The State of Texas and Kyle Madsen in His Official Capacity as Director of Right of Way v. Jrj Pusok Holdings, LLC

Supreme Court of Texas

══════════

No. 24-0447

══════════

The State of Texas and Kyle Madsen in his Official Capacity as

Director of Right of Way,

Petitioners,

v.

JRJ Pusok Holdings, LLC,

Respondent

═══════════════════════════════════════

On Petition for Review from the

Court of Appeals for the Fourteenth District of Texas

═══════════════════════════════════════

JUSTICE HAWKINS, joined by Justice Sullivan, and by Chief Justice

Blacklock except as to Part III(A), dissenting.

Since our State’s Founding, our Court has enforced the bedrock

principle, known as sovereign immunity, that private citizens may not

sue the State to redress private injuries unless the State has given its

express permission to be sued. That rule exists for good reason. A private

individual’s suit against the State is really a suit against all of us as

taxpayers. Every dollar the government spends defending a lawsuit or

paying a judgment to benefit one individual is a dollar that could have

been spent on schools, roads, and public safety to benefit us all.

Some individual claims vindicating certain rights are so

important to the fabric of our society that we accept that tradeoff. But

reasonable minds can and do disagree about where to draw the line—

which is why we insist that the Legislature, not courts, decides when to

waive sovereign immunity. In our system of democratic governance, our

elected representatives are the only ones authorized to balance the

competing interests of individuals seeking redress against the State and

the burdens their claims place on the public.

We have long insisted that to be operative, a legislative waiver of

sovereign immunity must be unmistakable. It need not be marked by

any particular magic words, but it must make unambiguous the

Legislature’s declaration that a suit against the State may proceed.

Anything less risks judicial overreach—the usurpation of the

Legislature’s exclusive prerogative to decide when private citizens may

bring suit against the State and burden us all to redress their private

individual grievances.

Today, the Court holds that Chapter 21 of the Property Code

waives immunity for repurchase claims. In reaching that decision, the

Court acknowledges that the statute contains none of the statutory

indicia that traditionally signal the waiver of immunity. It agrees that

the statute neither mentions “waiver” nor requires the State’s

participation—the two hallmarks of immunity waivers. It acknowledges

that the statutory scheme would have at least some effect without a

waiver. Nevertheless, the Court invokes various other considerations—

including the constitutional backdrop unique to property interests,

2

Chapter 21’s broader statutory context, and practical concerns—to

declare that immunity is waived.

Clear-statement rules exist in large part to obviate the need for

these searching multi-factor inquiries. When it comes to immunity

waivers, the analytical task should be simple: look for an unambiguous

declaration that the State may be sued. Is there such a declaration? If

so, then immunity is waived. Otherwise, it is not. That basic heuristic

should have made this case easy. This statute lacks an unmistakable

waiver of immunity, and so the only permissible answer is that no such

waiver exists.

The good news is that the Court’s approach today is expressly

limited to this particular statutory provision, which is born of unique

considerations (including a special constitutional backdrop) and

inherently distinct from virtually every other statutory right known to

our law. Today’s decision thus has no bearing on our sovereign-

immunity jurisprudence in any other context. Even still, I cannot join

the Court’s approach or outcome. Properly understood, the State is

immune to this suit. I would reverse the judgment below and render

judgment for the State.

I respectfully dissent.

I

I first review the principles of sovereign immunity that underlie

my analysis.

A

Sovereign immunity is “an established principle of jurisprudence

in all civilized nations.” Tooke v. City of Mexia, 197 S.W.3d 325, 331 (Tex.

3

2006) (quoting Beers v. Arkansas, 61 U.S. 527, 529 (1857)). Its origins

predate the ratification of the U.S. Constitution. See ALEXANDER

HAMILTON, THE FEDERALIST NO. 81, at 487 (Clinton Rossiter ed., 1961)

(“It is inherent in the nature of [a State’s] sovereignty not to be amenable

to the suit of an individual without its consent.”); see also WILLIAM

BLACKSTONE, 1 COMMENTARIES ON THE LAWS OF ENGLAND 214 (4th ed.

1876) (“[N]o suit or action can be brought against the sovereign, even in

civil matters, because no court can have jurisdiction over him.”). The

U.S. Supreme Court has described “States’ immunity from suit” as “a

fundamental aspect of the sovereignty which the States enjoyed before

the ratification of the Constitution, and which they retain today.” Alden

v. Maine, 527 U.S. 706, 713 (1999).

Our Court’s embrace of sovereign immunity goes back to our

State’s earliest days. As we recognized in 1847, “no state can be sued in

her own courts without her consent, and then only in the manner

indicated by that consent.” Hosner v. DeYoung, 1 Tex. 764, 769 (1847).

In the 180 years since, we have ratified this foundational aspect of our

sovereignty generation after generation. See, e.g., Taylor v. Hall, 9 S.W.

148, 149 (Tex. 1888) (“The state cannot be sued without its consent,

either directly or indirectly.”); State v. Isbell, 94 S.W.2d 423, 424 (Tex.

[Comm’n Op.] 1936) (“[A] state cannot be sued without its consent, and

then only in the manner, place, and court or courts designated.”); Tex.

Dep’t of Transp. v. Jones, 8 S.W.3d 636, 638 (Tex. 1999) (“[I]mmunity

from suit bars an action against the state unless the state expressly

consents to the suit.” (citing Fed. Sign v. Tex. S. Univ., 951 S.W.2d 401,

405 (Tex. 1997) and Mo. Pac. R.R. Co. v. Brownsville Navigation Dist.,

4

453 S.W.2d 812, 813 (Tex. 1970))); Nazari v. State, 561 S.W.3d 495, 500

(Tex. 2018) (“The common-law doctrine of sovereign immunity prohibits

suits against the state unless the state consents and waives its

immunity.” (citing Hall v. McRaven, 508 S.W.3d 232, 238 (Tex. 2017)).

We have adhered to this principle for our entire history for many

reasons, two of which bear emphasis here. The first involves the

separation of powers—the “preserv[ation] [of] executive and legislative

power from judicial encroachment.” Univ. of the Incarnate Word v.

Redus, 602 S.W.3d 398, 404 (Tex. 2020) (citing cases). The second

implicates public policy. Sovereign immunity “shield[s] the public from

the costs and consequences of improvident actions of their

governments.” Tooke, 197 S.W.3d at 332. In this sense, sovereign

immunity “protect[s] the public treasury.” Redus, 602 S.W.3d at 404.

After all, when the government is sued, we all collectively, through our

government, must expend resources litigating cases and paying

judgments to satisfy one individual’s grievance. See Brown & Gay Eng’g

v. Olivares, 461 S.W.3d 117, 121 (Tex. 2015).

Safeguarding the public fisc is an important part of sovereign

immunity, but the doctrine has never been limited to that purpose alone.

See Nazari, 561 S.W.3d at 507-08. The doctrine broadly preserves “the

relationship between the legislative and judicial branches of

government,” irrespective of purely financial concerns. Rusk State Hosp.

v. Black, 392 S.W.3d 88, 97 (Tex. 2012); Fed. Sign, 951 S.W.2d at 414-

5

16 (Hecht, J., concurring) (outlining modern political and financial

reasons for sovereign immunity). 1

In light of the important values sovereign immunity protects, our

State’s judiciary has been particularly careful to ensure that the State

has waived its inherent immunity anytime we encounter a suit naming

the State as a defendant. And we have insisted time and again that

because the decision to waive immunity requires the balancing of

competing policy interests, immunity waivers are the prerogative of the

Legislature, not courts. Our Legislature, as the branch that decides how

to allocate resources and money, is the branch that decides which harms

merit a judicial remedy. This Court may not, consistent with the

separation of powers, determine which claims waive sovereign

immunity and which do not—regardless of how we ourselves might

weigh the competing public-policy concerns. See City of Galveston v.

State, 217 S.W.3d 466, 469 (Tex. 2007) (“Th[e] heavy presumption in

favor of immunity arises [in part] from separation-of-powers

principles.”). When we set aside that rule, we usurp power allocated

elsewhere and commit our State’s resources in ways reserved to the

1 The Court rightly rejects Respondent’s argument that this suit against

the State of Texas does not implicate sovereign immunity at all because it does

not threaten the public fisc. We have long held that “immunity is implicated

by any suit that seeks to control governmental action.” Hays St. Bridge

Restoration Grp. v. City of San Antonio, 570 S.W.3d 697, 704 (Tex. 2019). Here,

Respondent seeks to “control government action” by forcing a sale of land.

Anyway, Respondent’s argument fails on its own terms. Its live petition

expressly asserts that “Plaintiff is entitled to damages.” And it expressly

demands the State pays its “reasonable attorneys’ fees and costs.” Moreover,

there is no denying that a suit against the State of Texas requires the

expenditure of the resources of the State of Texas, regardless of the relief

ultimately awarded.

6

political branches. See Rattray v. City of Brownsville, 662 S.W.3d 860,

865 (Tex. 2023) (observing that plaintiffs can overcome immunity “only

by demonstrating that the legislature, as the branch of government

constitutionally empowered to manage the State’s financial affairs, has

waived immunity by statute”); Seay v. Hall, 677 S.W.2d 19, 25 (Tex.

1984) (“[I]t would be a usurpation of our powers to add language to a law

where the legislature has refrained.”).

B

The State always retains the power to waive its sovereign

immunity if and when it so chooses. Relevant here, we have traditionally

recognized two paths through which the State can and does forego

immunity to suit: legislative pronouncement, and constitutional

implication.

1

The State’s standard path for waiving sovereign immunity is

through an unambiguous declaration by the Legislature. As we have

explained, the Legislature retains the power to consent to suits, but

courts will not give effect to that consent unless it is “clear and

unambiguous.” Sampson v. Univ. of Tex. at Austin, 500 S.W.3d 380, 384

(Tex. 2016) (quoting TEX. GOV’T CODE § 311.034)). Only “express[]

consent[]” suffices. Jones, 8 S.W.3d at 638.

This test is inflexible. Even when our Court might otherwise be

sympathetic to a type of claim or a particular plaintiff, we have declined

to find a waiver of immunity when the Legislature has not provided the

clear language our precedents require. In Hillman v. Nueces County, 579

S.W.3d 354, 361 (Tex. 2019), for example, we held that the Michael

7

Morton Act did not waive a county’s immunity for a wrongful-

termination claim brought by a former prosecutor who was fired after

he refused to withhold allegedly exculpatory evidence. However

disappointing some might find that result as a matter of policy, we held

firm: no express statutory language equals no waiver. See also Wichita

Falls State Hosp. v. Taylor, 106 S.W.3d 692, 693-94 (Tex. 2003) (holding

that the Texas Health and Safety Code did not waive sovereign

immunity for a wrongful-death claim brought against a hospital on

behalf of a patient who committed suicide).

Our jurisprudence is rife with examples of statutes that we

declared inadequate for purposes of an immunity waiver. For example,

in Taylor, we held that a statute providing that “[a] treatment facility

. . . is liable” for violating the statute and that “[a] person who has been

harmed by a violation may sue” does not waive immunity. Id. at 698

(emphases omitted) (quoting TEX. HEALTH & SAFETY CODE § 321.003(a),

(b)). In Texas Department of Transportation v. City of Sunset Valley, we

held that Section 203.058 of the Transportation Code—which provides

that “adequate compensation . . . shall be made” when the State’s

acquisition “deprive[s] the agency of a thing of value”—does not waive

immunity. 146 S.W.3d 637, 641 (Tex. 2004) (quoting TEX. TRANSP.

CODE § 203.058(a)). We explained that “[n]othing in the plain language

of section 203.058 indicates the Legislature intended to waive immunity

in situations like the one presented.” Id. at 642. And in Tooke, we held

that “sue and [or] be sued,” “[im]plead and [or] be impleaded,” and other

similar phrases, without more, did not waive immunity. 197 S.W.3d at

333-42 (quoting TEX. LOC. GOV’T CODE § 51.075).

8

What gets the job done? Consider one good example from the same

subject matter relevant here—property rights. The Property Rights Act

provides that “[s]overeign immunity to suit and liability is waived and

abolished to the extent of liability created by this chapter.” TEX. GOV’T

CODE § 2007.004(a). So does the Texas Tort Claims Act, in nearly

identical language. TEX. CIV. PRAC. & REM. CODE § 101.025(a)

(“Sovereign immunity to suit is waived and abolished to the extent of

liability created by this chapter.”). And so does a different provision of

the Property Code, which provides that “[t]he state’s immunity from suit

without consent is abolished with respect to suits brought under this

section”—namely, appeals of Comptroller decisions regarding claims for

unclaimed property. TEX. PROP. CODE § 74.506(a), (c). Where the

Legislature uses such explicit waiver language, “we have had little

difficulty recognizing a waiver of immunity from liability.” Taylor, 106

S.W.3d at 696.

2

Our precedents recognize an additional—albeit narrow—path by

which the State has surrendered its immunity to certain types of claims.

In a line of cases exemplified by Steele v. City of Houston, 603 S.W.2d

786, 791 (Tex. 1980), we have recognized that the Takings Clause itself

operates as a waiver of immunity for inverse-condemnation claims. I will

document that precedent in order to illustrate that this line of cases

offers no interpretive force as to the statutory-interpretation question

before us today.

Article I, Section 17 of the Texas Constitution provides, in

relevant part, that “[n]o person’s property shall be taken, damaged, or

9

destroyed for or applied to public use without adequate compensation

being made, unless by the consent of such person.” TEX. CONST. art. I,

§ 17(a). By prohibiting takings absent public use and just compensation,

our Takings Clause plays a critical role in protecting “the sacred and

inviolable rights of private property,” a natural right that preexists our

government. See Kelo v. City of New London, 545 U.S. 469, 505 (2005)

(Thomas, J., dissenting) (quoting WILLIAM BLACKSTONE, 1

COMMENTARIES ON THE LAWS OF ENGLAND 134-35 (1765)); Eggemeyer v.

Eggemeyer, 554 S.W.2d 137, 140 (Tex. 1977); see also Steele, 603 S.W.2d

at 789 (“Uncompensated governmental taking of property was unlawful

before Magna Carta.”).

The text is built around the core injuries “taken, damaged, or

destroyed.” TEX. CONST. art. I, § 17(a). It ties those injuries to a “public

use.” Id. And most critically, it provides that when a taking, damage, or

destruction is perpetrated, the property owner “shall” be entitled to

“adequate compensation.” Id. § 17(a), (d). In that sense, our Takings

Clause is foundational but commonplace, mimicking that of the federal

Constitution and most of our sister States. 2

2 U.S. CONST. amend. V (“[N]or shall private property be taken for public

use, without just compensation.”); see also, e.g., CONN. CONST. art. I, § 11 (“The

property of no person shall be taken for public use, without just compensation

therefor.”); LA. CONST. art. I, § 4(B)(1) (“Property shall not be taken or

damaged by the state or its political subdivisions except for public purposes

and with just compensation paid to the owner or into court for his benefit.”);

OKLA. CONST. art. II, § 24 (“Private property shall not be taken or damaged for

public use without just compensation.”); IOWA CONST. art. I, § 18 (“Private

property shall not be taken for public use without just compensation first being

made . . . .”); MICH. CONST. art. X, § 2 (“Private property shall not be taken for

public use without just compensation therefore being first made or secured in

a manner prescribed by law.”).

10

Virtually all of our Takings Clause jurisprudence turns on those

core elements that describe a constitutional injury. We have assessed

whether a taking was really for a “public use.” See, e.g., Hous. Auth. of

City of Dall. v. Higginbotham, 143 S.W.2d 79, 83-85 (1940); KMS Retail

Rowlett, LP v. City of Rowlett, 593 S.W.3d 175, 182 (Tex. 2019). We have

analyzed whether some regulatory action is really a taking at all. See,

e.g., Westgate, Ltd. v. State, 843 S.W.2d 448, 453 (Tex. 1992); City of

Baytown v. Schrock, 645 S.W.3d 174, 180-81 (Tex. 2022). We have

considered whether “damage” to property was significant enough to

implicate the Takings Clause. See, e.g., State v. Heal, 917 S.W.2d 6, 10-

11 (Tex. 1996). We have evaluated what constitutes “adequate

compensation.” See, e.g., Uselton v. State, 499 S.W.2d 92, 96-99 (Tex.

1973). These cases are multifaceted, but they share a common theme:

each turns on the core constitutional injury expressly identified in the

text of the Takings Clause.

While nothing in that text refers to immunity or causes of action,

our Court has interpreted the Takings Clause to waive the State’s

immunity for a certain narrow class of claims that implicate the

constitutional injury at the Clause’s heart. The rationale is set out most

clearly in Steele. See 603 S.W.2d at 791. There, police officers for the City

of Houston destroyed a home while pursuing a criminal suspect. Id. at

789. The landowner (and his renters) sued the City to recover costs for

the damage to the home and their belongings. Id. at 788.

The plaintiffs lacked an “enabling statute” for this claim, which

arose not as the result of the condemnation process, but rather from a

police action without a formal “transfer of property rights.” Id. at 790.

11

We nevertheless allowed the claim to proceed. We held that under these

circumstances, the Takings Clause “itself is the authorization for

compensation for the destruction of property and is a waiver of

governmental immunity for the taking, damaging or destruction of

property for public use.” Id. at 791. We specifically tied this waiver of

immunity to the constitutional text, noting that the destruction the

police inflicted on the home is exactly the injury the Takings Clause

describes. We summarized: “It was a claim for the destruction of

property, and governmental immunity does not shield the City of

Houston.” Id. In short, the plaintiffs presented a classic constitutional

injury, and we in turn found a constitutional remedy with a

constitutional waiver.

Various post-Steele cases ratified that approach—but only when

the government action at issue implicated the Takings Clause’s

constitutional injury. For example, in State v. Biggar, we held that the

State’s denial of an easement exchange—which caused a decrease in the

plaintiffs’ land value—resulted in compensable damages that were

actionable as an inverse-condemnation claim because “[i]f government

is able to use its power as sovereign to adjust the value of ‘just

compensation,’ the constitutional protection is rendered meaningless.”

873 S.W.2d 11, 12, 14 (Tex. 1994). In Town of Flower Mound v. Stafford

Estates Limited Partnership, a town conditioned its approval of a

subdivision development on the rebuilding of an asphalt road with

concrete. 135 S.W.3d 620, 623-24 (Tex. 2004). The town rejected the

developer’s request for an exception, and the developer ended up paying

nearly $500,000 to rebuild the road. Id. at 624. We held that the exaction

12

imposed by the town constituted a taking for which the developer was

entitled to compensation. Id. at 645.

Similarly, in Tarrant Regional Water District v. Gragg, we held

that a regional water district committed a taking when its actions—

releasing water through a reservoir’s floodgates—caused extensive

flooding to the plaintiffs’ land. 151 S.W.3d 546, 549-55 (Tex. 2004). In

Texas Department of Transportation v. Self, we allowed an inverse-

condemnation claim to proceed where the State cut down trees on the

plaintiffs’ property outside the scope of the State’s easement and refused

to compensate the plaintiffs for the cost of replacing the trees. 690

S.W.3d 12, 17-18, 32 (Tex. 2024). And most recently, in Commons of

Lake Houston, Ltd. v. City of Houston, we held that a regulatory taking

can occur even when the regulation at issue—an ordinance requiring

construction two feet above the floodplain—resulted from a valid

exercise of police power and was intended to ensure compliance with the

National Flood Insurance Program. 711 S.W.3d 666, 681, 684 (Tex.

2025).

All of these cases illustrate the Steele principle: when the

government denigrates the core prohibition of the Takings Clause and

inflicts a constitutional injury, the Constitution itself supplies a cause

of action and a remedy. We have declared that there is no need in these

situations for the Legislature itself to step in with its own statutory

immunity waiver. Whether out of respect for the fundamental rights the

Takings Clause protects or to exalt the social contract between

government and citizen, our jurisprudence consistently—but

narrowly—holds that the Constitution’s categorical exclusion of core

13

takings claims from sovereign immunity has always been limited to just

that: the taking, damaging, or destruction of property absent public use

or adequate compensation. By contrast, government actions that are

merely adjacent to or downstream of takings are subject to immunity

absent express legislative waiver. See, e.g., TEX. GOV’T CODE

§ 2007.021(a) (authorizing a property owner to bring suit “to determine

whether the governmental action . . . results in a taking under this

chapter”); id. § 2007.004(a) (“Sovereign immunity to suit and liability is

waived and abolished to the extent of liability created by this chapter.”).

II

With that backdrop in mind, I now turn to the statute before us.

I would hold it does not waive immunity because it lacks the “clear and

unambiguous” language necessary to effectuate an immunity waiver.

See Sampson, 500 S.W.3d at 38. I respectfully disagree with the Court’s

contrary conclusion.

A

Section 21.101 of the Property Code provides that a person “is

entitled to repurchase [] property” under certain circumstances, and

that “[a] district court may determine all issues in any suit regarding

the repurchase of a real property interest acquired through eminent

domain by the former property owner or the owner’s heirs, successors,

or assigns.” TEX. PROP. CODE § 21.101(a), (c). Additionally, Section

21.003 provides:

A district court may determine all issues, including

the authority to condemn property and the assessment of

damages, in any suit:

14

(1) in which this state, a political subdivision of this

state, a person, an association of persons, or a corporation

is a party; and

(2) that involves a claim for property or for damages

to property occupied by the party under the party’s

eminent domain authority or for an injunction to prevent

the party from entering or using the property under the

party’s eminent domain authority.

Id. § 21.003.

It is immediately obvious that this text lacks two critical

ingredients. First, there is no reference to waiver. This is an important

omission because when the Legislature wishes to waive immunity, it

says so by, for example, labeling the provision “Waiver of Immunity.”

E.g., TEX. GOV’T CODE § 554.0035; see also id. § 2007.004(a) (“Sovereign

immunity to suit and liability is waived and abolished to the extent of

liability created by this chapter.”); TEX. CIV. PRAC. & REM. CODE

§ 101.025(a) (“Sovereign immunity to suit is waived and abolished to the

extent of liability created by this chapter.”); TEX. PROP. CODE § 74.506(c)

(“The state’s immunity from suit without consent is abolished with

respect to suits brought under this section.”). Second, nothing in the text

makes the State a necessary party. True immunity waivers expressly

declare that “[t]he state shall be made a party.” E.g., TEX. HEALTH &

SAFETY CODE § 571.0167(b). This statute says no such thing.

The closest the text comes is a single phrase that refers to a

district court “determin[ing] all issues.” TEX. PROP. CODE § 21.003. But

that is a classic jurisdictional grant, not a waiver of immunity. The

“determine all issues” phrase simply confirms that a district court has

jurisdiction to decide issues in suits already properly before it—such as,

for example, a repurchase suit brought against a private entity that

15

acquired land through eminent domain, id. § 21.101(c), or an inverse-

condemnation claim, id. § 21.003(1); TEX. CONST. art. I, § 17(a). Our

precedents have always distinguished between a grant of jurisdiction

and a waiver of immunity, and they make clear that the language we

see here exemplifies the former. See, e.g., In re Nestle USA, Inc., 359

S.W.3d 207, 212 (Tex. 2012) (holding that the “grant of jurisdiction to

this Court” in the Texas Franchise Tax Act “is not a clear and

unambiguous waiver of immunity”); Harris Cnty. Hosp. Dist. v. Tomball

Reg’l Hosp, 283 S.W.3d 838, 843 (Tex. 2009) (explaining that “sue and

be sued” language “anticipates the district’s involvement in civil

proceedings of some nature at some point, but it does not address

immunity from suit”). 3

In short, the text is missing the key ingredients we expect to see

in an immunity waiver, and the only phrase it does supply addresses

something different. We should easily conclude, consistent with our

precedents, that Section 21.101 does not waive the State’s immunity.

B

The Court nevertheless holds that “the State has no immunity

from Chapter 21 [repurchase] claims.” Ante at 26. The Court concedes

that the traditional indicia of an immunity waiver are absent. It

nevertheless finds a waiver by weaving together three extraneous

3 The Court calls my invocation of Nestle “misplaced” because the

statutory provision in that case used the word “jurisdiction,” whereas

Section 21.003 uses the phrase “determine all issues.” But Section 21.003 is

situated in Subchapter A, which is entitled “Jurisdiction.” Cf. TEX. GOV’T CODE

§§ 311.023(7), .024.

16

considerations: constitutional background, context, and practical

concerns. I will unpack each.

1

The first thread in the Court’s reasoning is the constitutional

background of this statutory scheme. While the Court does not (and

could not) claim that the Constitution itself waives immunity for

statutory repurchase claims, it nevertheless insists that it “cannot

ignore Chapter 21’s constitutional overlay, requiring that the State take

property only for public use.” Ante at 20. The Court does not specify how

much work this consideration is doing. But because the Court gives it at

least some weight, I will explain why, in my view, the constitutional

background has no meaningful bearing on the particular question before

us today.

The injury Respondent identifies—a price difference between

current market value and original sale value in connection with the

attempted repurchase of land previously sold to the State—is distinct

from the type of injuries for which the Takings Clause itself waives

immunity, and it does not follow that one informs the other. See supra

Part I.B.2. Even under Respondent’s own telling, the right it seeks to

vindicate is a very different concept from the injury at the heart of the

Takings Clause and the circumstances our previous precedents have

encountered. Respondent’s operative petition alleges that the State

“refused to comply with Sections 21.101, 21.102, and 21.103 of the Texas

Property Code” by not selling the surplus land at the price purchased by

the State. Respondent admits the State sent an “offer letter” to purchase

its land. The State thereafter initiated condemnation proceedings, but

17

Respondent later “entered into rule 11 agreements with the State to

settle the condemnation.” The record confirms that the parties

negotiated and agreed upon a sale. Respondent even acknowledges that

the State is now willing to sell the “surplus land” back to Respondent.

The sole injury, according to Respondent, is that the State wants to

charge “current market value as opposed to the price paid by the State”

in the original sale contract. None of this describes one of the core

constitutional injuries the Takings Clause addresses. See supra Part

I.B.2. 4

That repurchase claim differs from the Takings Clause not only

textually, but also temporally. By design, the repurchase statute comes

into play after the work of the Takings Clause is complete. The State

initiated condemnation proceedings in order to take land for a public

use. See TEX. PROP. CODE § 21.101(a). Those condemnation proceedings

ended in dismissal, after the parties negotiated and agreed to a sale. The

landowner received adequate compensation—some $682,000. At that

moment, when the check cleared and the deed changed hands, the

constitutional process ended. See City of Dallas v. VSC, LLC, 347 S.W.3d

231, 236 (Tex. 2011) (“[I]t is not the taking of property, as such, that

raises constitutional concerns, but the taking of property without just

4 The Court points out that the State’s offer to sell the property at its

current market value aligns with the statute’s pre-Kelo amendments. Ante at

24. That changes nothing: Respondent’s suit does not implicate a Takings

Clause injury. Moreover, the current version of the statute requires the entity

“to sell the property interest to the person for the price paid to the owner by

the entity at the time the entity acquired the property through eminent

domain.” TEX. PROP. CODE § 21.103(b). As I will explain below, Respondent’s

land was not acquired “through eminent domain.”

18

compensation.”) (citation omitted); see also City of San Antonio v.

Grandjean, 41 S.W. 477, 479 (Tex. 1897) (noting that “when a dedication

[to public use] is offered and accepted, it is complete, even before any act

is done by the representative of the public, which changes the position for

the worse” (emphasis added)). That sets this claim apart from Steele and

the other cases discussed above, all of which confronted a constitutional

injury identified in the text of the Takings Clause. See supra Part I.B.2.

The Court implies that the Takings Clause’s “public use”

requirement impacts its analysis today, but I do not see why it would.

The Takings Clause presents a fixed-in-time requirement—not a

permanent constitutional lien on all government property. Even if one

could conceive of a negotiated purchase as a taking, the public use

inquiry occurs at the moment of the transaction, not indefinitely into the

future. Put another way, once a taking is executed for a legitimate public

use, the constitutional requirements are satisfied, and the constitutional

inquiry ends. A constitutional injury does not spring into being years

later when the government’s plans change. 5

Our Court confirmed as much shortly after the Constitution of

1876 was ratified. In 1897, we held that a plaintiff could not reclaim

property taken by the State for which she had received compensation.

See Grandjean, 41 S.W. at 480. In reaching this conclusion, we

emphasized that a taking is complete when the transaction is

5 I of course am not addressing the rare situations involving dishonesty,

bad faith, or pretext in which the government falsely claims a public use it does

not actually intend. Nothing like that is alleged here, and no one has doubted

that the State intended in good faith at the moment of acquisition to put

Respondent’s property to public use.

19

effectuated: “when a dedication [to public use] is offered and accepted, it

is complete, even before any act is done by the representative of the public,

which changes the position for the worse.” Id. at 479 (emphasis added).

In other words, “[t]he conditional paramount title of the government

becomes absolute when the compensation is assessed or agreed upon

and paid.” Id. When a landowner “accepts as compensation a sum of

money, whatever the amount, and in whatever manner arrived at, his

property, to the extent of the taking, is expropriated and appropriated

to the use of the public.” Id. We confirmed that taken land does not need

to be put to public use in perpetuity: for “a public use[] to be effectual,”

it need not “be evidenced by the use of it having been continued for any

particular time.” Id. (quoting Oswald v. Grenet, 22 Tex. 94, 99 (1858)).

Rather, “[i]t is enough that there has been some clear, unequivocal act

or declaration of the proprietor evidencing an intention to set it apart for

a public use, and that others have acted in reference to, and upon the

faith of, such manifestation of intention.” Id. (emphasis added).

The interests the Takings Clause’s “public use” requirement

protects were satisfied and extinguished at the moment of transaction.

I would ascribe to it no meaningful interpretive force in deciding

whether Chapter 21 contains an express and unambiguous statutory

waiver of immunity for repurchase claims long after that transaction

concluded.

2

The second strand in the Court’s analysis is statutory context.

The Court zooms out to Chapter 21 writ large: “the Legislature nested

the repurchase statute within existing law governing eminent domain—

20

proceedings from which the State enjoys no immunity.” Ante at 15. In

my view, the fact that the Court needs to consider context only proves

that the statute lacks the clear and unambiguous waiver our precedents

demand.

No one doubts that context can inform meaning. See ANTONIN

SCALIA & BRYAN A. GARNER, READING LAW: THE INTERPRETATION OF

LEGAL TEXTS 167 (2012). And we have noted that “[a]mong the core

contextual considerations that generate reliable constructions are the

surrounding provisions of a disputed text and how that text arises

within the statute's larger historical sweep.” Brown v. City of Houston,

660 S.W.3d 749, 754 (Tex. 2023) (citing READING LAW at 33). But as our

decision in Brown illustrates, we “proceed to examine the larger

statutory context” only after concluding that the “plain text” of the

particular provision at issue is indeterminate. Id. at 752. To resort to

context, as the Court does, is to concede that the text lacks the

unmistakably clear and unambiguous statement that our precedents

demand of an immunity waiver. See supra Part I.B.1. This is a sure sign

that the actual provision before us provides no clear statement waiving

immunity.

Even if we could draw some meaning from context, its

interpretive force here would be particularly weak. True, Chapter 21 is

nested among provisions about condemnation proceedings, which we

have recognized fall outside sovereign immunity. See Hidalgo Cnty.

Water Improvement Dist. No. 3 v. Hidalgo Cnty. Irrigation Dist. No. 1,

669 S.W.3d 178, 188 (Tex. 2023). But condemnation proceedings lack

immunity because the State is the plaintiff. Sovereign immunity is a

21

shield against being sued, not a barrier to suing. The absence of

immunity in condemnation proceedings tells us nothing about whether

immunity has been waived for a separate, subsequent claim brought by

a private party against the State. See generally id. The fact that the

Legislature, as a matter of good housekeeping or statutory hygiene,

placed the repurchase right near provisions involving eminent domain

cannot be read as an unmistakable immunity waiver.

Chapter 21’s other provisions prove exactly that. Section 21.0112,

for instance, requires a condemning entity to provide a “Landowner’s

Bill of Rights” to property owners before initiating condemnation. See

TEX. PROP. CODE § 21.0112. If a state agency fails to provide this

document, has immunity been waived for a private suit to enforce

compliance? Similarly, Section 21.0111 requires entities with eminent

domain authority to prepare certain reports. See id. § 21.0111. If TxDOT

files an incomplete report, can an injured party sue? And Section 21.023

requires condemning entities to disclose the former owner’s repurchase

rights in writing at the time of acquisition. See id. § 21.023. If TxDOT

fails to make this disclosure, has immunity been waived for a suit to

enforce the disclosure obligation? The answer to all three of these

questions is surely “no,” which confirms that mere placement in

Chapter 21 cannot provide the “clear and unambiguous” indicator our

precedents demand.

Moreover, when the Legislature wishes to waive immunity across

an entire statutory chapter, it can and does say so. See, e.g., TEX. CIV.

PRAC. & REM. CODE § 101.025(a) (“Sovereign immunity to suit is waived

and abolished to the extent of liability created by this chapter.” (emphasis

22

added)). And our precedents reject the wholesale importing of immunity

waivers across multiple provisions absent clear instructions. See Rusk,

392 S.W.3d at 94 (“The Legislature has waived governmental entities’

immunity from certain claims by means of the Tort Claims Act (TCA).

But the TCA embodies only limited waivers of sovereign immunity; it

does not abolish it.” (citations omitted)); Kerrville State Hosp. v. Clark,

923 S.W.2d 582, 585 (Tex. 1996) (rejecting a result that “would be

tantamount to abolishing governmental immunity, contrary to the

limited waiver the Legislature clearly intended”). All this is consistent

with the foundational and uncontroversial principle that waivers of

immunity must be construed narrowly. TEX. GOV’T CODE § 311.034.

3

Finally, the Court invokes practical concerns, claiming that the

repurchase statute would “serve[] virtually no purpose absent a waiver

of immunity.” Ante at 18. Of course, the broader logical consistency of a

statutory scheme is a relevant consideration in the interpretive task.

See, e.g., Kerrville State Hosp. v. Fernandez, 28 S.W.3d 1, 6 (Tex. 2000).

But here, the repurchase statute makes adequate sense on its own terms

absent a waiver of immunity.

First, Section 21.101 still directs governmental conduct, even if it

does not provide a back-end option to sue the State when it fails to allow

for a repurchase. The State is subject to innumerable statutory

commands—procurement rules, environmental regulations,

employment standards, records retention requirements—enforceable

through internal oversight, legislative appropriations, audits, and

political accountability rather than private litigation. See Redus, 602

23

S.W.3d at 411 (“Political accountability is a vital counterweight to

sovereign immunity[.]”); CPS Energy v. Elec. Reliability Council of Tex.,

671 S.W.3d 605, 629 (Tex. 2023) (Boyd, J., dissenting) (“[T]he political

process often serves as a substitute for private lawsuits to deter

arbitrary and imprudent governmental action.”). There is nothing

remarkable about our Legislature imposing rules on our government

that cannot be enforced through private litigation, and we have held

repeatedly that the mere creation of a substantive right does not

necessarily waive immunity to enforce it. See, e.g., Taylor, 106 S.W.3d

at 700, 702 (holding that the Texas Health and Safety Code did not

waive sovereign immunity, even though it provided a “meaningful cause

of action against private mental health care facilities”); Sunset Valley,

146 S.W.3d at 641-44 (holding that a statute did not waive immunity to

enforce right to “adequate compensation” found in the Transportation

Code).

Second, the repurchase right binds private condemnors who lack

sovereign immunity. Railroads, pipelines, electric utilities, and others

possess legislatively granted condemnation authority. See, e.g., TEX.

TRANSP. CODE § 112.002(b)(5). When these private entities acquire

property through eminent domain and later abandon the public use,

Section 21 grants former landowners a right to sue, and the condemnors

cannot invoke sovereign immunity. We have relied on this exact

consideration in the past to declare that a statutory right still carries

“meaning” when it is “construed against an asserted waiver of

immunity.” Rolling Plains Groundwater Conservation Dist. v. City of

Aspermont, 353 S.W.3d 756, 759 (Tex. 2011) (citation omitted).

24

***

Today the Court holds that in this particular statutory context,

the Legislature waived immunity not by saying “immunity is waived” or

“the State is liable” or “the State is a necessary party,” but by placing a

statutory right in a particular code chapter, granting courts jurisdiction

over disputes arising under that right, and using the phrase “may

determine all issues.” That is no “clear and unambiguous” waiver. I thus

would conclude that Chapter 21 does not waive sovereign immunity and

end the analysis there.

III

Even if Chapter 21 waived sovereign immunity, Respondent did

not properly invoke that waiver. This property was not acquired

“through eminent domain”—the necessary prerequisite for the statutory

scheme to take effect. And Respondent sued in the wrong court. I

respectfully disagree with the Court’s contrary conclusions on each

point.

A

Section 21.101 operates on properties acquired “through eminent

domain.” TEX. PROP. CODE § 21.101(c). Respondent’s property was

acquired through a privately negotiated and mutually agreed upon sale,

not eminent domain. The State, therefore, does not occupy the land

under its “eminent domain authority,” the statute’s necessary

prerequisite. Id. § 21.003(2).

Honoring our duty to construe every word in the provision, see

e.g., City of San Antonio v. Realme, 731 S.W.3d 342, 349-50 (Tex. 2026),

we must place due weight on the word “through,” which traditionally

25

means “passage into and out of some . . . process.” Through, WEBSTER’S

THIRD NEW INTERNATIONAL DICTIONARY, UNABRIDGED (2002) (emphasis

added)). Another dictionary notes that “[p]rimarily through expresses

the general relations of passage from boundary to opposite boundary of

anything, penetration to all parts, continuance to the end of a period

. . . . Into, for subjection to some treatment, and then out of[.]” Through,

WEBSTER’S SECOND NEW INT’L DICTIONARY, UNABRIDGED (1954). Those

definitions indicate that for property to be “acquired through eminent

domain,” the eminent domain process must have not only begun—but

also concluded.

Here, it would be most natural to say that the State acquired the

property at issue “through purchase.” True, the condemnation process

began. But the negotiated sale is what effectuated the change in the

property’s legal status. And the Legislature itself treats condemnation

and purchase as distinct. The Transportation Code, for instance,

authorizes the Texas Transportation Commission to “acquire [an

interest in real property] by purchase . . . or by the exercise of eminent

domain, in the name of the state.” See TEX. TRANSP. CODE § 203.051(a)(1)

(emphasis added). It could have done the same here, but it instead

limited the repurchase right to circumstances when eminent domain,

not “purchase,” does the actual work.

The distinction comports with standard usage, as a simple

analogy illustrates. Suppose the State charges a defendant with felony

bank robbery. The State gets an indictment and initiates criminal

proceedings. The jury is empaneled, and trial starts. The defendant

hears the government’s first witness—and realizes he is doomed. He

26

asks the court for a recess to work out a plea deal with the prosecutor.

The prosecutor offers a misdemeanor, which the defendant accepts, and

the court enters a corresponding judgment of conviction. Would any

ordinary speaker of English say that the defendant was “convicted

through a jury trial”? Of course not. The process started, the jury heard

some evidence, and the court rendered judgment. Any ordinary speaker

would say that the defendant was “convicted through a plea bargain.”

The jury in my hypothetical plays much the same role the State’s

eminent domain power plays in this case—a powerful incentive to settle,

but not the legal mechanism that effectuates the end result.

By consenting to the warranty deed, Respondent has effectively

conceded that the eminent domain process started but never finished.

The document itself expressly states that “[t]he consideration recited

herein represents a settlement and compromise by all parties as to the

value of the property herein conveyed in order to avoid ED proceedings

and the added expense of litigation.” (emphasis added). Respondent

agreed to a sale to prevent the very process it now claims occurred. The

State did not acquire the land “through” eminent domain—it acquired it

through a negotiated purchase. It thus was acting “within a color of right

under the contract”—the warranty deed—rather than within its

eminent domain powers, and therefore lacked “ ‘the requisite intent

under constitutional-takings jurisprudence.’ ” Kirby Lake Dev., Ltd. v.

Clear Lake City Water Auth., 320 S.W.3d 829, 844 (Tex. 2010) (quoting

Gen. Servs. Comm’n v. Little–Tex Insulation Co. Inc., 39 S.W.3d 591,

598-99 (Tex. 2001)).

27

For these reasons, I would conclude Section 21.101 does not

encompass the claim Respondent seeks to assert. 6

B

Irrespective of the above, the State is entitled to judgment

because Respondent filed suit in the wrong court. We have always held

that immunity waivers may prescribe the form of the waiver, and when

the Legislature waives immunity subject to conditions, the failure to

strictly adhere to those conditions negates the waiver. See Nestle, 359

S.W.3d at 208 (“[S]tatutory prerequisites are conditions on the

legislative waiver of the State’s immunity from suit.”); Mission Consol.

Indep. Sch. Dist. v. Garcia, 253 S.W.3d 653, 660 (Tex. 2008) (“[T]he

Legislature . . . has consented to suits brought under the TCHRA,

provided the procedures outlined in the statute have been met.”

(emphasis added)). That principle forecloses Respondent’s suit.

6 The Court asserts in passing that its “approach coheres with” this

Court’s reasoning in Texas A&M University-Kingsville v. Lawson, 87 S.W.3d

518 (Tex. 2002). Ante at 19. I do not take this observation to be doing any work.

In any event, the plurality decision in Lawson carries no relevance here. The

plaintiff, a former university faculty member, sued a university for violations

of the Whistleblower Act. Lawson, 87 S.W.3d at 518. The parties later settled,

and the plaintiff sued for breach of the settlement agreement. Id. at 519. The

university, invoking sovereign immunity, argued the suit was barred. Id. This

Court held that the university was not immune because “when a governmental

entity is exposed to suit because of a waiver of immunity, it cannot nullify that

waiver by settling the claim with an agreement on which it cannot be sued.”

Id. at 521. In this case, by contrast, the parties settled the State’s

condemnation lawsuit; sovereign immunity was not implicated in that initial

action because the State brought suit as the plaintiff. Moreover, unlike the

plaintiff in Lawson, Respondent has not alleged that the State breached any

contract.

28

The unheeded condition specifies what type of court may hear

these suits. Both Section 21.003 and Section 21.101, which provide the

statutory hooks for a purported immunity waiver, refer to the “district

court.” TEX. PROP. CODE §§ 21.003, 21.101(c). The Legislature has

proclaimed, in other words, that any immunity waiver that might exist

is limited to suits brought in district court. It is entirely reasonable for

the Legislature to route these important and complex suits that

implicate significant sovereign interests to district courts rather than

county courts at law. When a plaintiff tries to override that decision and

sue elsewhere, he falls outside the waiver.

Respondent sued in a county court at law, not a district court. See

id. The waiver does not extend to county courts at law. Respondent did

not adhere strictly to the “conditions on the legislative waiver.” Nestle,

359 S.W.3d at 208. Its claim thus necessarily fails.

Neither the Court nor Respondent offers a plausible solution to

this dispositive defect. The Court’s analysis is limited to whether, as a

matter of the jurisdictional limits in our Constitution’s Article V, a

county court at law has authority to hear this type of claim. But that

assumes the premise that immunity is waived. We need not reach the

question of constitutional limits on county courts at law because

Respondent did not comply with a prerequisite necessary for the waiver

to be effective. See TEX. PROP. CODE §§ 21.003, 21.101(c); Nestle, 359

S.W.3d at 208. Any legislative waiver is effective only when the plaintiff

complies strictly with its terms, including filing in the court the

Legislature has identified—an issue unrelated to the abstract question

of what types of claims can proceed in a county court at law. Respondent

29

did not comply with the terms of any waiver, and so its claim fails at the

outset.

Neither of Respondent’s proposed solutions is satisfactory. First,

Respondent claims that Sections 21.003 and 21.101(c) use a permissive

“may,” and thus permit—but do not require—district courts to hear

repurchase suits. But the State “cannot be sued without its consent, and

then only in the manner, place, and court or courts designated.” Isbell,

94 S.W.2d at 424 (emphasis added) (citation omitted); see also City of

Madisonville v. Sims, 620 S.W.3d 375, 379 (Tex. 2020) (explaining that

even where a statute waives immunity, “strict compliance with the

[statute’s] procedural requirements is required to bring a claim”). Even

if the repurchase statute “permits” the State to be sued in a district

court, the statute does not affirmatively permit such suits to be brought

in a county court at law.

Respondent next invokes Government Code Section 25.1032,

which grants “[a] county civil court at law in Harris County”

“jurisdiction” to “hear a suit for the recovery of real property.” TEX. GOV’T

CODE § 25.1032(a), (d)(6). But as its text makes obvious,

Section 25.1032(d)(6) is a run-of-the-mill jurisdictional grant. See AIC

Mgmt. v. Crews, 246 S.W.3d 640, 643 (Tex. 2008) (labeling Section

25.1032 as a “jurisdictional grant to statutory county courts in Harris

County”). It does not even purport to (much less actually) modify or

excuse compliance with the separate provision in which Respondent

claims to find a waiver. Indeed, a neighboring provision allows county

civil courts at law to “hear a suit to recover damages for slander or

defamation of character,” TEX. GOV’T CODE § 25.1032(d)(2), but this

30

language obviously does not waive immunity for slander suits brought

against the State.

Nor does Property Code Section 21.001—which provides that

“[d]istrict courts and county courts at law have concurrent jurisdiction

in eminent domain cases”—waive immunity. TEX. PROP. CODE § 21.001.

As the Court correctly observes, Section 21.001 is a jurisdictional grant

that vests county courts at law with the power to hear “eminent domain

cases.” Ante at 25. However, as with Section 25.1032(d)(6), such

jurisdictional grants are not waivers of immunity. See supra Part II.A.

For these reasons, I would hold that even if Chapter 21 could be

read to waive immunity for repurchase suits, that waiver does not

extend to suits—like Respondent’s—brought in a county court at law.

IV

I would reverse the judgment below and render judgment for the

State. I respectfully dissent.

Kyle D. Hawkins

Justice

OPINION FILED: June 12, 2026

31