Imperfect Vehicles and Certified Questions

The Supreme Court continues to clear its docket for the term, almost certainly meeting its goal to release every merits opinion by the end of June—only five causes submitted on argument remain. On the June 13 orders, the Court issued its decision in Perez v. City of San Antonio. I’ve previously written about this certified question from the Fifth Circuit, asking the Supreme Court to decide the meaning of the newly enacted religious liberty protections in Article I, Section 6-a of the Texas Constitution. The substance of the decision is interesting, and I’ll get to that towards the end of this post. But the colloquy between the majority and the dissent about whether and how to answer certified questions is perhaps even more interesting and more broadly useful for readers of this Substack.

Certified questions make up a small fraction of the Supreme Court’s docket—just a few cases every term. But they can present interesting and significant procedural issues that are worth paying attention to. For example, as Justice Young noted in another of the Court’s recent decisions (addressing mootness exceptions), much like federal courts, the judicial power of Texas courts does not include the rendition of advisory opinions. “So unyielding is this principle that only by separate constitutional authorization may this Court answer certified questions from federal appellate courts. See [Tex. const.] art. V, § 3-c.” All other advisory opinions remain prohibited. The issue of advisory opinions animated the Justices’ disagreement in Perez.

As a reminder, the Fifth Circuit certified the following question:

Does the “Religious Service Protections” provision of the Constitution of the State of Texas—as expressed in Article 1, Section 6-a—impose a categorical bar on any limitation of any religious service, regardless of the sort of limitation and the government’s interest in that limitation?

As I’ve written about before, for many years the Supreme Court has almost reflexively accepted certified questions from the Fifth Circuit. I reviewed about a decade of certified questions (TAMES “event reports” makes this a far easier to do than it sounds). On average, the Court takes about a week to ten days to accept certified questions. And the Court effectively accepts 100 percent of the cases certified—in contrast to many state supreme courts around the country (more on that below).

Justice Boyd, writing for the majority, notes that the Courts’ willingness to accept cases means that unlike the ordinarily highly selective “conveyor belt” process for choosing cases, through which the Court ultimately grants about 9 percent of petitions for review, certified questions do not arise from the “normal litigation process” in Texas courts. That means the Court takes the cases, warts and all, often “without the benefit of careful consideration of the issues by [the] lower courts,” and with little consideration for whether the case is a good vehicle for deciding what can be very important issues of Texas law. Echoing Justice Young, Justice Boyd noted that the Court even lacks subject-matter jurisdiction in the usual sense. Thus, while certified questions present an excellent opportunity for “cooperative federalism,” “[c]ertified questions thus create the risk that [the Court] might answer a question in the abstract, divorced from a factual record to illuminate the legal question by grounding it in a real-world dispute.”

Indeed, as Justice Young noted in a statement respecting the denial of a motion to participate in oral argument earlier in the case, some of these pitfalls applied here. No Court has addressed or even mentioned the provision (save for then-Justice Blacklock’s off-hand mention in State v. Loe last term) the meaning of this new constitutional provision since it was ratified in 2021. And, at first, the Fifth Circuit forfeited the issue: “Perez did ‘not sufficiently brief the question of whether’ the new Texas Religious Services Clause provides him with additional protections. . . . The court thus concluded that Perez failed ‘to meet [his] burden to show a likelihood of success on the merits’ of that claim.” That is, according to the panel, Perez had given the issue insufficient attention to even preserve it for review. After vacating its initial decision, the Fifth Circuit certified the question on rehearing (despite again acknowledging it had previously held the issue was inadequately briefed). True, the Supreme Court had the benefit of full briefing in the case before it, including several amicus briefs. But it wasn’t enough for the Court to feel comfortable providing a comprehensive interpretation of the provision.

It should therefore come as little surprise that the majority ultimately decided to offer only a cagey, narrow answer to the Fifth Circuit’s question. And that’s ok by the Fifth Circuit. As a matter of course, the Fifth Circuit acknowledges the possibility that the Supreme Court may not provide an answer to the exact question asked:

“We disclaim any intention or desire that the Supreme Court of Texas confine its reply to the precise form or scope of the question certified.” See, e.g., Janvey v. Golf Channel, Inc., 792 F.3d 539, 547 (5th Cir. 2015).

As Justice Boyd noted, this is not the first time the Supreme Court has restated a certified question. And the “question asks simply whether the Religious Services Clause forbids ‘any limitation of any religious service, regardless of the sort of limitation’—a question to which, as everyone agrees, the answer is ‘No.’ ” Thus, to provide helpful guidance, as the question inherently asks the Court to do, the Supreme Court limited its answer here to what the law does not protect. Put simply, the court “conclude[d] it does not extend to the government’s preservation and management of publicly owned lands.” Thus, the City of San Antonio does not offend the constitution by proceeding with a park restoration that would destroy the trees necessary for the Perez’s religious rituals.

In dissent, Justice Sullivan thought that the whole endeavor was improper and the Court simply should refuse to answer the certified question. He focuses on the advisory nature of opinions in certified questions and the majority’s less-than-complete response. In what is likely a first for this Court, Justice Sullivan quotes Will Ferrell and Christopher Walken’s “More Cowbell” skit from Saturday Night Live: The majority “refuses this certified request to take Section 6-a and ‘really explore the studio space.’ ” Despite the Fifth Circuit’s admonition quoted above, Justice Sullivan says that the majority opinion doesn’t “confine [its] answer to the question propounded by the certifying court. Amberboy v. Societe de Banque Privee, 831 S.W.2d 793, 798 (Tex. 1992).” And “[t]hat’s a problem, because [the Court] lack jurisdiction to issue an advisory opinion that answers a certified question the Fifth Circuit didn’t ask.” But perhaps Justice Sullivan’s bigger problem is the briefing-forfeiture issue I noted above:

Given that our rules insist on answering “determinative questions of Texas law,” Tex. R. App. P. 58.1 (emphasis added), we shouldn’t go out of our way to torpedo a Section 6-a claim that might sink anyway due to forfeiture when the plaintiffs return to the Fifth Circuit.

So it seems that for Justice Sullivan the failure to “explore the studio space” is only partly responsible for the Court exceeding its jurisdiction. That problem is compounded by the fact that this certified question is (as he put it) “academic”—the plaintiffs are likely to lose their state constitutional claim on forfeiture grounds anyway. Thus, the Court’s decision exceeds Rule 58.1’s mandate to decide “determinative” questions, and (in his view) for no good reason because the Court provides an answer that is not-quite good for one ride only, but pretty close.

He notes, and appreciates the Court’s eagerness to engage in cooperative federalism. And it is this willingness that makes it all the more unnecessary to answer this question.

Like the majority, I’m not convinced the Court’s narrow decision by itself exceeds its jurisdiction. In Amberboy, then-Justice Cornyn is explaining why the Court doesn’t apply its answer to the facts of the case, as the concurring and dissenting justices, led by then-Justice Doggett, wanted to. And Amberboy relied on Moreno v. Sterling Drug, Inc., 787 S.W.2d 348, 349 (Tex. 1990), which also deals with whether the Court should take its answer and apply it to the facts:

For clarity, we emphasize that these issues are before us on certified questions from the Fifth Circuit. This is a very limited procedural device; we answer only the questions certified and nothing more. See Tex.R.App.P. 114. Thus, the whole case is not before this court as it would be in an ordinary appeal.

The majority here answered the question, particularly given the Fifth Circuit’s admonition (or recognition) that the Court may not answer in the precise form posed. As the majority noted, “[the Court] need not endeavor to comprehensively define the Clause’s scope to provide a helpful answer here—nor should any court undertake to make such comprehensive pronouncements about the contours of a provision such as the Religious Services Clause beyond those that are necessary to decide the case before it.”

Justice Sullivan’s point about Rule 58.1 is perhaps more interesting—does the rule strip the Supreme Court of jurisdiction to decide any question of law that is not determinative of the merits in a case? The Constitution does not so limit the Court’s jurisdiction. It simply states that “[t]he supreme court and the court of criminal appeals have jurisdiction to answer questions of state law certified from a federal appellate court.” Certainly an interesting issue to ponder.

So does this case portend any change in how often the Court takes certified questions? It seems unlikely. It was an 8-1 decision. And it has been widely reported that Justice Sullivan is a candidate for the federal district bench in Austin. So he may not have long to convince his colleagues to change their minds. But, as I mentioned in a previous post, Umphress may be a sign that the Court isn’t going to reflexively accept everything that is certified to it. The appellant’s brief was filed in that case. But we still have a bit to go before we see how it plays out.

These last few certified questions prompted me to look at the statistics for other states. And the Supreme Court of Texas is way ahead of the curve here. For example, a recent article highlights the significantly lower percentage of questions accepted in several other state supreme courts: Ohio, Nevada, and Alabama. The article found that these three state supreme courts declined (in full) to accept around 30 percent of the cases certified to them. Another small fraction were declined in part. Thus, in total, these courts declined around one third of cases certified over a 23 year period from January 2000 to June 2023. From the federal courts’ perspective, around 80 percent of certifications were accepted by state high courts within the Ninth Circuit from 2010 to 2018, 87 percent accepted within the Third Circuit in the same period, and only 60 percent accepted within the Sixth Circuit. So, whether the Supreme Court rejects a case or two here or there, it still leads the way in “cooperative federalism,” ensuring that litigants in the Fifth Circuit get authoritative answers on issues of Texas law and the orderly development of Texas jurisprudence.

Again, that issue is far from clear. We know what it does not protect, but how did the Court get there? The litigants, and the Court, recognized that there were two key questions for the Court to address: the force and the scope of the provision. Ultimately, the Court concluded that

(1) when the clause applies, its force is absolute and categorical, meaning it forbids governmental limitations on religious services regardless of the government’s interest in the limitation or how tailored the limitation is to that interest, but (2) the scope of the clause’s applicability is not unlimited.

The clause’s force

Perez argued that, when the clause applies, it applies with “absolute force” and “categorically bars” a prohibited limitation on religious services regardless of the government’s interest in that limitation. Amici largely agreed. The City disagreed, taking the position that the clause does not forbid laws that are narrowly tailored to promote a compelling governmental interest. Put another way, the City suggested that the Court apply a strict scrutiny standard and hold that the clause passes muster under that test. The Court agreed with Perez and the amici—the language “may not” is categorical. And the context supported the plain text reading. This provision was ratified in the wake of COVID-19 lockdowns, where some cities were shutting down religious gatherings but permitting secular activities to continue. The City argued that every constitutional right is subject to some limit and the Free Exercise clause and Freedom of Worship clause of the First Amendment are subject to strict scrutiny, despite their language seeming absolute. The Court rejected this argument in light of the context—neither those nor the Texas RFRA was enough to ensure protections for religious activities during COVID. So this new provision must have been intended to go farther. So “[b]ased on the Clause’s text and context, [the Court] conclude[d] that, when it applies, [the clause] categorically bars a governmental prohibition or limitation on religious services without regard to whether it passes strict scrutiny or any other test that balances the right against the government’s interests.” But the Court concluded there was some limit to the scope of the clause—it doesn’t apply in every circumstance.

The clause’s scope

Every party and amici agreed that the answer to the scope question is “No”—the clause is not absolute. The difficulty, however, is in identifying the boundaries of the Clause’s scope. This is where the Court’s colloquy about certified questions came into play. Ultimately, the Court acknowledged that the text expressly limits the scope of the clause’s applicability in at least five ways.

First, the clause forbids only actions by “this state or a political subdivision of the state.” Tex. Const. art. I, § 6-a. Second, it forbids only actions through which the state or a political subdivision may “enact, adopt, or issue a statute, order, proclamation, decision, or rule.” Id. Third, the Clause protects only “religious services”; it does not, for example, purport to protect the broader concept of the “free exercise of religion.” Id. Fourth, it protects only religious services “conducted . . . in this state by a religious organization established to support and serve the propagation of a sincerely held religious belief.” Id. And fifth, it forbids only government actions that “prohibit[] or limit[]” such services. Id.

What is clear, the Court says, is that the clause protects religious gatherings. The Court thus provided some affirmative limitations: the clause generally forbids governmental enactments that prohibit people from gathering for a religious service, restrict the number or relationships of people who can gather for a religious service, or regulate the activities in which people may engage when they gather. All of these are in reference to various orders during COVID. Then the Court dealt specifically with this case.

After walking through the history and context, the Court ultimately concludes that the clause does not encompass “limitations” on religious services that result from the government’s preservation and maintenance of the natural features of public lands.

I’ll be keeping an eye out on Friday for what is likely to be the Court’s final merits decisions of the term, and I’ll have more for you soon.