreadth
doctrine. Beyond necessity, the Court should consider whether
GONZALEZ – 4
retroactive application through post-conviction 11.07 collateral attack is
fair, given the windfall we are repeatedly granting applicants who have
not shown or even claimed that their own First Amendment rights have
been harmed. And it is not necessary.
Even if retroactive relief from convictions under overbroad
statutes were appropriate at some other procedural stage, such a claim
should not be entertained by this Court in a post-conviction 11.07
collateral attack. As a foundational rule, “a post-conviction writ of
habeas corpus is reserved for those instances in which there was a
jurisdictional defect in the trial court which renders the judgment void
or for denials of fundamental or constitutional rights.” Ex parte Drake,
883 S.W.2d 213, 215 (Tex. Crim. App. 1994). Granting applicants post-
conviction retroactive relief, without first requiring the applicants to
show that their own constitutional rights have been violated, goes
against this threshold requirement.
I. GENERAL RULE FOR FACIAL UNCONSTITUTIONALITY
When bringing a claim that a statute is unconstitutional on its
face, the starting point is to show that the statute is “unconstitutional
in every conceivable application[.]” Members of City Council of City of
Los Angeles v. Taxpayers for Vincent, 466 U.S. 789, 796 (1984). In other
words, the most basic claim of facial unconstitutionality is that none of
a statute’s possible applications, in the past, present, or future, has been
or ever could be valid. From this understanding of facial
unconstitutionality, Texas courts logically recognize that a statute with
no conceivable valid application is “void from its inception,” “stillborn,”
and “amounts to nothing and accomplishes nothing[.]” Reyes v. State,
GONZALEZ – 5
753 S.W.2d 382, 383 (Tex. Crim. App. 1988).
The propriety of retroactive relief from convictions under these
facially unconstitutional statutes is uncontroversial and apparent.
Fournier, 473 S.W.3d at 801 (Yeary, J., dissenting) (“Given this general
rule—that an unconstitutional statute is inoperable ‘from its
inception’—it may seem redundant or even pointless to engage in an
analysis of retroactivity of the judicial decision that declared the statute
to be unconstitutional.”) (quoting Reyes 753 S.W. at 383). Such a statute
is lifeless not just from the date a court held it to be unconstitutional,
but from the moment of its enactment. Id. Because there are no
circumstances in which the statute can be or could have been validly
applied, any person convicted under the statute at any point in time
necessarily has had his rights infringed upon.
II. FIRST AMENDMENT SUBSTANTIAL OVERBREADTH DOCTRINE
In Ex parte Thompson, this Court did not hold the improper
photography statute unconstitutional in the sense that every
conceivable application of the statute was invalid. Instead, the Court
decided that the statute was facially unconstitutional because it was
“overbroad.” Thompson, 442 S.W.3d at 349–51. As explained by the
United States Supreme Court in Taxpayers for Vincent, “[t]here are two
quite different ways in which a statute or ordinance may be considered
invalid ‘on its face’—either because it is unconstitutional in every
conceivable application, or because it seeks to prohibit such a broad
range of protected conduct that it is unconstitutionally ‘overbroad.’”
Taxpayers for Vincent, 466 U.S. at 796.
For a statute to be unconstitutionally overbroad, it does not have
GONZALEZ – 6
to be the case that every imaginable application has been or would be
unconstitutional. Rather, under the overbreadth doctrine, a statute
must “‘prohibit[ ] [only] a substantial amount of protected expression[,]’”
Thompson, 442 S.W.3d at 350 (quoting Ashcroft v. Free Speech Coal.,
535 U.S. 234, 244 (2002)), in relation to the statute’s “‘plainly legitimate
sweep.’” Id. at 349 (quoting Broadrick v. Oklahoma, 413 U.S. 601, 615
(1973)).
The propriety of retroactive relief from final convictions under an
overbroad statute should be more controversial; it is certainly less
apparent than retroactive relief from convictions under a statute with
only unconstitutional applications. See Fournier, 473 S.W.3d at 803(Yeary, J., dissenting) (“A defendant finally convicted pursuant to a
statute that is found only to be constitutionally overbroad is unlike a
defendant who was convicted pursuant to a statute that has no
conceivable constitutional applications. * * * [R]etroactively declaring
his conviction void ab initio would seem arguably to be neither necessary
nor just.”). Because the doctrine frankly concedes that an overbroad
statue has at least some legitimate sweep, it may be that many of the
prior convictions under the statute were constitutional. Id.Another way to understand the lighter burden attached to
overbreadth claims is in terms of standing. Ordinarily, litigants do not
have standing to come before a court to raise others’ rights. Parties must
have their own concrete stake in the outcome of the case. See Heckman
v. Williamson Cty., 369 S.W.3d 137, 154 (Tex. 2012) (“In Texas, the
standing doctrine requires a concrete injury to the plaintiff and a real
controversy between the parties that will be resolved by the court. This
GONZALEZ – 7
parallels the federal test for Article III standing[.]”). Carefully
considered principles of judicial restraint justify these strict standing
requirements.
When a court hears First Amendment overbreadth claims, it
cautiously strays from established standing requirements and allows
parties to appear before it who have not personally been harmed. As the
doctrine of overbreadth accepts, the party before the court may have
been punished under the statute’s plainly legitimate sweep. The United
States Supreme Court describes the doctrine as an “exception” by which
statutes “should be subject to challenge even by a party whose own
conduct may be unprotected.” Taxpayers for Vincent, 466 U.S. at 798.
III. UNDERLYING CONCERN OF THE OVERBREADTH DOCTRINE: THE
CHILLING EFFECT
The specific danger presented by overbroad statutes that justifies
both the lighter burden a claimant must meet, and also the relaxation
of ordinary standing requirements, is the so-called “chilling effect.” As
explained by the United States Supreme Court, “[t]his exception from
the general rule is predicated on ‘a judicial prediction or assumption that
the statute’s very existence may cause others not before the court to
refrain from constitutionally protected speech or expression.’” Id. at 799(quoting Broadrick, 413 U.S. at 612). As reiterated by this Court, the
doctrine “reflects the judgement that ‘the possible harm to society in
permitting some unprotected speech to go unpunished is outweighed by
the possibility that protected speech of others may be muted[.]’” Ex parte
Lo, 424 S.W.3d at 18 (quoting Broadrick, 413 U.S. at 612). Flipping
standing on its head, the overbreadth doctrine’s primary concern is not
to protect the party standing directly before a court. Instead, the
GONZALEZ – 8
doctrine protects the outside world of potential speakers who would have
engaged in First Amendment activity but for reading the overbroad
statute and fearing retribution.
Today, the Court follows a pattern of cases in which it goes farther
than is necessary, or even justified, to abate the concern that motivates
the overbreadth doctrine. Going beyond allowing an uninjured party to
initially claim a statute is overbroad, the Court takes the additional step
of allowing presumptively uninjured habeas applicants to nullify final
convictions, even in post-conviction collateral attacks. The Court takes
this gratuitous and socially costly step despite its own admonishment
that the overbreadth doctrine is a “‘strong medicine’ to be employed with
hesitation and only as a last resort[.]” Thompson, 442 S.W.3d at 349
(quoting New York v. Ferber, 458 U.S. 747, 769 (1982)).
This Applicant’s claim illustrates the need for us to think more
critically about retroactive relief from convictions under overbroad
statutes. Applicant does not claim that the chilling effect of the law
under which he was prosecuted remains a threat to the outside world.
He does not claim that the chilling effect of the overbroad statute muted
any of his own speech that the First Amendment would have protected.
Instead, in testimony before the Houston Immigration Court, Applicant
explained that he was convicted because he “used his cell phone to take
a photograph up a woman’s skirt.” Matter of Gonzalez-Cazares, Jose
Domingo, Houston, Texas Immigration Court Decision File Number
A058-079-828, at 3.
When explaining what kind of potentially protected expression
the Court believed might fall under the illegitimate sweep of Texas
GONZALEZ – 9
Penal Code Section 21.15(b)(1), the Thompson Court pointed to
indictments charging defendants with taking pictures “of people in a
public place (a water park) and of areas of the person that were exposed
to the public (wearing swimsuits).” Thompson, 442 S.W.3d at 350. But
in identifying the clearest examples of conduct falling within the
statute’s plainly legitimate sweep, the Court explained: “[Section]
21.15(b)(1) does apply to the situation in which a non-consensual
photograph is taken of a person in a private place, such as the home,
and the situation in which a photograph is taken of an area of a person’s
body that is not exposed to the public, such as when a photograph is
taken up a woman’s skirt.” Id. at 349 (emphasis added).
This is not the first case in which examining the actual conduct
of an applicant highlights the absurdity of granting retroactive relief
without asking whether the particular conviction fell within the plainly
legitimate sweep of the overbroad statute. I pointed out the same
scenario in Ex parte Shay, in which the applicant was convicted after
taking a video of others having sex. Shay, 507 S.W.3d at 739 (Yeary, J.,
dissenting). There, I reasoned from this Court’s discussion in Thompson
that such a conviction “hardly seems to establish that the statute
operated unconstitutionally as applied to Applicant’s particular
conduct.” Id. Here, it is even more apparent that the statute operated
constitutionally as applied to this Applicant; he was convicted for
engaging in the exact conduct the Court itself, in Thompson, made a
point to identify as easily fitting within the statute’s plainly legitimate
sweep.
Gifting this Applicant with retroactive relief does nothing to serve
GONZALEZ – 10
the chilling-effect concern underlying the overbreadth doctrine. The risk
of inhibiting First Amendment activity ends once a court holds that an
overbroad statute is unconstitutional. Such a decision eliminates the
chilling message previously broadcast to potential speakers, halts all
future convictions under the statute, and notifies the Legislature of its
risky drafting. Fournier, 473 S.W.3d at 803 (Yeary, J., dissenting). The
legitimate purposes of the overbreadth doctrine, if any, have already
been served by our decision in Thompson. Id. Prospective application is
enough. Id. Allowing retroactive relief at this stage—in no less than
post-conviction collateral habeas corpus proceedings⸻causes societal
harm by allowing unprotected speech to go unpunished and by letting
defendants escape final convictions well after the fact. The benefits of
the overbreadth doctrine are no longer outweighed by the possibility
that protected speech of others may be chilled or circumscribed.
IV. CONCLUSION
In Fournier, I expressed that I would “hesitate” to grant post-
conviction retroactive relief on the ground that an applicant was
convicted under a statute later held to be overbroad without requiring
the applicant to show that the statute was unconstitutional as it applied
to their conviction. Id. at 800. I no longer hesitate. See Ex parte
Mitcham, 542 S.W.3d 561, 567 n.1 (Tex. Crim. App. 2018) (“In Fournier,
I argued that the Court should not grant retroactive relief on the
grounds that the statute was later declared unconstitutionally
overbroad without first deciding whether an applicant needed to show
that the statute was unconstitutional as it applied to him. * * * I have
since concluded that this question should be answered in the
GONZALEZ – 11
affirmative.”); Chang, 485 S.W.3d at 918–19 (Yeary, J., dissenting)
(dissenting to the Court’s grant of retroactive relief based on Thompson);
Shay, 507 S.W.3d at 738–40 (Yeary, J., dissenting) (dissenting to the
Court’s grant of retroactive relief based on Thompson).
Final convictions under a statute later declared overbroad should
not receive retroactive nullification from this Court in every case.
Retroactive relief does not naturally follow when the doctrine itself
admits that at least some of the convictions that occurred prior to the
overbreadth holding may have been perfectly legitimate and
constitutional. Additionally, the sweep we give to the overbreadth
doctrine should mirror the justification for the doctrine itself. The risk—
that the very existence of a particular law might suppress First
Amendment activity—justifies straying from ordinary standing
requirements only to the extent that an uninjured party may bring the
initial claim that a statute is overbroad.
The chilling effect should no longer be feared once a court like
ours has explained that a statute is unconstitutionally overbroad. Once
that has happened, we should return to traditional principles of
standing for applicants challenging their prior, final convictions under
the overbroad statute. Instead of weaponizing our overbreadth decisions
with the power of indiscriminate, retroactive habeas relief, we should
require post-conviction applicants to show that their convictions did not
fall within the plainly legitimate sweep of the overbroad statute.
FILED: October 19, 2022
PUBLISH