In RE MARIA TERESA RAMIREZ MORRIS, AND TEXAS ALLIANCE FOR LIFE, INC. v. the State of Texas

Supreme Court of Texas
══════════
No. 23-0111
══════════
In re Maria Teresa Ramirez Morris, and
Texas Alliance for Life, Inc.,
Relators
═══════════════════════════════════════
On Petition for Writ of Mandamus
═══════════════════════════════════════
JUSTICE YOUNG, joined by Justice Devine and Justice Blacklock,
dissenting.
Today the Court refuses to remedy a clear violation of the Election
Code. It offers a host of salutary reasons. I agree with them all—when
they apply. But none of the Court’s stated reasons apply here because
they all depend on the same mistaken premise: the existence of a lawfully
ordered special election. If such a special election had been properly
called, each of these hoary principles would be relevant:
• “our longstanding commitment to avoid undue interference
with elections,” ante at 2;
• our preference to act “without disturbing [an] election from
going forward,” id. at 5;
• our historical practice of not “enjoin[ing] elections altogether,”
id. at 6; see also id. at 11;
• the obligation of the judiciary never to “deprive the voters of
an election,” id. at 6; and
• the need “to facilitate elections, not to stymie them,” id. at 16.
Each of those important values, however, depends on a common premise
that is missing here: a lawfully ordered election. The Court’s reliance
on those values requires it to assume the answer to the very issue that
is in dispute—whether any special election for May has been validly
ordered at all. Using doctrines that protect elections to protect events
that are not elections is bootstrapping. When a purported election is not
a real election, saying so does not “interfere” with an election; stating
the correct date to hold an election does not “disturb” that election;
telling the City when the law allows it to hold an election does not
“enjoin” or “stymie” or “deprive” anyone of an election. These steps
instead “facilitate” holding the election. Following the law of Texas can
never be regarded as against the interests of the People of Texas.
Texas law provides strict requirements for calling special elections,
including how to order them and when they may or must include
petition-initiated proposed city-charter amendments. One such
requirement is that, to hold a special election, a city council must order
it at least 78 days beforehand. The city council clearly failed to follow that
binding legal requirement here. The legislature expressly authorizes
Texas courts to enforce such election requirements, which protect all
voters. The courts should provide pre-election relief when, as here, the
core of the protection will be lost forever if not vindicated before the
election and when the request for relief is timely. The need for that relief
is amplified in this case because other serious issues are also present,
2
such as when, whether, and how the courts can address the contention
that the proposed charter amendment violates the Local Government
Code’s single-subject requirement. All the Court needs to address now,
however, is the timing violation. It can and should do so by granting
partial relief that directs the City to hold the special election in
November, not May.
In my judgment, the Court’s refusal to do so is mistaken and lacks
substantial legal support. My disagreement does not connote disrespect.
The Court’s decision does not follow from any improper motive. To the
contrary, it expresses a most praiseworthy motivation—the determination
to allow the People to govern themselves through the electoral process. I
share that commitment and disagree only as to how the judiciary properly
serves that foundational principle. I cannot agree that the judiciary is
limited to post-election relief in the narrow circumstances before us. I
therefore must respectfully dissent.
I
I begin with how this case came to us and why the law warrants
granting partial relief.
A
As the Court describes, a lengthy citizen-initiated petition in the
City of San Antonio—dubbed the “Justice Policy Charter Amendment”—
seeks to amend the city’s charter through a variety of mechanisms
concerning a host of policy issues, all related in some way to the
enforcement of local, state, and federal law within the city.1 This
1 The Court briskly summarizes the proposed amendment, see ante at
2, which is even more dramatic than the brief summary suggests. Broadly
speaking, the proposed amendment seeks to regulate or outright prohibit the
3
litigation commenced based on relators’ contention that the proposed
amendment’s scope violated the single-subject requirement of Texas
Local Government Code § 9.004(d).2 Whether the proposed amendment
is otherwise substantively or procedurally improper, however, is
immaterial to the question before us now: whether there is any lawfully
called special election at all.
The timeliness issue emerged after the city council scheduled a
vote on February 16 for the purpose of ordering the May 6 election. That
scheduling choice left only one extra day before the last possible day to
comply with state law, which requires special elections to “be ordered
enforcement of abortion crimes, drug offenses, and theft (to name a few). More
specifically, it would bar sharing specified information with other governmental
agencies unless doing so would advance the proposed amendment’s preferred
policy goals (like “defend[ing] the patient’s right to abortion care or the
healthcare provider’s right to provide that care”). It would regulate how the
police enforce warrants; restrict police authority to seek so-called “no knock”
warrants from a court; and limit what the police may take as evidence incident
to a search. The proposed amendment would bar the use of “chokeholds” by
the police and require that police release (rather than arrest) certain categories
of alleged offenders—those charged with theft of property worth less than
$750, those who bring contraband into a jail or prison, and those charged with
various other crimes. The proposed amendment would also have its
non-enforcement scheme overseen by a newly appointed “Justice Director,”
who would work with the city council and provide impact statements for any
decisions affecting the city’s budget or law enforcement. All that and far more.
2 That contention, and its companion argument that such a multifaceted
proposed amendment violates the requirements of Local Government Code
§ 9.004(e), warrant more analysis—including whether those provisions entail
judicially enforceable rights and, if so, whether they protect voters at the polls
(so that pre-election review is proper) or only those affected by enforcement of
a voter-approved charter amendment (so that only post-election review is
proper). See infra Part III (discussing how granting partial relief based on
timing would have allowed the Court to consider these questions without the
rush of emergency litigation).
4
not later than the 78th day before election day.” Tex. Elec. Code
§ 3.005(c). Shortly before the vote, this Court ordered the parties to
provide status reports regarding what action, if any, the city council took
with respect to the proposed amendment, and what effect the city
council’s action had on the pending request for relief. The parties
subsequently informed the Court that the ordinance ordering the special
election passed by a vote of 7-0-3: seven voted to order the election; none
voted against doing so; and three abstained from the vote.
All parties agree that, under the city charter, the ordinance could
not take effect until February 26—ten days from February 16, when it
passed—because it received fewer than eight votes. See San Antonio,
Tex., City Charter art. II, § 15; ante at 3 & n.2.3 Thus, the ordinance
could only effectively “order” an election on February 26—69 days before
the scheduled election.
B
Yet the law requires 78 days, not 69. That 78-day requirement is
clear and unambiguous. Tex. Elec. Code § 3.005(c).4 Likewise clear and
3 Delaying the effective date of a newly enacted law is common practice
and courts must respect that aspect of legislation as much as any other. Cf.,
e.g., Landgraf v. USI Film Prods., 511 U.S. 244, 257 (1994) (“A statement that
a statute will become effective on a certain date does not even arguably suggest
that it has any application to conduct that occurred at an earlier date.”).
4 The legislature expressly eliminated any conceivable doubt that might
be generated from other sources of law: “This section supersedes a law outside
this code to the extent of any conflict.” Id. § 3.005(b). The legislature’s focus
on timeliness is illustrated by its repeated amendments to lengthen the time
required. See, e.g., Act of Sept. 1, 2015, 84th Leg., R.S., ch. 84, § 3 (expanding
the requirement from 70 days to 78 days); In re Woodfill, 470 S.W.3d 473, 480
n.11 (Tex. 2015) (noting this recent statutory elongation).
5
unambiguous is the city council’s violation of it—a point the Court does
not dispute. The only question, then, is what to do about it.
Nothing is one possible answer. It is the answer that the Court
adopts by relying on the general principle of restraint in the context of
elections as its justification. See ante at 5–6. I readily agree that election-
related claims warrant a jaundiced judicial eye, especially when parties
ask the courts to disrupt an ongoing election. See, e.g., In re Khanoyan,
637 S.W.3d 762, 764–65 (Tex. 2022). But traditional judicial reluctance
to disrupt an ongoing election is limited to just that: an ongoing election.
There has been no lawfully ordered election in this case, so the principles
of restraint that the Court ably summons today have no relevance here.
Thus, far from merely applying our precedents and the separation-of-
powers principles on which they are built, the Court unduly expands them
to hold that we must idle by during an election that is no election at all.
Indeed, we have said that “[i]n some circumstances, litigants could
present the courts with a clear violation of ministerial duties imposed
by law, which—especially if brought early enough to avoid harm to the
larger election—could lead to prompt judicial correction.” Id. at 767.
The present case surely qualifies. In the face of an objective basis to
deem a special election invalid ab initio because of something as basic
and clear-cut as the calendar,5 the Court should say so without delay.
5 I agree with the Court that the general election would proceed as
planned regardless of what happens with respect to the special election. See
ante at 12 n.30. A general election is an election, “other than a primary election,
that regularly recurs at fixed dates,” Tex. Elec. Code § 1.005(6), whereas a
special election is simply one “that is not a general election,” id. § 1.005(18). A
general election will elect the mayor and council members, for example. Special
elections include discretionary or irregular matters, like proposed charter
6
The violation is holding the election prematurely, not enforcing the
election’s results. The proper relief when a voter is asked to vote at the
wrong time is to direct the city to hold the election at the right time, a
pre-election remedy. By contrast, when a voter fears the outcome of an
election, the voter must first await the election to see what the outcome
even is before challenging it as legally infirm.6
Consider an analogous situation: violations of the double-
jeopardy clause. The core right protected by that clause is immunity
from standing trial a second time. That right will be destroyed if the
courts force a defendant through a second trial on the theory that any
conviction is speculative and can be reversed after the fact. That is why
both federal and Texas courts allow extraordinary pre-trial relief through
amendments. The planned May 6 election is, as is common, a joint general and
special election.
6 The Court contends that post-election relief would allow relators to
“fully present their claims in a trial court” and notes that “the law is typically
better served when the lower courts review a legal issue before this Court does.”
Ante at 8 (quoting Rattray v. City of Brownsville, ___ S.W.3d. ___, ___, 2023 WL
243892, at *6 (Tex. Mar. 10, 2023)). I certainly vouch for the general proposition.
But the Court places far too much weight on the word “typically.” For one
thing, in exceptional and time-sensitive circumstances like these (in which our
rules expressly permit parties to seek relief directly from this Court, see Tex.
R. App. P. 52.3(e)), the law certainly cannot be “better served” by condoning a
violation of it. And for another, post-election relief is by definition inadequate
for an injury that is complete at the time of the election. Cf. In re Petricek, 629
S.W.3d 913, 917 (Tex. 2021) (holding that a post-election remedy is inadequate
“[i]f the ballot can be corrected before the election”). The Court fails to
distinguish between (a) a challenge to the substance, in which—if a ballot item
passes and would otherwise become enforceable—a party who objects to it can
contend that the ballot item is nonetheless bereft of force and (b) a challenge
to how the election is to be held, which has nothing to do with whether the
ballot item passes or not and indeed could be brought by a voter regardless of
that voter’s support for or opposition to the underlying substantive proposition.
7
an appeal authorized by the collateral-order doctrine, see, e.g., United
States v. Hollywood Motorcar Co., 458 U.S. 263, 266 (1982), or through
a pre-trial writ of habeas corpus, see, e.g., Ex parte Robinson, 641 S.W.2d
552, 555 (Tex. Crim. App. 1982). Likewise here: voters’ core statutory
right to be called only to properly authorized elections will be irretrievably
harmed if we await the election before vindicating the right.
The Court seems to suggest that the 78-day requirement is no
match for a petition-generated initiative. I cannot see why. The reason
a proposed charter amendment is placed on the ballot is because state
law requires it. The law channels ballot access in many ways. For
example, for a proposed charter amendment to be placed on the ballot,
the law has long required petition signatures only of the smaller of 5%
or 20,000 of the qualified voters, even in our largest cities. See Tex. Loc.
Gov’t Code § 9.004(a). If the only basis for including a proposed
amendment on the San Antonio ballot was that 19,000 signatures had
been obtained, I find it hard to believe that the Court would not intervene
to protect the voters from being convened to exercise their franchise
without legal authority. I thus fail to see why 19,000 voters instead of
20,000 would be dramatically different than 69 days instead of 78. Both
are necessary conditions for calling this special election.
Other existing requirements that channel authority to amend a
charter include the frequency of doing so—a city may not amend its
charter “oftener than every two years.” Tex. Const. art. XI, § 5. This
Court (by refusing a writ of error) has enforced that restriction on voting.
See State v. City Comm’n of San Angelo, 101 S.W.2d 360 (Tex. Civ.
App.—Austin 1937, writ ref’d). Despite the requisite number of
8
signatures for a proposed charter amendment, the Court agreed that it
was not lawful to hold an election that would violate the frequency
requirement. Id. at 360–61.
The 78-day mandate is a comparable requirement that we should
enforce when it has clearly been violated. The Court implies that we
should simply ignore the requirement because the city council was
“arguably” bound to call the election. Ante at 8. I doubt that the casting
of such votes was purely ministerial, and I think that the law confirms
as much by providing for how to achieve ballot access if a city council,
for reasons of principle or otherwise, proves unwilling to call an election
or put a proposed amendment on the ballot. But regardless of any duty
to have acted timely, the city council did not do so. In such instances,
may courts retroactively deem the 78-day requirement to be satisfied
even when it is not, thus judicially striking the 78-day requirement from
the statute?
A better way to approach a situation like this one is to simply
apply the provisions of the law, which are readily harmonized. Both
state law and the San Antonio city charter provide for a two-step process
in which the city council retains a proper but limited role. Under the
law, the city council must have the first crack, but if it fails to act, the
proposed charter amendment will be presented at the next available
election—a specific right that the courts may enforce. See Tex. Loc.
Gov’t Code § 9.004(b); San Antonio, Tex., City Charter art. IV, § 41
(providing that if the city council “fails to pass an ordinance proposed by
initiative petition . . . the proposed or referred ordinance shall be
submitted to the electors on the next authorized uniform election date
9
that allows enough time to hold the election in the manner required by
law” (emphasis added)).
In other words, we need not regard the city council as a potted
plant or an automaton, as the Court does by recognizing nothing but a
ministerial role for the council.7 If there were such an utter lack of
discretion, then mandamus relief certainly should have been forthcoming
in San Angelo, where the city refused to place a proposed charter
amendment on the ballot. 101 S.W.2d at 360–61. The legislature could
have eliminated any role for city councils but instead has left it to city
councils both to call special elections and to “prescribe the wording of a
proposition that is to appear on the ballot.” Tex. Elec. Code § 52.072(a).
These are not examples of pointless political theater; they are part of the
process. A city council’s failure or refusal to act leads to a different
remedy, which is also part of the process: granting the proposed
amendment automatic access to the ballot in the “next” available
election, enforceable if necessary by a court.
For this reason, we need only follow the law to preserve the rights
of all parties (and, most importantly, the people of San Antonio) by
recognizing that May 6 is no longer available for a special election.
Contrary to the Court’s assumption, May 6 is not “the earliest lawful
opportunity” to hold a special election. Ante at 8 (emphasis added). That
ship has sailed; it is no longer possible to comply with the 78-day
7 Remarkably, despite one of the Court’s bases for denying relief—that
it may not intrude upon the prerogatives of the political branches—the Court
castigates the three abstaining members of the city council as law-breakers
because they did not cast their votes the way the Court concludes they were
bound to do. Ante at 8–9 (suggesting that mandamus could lie for a court to
order members to vote in a particular way).
10
requirement (or to order the city council to do so). Absent any other
legally cognizable objection to the proposed charter amendment, however,
the city council will no longer be in the position to foreclose the
amendment from appearing on the November 2023 ballot.
The Court, however, denies all mandamus relief, casting the city
council’s ordinance ordering the special election as pure ceremony and
conscripting San Antonio’s voters into the same playbill—but at a cost.
Allowing manifest violations of the law is bad enough. Inviting a costly
and corrosive post-election challenge no matter what happens is
unfortunate, too.8 But perhaps most serious of all, if the charter
amendment passes, San Antonians will be blocked from making any
other amendments to their charter until 2025. See Tex. Const. art. XI,
§ 5(a). As a result of the special election that should not have happened,
other opportunities that would have been available for charter
amendments will be gone forever. Ironically, the Court’s resolution of
this petition creates the very risk it purports to eliminate: the risk that
the voters of San Antonio will be unable to vote on charter amendments
they desire. I fear that today’s decision will undermine the very right to
vote the Court seeks to protect.
8 Suppose that the proposed charter amendment fails. While those who
oppose it, like relators, may be content to let bygones be bygones, what about
someone who supports it? I take no position on whether such a person could
bring a post-election challenge on the ground that the charter amendment
should have been placed on the November ballot instead of the May ballot—
but failing to address the timing problem now makes future litigation more
likely even if the voters reject it.
11
C
The Court’s opinion portrays denial of relief here as tantamount
to preserving the right to vote. If that truly were the question, I would
gladly join the Court. The opposite is more nearly true, however,
because the real question is whether there even is any valid election to
which the voters can be summoned. Requiring voters to exercise their
electoral authority when the law does not provide for it undermines our
system of elections. Setting a special election is—or at least should be—
a solemn act of self-government, not a casual event. If the Court is
unwilling to protect electoral regularity even in the face of an open,
obvious, and objective violation of legal standards, it will be hard for the
Court to stop lawlessness of other sorts in future abuses of our system.
Denying relief today does not protect that system but makes it more
vulnerable. Forcing unlawful elections diminishes the respect we have
for the rules governing them. Granting limited relief would have
prevented these real and unfortunate consequences.
I now turn to the Court’s rationale for why it refuses to act.
II
As I see it, the Court invokes three main grounds to justify denying
relators the pre-election relief they seek: two based on precedent and one
based on statutory construction. None of the three, in my view, supports
the Court’s decision.
A
First, the Court embraces as a chief authority our decision in City
of Austin v. Thompson, 219 S.W.2d 57 (Tex. 1949). After citing the
statutory provisions that empower courts to grant mandamus and
12
injunctive relief to prevent Election Code violations, the Court invokes
Thompson to claim that, “[h]istorically . . . the Court has not enjoined
elections altogether, even elections ‘called without authority and
therefore absolutely void.’” Ante at 6 (quoting Thompson, 219 S.W.2d at
59). “Historically” is the correct adverb, because Thompson relied
exclusively on the traditional “power of a court of equity.” 219 S.W.2d at
59. Neither the majority nor the dissent in that 5–4 decision mentioned
any statutes at all. For good reason: not until two years after this Court
declined to exercise its ordinary equitable authority in Thompson did
the legislature confer the express statutory authority to grant either
mandamus or injunctive relief to prevent Election Code violations.9 See
Act of May 30, 1951, 52d Leg., R.S., ch. 492, § 1, art. 7, 1951 Tex. Gen.
Laws 1097, 1098.
The statutory authority is no makeweight. As the Court notes
(ante at 10 n.30), we recently observed in In re Stetson Renewables
Holdings, LLC that “an express statutory deadline . . . does not
necessarily mean that the legislature intended for courts to enforce the
deadline.” 658 S.W.3d 292, 296 (Tex. 2022) (internal quotations omitted).
But the Court does not complete the narrative, because we specifically
9 One wonders if one (or all) of the five members of the Thompson
majority, if armed with the statutory authority we now possess, would have
joined the four dissenters to direct the City of Austin not to proceed with its
“election.” Regardless, that case was a strange one, turning purely on the
Court’s construction of Austin’s city charter, which the Court unanimously
read to forbid elections to fill unexpired city council positions. The election was
not even really an election because the council used it as a way to invite the
public to “advise” the council on whom the council should choose to fill the open
spot. Whatever else Thompson means, it does not mean anything about Texas
courts’ authority after the legislature has acted.
13
juxtaposed the area of law at issue in Stetson with “election cases,”
where (again, “with extreme caution”) courts indeed “may play a more
active part” because “there is ample and express statutory authority for
a judicial role.” Id. (citing Tex. Elec. Code § 273.061).
B
The Court’s other lead authority is more recent and relevant:
Coalson v. City Council of Victoria, 610 S.W.2d 744 (Tex. 1980). See ante
at 9–10. The Court cites Coalson for two main propositions that only
show how different this case is. Reaffirming Coalson does not remotely
require denying relief today.
First, Coalson “held that any opinion on the constitutionality of
the amendment before the election was held would be purely advisory
because voters may disapprove the amendment.” Ante at 10; see id. at 7
n.16. Quite right. The whole basis for the city council in Victoria refusing
to put a proposed amendment on the ballot was the city council’s view of
whether the amendment, if passed, would have been legal. This is why
the Court emphasized that the council members’ declaratory-judgment
action “seeks an advisory opinion” and why “[t]he election [would]
determine whether there is a justiciable issue.” Coalson, 610 S.W.2d at
747. By definition, such an unripe claim must await the election’s outcome.
No such claim appears in this case, of course. True, relators (and
apparently the city attorney and several members of the council) believe
that, if passed, the proposed charter amendment would violate state
law—but that is not the basis of their petition and not the basis on which
I would grant relief. Instead, relators complain of a procedural injury
14
that materializes whether or not the proposed amendment passes: being
forced to vote in an untimely ordered special election.
Second, the Court quotes Coalson for this accurate statement:
“The City Council’s duty [to order an election on a citizen-initiated
amendment] is clear, and its compliance with the law is ministerial in
nature.” Ante at 8 n.23 (quoting 610 S.W.2d at 747). Again, quite right.
In Coalson, a timely and properly ordered special election was ongoing.
“The election process had been lawfully put in motion,” the Court
expressly observed. 610 S.W.2d at 747 (emphasis added). Unlike the
special election here, the special election there contained other ballot
items. See id. at 746. The only question was whether the city council
could order a special election but exclude a validly proposed charter
amendment that it disliked. Here, of course, that antecedent question,
which was not at issue in Coalson, is precisely what the parties dispute:
whether any “lawful” special election has been called, given the open
violation of Election Code § 3.005(c).
The Court in Coalson was thus certainly correct to have held that
the city council members violated their ministerial duties by attempting
to block a proposition from the special election—after all, the election
had been ordered without any challenge to any statutory (or other)
prerequisite for doing so. Id. at 747. Likewise, it is obvious that the
properly ordered special election was the “next” such election. The city
council did not have a leg to stand on.
Worse yet, the very fact that (unlike here) the special-election
order in Coalson was entirely valid explains why this Court was so
focused on the detrimental consequence that would flow from excluding
15
the ballot initiative: the voters would have faced “a two-year delay before
another charter election” could be held on it or any other proposed
amendment. Id. at 746 (citing Tex. Const. art. XI, § 5). In a touch of
irony, then, today’s denial of relief yields the exact result Coalson sought
to prevent: the threat to voters’ ability to bring charter amendments
within two years.
Coincidentally, the Court in Coalson cited State v. City
Commission of San Angelo—the case I mentioned in Part I.B, supra—
for this very point. San Angelo likewise involved a proposed charter
amendment, but there “the city officials refused to call the election, on
the sole ground that the charter of said city had [already] been amended
by an election” less than two years ago, in violation of the frequency
limitation on such amendments. 101 S.W.2d at 361. In the ensuing
mandamus petition, supporters of the proposed amendment—observing
the requisite number of signatures—sought to compel the calling of a
special election despite the timing problem. Unlike the Court’s ruling
today, the holding in San Angelo was that officials did not have the
authority, much less obligation, to call an unlawfully timed special
election. If such a ministerial duty on the part of the council existed,
then it would not have been “an example of typical judicial restraint,”
ante at 12 n.35, but an example of judicial unwillingness to protect the
core rights of the People that the Court today so frequently invokes.
Said differently, the demanded election in San Angelo—just like
today’s—suffered from a threshold requirement that made it unlawful.
In San Angelo, it was inconsistency with the Texas Constitution’s two-
year requirement; here, it is inconsistency with the Texas Election Code’s
16
78-day requirement. San Angelo did not say “well, who knows, maybe
it won’t even pass—if it does, we’ll take it up then.” Coalson cited San
Angelo approvingly: it was because of San Angelo’s correct holding (that
another election with charter amendments would be improper even with
sufficient signatures) that Coalson deemed it so important to allow the
proposed Victoria amendment onto the ballot at once. See Coalson, 610
S.W.2d at 746.
But the sum and substance of today’s decision is to invert San
Angelo by holding, counterintuitively, that we must nevertheless press
ahead in the face of such procedural infirmities. This approach is simply
the election equivalent of the “certify now and worry later” approach we
have forcefully rejected in the class-action context. E.g., Sw. Refin. Co.
v. Bernal, 22 S.W.3d 425, 435 (Tex. 2000).
C
The Court’s third key argument for denying relief is, in essence,
a misunderstanding of the Election Code. That code simultaneously
imposes deadlines for calling special elections and denies special elections
the immunity from validity challenges granted to general elections. See
Tex. Elec. Code § 3.007 (“Failure to order a general election does not
affect the validity of the election.” (emphasis added)).
Under our ordinary rules of interpretation, the very fact that the
legislature went out of its way to say that an untimely order would not
affect a general election’s validity quite clearly implies that it would do
so for a special election. Otherwise, we would have both expressio unius
17
and surplusage problems—canons of construction we have applied in
case after case.10
But, in addition to its unduly narrow view of election deadlines
generally,11 the Court dismisses this consequence of § 3.007 (the “savings
clause”) by contending that “the savings clause [for general elections]
predates statutory deadlines for special elections.” Ante at 10 n.30. I
fail to see how that matters. Looking to statutory history to see how a
word was used is one thing; slicing up an integrated code to deny its
provisions the usual semantic cohesion is quite another. We have always
construed statutory text “as a whole,” In re Ford Motor Co., 442 S.W.3d
265, 280 (Tex. 2014) (quoting TGS-NOPEC Geophysical Co. v. Combs,
340 S.W.3d 432, 439 (Tex. 2011)), and presumed that the legislature acts
“with complete knowledge of the existing law and with reference to it,”
Acker v. Tex. Water Comm’n, 790 S.W.2d 299, 301 (Tex. 1990). Those
two principles have practical force here, where the legislature—when
codifying the Election Code—added the deadline for ordering elections
literally just two sections before it incorporated the preexisting savings
clause for general elections—and maintained that it would be only for
10For surplusage, see, e.g., In re CenterPoint Energy Hous. Elec., LLC,
629 S.W.3d 149, 159 (Tex. 2021); Waak v. Rodriguez, 603 S.W.3d 103, 108 (Tex.
2020); Randol Mill Pharmacy v. Miller, 465 S.W.3d 612, 617 (Tex. 2015). For
expressio unius, see, e.g., ConocoPhillips Co. v. Koopman, 547 S.W.3d 858, 877
(Tex. 2018); Johnson v. Second Inj. Fund, 688 S.W.2d 107, 109 (Tex. 1985);
James v. Brown, 637 S.W.2d 914, 918 (Tex. 1982); State v. Mauritz-Wells Co.,
175 S.W.2d 238, 241 (Tex. 1943); Collingsworth County v. Allred, 40 S.W.2d
13, 15 (Tex. 1931).
11 As noted above, the Court cites Stetson (rather expansively) for the
proposition that courts need not enforce the Election Code’s statutory deadlines.
See ante at 10 n.30; supra Part II.A (explaining why Stetson said the opposite
of this in the election context).
18
general elections. See Act of May 13, 1985, 69th Leg., R.S., ch. 211, § 1,
1985 Tex. Gen. Laws 802, 809 (adding newly codified §§ 3.005–.007).
In short, the Court’s chief authorities do not support the denial of
partial relief. I fear instead that the deployment of these authorities
will harm the coherence of our election law and indeed of our general
approach to statutory construction.
III
The issue that originally divided—and still divides—the parties
is not the election’s timeliness. Rather, it is whether, when, and how
the courts can vindicate the requirement that a charter amendment
“may not contain more than one subject,” along with the corresponding
mandate that “[t]he ballot shall be prepared so that a voter may approve
or disapprove any one or more amendments without having to approve
or disapprove all of the amendments.” Tex. Loc. Gov’t Code § 9.004(d),
(e). The parties vehemently disagree about whether the lengthy proposed
charter amendment violates these statutory requirements.12
We ought to prefer not to resolve on an emergency basis such a
fraught question. Nor would it be necessary: granting partial relief on
the ground of untimeliness would carry the added benefit of allowing the
Court adequate time to consider the single-subject question. After
12 The City repeatedly asserts in its filings that mandamus relief is
unnecessary and improper because of an available post-election remedy should
the amendment pass. Given the Court’s resolution, the City certainly could
not challenge the procedural availability of a post-election remedy with respect
to the single-subject requirement. See Ferguson v. Bldg. Materials Corp. of
Am., 295 S.W.3d 642, 643 (Tex. 2009) (“Judicial estoppel precludes a party who
successfully maintains a position in one proceeding from afterwards adopting
a clearly inconsistent position in another proceeding to obtain an unfair
advantage.”).
19
directing the election to be held in November, as the law requires, I
would direct merits briefing so that our ultimate resolution of the single-
subject question could deliver on two promises that our law makes to
voters: that they can vote on what they wish and that they will not be
asked to do so in the face of noncompliance with important legal
requirements.
Taking that opportunity now would be sensible. For one thing,
the single-subject question is unlikely to disappear no matter what
happens in the May election because the timing issue will likely keep it
alive.13 For another, the question of whether a single-subject claim is
properly brought before an election will be hard to answer after an
election, and granting partial relief now would give us an unusual
opportunity to do so with comparative leisure. There are at least three
important questions we will have to address at some point.
First, is the Local Government Code’s single-subject requirement
judicially cognizable? Even if courts properly can reach it, we must be
persuaded, among other things, that we have access to judicially
administrable standards. Otherwise, the dispute might constitute a
nonjusticiable political question. See, e.g., Van Dorn Preston v. M1
Support Servs., L.P., 642 S.W.3d 452, 465 (Tex. 2022) (concluding that
the presence of judicially manageable standards meant that courts could
adjudicate the case at issue); see also Jeffrey S. Sutton, Who Decides?
States as Laboratories of Constitutional Experimentation 258–60 (2022)
13I agree that, if the special election had been timely called, and if the
proposed amendment were to be defeated at the polls, then the single-subject
question would become moot. With the untimely special election, though, any
hope of mootness will likely be, at best, long deferred by litigation.
20
(“The nature of the single-subject [requirement] does not demand
judicial enforcement,” id. at 258, because it “faces a soaring level of
generality problem,” id. at 260.). The parties’ comparatively cursory
briefing on this point reflects a divergence about what the single-subject
requirement means.14
14 Notably, the Texas Constitution includes a single-subject provision
similar to that of Local Government Code § 9.004(d). Compare Tex. Const. art.
III, § 35(a) (“No bill . . . shall contain more than one subject.”), with Tex. Loc.
Gov’t Code § 9.004(d) (“An amendment may not contain more than one
subject.”). Chief Judge Sutton’s observation quoted above was primarily
directed to such constitutional requirements, which are common among
American states. Whether our Constitution allows courts to enforce the
single-subject requirement against the legislature, a coordinate branch of state
government, asks a very different question than whether a statute requires the
courts to enforce its provisions against cities, which are administrative units
of the State.
In addition to this structural difference, there is a textual distinction:
the Constitution says that a “bill” (not a statute) may not contain more than
one subject. That distinction at least suggests that the constitutional single-
subject requirement is “directed at the legislature’s own process for enacting
laws,” Sutton, supra, at 258, so the legislature itself (especially, perhaps, the
other house) and not the judiciary must enforce it. By contrast, the Local
Government Code focuses both on the end product, as § 9.004(d) makes clear,
and on the process leading to it, as § 9.004(e) shows. The latter provision states
that “[t]he ballot shall be prepared so that a voter may approve or disapprove
any one or more amendments without having to approve or disapprove all of
the amendments.” No comparable constitutional provision exists to limit the
legislature.
So perhaps the legislature borrowed the Constitution’s language but
imposed a different single-subject standard, at least with respect to justiciability,
on municipalities. Similar language can convey different meanings in the
constitutional context. The U.S. Supreme Court has long given markedly
different meanings to the nearly identical text of Article III and (what is now)
28 U.S.C. § 1331, both of which describe “arising under” jurisdiction. Compare
Osborn v. Bank of the U.S., 22 U.S. (9 Wheat.) 738, 823–25 (1824) (construing
the constitutional grant expansively enough to reach any case with a mere
federal “ingredient,” including a potential ingredient), with Louisville &
Nashville R.R. Co. v. Mottley, 211 U.S. 149, 152 (1908) (construing the same
21
Second, if the question is justiciable, what are its doctrinal
boundaries, and what is the proper level of deference that will allow
courts to distinguish between (a) a single issue with related subparts
and (b) multiple separate issues? Perhaps § 9.004(e) will assist. If the
legislature aims to protect voters from having to make impossible or
unfair choices, as § 9.004(e) provides, then § 9.004(d)’s single-subject
requirement might still require substantial deference to city councils (or
citizens promoting initiatives). Perhaps further analysis would confirm
(or refute) the instinct thus far reflected in some lower-court opinions.15
Third, if the courts can address the question, may they do so
before or only after an election? There is at least a plausible argument
that single-subject scrutiny, if permissible at all, should be done as soon
as possible. The requirements of § 9.004(d)–(e) appear designed to protect
the very act of voting. If the goal is to ensure that voters spend their
time and civic energy only on properly framed ballot items, it seems
language as a statutory matter to authorize jurisdiction only when a federal
question is on the face of a well-pleaded complaint).
15 See, e.g., City of Galena Park v. Ponder, 503 S.W.3d 625 (Tex. App.—
Houston [14th Dist.] 2016, no pet.). I express no view regarding whether any
part of that opinion was correct but note that the case suggested that a “single
subject” could encompass integrated policy changes that would not be feasible
to pursue except as a unit. At issue there were proposed “rules governing the
appointment and qualifications for police chief and fire chief [that] are part of
the subject of how the city’s emergency services departments are to be
organized.” Id. at 635. When reticulated provisions depend on one another,
such that courts must defer to municipalities, the question may be whether the
proposed amendment’s provisions here are in any real sense mutually
dependent or are only vaguely related. If the latter, providing guidance to
reliably inform local governments about how attenuated a relationship can be
before § 9.004(d) is implicated may be challenging. Further briefing would
clearly assist in devising a proper starting point.
22
suitable—when possible, see Khanoyan, 637 S.W.3d at 764–66—to
provide pre-election relief.16 Indeed, if § 9.004(d)’s single-subject
requirement aims to facilitate the voter’s right in § 9.004(e), then the
single-subject requirement would logically be one that, when timely
presented, should be vindicated before votes are cast.
I take no position on any of those questions yet, much less
determine how the answers to those questions would apply here. But it
should be self-evident that answering them will be a very serious
endeavor. These thoughts, I hope, explain why I believe that we should
have accepted the opportunity that partial relief would have afforded us
to consider each question without undue haste.
IV
I could better understand the Court’s denial today if it was
predicated on the concern that granting this mandamus petition would
break the dam and create an influx of similar pre-election challenges
across the second largest State in the Union. This Court cannot sit as a
court of first instance whenever someone is aggrieved. There are only
nine of us in a State with 254 counties and 30 million residents. We sit
en banc in every case. But given the unusual posture of this dispute—
saddled with not only a single-subject issue but also with a violation of
the 78-day election-order requirement—we should not be worried about
the limited ramifications that might follow from a partial grant of relief.
If the city council has indeed violated the 78-day election-order
16 By contrast, a challenge to the legal substance of a properly framed
ballot item could not be challenged ex ante; no one asks the Court to reject the
proposed amendment here on the ground that its provisions might be unlawful,
even though some parties clearly believe that to be true.
23
requirement, which it clearly has, I suspect our saying so would lead
more municipalities to dutifully observe their deadlines. Even the
occasional blown deadline in the future would lead to relief in the lower
courts in light of an unequivocal and easily followed precedent from this
Court. Granting partial relief today would open no floodgates.
* * *
The right to vote is unquestionably fundamental. It is how we
remain a self-governing people. See Khanoyan, 637 S.W.3d at 763. But
the People of Texas also have chosen to guard their electoral role by
enacting comprehensive legislation that ensures the propriety of elections
that are called. The People also have given their courts broad authority
to enforce compliance with “any duty imposed by law in connection with
the holding of an election” through the issuance of writs of mandamus.
Tex. Elec. Code § 273.061(a). We should use that authority sparingly—
but we should use it when it is warranted, as it is here.
Respondents are wrong to suggest that following the law would
“thwart the will of thousands of San Antonio voters.” Following the law
would implement the will of all the voters of San Antonio—and of Texas.
Likewise, I disagree with the Court that following the law would
“disrupt[] the settled expectations of the people of San Antonio.” Ante
at 2. The People’s settled expectations are that the law will be
followed—that they will be called to the polls when the law so requires
but not otherwise.
The Court summarizes its holding with a troubling statement:
that “[c]ourts must not lightly usurp [the] power” of placing popular
initiatives on local ballots. Ante at 16. I propose a different rule: courts
24
should never usurp any power in any context—not lightly, not
hesitantly, but never. When power does belong to the courts, as the
authority to enforce election law unquestionably does, we should
cautiously exercise that power. Cautiously discharging the judicial role
here requires granting modest partial relief: holding that there is no
lawful authority for a special election in May.
I respectfully dissent.
Evan A. Young
Justice
OPINION FILED: March 17, 2023
25