Keller, P.J., delivered the opinion of the Court in which Keasler, Hervey, Richardson, Keel, and Walker, JJ., Joined.

A provision of the Texas Open Meetings Act (TOMA) makes it a crime if a member or group of members of a governmental body "knowingly conspires to circumvent this chapter by meeting in numbers less than a quorum for the purpose of secret deliberations in violation of this chapter."1 We conclude that this provision is unconstitutionally vague on its face. Consequently, we reverse the judgment of the court of appeals and affirm the trial court's judgment dismissing the prosecution.

I. BACKGROUND

Appellee was the Montgomery County Judge, and as such, he was a member of the Montgomery County Commissioners Court. He was indicted for violating TOMA's § 551.143, the statute described above. The indictment alleges that Appellee did

Appellee filed a motion to dismiss on the basis that § 551.143 was overbroad in violation of the First Amendment and was unconstitutionally vague. The trial court granted the motion and dismissed the indictment.

On appeal, the State contended that the statute did not violate the Constitution. The court of appeals agreed, concluding that the statute did not violate the First Amendment and was not unconstitutionally vague.2 In response to Appellee's First Amendment claims, the court of appeals held that § 551.143 was a content-neutral law because it was "directed at conduct, i.e., the act of conspiring to circumvent TOMA by meeting in less than a quorum for the purpose of secret deliberations in violation of TOMA."3 The court further concluded that the strict-scrutiny standard was inapplicable because the prohibition in TOMA "is applicable only to private forums and is designed to encourage public discussion."4

With respect to vagueness, the court of appeals concluded that the statutory terms "conspire," "circumvent," and "secret," although undefined, have commonly understood meanings.5 Relying on an opinion of the Texas Attorney General, the court further concluded that the statute applies to "members of a governmental body who gather in numbers that do not physically constitute a quorum at any one time but who, through successive gatherings, secretly discuss a public matter with a quorum of that body."6 Under this construction, the court concluded that the statute "describes a criminal offense with sufficient specificity that ordinary people can understand what conduct is prohibited."7

Consequently, the court of appeals reversed the trial court's order dismissing the indictment and remanded the case for further proceedings.8 We granted Appellee's petition for discretionary review, which complained, inter alia, that § 551.143 is void for vagueness.9 We agree that the statute is unconstitutionally vague on its face.

II. ANALYSIS

A. The Statutory Scheme

TOMA generally requires that meetings of a governmental body be open to the public.10 "Meeting" is defined in two ways, both of which require that a quorum be present:

A "quorum" is defined as "a majority of a governmental body, unless defined differently by applicable law or rule or the charter of the governmental body."12 "Deliberation" is defined as "a verbal exchange during a meeting between a quorum of a governmental body, or between a quorum of a governmental body and another person, concerning an issue within the jurisdiction of the governmental body or any public business."13

The main TOMA provision, § 551.144, makes it a crime to engage in conduct that calls, facilitates, or participates in a closed meeting.14 A "closed meeting" is "a meeting to which the public does not have access."15

Appellee was not charged under the main provision though. Instead, he has been prosecuted under, § 551.143, which provides:

B. Implicating the First Amendment

As we shall explain more fully below, more clarity is required of a criminal law when that law implicates First Amendment freedoms.17 Consequently, we first address whether § 551.143 implicates the First Amendment's freedom of speech.18

We have recognized that the First Amendment is implicated when the government seeks to impose criminal sanctions on an elected official for communications made in his official capacity.19 As a Fifth Circuit panel once stated, "[T]he Supreme Court's decisions demonstrate that the First Amendment's protection of elected officials' speech is robust and no less strenuous than that afforded to the speech of citizens in general."20 The Fifth Circuit decision of Asgeirsson v. Abbott, relied upon by the State in the present case, held that TOMA's § 551.144 was "a content-neutral time, place, or manner restriction."21 Calling a statute a reasonable time, place, or manner restriction is an implicit acknowledgment that some of the activity regulated by the statute is protected speech.22

The State contends that § 551.143 reaches only conduct rather than speech. At oral argument, the State's attorney maintained that the statute punishes the conduct of "meeting" rather than what might be said during that meeting.23 But both of TOMA's definitions of "meeting" incorporate communications, either through "deliberations," the passing of "information" from one person to another, or the asking of questions. The State contends that these definitions do not control because they define "meeting" as a noun and § 551.143 uses "meeting" as a verb.24 Even if the State is correct that the definitions are not controlling,25 the statute does not proscribe "meeting" in the abstract but proscribes a particular kind of meeting—one that is for the purpose of "deliberations." This purpose makes the statutory act of "meeting" communicative, even if the bare fact of meeting would not be so. The Supreme Court has observed that a parade could be non-communicative "[i]f there were no reason for a group of people to march from here to there except to reach a destination" but that "[r]eal" parades are in fact "public dramas of social relations" and, as such, are "a form of expression."26 For the same reason, TOMA's punishment of meeting for the purpose of deliberations reaches speech, and not just conduct.

The State also contends that any speech that is implicated by the statute is unprotected because it constitutes "speech integral to criminal conduct." But the cases that involve this form of unprotected speech involve speech that furthers some other activity that is a crime.27 Examples of this include picketing designed to coerce a company to sign an illegal contract or solicitation to facilitate a sex crime.28 The statute before us proscribes activity designed to "circumvent" TOMA, but circumventing TOMA is not a crime apart from § 551.143.29

C. Nature of a Facial Vagueness Challenge

We next turn to whether the facial vagueness challenge advanced here requires a showing that there are no possible instances of conduct that it is clear would fall within the statute's prohibitions. If such a showing is required, and if at least one such instance of conduct can be imagined, then we would have to address whether a trial would be needed to develop a record to substantiate an as-applied challenge.30 In Long v. State, we concluded, "[W]hen a vagueness challenge involves First Amendment considerations, a criminal law may be held facially invalid even though it may not be unconstitutional as applied to the defendant's conduct."31 The Supreme Court more recently suggested that such a conclusion might be incorrect: "Even assuming that a heightened standard applies because the ... statute potentially implicates speech, the statutory terms are not vague as applied to plaintiffs."32 But in an even more recent case, Johnson v. United States, the Supreme Court stated, "[A]lthough statements in some of our opinions could be read to suggest otherwise, our holdings squarely contradict the theory that a vague provision is constitutional merely because there is some conduct that clearly falls within the provision's grasp."33 The Court's statements in Johnson do not appear to be limited to vagueness challenges that implicate First Amendment freedoms, but to the extent that more clarity is required in the law, those statements would seem to apply with even greater force when First Amendment freedoms are implicated.34 We conclude that a facial vagueness challenge to a statute that implicates First Amendment freedoms does not require a showing that there are no possible instances of conduct clearly falling within the statute's prohibitions.35 What is required to establish a facial vagueness violation is addressed below.

D. Vagueness

1. Standard

To pass constitutional muster, a law that imposes criminal liability must be sufficiently clear (1) to give a person of ordinary intelligence a reasonable opportunity to know what is prohibited and (2) to establish determinate guidelines for law enforcement.36 When the law also implicates First Amendment freedoms, it must also be sufficiently definite to avoid chilling protected expression.37 Greater specificity is required when First Amendment freedoms are implicated because "uncertain meanings inevitably lead citizens to steer far wider of the unlawful zone than if the boundaries of the forbidden areas are clearly marked."38 Nevertheless, "perfect clarity and precise guidance have never been required even of regulations that restrict expressive activity."39 A scienter requirement in the statute may sometimes alleviate vagueness concerns40 but does not always do so.41

What renders a statute vague is the "indeterminacy of precisely what" the prohibited conduct is.42 Statutes have been struck down as vague when they tied the defendant's criminal culpability to conduct that was "annoying" or "indecent" because those terms encompassed "wholly subjective judgments without statutory definitions, narrowing context, or settled legal meanings."43 The Supreme Court has also found a statute to be void for vagueness when it prohibited the charging of an "unjust or unreasonable rate," without further defining what "unjust or unreasonable" in this context meant.44 And in a First Amendment case involving concerns about the indeterminacy of a law, the Court has struck down a statute that prohibited the wearing of a "political badge, political button, or other political insignia" when "political" was not defined and had been haphazardly construed by the state courts.45

In Johnson, and in the subsequent case of Dimaya, the Supreme Court found "hopeless indeterminacy" in statutes that required a judge to determine whether the "ordinary case" of a particular statutory offense posed a "serious potential risk" of physical injury or "substantial risk" of physical force.46 The Court characterized this as the application of a "qualitative standard" of risk assessment to the "judge-imagined abstraction" of an "idealized ordinary case of the crime."47 The Court criticized this sort of assessment for not being tied to "real-world facts or statutory elements."48

2. Application

We conclude that the statute before us is vague in much the same way as the statutes in Johnson and Dimaya. Like those statutes, the statute before us is hopelessly indeterminate by being too abstract. As we shall see, the statute has little in the way of limiting language and notably lacks language to clarify its scope.

An offense is committed under § 551.143 if a member or group of members of a governmental body "knowingly conspires to circumvent this chapter by meeting in numbers less than a quorum for the purpose of secret deliberations in violation of this chapter."49 Viewed in isolation, the phrase "less than a quorum" could seem to serve a limiting function by carving out a subset of fact situations to which the statute applies, but an examination of this language in light of TOMA as a whole shows otherwise. Aside from the statute at issue here, TOMA's public-meeting provisions apply only when a governmental body meets as a "quorum."50 In specifying that an offense is committed when members meet in "less than a quorum," § 551.143 signifies a residual or catch-all provision, designed to enlarge TOMA's reach.51 Because the phrase "numbers less than a quorum" is catch-all language that expands the reach of TOMA, it does not serve a limiting function in the statute.

The words "meeting" and "deliberation" are defined in TOMA, but both definitions require a quorum,52 which seems to contradict § 551.143's use of these words in connection with the phrase "less than a quorum." As we explained earlier, the State claims that the definition of "meeting" is inapplicable because the definition is of "meeting" as a noun while § 551.143 uses the word as a verb. Even if we accept the State's contention in that regard, "deliberation" is used in § 541.143 as the same part of speech—a noun—for which it is defined.53 In any event, applying the statutory definitions literally to these words as they appear in § 551.143 would result in an internally inconsistent statute, so the definitions cannot serve to limit or clarify that provision.

Likewise, the words "in violation of this chapter" cannot also be construed literally because, aside from § 551.143, TOMA applies only when there is a quorum. If the requisite violation of TOMA requires meeting in a quorum and the person does not contemplate meeting in a quorum, then the person cannot literally have the purpose of violating TOMA.

The word "secret" indicates that § 551.143, like other parts of TOMA, is aimed at preventing meetings that are not open to the public. As such, the word serves a limiting function but, given the wide array of possible interactions between public officials, is not sufficient by itself to supply the requisite clarity to the statute.

What remains is probably the crucial part of the statute: "knowingly conspires to circumvent this chapter." In the past, the Supreme Court has warned against the potential breadth and vagueness of the doctrine of conspiracy and of the need to restrict its application.54 A conspiracy to violate a law55 would not ordinarily present a vagueness problem. But a conspiracy to "circumvent" a law is another matter.

What does it mean to "circumvent" a law? The court of appeals concluded that "circumvent" means "to overcome or avoid the intent, effect, or force of: anticipate and escape, check, or defeat by ingenuity or stratagem: make inoperative or nullify the purpose or power of esp. by craft or scheme."56 We accept that definition, but it does not really answer the question. What constitutes "avoiding or overcoming" the effect of the law or "nullifying the purpose" of the law? Consistent with our observation regarding other portions of § 551.143, the "circumvent" language necessarily requires something other than a literal violation of some other provision of TOMA. But proscribing a non-literal violation of TOMA does not set forth a clear standard. That is true even with the culpable mental state of "knowing." If it is unclear what it means to circumvent a law, one cannot "know" that he is circumventing the law.

And that is what makes this case like Johnson and Dimaya. Like the statutes in those cases, the statute in this case is hopelessly abstract. The present statute does not focus on real-world conduct other than catch-all conduct that expands the scope of TOMA. And § 551.143 does not focus on the elements of some other offense in TOMA. Rather, § 551.143 imposes criminal punishment for doing something that conflicts with the purpose of TOMA. It requires a person to envision actions that are like a violation of TOMA without actually being a violation of TOMA and refrain from engaging in them.

The statutory language here requires a sort of extratextual-factor inquiry that is unmoored to any statutory text. Ordinarily, we are limited to the text in construing a statute, but we have latitude to address extratextual factors when a statute is ambiguous or the literal text would lead to absurd results.57 Extratextual factors can include the object of the legislation and the consequences of a particular construction.58 Language that appears vague on its face "may derive much meaningful content from the purpose of the Act, its factual background, and the statutory context."59 However, even when a statute is ambiguous, it is ordinarily because the actual text is reasonably susceptible to more than one interpretation.60 It is one thing to use extratextual factors to help determine which of two or more competing interpretations of the text is probably the right one. It is quite another to engage in a free-floating extratextual inquiry to determine what a statute probably means. Even assuming that we could engage in the latter sort of inquiry under some circumstances, we could not do so for a statute that proscribes a criminal offense and that implicates protected expression under the First Amendment.

The State contends, however, that there is only one possible interpretation of the statute, and that it is the interpretation found in a 2005 attorney general opinion. That attorney general opinion concluded that § 551.143 applied to "members of a governmental body who gather in numbers that do not physically constitute a quorum at any one time but who, through successive gatherings, secretly discuss a public matter with a quorum of that body."61 The attorney general opinion referred to this as "a daisy chain of members the sum of whom constitute a quorum" or a "walking quorum."62

Even if the statute could be limited to a "daisy chain" of meetings or a "walking quorum," there are a number of different ways in which those concepts could be defined, and there is disagreement on whether certain situations qualify. A Louisiana court of appeals has described a "walking quorum" as a meeting "where different members leave the meeting and different members enter the meeting so that while an actual quorum is never physically present an actual quorum during the course of the meeting participates in the discussion."63 The Wisconsin Court of Appeals described a "walking quorum" as "a series of gatherings among separate groups of members of a governmental body, each less than quorum size, who agree, tacitly or explicitly, to act uniformly in sufficient number to reach a quorum."64 The Supreme Court of Ohio found an improper game of "legislative musical chairs" when a city manager called three series of back-to-back non-quorum meetings with groups of council members."65 A California appellate court concluded that one-on-one telephone calls with members of the governing body would suffice if the calls were essentially a poll to arrive at the collective agreement of the governing body.66 Hawaii's intermediate appellate court has held that "a series of one-on-one conversations relating to a particular item of Council business" circumvented the spirit of the state's open meeting law.67

Nevada's Supreme Court has held, however, that a "constructive quorum" is not necessarily established by back-to-back briefings conducted with agency members, that, taken as a whole, would add up to a quorum.68 That court further concluded that, in the absence of a quorum, it was not improper for members of a public body to "privately discuss issues or even lobby for votes."69 And Montana's Supreme Court declined to adopt a "constructive quorum" rule that would encompass "serial one-on-one discussions."70

Although these cases involve a variety of statutory schemes,71 their various conclusions point to the fact that there can be different ideas about what constitutes a "walking" or "constructive" quorum. Those ideas range from the narrow conception articulated by the Louisiana court of appeals—a single meeting at which a quorum is defeated by the mere expediency of different members stepping out of the room for a period of time—to the broad conception articulated by Hawaii's intermediate court—to include serial one-on-one communications with enough members to reach a quorum.

A broad view of what constitutes a "walking quorum" would constrain one-on-one lobbying for votes or even one-on-one discussions. Suppose a person is a member of a nine-member board, and he wishes a certain rule to be adopted, and he approaches another board member one-on-one to lobby that member to vote for his preferred rule. A discussion between two board members is not enough to make a quorum. But if the person then repeats that procedure with three other board members, individually approaching each one at different times, he has now approached a total of four members, which, with himself, constitutes a majority of the board. Whether that constitutes a "walking quorum" depends on how broad the concept really is. Under the "circumvents" language of § 551.143, this could be illegal, but it's not certain that it is.

But the "circumvents" language potentially sweeps even more broadly. If lobbying other members to achieve a majority vote is a "circumvention" under § 551.143, it may not even be necessary for a member to actually communicate with a majority-forming number72 of the members. Suppose, in the nine-member-board hypothetical, that the member who wants a certain rule passed knows that one of the other members already intends to vote for the rule. To get a majority vote for his preferred rule, the first member need only persuade three other members. If he lobbies those three members, he has not communicated with a quorum, but his purpose is to ensure that a majority—which is a quorum—votes his way.

Suppose, instead, that the member who wants a certain rule passed knows that three other members already intend to vote for the rule. To get a majority, he need persuade only one other member. He communicates with only that one member in an attempt to sway that person's vote. The purpose of his communication is still to ensure that a majority—again a quorum—votes his way. To the protest that this scenario strays beyond any recognized concept of "walking quorum," the answer is that, contrary to the State's contention and the Attorney General's opinion, the "circumvents" language in § 551.143 is not necessarily limited by the concept of a "walking quorum." If lobbying other members to get a majority vote circumvents TOMA, then lobbying even a single member of a more-than-three-member board could do so.73

But it gets worse, because the "circumvents" language can conceivably reach even further. Suppose, in the nine-member board hypothetical, that seven of the members have decided how they will vote on the rule at issue, with the vote split four to three. The two remaining undecided members discuss the issue between themselves to decide how they stand on it. That discussion could be viewed as a circumvention because the two undecided members hold the votes that would resolve the issue one way or another.

What if one member knows enough about other members to be reasonably sure how they will vote on a given issue, even if they have not yet expressed their thoughts? How sure does one have to be that communicating with another member will ultimately be decisive on a matter of official business before one runs afoul of the law? And the net that the word "circumvents" casts may be even wider. If part of the purpose of having an open meeting is for the public to see all of the information received by the public officials, then receiving information in a one-on-one session might itself be viewed as a "circumvention" of TOMA. All of this discussion reinforces our conclusion that the language in § 551.143 is potentially very broad and lacks any reasonable degree of clarity on what it covers. We also conclude that protected speech is likely to be chilled because of the great degree of uncertainty about what communications government officials may engage in.

E. Narrowing Construction

We have a duty to employ a reasonable narrowing construction to avoid a constitutional violation, but we can employ such a construction only if the statute is readily susceptible to one.74 We may not rewrite a statute that is not subject to a narrowing construction, because such a rewriting "constitutes a serious invasion of the legislative domain."75 A statute is readily subject to a narrowing construction only "if the language already in the statute can be construed in a narrow manner. Adding language to a statute is legislating from the bench."76 Even when faced with a vague statute, we will not impose a narrowing construction when one "would add significant content not now present in the statute and could be fashioned in a number of different ways."77 In considering a narrowing construction, we should take into account that vague laws, even when not overtly invidious, "invite the exercise of arbitrary power . . . by leaving the people in the dark about what the law demands and allowing prosecutors and courts to make it up."78

We do not doubt the legislature's power to prevent government officials from using clever tactics to circumvent the purpose and effect of the Texas Open Meetings Act. But the statute before us wholly lacks any specificity, and any narrowing construction we could impose would be just a guess, an imposition of our own judicial views. This we decline to do.

F. Conclusion

In light of the above discussion, we conclude that § 551.143 is unconstitutionally vague on its face. We reverse the judgment of the court of appeals and affirm the judgment of the trial court.

Slaughter, J., filed a concurring opinion.

Yeary, J., filed a dissenting opinion.

Newell, J., dissented.

CONCURRING OPINION

Slaughter, J., filed a concurring opinion.

I agree with the Court's conclusion that the indictment against Craig Doyal, Appellee, must be dismissed because Government Code Section 551.143 is unconstitutional. But I disagree with the Court's reasoning in reaching that decision. I do not believe that the statute is impermissibly vague. Rather, I believe that the statute "abridg[es] the freedom of speech" in violation of the First Amendment of the United States Constitution. As such, I respectfully concur in this Court's decision to reverse the judgment of the court of appeals and uphold the trial court's order dismissing the indictment against Appellee, but I do not join the Court's opinion.

I. Section 551.143 is not impermissibly vague.

This Court's opinion holds that Government Code Section 551.143 is unconstitutionally vague on its face, in that it fails to provide adequate notice of the prohibited conduct and/or fails to provide sufficiently definite guidelines for law enforcement. I disagree. The statutory language, viewed as a whole and in the context of the remaining provisions in the Open Meetings Act, is adequate to place an ordinary officeholder on notice of the prohibited conduct and to prevent arbitrary enforcement. Further, the requirement of mandatory training for public officials on the provisions of the Open Meetings Act and the resources provided by the Attorney General's Office substantially reduce the likelihood that officials will be deprived of a reasonable opportunity to know what the law prohibits.

The statute at issue here, Government Code Section 551.143, provides:

TEX. GOV'T CODE § 551.143(a).

As shown below, this statute is not vague because it gives a person of ordinary intelligence a reasonable opportunity to know what is prohibited, and establishes definite guidelines for law enforcement. See Scott v. State, 322 S.W.3d 662, 665 n. 2 (Tex. Crim. App. 2010).

It is a basic principle of due process that an enactment is void for vagueness if its prohibitions are not clearly defined. Grayned v. City of Rockford, 408 U.S. 104, 108, 92 S.Ct. 2294, 33 L.Ed.2d 222 (1972). The vagueness doctrine prohibits the government from "taking away someone's life, liberty, or property under a criminal law so vague that it fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement." Johnson v. United States, ___ U.S. ___, 135 S.Ct. 2551, 2556, 192 L.Ed.2d 569 (2015) (citing Kolender v. Lawson, 461 U.S. 352, 357-58, 103 S.Ct. 1855, 75 L.Ed.2d 903 (1983)).1 Thus, "[a] statute may be challenged as unduly vague, in violation of the Due Process Clause of the Fourteenth Amendment, if it does not: (1) give a person of ordinary intelligence a reasonable opportunity to know what is prohibited, and (2) establish definite guidelines for law enforcement." Scott, 322 S.W.3d at 665 n.2.

A. By including the culpable mental state of "knowingly conspiring to circumvent" TOMA, the statute provides fair notice of what is prohibited.

As held by the United States Supreme Court, a statute that may otherwise be found impermissibly vague may be saved by including a culpable mental state. See McFadden v. United States, ___ U.S. ___, 135 S.Ct. 2298, 2307, 192 L.Ed.2d 260 (2015) ("Under our precedents, a scienter requirement in a statute `alleviate[s] vagueness concerns,' `narrow[s] the scope of the [its] prohibition[,] and limit[s] prosecutorial discretion.'") (quoting Gonzales v. Carhart, 550 U.S. 124, 149, 150, 127 S.Ct. 1610, 167 L.Ed.2d 480 (2007) (modifications in original)).2 Here, the statute conditions criminal liability on proof of the actor's guilty mind. That is, an actor is subject to prosecution only if he "knowingly conspires to circumvent" the Open Meetings Act by engaging in the specified conduct. See § 551.143(a).

As noted in the Court's opinion, a statute's inclusion of a culpable mental state does not invariably alleviate vagueness concerns. But this particular statute's wording—requiring proof of the actor's awareness that he is making a secret agreement with others to overcome or avoid the requirements of the Open Meetings Act3—makes it highly unlikely that the statute will be applied to individuals who genuinely believe their conduct to be lawful or who are utterly unaware that their conduct is prohibited. If one of the principal concerns underlying the vagueness doctrine is that vague laws will "trap the innocent by not providing fair warning," see Grayned, 408 U.S. at 108, 92 S.Ct. 2294, then that concern is substantially mitigated by the statute's inclusion of this language requiring proof of the actor's guilty mind in knowingly conspiring to circumvent the Open Meetings Act.4

B. The language at issue, when viewed in context of the other Chapter 551 provisions, provides clarity for what is prohibited.

The Court concludes that, in spite of Section 551.143's culpable mental state requirement, the statute is nevertheless vague on its face because the conduct for which the actor must possess the requisite culpability—"conspires to circumvent this chapter by meeting in numbers less than a quorum for the purpose of secret deliberations in violation of this chapter"—is hopelessly abstract. This language may appear vague when each term is viewed in isolation, but its meaning becomes sufficiently clear when it is considered holistically and in the broader context of the other provisions in Chapter 551.

Section 551.002, entitled "Open Meetings Requirement," broadly provides that, subject to limited exceptions, all governmental meetings must be open to the public. TEX. GOV'T CODE § 551.002 ("Every regular, special, or called meeting of a governmental body shall be open to the public, except as provided by this chapter."). The governmental body must give written notice of the date, hour, place, and subject of each meeting held by the governmental body. Id. § 551.041. The noun form of "meeting" is statutorily defined, and its meaning is quite sweeping as it encompasses all deliberations by a quorum of a governmental body, or between a quorum of a governmental body and a third person, during which public business is discussed, considered, or the subject of formal action. Id. § 551.001(4)(A).5 The word "deliberation" is also broadly defined as "a verbal exchange during a meeting between a quorum of a governmental body, or between a quorum of a governmental body and another person, concerning an issue within the jurisdiction of the governmental body or any public business." Id. § 551.001(2). Quorum is separately defined as "a majority of a governmental body, unless defined differently by applicable law or rule or the charter of the governmental body." Id. § 551.001(6).

Viewing these provisions in conjunction, it becomes apparent that Chapter 551's overarching purpose is to broadly require that all deliberations by a quorum of a governmental body be conducted in a manner that is accessible to the public through an open meeting; in other words, a quorum may not discuss any issue over which it has jurisdiction without holding a properly-called meeting that is open to the public. Interpreting the language of Section 551.143 in this particularized context, the statute's meaning becomes sufficiently clear such that an ordinary person would be fairly placed on notice of the prohibited conduct: a member or members of a governmental body may not knowingly conspire to circumvent Chapter 551's clear requirement of open deliberations by meeting in numbers less than a quorum under circumstances that reflect a purpose to secretly discuss public business with a quorum of that body without holding a properly-called meeting. A similar interpretation of the statute was embraced more than ten years ago in an advisory opinion by former Attorney General Greg Abbott, which stated:

Tex. Att'y Gen. GA-0326 (2005), at 3-4 (citations omitted).6 This manner of forming a quorum has become commonly known as a "walking quorum."

Appellee rejects this interpretation and instead suggests that the statutory language is self-contradictory and thus defective on its face. Specifically, Appellee contends that the statutory definitions of "meeting" and "deliberations" expressly refer to conduct by a quorum, but Section 551.143 refers to "meeting in numbers less than a quorum." TEX. GOV'T CODE §§ 551.001(2), (4); 551.143. Appellee asserts that this apparent contradiction renders the statute fatally flawed because, in light of the statutory definitions, it is impossible for a governmental body to "meet" or "deliberate" in numbers less than a quorum.7

In analyzing the word "meeting" in the statute, Appellee ignores the fact that only the noun form of that word is statutorily defined, whereas Section 551.143 uses "meeting" as a verb. As such, the technical definition of "meeting" as a noun is not implicated by Section 551.143. See Tex. Att'y Gen. GA-0326 (2005), at 3 ("[T]he section 551.001 definition of `meeting' as a noun does not apply here because section 551.143 employs the word as a verb."). "Thus, the phrase `meeting in numbers less than a quorum' does not present a legal dilemma because the plain meaning of `meeting' as a verb does not require a quorum." Id. ("Furthermore, we read `meeting in numbers less than a quorum' to have a particular meaning that does not render the provision circular.").

The word "deliberations," however, is used as a noun in both the statutory definition and in Section 551.143. The statutory definition does refer to a verbal exchange during "a meeting between a quorum of a governmental body," whereas Section 551.143 expressly refers to "meeting in numbers less than a quorum for the purpose of secret deliberations in violation of this chapter." See TEX. GOV'T CODE §§ 551.001(2), 551.143 (emphasis added). Given this, it appears that a literal application of the statutory definition of "deliberations" in the context of Section 551.143 would create an absurdity by rendering the statutory language internally self-contradictory.8 But this Court is not bound to adopt an interpretation of the statute that would render its language illogical or absurd. Instead, this Court is obligated to seek out any reasonable construction that upholds the statute while giving effect to the Legislature's intent. Boykin v. State, 818 S.W.2d 782, 785-86 (Tex. Crim. App. 1991) (in construing statutes, we must "seek to effectuate the `collective' intent or purpose of the legislators who enacted the legislation"; if a statute's plain language leads to an absurd result, this Court "should not apply the language literally," but should seek to arrive at a "sensible interpretation" consistent with the Legislature's purpose); see also Skilling v. United States, 561 U.S. 358, 406, 130 S.Ct. 2896, 177 L.Ed.2d 619 (2010) ("`The elementary rule is that every reasonable construction must be resorted to, in order to save a statute from unconstitutionality.'") (quoting Hooper v. California, 155 U.S. 648, 657, 15 S.Ct. 207, 39 S.Ct. 297 (1895)).

If one considers the entire statutory phrase, "meeting in numbers less than a quorum for the purpose of secret deliberations in violation of this chapter," it becomes apparent that the Legislature could not have intended for the statutory definition of "deliberations" to apply literally here. Deliberations "between a quorum" that take place "during a meeting" that is open to the public are not "secret," and thus they are not in violation of Chapter 551. See TEX. GOV'T CODE §§ 551.001(2), 551.002, 551.143. Looking beyond the technical statutory definition of "deliberations," the ordinary definition of that word is "a discussion and consideration by a number of persons of the reasons for and against a measure." WEBSTER'S NEW INTERNATIONAL DICTIONARY 596 (3d ed. 2002). Placing this ordinary definition in the context of the phrase in which it appears ("secret deliberations in violation of this chapter"), it becomes sufficiently clear what the core conduct targeted by the statute is—knowingly evading the requirement of open meetings by gathering in smaller groups that do not comprise a quorum for the ultimate purpose of secretly discussing government business with a quorum without holding a proper meeting. Such secret deliberations by a quorum are clearly "in violation of" the remaining provisions in Chapter 551. See, e.g., TEX. GOV'T CODE § 551.002, 551.041. This understanding of the statutory language has been embraced by other courts,9 as well as by the Attorney General as discussed above. Because this interpretation is a reasonable one and is consistent with the Legislature's intent in broadly requiring open deliberations by governmental bodies, this Court is obligated to adopt such a construction before striking the statute on vagueness grounds. Skilling, 561 U.S. at 406, 130 S.Ct. 2896.

Moreover, to the extent that the Court's opinion suggests the statutory language is so ambiguous as to render the statute vague, I disagree with this reasoning because it conflates the concept of ambiguity with vagueness. See Antonin Scalia & Bryan A. Garner, READING LAW: THE INTERPRETATION OF LEGAL TEXTS 31-32 (2012) ("A word or phrase is ambiguous when the question is which of two or more meanings applies; it is vague when its unquestionable meaning has uncertain application to various factual situations.").10 Because any ambiguity here may be resolved in such a way as to effectuate the clear legislative intent to prohibit knowing evasions of open-meetings requirements, it cannot be the basis for finding the statute void for vagueness. Further, although the Court's opinion uses numerous hypothetical scenarios and suggests that the statute is hopelessly unclear as to each, each of these examples is flawed for various reasons. Most do not take into account the required mental state, and the actors involved would fail to possess that required mental state. Other examples are flawed because they focus on outlier situations. Many are likely not encompassed within the statutory prohibition because they do not involve deliberations by a quorum. While several of these hypotheticals admittedly show that it may be difficult to discern whether Section 551.143 would apply to certain marginal cases, that fact alone does not prove the statute's vagueness. As the Supreme Court has explained, the mere fact that it may occasionally be difficult to determine how a statute applies to a particular fact pattern does not render the statute facially vague, for "even clear rules `produce close cases.'" Salman v. United States, ___ U.S. ___, 137 S.Ct. 420, 429, 196 L.Ed.2d 351 (2016) (quoting Johnson, 135 S.Ct. at 2560).11

C. The Legislature's statutorily-mandated training and resources ensures that a person of ordinary intelligence has a reasonable opportunity to know what is prohibited.

Setting aside the statutory language of Section 551.143, the mandatory training provisions in Chapter 551 further serve to alleviate concerns about the statute's potential for vagueness. Section 551.005 mandates that "[e]ach elected or appointed public official who is a member of a governmental body subject to this chapter shall complete a course of training of not less than one and not more than two hours regarding the responsibilities of the governmental body and its members under this chapter[.]" TEX. GOV'T CODE § 551.005(a).12 The statute provides that the Office of the Attorney General is responsible for ensuring that the training is made available. Id. § 551.005(b).13 The current one-hour video training provided by the Attorney General discusses the issue of criminal penalties for conduct amounting to a "walking quorum." To illustrate what that concept means, the video discusses the facts of a federal district court decision in Esperanza Peace & Justice Ctr. v. City of San Antonio, in which that court determined that the San Antonio City Council had violated Section 551.143 by gathering successively in small groups the night before a budget vote for the purpose of forming a consensus about how to vote on the budget. 316 F.Supp.2d 433, 472 (W.D. Tex. 2001).14

In addition to the mandatory training video that discusses the concept of a walking quorum, the corresponding "Open Meetings Handbook" also references this concept in multiple places, observing that a walking quorum may arise when members of a governmental body try to "avoid complying with the Act by deliberating about public business without a quorum being physically present in one place."15 The Office of the Attorney General also operates an Open Meetings Hotline, staffed by attorneys, for the purpose of answering public officials' questions about the scope of the law.

With such an abundance of readily-available resources providing public officials with an understanding of the law (some of which is required viewing), this situation is distinguishable from a typical vagueness challenge in which an innocent actor may be forced to roll the dice by guessing at the meaning of indeterminate or abstract statutory terms, only finding out later if his actions are prohibited. Having statutorily-mandated training and readily-available informational resources is surely a factor that we may take into consideration in evaluating whether the ordinary officeholder is afforded a fair opportunity to know whether a particular course of conduct is prohibited. These training requirements have been in place since 2006,16 and information regarding Section 551.143's prohibition on walking quorums has been included in official training materials for more than a decade.17 If the Legislature believed that the resources currently provided were misleading or constituted an incorrect statement of the law, the Legislature would have amended the statute to clarify its intent. That it has not done so reflects implied approval of the foregoing interpretation by the Legislature.

Further, the fact that training is required under Chapter 551 is not an indication that Section 551.143 is vague. Although the requirements of the Open Meetings Act are admittedly complex, complexity is not the equivalent of vagueness. See Asgeirsson v. Abbott, 696 F.3d 454, 466 (5th Cir. 2012) (rejecting vagueness challenge to TOMA, and noting that plaintiffs had "point[ed] to no section of TOMA that is vague on its face"). As the Fifth Circuit explained in Asgeirsson,

Id.18

In sum, because the statutory language is susceptible to a reasonable interpretation that renders the statute's meaning sufficiently definite, and because ample training resources help to define the scope of the law, Section 551.143 is not void on its face for vagueness. While the statutory language should be clarified in some respects, the statute is not so indeterminate, abstract, or subjective that government officials will be deprived of fair notice of the conduct that it prohibits. See Williams, 553 U.S. at 306, 128 S.Ct. 1830 (upholding federal child pornography statute over vagueness challenge where statute involved "clear questions of fact" and "true-or-false determinations," rather than "wholly subjective judgments without statutory definitions, narrowing context, or settled legal meanings"). Nor is the statute so standardless that it invites arbitrary enforcement. Johnson, 135 S.Ct. at 2556. Because it is clear what the statute as a whole prohibits, I reject Appellee's facial vagueness challenge and disagree with the Court's analysis.

II. Section 551.143 violates the First Amendment because it abridges the freedom of speech.

While Section 551.143 is not void for vagueness, it "abridg[es] the freedom of speech" in violation of the First Amendment of the United States Constitution. By criminalizing all policy discussions by a quorum of members of a governmental body outside the context of a formal meeting, the statute significantly infringes on the rights of governmental officials to engage in the free exchange of ideas that are essential to effective governance. The State has not established that this sweeping regulation prohibiting even informal policy discussions outside of a formal meeting is necessary to achieve its interest in maintaining an open and transparent government. Therefore, I would hold that the statute fails to pass constitutional muster and violates the First Amendment.

A. Section 551.143 is a content-based restriction on speech.

The First Amendment commands that the government "shall make no law . . . abridging the freedom of speech." U.S. Const. amend. I. Applicable to the States through the Fourteenth Amendment, the Free Speech Clause prohibits the government from restricting speech "because of its message, its ideas, its subject matter, or its content." Reed v. Town of Gilbert, ___ U.S. ___, 135 S.Ct. 2218, 2226, 192 L.Ed.2d 236 (2015) (internal citations and quotations omitted).

When a statute is attacked as unconstitutional, courts usually begin with the presumption that the law is valid and that the Legislature has not acted unreasonably or arbitrarily. Ex parte Lo, 424 S.W.3d 10, 15 (Tex. Crim. App. 2013). However, when the government seeks to restrict and punish speech based on its content, the usual presumption of constitutionality is reversed; the content-based regulation is presumptively invalid, and the State bears the burden to rebut this presumption. Id. A law is content based "[i]f it is necessary to look at the content of the speech in question to decide if the speaker violated the law." Ex parte Thompson, 442 S.W.3d 325, 345 (Tex. Crim. App. 2014). This commonsense meaning of the phrase "content based" requires a court to consider whether a regulation of speech "on its face" draws distinctions based on the message a speaker conveys. Reed, 135 S.Ct. at 2227.19 Such a regulation may be upheld only if it is necessary to serve a compelling state interest and employs the least restrictive means to achieve its goal. Lo, 424 S.W.3d at 15; see also Reed, 135 S.Ct. at 2226 ("Content-based laws—those that target speech based on its communicative content—are presumptively unconstitutional and may be justified only if the government proves that they are narrowly tailored to serve compelling state interests.").

"The Constitution gives significant protection from overbroad laws that chill speech within the First Amendment's vast and privileged sphere." Ashcroft v. Free Speech Coal., 535 U.S. 234, 244, 122 S.Ct. 1389, 152 L.Ed.2d 403 (2002). Under the First Amendment's "overbreadth" doctrine, "a statute is facially invalid if it prohibits a substantial amount of protected speech" in relation to its plainly legitimate sweep. Williams, 553 U.S. at 292, 128 S.Ct. 1830; Ex parte Perry, 483 S.W.3d 884, 902 (Tex. Crim. App. 2016). "The person challenging the statute must demonstrate from its text and from actual fact `that a substantial number of instances exist in which the Law cannot be applied constitutionally.'" Perry, 483 S.W.3d at 902 (citing New York State Club Ass'n v. City of New York, 487 U.S. 1, 14, 108 S.Ct. 2225, 101 L.Ed.2d 1 (1988)). In conducting an overbreadth analysis, the first step "is to construe the challenged statute," as "it is impossible to determine whether a statute reaches too far without first knowing what it covers." Williams, 553 U.S. at 293, 128 S.Ct. 1830; Ex parte Perry, 483 S.W.3d at 902.

In construing Section 551.143, the court of appeals held that the provision targets conduct rather than speech, and thus does not implicate First Amendment protections. State v. Doyal, 541 S.W.3d 395, 401 (Tex. App.-Beaumont 2018). As discussed in the Court's opinion, this conclusion by the court of appeals was erroneous because Section 551.143 reaches speech and not merely conduct. See Maj. Op. at 143 (explaining that "the statute does not proscribe `meeting' in the abstract but is directed at a particular kind of meeting—one that is for the purpose of `deliberations'"). As discussed supra, Section 551.143 only prohibits meeting for the purpose of "secret deliberations in violation of this chapter"—that is, secret policy discussions by a quorum of a governmental body outside the proper forum of a public meeting. A discussion necessarily requires speech. Therefore, I agree with the Court's opinion that Section 551.143 regulates speech rather than conduct. I also agree with the Court's determination that the statute reaches protected speech, for "the First Amendment's protection of elected officials' speech is robust and no less strenuous than that afforded to the speech of citizens in general." Maj. Op. at 142 (citing Rangra v. Brown, 566 F.3d 515, 524 (5th Cir. 2009), vacated as moot en banc by 584 F.3d 206 (5th Cir. 2009)). Thus, the First Amendment is implicated, and the next question becomes the level of scrutiny to apply, which in turn depends on whether the statute is content based or content neutral.

B. Because the statute is content-based, strict scrutiny applies.

In 2015, the United States Supreme Court issued its opinion in Reed. In Reed, the Supreme Court emphasized that a statute that is content based on its face is subject to strict scrutiny regardless of the government's justification for the law. The Court stated:

Reed, 135 S.Ct. at 2228 (internal citations and quotations omitted).20

Reed has shifted the landscape for analyzing First Amendment claims by holding that all content-based regulations, even if they appear to be viewpoint neutral, are subject to strict scrutiny.21 The court of appeals failed to rely on Reed and instead cited the United States Fifth Circuit Court of Appeals's decision in Asgeirsson v. Abbott, 696 F.3d at 459-60, for the proposition that "[a] statute that appears content-based on its face may still be deemed content-neutral if it is justified without regard to the content of the speech." Doyal, 541 S.W.3d at 400. Reed, however, was decided three years after Asgeirsson and calls into question Asgeirsson's reasoning.

Under Reed, Section 551.143 is plainly content based on its face. The statute prohibits conspiring to circumvent the act by "meeting in numbers less than a quorum for the purpose of secret deliberations." TEX. GOV'T CODE § 551.143 (emphasis added). Whether one looks to the common or the statutory definition of "deliberations," a specific category of speech is implicated—matters within the jurisdiction of the governmental body or matters pertaining to public business. The common definition of "deliberations" involves a discussion regarding the "reasons for and against a [governmental] measure." WEBSTER'S NEW INTERNATIONAL DICTIONARY 596 (3d ed. 2002). The statutory definition of "deliberations" involves a "verbal exchange . . . concerning an issue within the jurisdiction of the governmental body or any public business." See TEX. GOV'T CODE § 551.001(2) (emphasis added). Thus, it is a violation to discuss a particular subject matter. If members of a governing body subject to TOMA met privately in numbers less than a quorum to discuss golfing or the latest fashion trends, there would be no violation of Section 551.143 because discussing those topics could never amount to a deliberation within the meaning of the statute. However, if during that same meeting, members informally discussed an issue within the jurisdiction of the governing body or public business while being aware that such discussions are prohibited by the Act, then there likely would be a violation of the statute. It is not until the content of the discussion is examined that a determination can be made as to whether the provision was violated. As such, Section 551.143 discriminates against an entire subject matter of speech and falls squarely under Reed. See Reed, 135 S.Ct. at 2227 ("[A] speech regulation targeted at specific subject matter is content based even if it does not discriminate among viewpoints within that subject matter."). Therefore, Section 551.143 is subject to strict scrutiny.

C. Section 551.143 is not narrowly tailored to further the compelling state interest.

Because strict scrutiny applies, the State must demonstrate that Section 551.143 furthers a compelling state interest and is narrowly tailored to achieve that interest. Lo, 424 S.W.3d at 15. A narrowly-tailored regulation "employs the least restrictive means to achieve its goal." Id. "If a less restrictive means of meeting the compelling interest could be at least as effective in achieving the legitimate purpose that the statute was enacted to serve, then the law in question does not satisfy strict scrutiny." Id. at 15-16. Further, the government's chosen restriction on the speech at issue must be "`actually necessary' to achieve its interest"; in other words, there must be a "direct causal link between the restriction imposed and the injury to be prevented." United States v. Alvarez, 567 U.S. 709, 725, 132 S.Ct. 2537, 183 L.Ed.2d 574 (2012) (plurality op.) (quoting Brown v. Entertainment Merchants Ass'n, 564 U.S. 786, 799, 131 S.Ct. 2729, 180 L.Ed.2d 708 (2011)).

The State asserts compelling interests of transparency, faith in government, creating an environment where corruption cannot thrive, and protecting the rights of public officials to observe and participate in the public policy making for which they were elected. While these certainly are compelling goals, the State has failed to meet its burden in proving that Section 551.143 is necessary to achieve these interests and constitutes the least restrictive means of accomplishing these goals.

Broadly speaking, courts have recognized that the overarching purpose of TOMA is to ensure the transparency of governmental actions and to protect the public's interest in knowing how governmental decisions are made. See Acker v. Texas Water Comm'n, 790 S.W.2d 299, 300 (Tex. 1990) (recognizing purpose of TOMA is to ensure "that the public has the opportunity to be informed concerning the transactions of public business"); Finlan v. City of Dallas, 888 F.Supp. 779, 783 (N.D. Tex. 1995) ("The purpose of the Texas Open Meetings Act is to protect the public's interest in knowing the workings of its governmental bodies. . . . Thus, the public policy embodied in the TOMA is that, absent compelling reasons to the contrary, the public business should be conducted in public.").22 TOMA seeks to achieve this purpose by requiring "openness at every stage of a governmental body's deliberations." Acker, 790 S.W.2d at 300. Pursuant to the regulations in TOMA, citizens are "entitled not only to know what government decides but to observe how and why every decision is reached." Id. Thus, the Act requires that "[t]he executive and legislative decisions of our governmental officials as well as the underlying reasoning must be discussed openly before the public rather than secretly behind closed doors." Id.

How does Section 551.143 fit into this broader asserted goal of ensuring an open and transparent government? As noted above, Section 551.143 operates to prohibit public officials from conducting "secret deliberations" and "walking quorums" that would permit governmental bodies to engage in policy discussions outside the public eye. As the Texas Supreme Court noted in Acker, TOMA strictly prohibits this type of informal discussion amongst governmental officials—"When a majority of a public decisionmaking body is considering a pending issue, there can be no `informal' discussion. There is either formal consideration of a matter in compliance with the Open Meetings Act or an illegal meeting." Id. But why is such a sweeping regulation necessary to ensure public knowledge of and access to governmental affairs? Must the public be privy to every single conversation amongst members of a governmental body in order to prevent corruption and to ensure access and transparency? The State has not put forth any evidence, aside from sweeping generalizations and speculation, to support the notion that even informal discussions by public officials prior to a formal meeting will frequently result in corruption, secret decision-making, or a lack of transparency. Because other provisions in TOMA require the holding of formal meetings to conduct actual business, such as casting votes, it cannot be said that the public would be denied access to governmental decision-making in the absence of Section 551.143.23 Moreover, the remedy for a governmental decision made in the absence of a public meeting is to invalidate the decision. Even without the provisions in Section 551.143, the government cannot take any binding formal action absent a properly-held public meeting. See Esperanza Peace, 316 F.Supp.2d at 477 ("Governmental actions taken in violation of the Act are subject to judicial invalidation.") (citing Smith County v. Thornton, 726 S.W.2d 2, 3 (Tex. 1986)). Further, voters can determine whether their elected officials are providing enough transparency.

By prohibiting all informal discussions amongst officeholders—even those that are not aimed at secrecy or corrupt decision-making—Section 551.143 sweeps too broadly to include innocent speech that has no bearing on TOMA's purposes. While everyone seems to agree that the government may validly regulate conduct that would amount to secret and/or corrupt decision-making outside the public eye, Section 551.143 goes far beyond that by prohibiting even informal deliberations which might aid governmental officials in learning about issues and perspectives ahead of a formal vote. Because of this, rather than advancing the government's interests in effective government, Section 551.143 arguably undermines the broader purpose of TOMA to ensure effective representation for all citizens.24

Further, due to the significant threat of criminal sanctions, Section 551.143 operates to chill even more speech than is already encompassed within the statute's broad scope. Many public officials, out of fear of even just being accused of a TOMA violation, avoid communicating with each other or even being seen together outside of an official meeting.25 This chilling effect results in a significant infringement upon the rights of public officials to communicate one-on-one regarding policy issues.

While of course our Texas Legislature has the right to criminalize open-meeting violations if the statute is otherwise constitutional, most states find that criminalizing open-meetings law violations is unnecessary. Legal scholars note that all fifty states have open meeting laws, but the majority of those state laws do not impose criminal sanctions for violating the open meeting requirements. See Devon Helfmeyer, Do Public Officials Leave Their Constitutional Rights at the Ballot Box — A Commentary on the Texas Open Meetings Act, 15 TEX. J. ON CIVIL LIBERTIES & CIVIL RIGHTS 205, 227-30 (2010) (noting that only nineteen states impose criminal sanctions for violating open-meetings laws, and of those, only twelve—Texas included—allow prison as a punishment option); Christopher J. Diehl, Open Meetings and Closed Mouths: Elected Officials' Free Speech Rights After Garcetti v. Ceballos, 61 CASE W. RES. L. REV. 551, 593 (2010). Stated plainly, "[t]his shows that criminal penalties, particularly imprisonment, are not necessary to the proper and effective functioning of open meeting laws." Diehl, supra, at 593; see also Helfmeyer, supra, at 231 ("The federal and numerous state open meetings laws that lack criminal provisions indicate that less restrictive means are available to advance the goal of open government and access to information.").

The State puts forth a number of arguments in support of its position that Section 551.143 is compliant with the First Amendment, including that the statute may be upheld as a reasonable time, place, or manner restriction that is subject to intermediate scrutiny. Judge Yeary similarly posits that the statute may be upheld as a reasonable time, place, or manner regulation on speech in a nonpublic forum.26 But, even assuming that some lesser level of scrutiny would apply here, for all the reasons I have just described, the statute fails to meet even these less rigorous standards of review. To pass constitutional muster, even these lesser standards would require a reasonably precise fit between the government's asserted goal and its chosen means of advancing these goals. By prohibiting significantly more speech than is necessary to ensure transparency and a lack of corruption, and by infringing upon an even greater amount of speech due to a chilling effect on public officials' speech, the statute fails to pass constitutional muster under any formulation of First Amendment scrutiny.

In sum, the State has not shown how criminalizing informal, initial discussions by a governmental body prior to a formal meeting is necessary to ensure its interest in transparency and public access to governmental deliberations. In other words, the State has failed to meet its burden of demonstrating how prohibiting any and all policy discussions by a quorum of a governmental body outside of the public eye is narrowly tailored to serve the compelling state interests of promoting public trust, public access in government, and transparency. Therefore, because the statute sweeps up more speech than is required to fulfill the government's asserted purpose, the statute fails to comport with the First Amendment and is fatally overbroad.

III. Conclusion

I would hold that Section 551.143 unduly infringes upon the First Amendment rights of officeholders in this state to engage in informal discourse amongst themselves regarding matters within the jurisdictions of their offices. Although I recognize that having an open and transparent government is a compelling interest which justifies some regulation in this area, the government has failed to show that this sweeping regulation that prohibits essentially all private policy discussions amongst officeholders is narrowly tailored to achieve that interest. On this basis, I respectfully disagree with the Court's reasoning but concur in its judgment reversing the judgment of the court of appeals and upholding the trial court's order dismissing the indictment.

DISSENTING OPINION

Yeary, J., filed a dissenting opinion.

Yet another perfectly good statute falls today, adding fuel to the claims that this Court is often too quick to reject the considered will of our state's Legislative Department.1 In my opinion, striking this law is unnecessary. The Court's decision to strike the law relies on opinions from the United States Supreme Court that are, in the first place, less than a model of clarity, and that, in any event, are not at all like the case before us. It is also a product of the Court's failure to perceive the rather plain import of the Legislature's choice of words establishing a very simple prohibition: "conspiring to circumvent the Open Meetings Act by meeting in numbers less than a quorum for the purpose of secret deliberations [that would otherwise violate the Act]."

Relying on Johnson v. United States, ___ U.S. ___, 135 S.Ct. 2551, 192 L.Ed.2d 569 (2015), and Sessions v. Dimaya, ___ U.S. ___, 138 S.Ct. 1204, 200 L.Ed.2d 549 (2018), the Court concludes that "a vagueness challenge to a statute that implicates First Amendment freedoms does not require a showing that there are no possible instances of conduct clearly falling within the statute's prohibitions." Then, relying on its own opinion in Long v. State, 931 S.W.2d 285 (Tex. Crim. App. 1996), the Court refuses even to require a showing that the statute is vague as applied to Appellee. I am unconvinced that Appellee ought to be able to prevail in his facial vagueness challenge if he cannot make these showings.

I would hold (for some, but not all, of the reasons identified in Judge Slaughter's concurring opinion) that Section 551.143(a) of the Government Code, the Texas Open Meetings Act, is not unconstitutionally vague. TEX. GOV'T CODE § 551.143(a). But I disagree with Judge Slaughter that it nevertheless violates the First Amendment to the United States Constitution—an issue that the Majority need not address, having struck the statute on vagueness grounds. I write further to explain the reasons for my dissent.

I. VAGUENESS

Today the Court allows Appellee to prevail in a facial challenge to the constitutionality of Section 551.143(a) without having to demonstrate that it would be impermissibly vague in all of its applications. Majority Opinion at 143-46. I am unconvinced that this reflects an accurate assessment of the law. Moreover, why should Appellee be permitted to prevail in a facial vagueness claim to dismiss the prosecution against him when we do not even know what the facts of his case may show? Indeed, the Court today affirms a judgment granting Appellee's motion to dismiss under circumstances in which it is entirely possible he would not even be able to prevail in an as-applied challenge. I cannot go along with this.

A. In a Facial Challenge, Must Appellee Show That the Statute is Vague in All of its Applications?

When a litigant raises a facial challenge to a statute on ordinary vagueness grounds, based on the Due Process Clause of the United States Constitution, a court should uphold the challenge only if the enactment is impermissibly vague in all of its applications. A [litigant] who engages in some conduct that is clearly proscribed cannot complain of the vagueness of the law as applied to the conduct of others.

Village of Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489, 495, 102 S.Ct. 1186, 71 L.Ed.2d 362 (1982).2 It is true that the Supreme Court has held that when First Amendment rights are implicated, a "more stringent vagueness test should apply." Hoffman, 455 U.S. at 495, 102 S.Ct. 1186. But, even so, the United States Supreme Court held in 2010 that, "even to the extent a heightened vagueness standard applies, a plaintiff whose speech is clearly proscribed cannot raise a successful vagueness claim under the Due Process Clause of the Fifth Amendment for lack of notice." Holder v. Humanitarian Law Project, 561 U.S. 1, 20, 130 S.Ct. 2705, 177 L.Ed.2d 355 (2010).

Humanitarian Law Project involved a lawsuit in which the plaintiffs attempted to block any application of a criminal provision of the Antiterrorism and Effective Death Penalty Act (AEDPA) to their conduct on grounds that the provision was unconstitutionally vague and that it criminalized the enjoyment of their First Amendment rights. Id. at 10-11, 130 S.Ct. 2705. The Supreme Court held that the Court of Appeals, in conducting a faulty vagueness analysis, had "contravened the rule that `[a] plaintiff who engages in some conduct that is clearly proscribed cannot complain of the vagueness of a law as applied to the conduct of others.'" Id. at 20, 130 S.Ct. 2705 (citing Hoffman Estates, 455 U.S. at 495, 102 S.Ct. 1186). The Supreme Court then continued, "That rule makes no exception for conduct in the form of speech." Id. Chief Justice Roberts, who authored the opinion for the Court, explained further:

Id. He then concluded:

Id. at 21, 130 S.Ct. 2705.

I am aware that this Court has held that, "when a vagueness challenge involves First Amendment considerations, a criminal law may be held facially invalid even though it may not be unconstitutional as applied to the defendant's conduct." Long v. State, 931 S.W.2d at 288. But it is not clear to me that our holdings in that regard could survive Humanitarian Law Project, which declined to treat First-Amendment-implicated vagueness claims any differently than ordinary vagueness claims.

The Court today relies upon two more recent Supreme Court opinions to hold that Appellee may nevertheless challenge Section 551.143(a) on facial vagueness grounds: Johnson v. United States, 135 S.Ct. at 2560-61, and Sessions v. Dimaya, 138 S.Ct. at 1214 n.3. Majority Opinion at 143-46 & n.33. Neither opinion cites, much less explicitly overrules, Humanitarian Law Project, however. And the subsequent Ninth Circuit case that the Court cites—for the proposition that Humanitarian Law Project and its many precedents have now been rejected—did no more than tentatively observe that they "may not reflect the current state of the law." Id. at 144-45 n.33 (citing Henry v. Spearman, 899 F.3d 703, 709-10 (9th Cir. 2018)). Until the Supreme Court plainly proclaims its demise, I will continue to rely on the clear holding of Humanitarian Law Project.

B. Even If He Need Not Show the Statute is Vague in All of its Applications, Must Appellee Still Show That the Statute is Vague as Applied to His Own Conduct?

There is another—even more compelling—reason to find that neither Johnson nor Dimaya should be relied upon to control our conclusion relating to the propriety of granting Appellee relief on a facial challenge to Section 551.143(a) in a pre-trial setting. Even if Johnson and Dimaya stand for the proposition that it is no longer necessary to the success of a facial vagueness challenge to establish that the statute is vague in all of its applications, it is still necessary, according to Hoffman and Humanitarian Law Project, to show that the scope of the statute's vagueness extends to the litigant's own conduct. See Hoffman, 455 U.S. at 495, 102 S.Ct. 1186 (holding that, in the context of a facial challenge, a "plaintiff who engages in some conduct that is clearly proscribed cannot complain of the vagueness of the law as applied to the conduct of others"); Humanitarian Law Project, 561 U.S. at 20, 130 S.Ct. 2705 (holding that this rule applies equally to vagueness claims implicating First Amendment speech).3 Appellee has not made that showing.

Indeed, the fact that Appellee raises his facial claim in a pre-trial proceeding distinguishes this case from both Johnson and Dimaya. In both of those cases, appeals were taken after a trial court judgment had already been obtained. As a result, the facts underlying those cases were well known and, consequently, the courts were in a position to judge whether the vagueness of the law at issue reached as far as the cases that were presented. Here, in contrast, we address Appellee's claims in a pre-trial posture, not knowing whether the evidence at trial might show that Appellee committed a clear incursion upon the requirements of the law. Humanitarian Law Project at least established that a plaintiff whose speech is clearly proscribed cannot raise a successful vagueness claim under the Due Process Clause of the Fifth Amendment for lack of notice. And he certainly cannot do so based on the speech of others.

561 U.S. at 20, 130 S.Ct. 2705. Even to the extent that Johnson and Dimaya might evidence a limitation on the principle that "a statute is void for vagueness only if it is vague in all of its applications," neither of those cases had occasion to examine whether a person challenging a statute's facial constitutionality for vagueness must first establish that the law is vague as applied to his own conduct.4

C. Is Section 551.143(a) Either: (1) Vague In All of Its Applications or (2) Vague With Respect to Appellee's Conduct?

Courts are obliged to construe a statutory provision in such a manner as to avoid constitutional infirmity whenever such a reading is at least plausible—even if it is not necessarily the most evident construction. See, e.g., United States v. Harriss, 347 U.S. 612, 618, 74 S.Ct. 808, 98 S.Ct. 989 (1954) ("[I]f the general class of offenses to which the statute is directed is plainly within its terms, the statute will not be struck down as vague even though marginal cases could be put where doubts might arise. And if this general class of offenses can be made constitutionally definite by a reasonable construction of the statute, this Court is under a duty to give the statute that construction.") (citations omitted); Johnson v. United States, 135 S.Ct. at 2578 (Alito, J., dissenting) ("Whether [a constitutional construction] is the best interpretation [of a statute] is beside the point. What matters is whether it is a reasonable interpretation of the statute."). It is certainly possible to construe Section 551.143(a) of the Government Code as definite and specific enough to embrace certain core conduct, even if its application to other "marginal" conduct seems less certain.5 If construing the statute in this way saves it from a claim of facial invalidity on vagueness grounds, then precedent directs that we should take that approach.

Section 551.143(a) provides:

Under the plain language of this provision, an offense is shown by evidence that the actor "knowingly conspire[d.]" Black's Law Dictionary defines "conspire" to be to "engage in a conspiracy; to join in a conspiracy." BLACK'S LAW DICTIONARY at 376 (10th ed. 2014). "[C]onspiracy," in turn, is defined as "[a]n agreement by two or more persons to commit an unlawful act, coupled with an intent to achieve the agreement's objective, and (in most states) action or conduct that furthers the agreement[.]" Id. at 375.

Just what is the "unlawful act" or "objective" that the actor must knowingly conspire to do before he may be convicted under this provision? He must conspire to "circumvent" the Open Meetings chapter of the Government Code.6 Chapter 551 of the Government Code affirmatively requires (with certain exceptions): (1) that government business be transacted in a "meeting" (defined as a "deliberation" involving a "quorum"—that is, a majority—of the governmental body, during which public business or public policy are discussed or considered or during which formal action is taken, TEX. GOV'T CODE §§ 551.001(4) & (6)); (2) that such meetings must be preceded by notice to the public, and must be "open to the public[,]" TEX. GOV'T CODE §§ 551.041 & 551.002; and (3) that such meetings must be duly and fully documented for public consumption by minutes or recording, TEX. GOV'T CODE §§ 551.021 & 551.022.

To be guilty under Section 551.143(a), then, it is necessary for an actor to "knowingly conspire" to "circumvent" these easily identified, manifest requirements of the Open Meetings Act. But that is not all. The actor must also "knowingly conspire" to "circumvent" these requirements of the Open Meetings Act in a particular way. The object of the conspiracy must be to circumvent those requirements "by meeting in numbers less than a quorum" and doing so "for the purpose of" conducting "secret deliberations" that would constitute "a violation of this chapter." On its face, this lengthy adverbial phrase does pose a certain dilemma. It criminalizes the act of "meeting in numbers less than a quorum[,]" but only "for the purpose of secret deliberations[.]" And yet, Section 551.001(2) defines "deliberation" for purposes of the Open Meetings Act to be a "verbal exchange during a meeting" of the governmental body, and Section 551.001(4) defines a "meeting" to require a quorum of the governmental body. This being the case, for Section 551.143(a) to speak in terms of a "meeting" of less than a quorum for the purpose of deliberations (secret or otherwise) would seem to be nonsense, a non-sequitur, a paradox—a literal absurdity. If "deliberations" in Section 551.143(a) requires a quorum, how can one deliberate in the presence of less than a quorum?

Here, what may be considered by some to be an absurdity is readily resolved when it is considered in context of the balance of the statutory language and the evident purpose of the overall statutory scheme. It is possible to make perfectly good sense of the statute when we consider that, by use of the qualifier "secret," the Legislature delineated an understanding of "deliberations" slightly different than the definition set out in Section 551.001(2). It is evident enough that the statute is designed to proscribe "verbal exchanges" between members of a governmental body "concerning an issue within the jurisdiction of the governmental body" (or, for that matter, "any public business"), TEX. GOV'T CODE § 551.001(2), that are conducted by a majority of the governmental body—but in a way that is in "secret," so as to avoid the manifest requirements of an actual quorum, an announced and open meeting, and full documentation. See Acker v. Texas Water Commission, 790 S.W.2d 299, 300 (Tex. 1990) ("When a majority of a public decisionmaking body is considering a pending issue, there can be no `informal' discussion. There is either formal consideration of a matter in compliance with the Open Meetings Act or an illegal meeting.").

Then-Attorney General Greg Abbott construed Section 551.143(a) in a way similar to this, in a 2005 Attorney General Opinion. He reached the same construction of the statute by interpreting "quorum" to reach the concept of a so-called "walking quorum," whereby a majority of a governmental body meets, not all at once, but serially. TEX. ATT'Y GEN. OP. GA-0326, at 2 (2005).7 By this reasoning, he construed Section 551.143(a) "to apply to members of a governmental body who gather in numbers that do not physically constitute a quorum at any one time but who, through successive gatherings, secretly discuss a public matter with a quorum of that body." Id. To illustrate judicial support for this construction, he cited a case that clearly illustrates a violation of Section 551.143(a): Esperanza Peace and Justice Center v. City of San Antonio, 316 F.Supp.2d 433 (W.D. Texas 2001). Id. at 3. As United States District Judge Orlando Garcia described the offense that occurred in Esperanza:

316 F.Supp.2d at 476-77. I second Judge Garcia's observation that, whatever questions may be raised about the potential reach of Section 551.143(a), there can be little doubt it embraces at least these core facts.

The Court spins a number of hypothetical scenarios in an effort to illustrate a lack of pellucidity at the margins—as if the breadth of application necessarily translates into fatal vagueness. Majority Opinion at 149-52. Many of these scenarios strike me as falling within the plain ambit of the statute as I have construed it, pursuant to our duty to preserve its constitutionality. Others may illustrate arguable incursions upon the statute as I have construed it—depending upon whatever evidence may be offered to establish the requisite intent. And still others seem to me not to violate the statute at all because they do not involve an agreement to circumvent the Open Meetings Act by specifically involving a majority of the governing body in "secret deliberations." In any event, I agree with Chief Justice Roberts' observation in Humanitarian Law Project that, "[w]hatever force these arguments might have in the abstract, they are beside the point here." 561 U.S. at 22, 130 S.Ct. 2705. The statute is susceptible to a construction that would render any number of obvious applications to be clear, and under those circumstances, Appellee should not have been permitted to prevail in a due process void-for-vagueness attack on its facial validity. Appellee has failed to show that the statute is vague in all of its applications.

Indeed, granting Appellee relief on his First-Amendment-enhanced due process void-for-vagueness argument, when the statute can readily be construed to admit of many valid applications, is to confuse the due-process vagueness analysis with the First Amendment overbreadth doctrine. See id. at 19, 130 S.Ct. 2705 ("By deciding how the statute applied in hypothetical circumstances, the Court of Appeals' discussion of vagueness seemed to [erroneously] incorporate elements of First Amendment overbreadth doctrine."). And in doing so, the Court essentially grants Appellee relief on overbreadth grounds without inquiring whether he has satisfied his burden to establish an indispensable facet of such a claim—"that the overbreadth of a statute must not only be real, but substantial as well, judged in relation to the statute's plainly legitimate sweep." Broadrick v. Oklahoma, 413 U.S. 601, 615, 93 S.Ct. 2908, 37 L.Ed.2d 830 (1973).

Finally, even if the Court is correct that it is unnecessary for Appellee to show vagueness in all possible applications of the statute before he may succeed in a facial challenge, we should still deny relief. To assert a successful facial challenge, he must at least show that whatever vagueness infects the statute makes it unclear whether his own conduct is proscribed. Hoffman, 455 U.S. at 495, 102 S.Ct. 1186; Humanitarian Law Project, 561 U.S. at 20, 130 S.Ct. 2705. Because the trial court granted Appellee's motion to dismiss in a pre-trial setting, we know nothing about the State's theory of the case, much less what its evidence may have revealed.8 For all we know, whatever conduct Appellee engaged in falls within the clear ambit of the statute, whatever its murkiness at the margins. He has not shown otherwise. For this reason, if no other, the trial court erred to grant Appellee's motion to dismiss. The Court errs to reverse the judgment of the court of appeals with respect to Appellee's vagueness claim.

II. THE FIRST AMENDMENT

My construction of the statute also preserves it, I believe, from First Amendment attack. As thus circumscribed, Section 551.143(a) represents a reasonable time, place, or manner restriction upon nonpublic, not public, speech. For this reason, I disagree with Judge Slaughter's conclusion that it must be invalidated as an unconstitutional encroachment upon the free speech rights of public decisionmakers. Moreover, even if I agreed that strict scrutiny represented the appropriate standard for gauging the constitutionality of the statute for First Amendment purposes, I would hold that the legislative will should prevail.

Opinions that delineate the First Amendment restrictions on criminal proscriptions tend to be somewhat sui generis. We often find ourselves trying to force the square peg of a new statutory regulation implicating speech within the round hole of prior First Amendment precedent. This is such a case. The United States Supreme Court has not weighed in on the First Amendment implications of open meetings legislation, so we have yet to obtain that Court's guidance as to the appropriate standard to apply.

Judge Slaughter believes that the appropriate standard is strict scrutiny because Section 551.143(a) places criminal restrictions on speech based on its "subject matter," which the Supreme Court has lately identified as "content-based" speech. Concurring Opinion at 162-64 (taking the position that strict scrutiny applies because the statute regulates speech according to it subject matter). For this proposition, she relies upon Reed v. Town of Gilbert, Ariz., ___ U.S. ___, 135 S.Ct. 2218, 192 L.Ed.2d 236 (2015). Reed indeed involved the suppression of speech (street signs advertising church services) on the basis not of its message, but simply because of its subject matter. But because it involved speech in a public forum, it may not represent the best analogy to open meetings legislation.

Since Reed was decided, the Supreme Court has reiterated that the standard for measuring regulations on nonpublic speech is different—the so-called nonpublic forum standard, which will tolerate reasonable restrictions based upon time, place, or manner, so long as the restrictions are viewpoint neutral. See Minnesota Voters Alliance v. Mansky, ___ U.S. ___, 138 S.Ct. 1876, 1885-86, 201 L.Ed.2d 201 (2018) ("[O]ur decisions have long recognized that the government may impose some content-based restrictions on speech in nonpublic forums[.]"). In Mansky, the issue was whether a state could impose reasonable time, place, and manner restrictions upon political paraphernalia worn within a polling place—a place that, at least for the duration of its function as a polling place, was regarded by the Supreme Court as a nonpublic forum. Id. at 1886. The Supreme Court therefore held that the nonpublic forum standard applied, even though it nevertheless struck down the specific regulation at issue in Mansky as insufficiently precise to satisfy even that standard. Id. at 1885, 1888-92.

While the fit is not perfect, I would apply the nonpublic forum standard to gauge the First Amendment tolerableness of Section 551.143(a). That the Open Meetings Act regulates only the private speech of governmental body members has previously been recognized. See Asgeirsson v. Abbott, 696 F.3d 454, 461 (5th Cir. 2012) ("The prohibition in TOMA is applicable only to private forums and is designed to encourage public discussion."). Though it may be "content-based" in contemplation of Reed, the Open Meetings Act is plainly viewpoint neutral—it bans "walking quorums" without reference to a governmental body member's particular view of whatever public business he may wish to debate or discuss outside of the Act's requirements. Indeed, as Asgeirsson recognized, the Open Meetings Act does not prohibit public speech at all—it requires that the specified speech, regardless of viewpoint, be conducted in public. Id. As Asgeirsson went on to observe, "the requirement to make information public is treated more leniently than are other speech regulations." Id. at 463.

As I have construed Section 551.143(a), it constitutes a reasonable time, place, or manner restriction. "Although there is no requirement of narrow tailoring in a nonpublic forum, the State must [still] be able to articulate some sensible basis for distinguishing what may come in from what must stay out." Mansky, 138 S.Ct. at 1888. If we limit our construction of the statute to apply only to the core "walking quorum" conduct, as illustrated by cases such as Esperanza and Hitt v. Mabry, 687 S.W.2d 791, 793 (Tex. App.-San Antonio 1985, no pet.),9 then the statute should readily survive a First Amendment attack. See Boos v. Barry, 485 U.S. 312, 331, 108 S.Ct. 1157, 99 L.Ed.2d 333 (1988) (holding that a statute challenged under the First Amendment overbreadth doctrine may be saved by a judicial narrowing construction). So construed, it plainly achieves the legitimate policy objectives of open meeting legislation—transparency, public involvement, and anti-corruption—by assuring that the affirmative requirements of the statutory scheme—openness, notice, and documentation of a governmental body's official business—are not thwarted by artifice and stratagem. And it does so without unnecessarily restricting the private speech rights of government body members so long as their private interactions do not rise to the level of knowingly conducting their official business as a governmental body outside the glare of public scrutiny. For this reason, I would hold that Section 551.143(a) constitutes a reasonable time, place, and manner restriction under the nonpublic forum standard.

But, even if I believed that Reed identified the appropriate standard by which to measure Section 551.143(a), I would hold that the statute survives strict scrutiny analysis. Like Judge Slaughter, I have no doubt that the interests underlying the Open Meetings Act are compelling ones. Concurring Opinion at 164-65. The statute, as the reasonable construction I have outlined above would narrow it, would also extend only so far as to serve those compelling interests, and would not otherwise restrict the legitimate private speech of governmental body members. Such members would remain free to discuss among themselves, in whatever numbers they desire, any topic that does not involve "an issue within the jurisdiction of the governmental body or any public business." TEX. GOV'T CODE § 551.001(2). They may even discuss official business among themselves, in numbers less than a quorum, so long as those discussions do not take place as part of a knowing conspiracy ultimately to conduct official business as a de facto quorum without adhering to the affirmative requirements of the Open Meetings Act.

I also do not agree that the imposition of criminal penalties for violations of the act equates to a failure on the part of the Legislature to narrowly tailor its terms. Civil remedies for violations of the act are just that—remedial only. See TEX. GOV'T CODE §§ 551.141 & 551.142 (providing that an action taken by a governmental body in violation of the open meeting chapter "is voidable" and that violations may be vindicated by way of mandamus and injunctive remedies). They provide no real disincentive to members of governmental bodies to try to conduct business in secret. The worst that could happen under that type of regime is that civil remedies may be imposed and that efforts to avoid the requirements of the Open Meetings Act could be thwarted. To provide a true disincentive, the stigma of a criminal penalty is necessary. Besides, the fact that a violation is only a misdemeanor shows that even the criminal penalty has been narrowly tailored. Misdemeanors are the least restrictive criminal stigma available and adequate to do the job. Section 551.143(a) is therefore, in my view, sufficiently narrowly tailored to achieve the State's compelling interests.

III. CONCLUSION

Because the Court strikes down a statute that is plainly salvageable, I respectfully dissent.