Reed v. Goertz, 598 U.S. ___ (2023)

NOTICE: This opinion is subject to formal revision before publication in the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, pio@supremecourt.gov, of any typographical or other formal errors.

SUPREME COURT OF THE UNITED STATES

_________________

No. 21–442

_________________

RODNEY REED, PETITIONER v. BRYAN GOERTZ

on writ of certiorari to the united states court of appeals for the fifth circuit

[April 19, 2023]

Justice Kavanaugh delivered the opinion of the Court.

In many States, a convicted prisoner who still disputes his guilt may ask state courts to order post-conviction DNA testing of evidence. If the prisoner’s request fails in the state courts and he then files a federal 42 U. S. C. §1983 procedural due process suit challenging the constitutionality of the state process, when does the statute of limitations for that §1983 suit begin to run? The Eleventh Circuit has held that the statute of limitations begins to run at the end of the state-court litigation denying DNA testing, including the state-court appeal. See Van Poyck v. McCollum, 646 F.3d 865, 867 (2011). In this case, by contrast, the Fifth Circuit held that the statute of limitations begins to run when the state trial court denied DNA testing, notwithstanding a subsequent state-court appeal. See 995 F.3d 425, 431 (2021). We conclude that the statute of limitations begins to run at the end of the state-court litigation.

I

In 1996, Stacey Stites was strangled to death in Bastrop County, Texas. The State charged Rodney Reed with murdering Stites. At trial, Reed claimed that he was innocent and that Stites’s fiancé or another acquaintance had committed the murder. A jury rejected that defense theory and found Reed guilty. Reed was sentenced to death. The Texas Court of Criminal Appeals affirmed the conviction and death sentence. Reed’s state and federal habeas petitions were unsuccessful.

Then in 2014, Reed filed a motion in state court under Texas’s post-conviction DNA testing law. See Tex. Code Crim. Proc. Ann., Arts. 64.01–64.05 (Vernon 2018). Reed requested DNA testing on more than 40 pieces of evidence, including the belt used to strangle Stites. Reed contended that DNA testing would help identify the true perpetrator. The state prosecutor, respondent Bryan Goertz, agreed to test several pieces of evidence, but otherwise opposed the motion and refused to test most of the evidence.

The state trial court denied Reed’s motion. The court reasoned in part that (i) many items Reed sought to test—including the belt—were not preserved through an adequate chain of custody and (ii) Reed did not demonstrate that he would have been acquitted if the DNA results were exculpatory. On appeal, the Texas Court of Criminal Appeals affirmed the trial court and later denied Reed’s motion for rehearing.

Reed next sued in federal court under 42 U. S. C. §1983, asserting that Texas’s post-conviction DNA testing law failed to provide procedural due process. Among other things, Reed argued that the law’s stringent chain-of-custody requirement was unconstitutional and in effect foreclosed DNA testing for individuals convicted before “rules governing the State’s handling and storage of evidence were put in place.” App. 39.

The U. S. District Court for the Western District of Texas dismissed Reed’s complaint. The U. S. Court of Appeals for the Fifth Circuit affirmed on the ground that Reed’s §1983 suit was filed too late, after the applicable 2-year statute of limitations had run. The Fifth Circuit ruled that the statute of limitations began to run when the Texas trial court denied Reed’s motion (which occurred more than two years before Reed filed his §1983 suit in federal court), not when the Texas Court of Criminal Appeals denied rehearing.

Because the federal Courts of Appeals disagree about when the statute of limitations begins to run for a §1983 suit regarding a State’s post-conviction DNA testing procedures, we granted certiorari. 596 U. S. ___ (2022).

II

Texas raises three threshold arguments.

First, Texas argues that Reed lacks standing. We disagree. Reed sufficiently alleged an injury in fact: denial of access to the requested evidence. The state prosecutor, who is the named defendant, denied access to the evidence and thereby caused Reed’s injury. And if a federal court concludes that Texas’s post-conviction DNA testing procedures violate due process, that court order would eliminate the state prosecutor’s justification for denying DNA testing. It is “substantially likely” that the state prosecutor would abide by such a court order. Utah v. Evans, 536 U.S. 452, 464 (2002) (internal quotation marks omitted). In other words, in “terms of our ‘standing’ precedent, the courts would have ordered a change in a legal status,” and “the practical consequence of that change would amount to a significant increase in the likelihood” that the state prosecutor would grant access to the requested evidence and that Reed therefore “would obtain relief that directly redresses the injury suffered.” Ibid.

Second, Texas invokes the State’s sovereign immunity. But the Ex parte Young doctrine allows suits like Reed’s for declaratory or injunctive relief against state officers in their official capacities. 209 U.S. 123, 159–161 (1908).

Third, Texas contends that Reed’s procedural due process claim contravenes the Rooker-Feldman doctrine. See Rooker v. Fidelity Trust Co., 263 U.S. 413 (1923); District of Columbia Court of Appeals v. Feldman, 460 U.S. 462 (1983). That doctrine prohibits federal courts from adjudicating cases brought by state-court losing parties challenging state-court judgments. But as this Court explained in Skinner v. Switzer, even though a “state-court decision is not reviewable by lower federal courts,” a “statute or rule governing the decision may be challenged in a federal action.” 562 U.S. 521, 532 (2011). Here, as in Skinner, Reed does “not challenge the adverse” state-court decisions themselves, but rather “targets as unconstitutional the Texas statute they authoritatively construed.” Ibid.

III

This Court’s case law “severely limits the federal action a state prisoner may bring for DNA testing.” Skinner v. Switzer, 562 U.S. 521, 525 (2011). The Court has “rejected the extension of substantive due process to this area, and left slim room for the prisoner to show that the governing state law denies him procedural due process.” Ibid. (citation omitted); see District Attorney’s Office for Third Judicial Dist. v. Osborne, 557 U.S. 52, 69, 72 (2009).

Seeking to fit his §1983 suit within the “slim room” left by this Court’s precedent, Reed raised a procedural due process challenge to Texas’s post-conviction DNA testing law. The sole question now before this Court is whether Reed’s §1983 suit was timely. The parties agree that the statute of limitations for Reed’s claim is two years. But the parties disagree about when that 2-year limitations period began to run. That question is one of federal law. See Wallace v. Kato, 549 U.S. 384, 388 (2007).

As a general matter, the statute of limitations begins to run when the plaintiff has a “complete and present cause of action.” Bay Area Laundry and Dry Cleaning Pension Trust Fund v. Ferbar Corp. of Cal., 522 U.S. 192, 201 (1997) (internal quotation marks omitted). To determine when a plaintiff has a complete and present cause of action, the Court focuses first on the specific constitutional right alleged to have been infringed. See McDonough v. Smith, 588 U. S. ___, ___ (2019) (slip op., at 4).

Here, the specific constitutional right allegedly infringed is procedural due process. A procedural due process claim consists of two elements: (i) deprivation by state action of a protected interest in life, liberty, or property, and (ii) inadequate state process. See Zinermon v. Burch, 494 U.S. 113, 125 (1990). Importantly, the Court has stated that a procedural due process claim “is not complete when the deprivation occurs.” Id., at 126. Rather, the claim is “complete” only when “the State fails to provide due process.” Ibid.

Reed contends that the State’s process for considering his DNA testing request was fundamentally unfair in violation of the Due Process Clause. Texas’s process for considering a request for DNA testing in capital cases includes not only trial court proceedings, but also appellate review by the Court of Criminal Appeals. Tex. Code Crim. Proc. Ann., Art. 64.05. And under longstanding Texas rules of appellate procedure, the Court of Criminal Appeals’s appellate review process encompasses a motion for rehearing. Tex. Rule App. Proc. 79.1 (2022).

In Reed’s case, the State’s alleged failure to provide Reed with a fundamentally fair process was complete when the state litigation ended and deprived Reed of his asserted liberty interest in DNA testing. Therefore, Reed’s §1983 claim was complete and the statute of limitations began to run when the state litigation ended—when the Texas Court of Criminal Appeals denied Reed’s motion for rehearing.

The soundness of that straightforward conclusion is “reinforced by the consequences that would follow” from a contrary approach. McDonough, 588 U. S., at ___ (slip op., at 9). If the statute of limitations for a §1983 suit like Reed’s began to run after a state trial court’s denial of a plaintiff ’s motion for DNA testing (or even after the appeal before the plaintiff ’s rehearing proceedings), the plaintiff would likely continue to pursue relief in the state system and simultaneously file a protective federal §1983 suit challenging that ongoing state process. That parallel litigation would “run counter to core principles of federalism, comity, consistency, and judicial economy.” Id., at ___ (slip op., at 10). We see no good reason for such senseless duplication.

Moreover, significant systemic benefits ensue from starting the statute of limitations clock when the state litigation in DNA testing cases like Reed’s has concluded. If any due process flaws lurk in the DNA testing law, the state appellate process may cure those flaws, thereby rendering a federal §1983 suit unnecessary. And if the state appellate court construes the DNA testing statute, that construction will streamline and focus subsequent §1983 proceedings.

In sum, when a prisoner pursues state post-conviction DNA testing through the state-provided litigation process, the statute of limitations for a §1983 procedural due process claim begins to run when the state litigation ends. In Reed’s case, the statute of limitations began to run when the Texas Court of Criminal Appeals denied Reed’s motion for rehearing. Reed’s §1983 claim was timely.[1]

We reverse the judgment of the U. S. Court of Appeals for the Fifth Circuit.

It is so ordered.

SUPREME COURT OF THE UNITED STATES

_________________

No. 21–442

_________________

RODNEY REED, PETITIONER v. BRYAN GOERTZ

on writ of certiorari to the united states court of appeals for the fifth circuit

[April 19, 2023]

Justice Thomas, dissenting.

The Texas Court of Criminal Appeals (“CCA”) affirmed the denial of petitioner Rodney Reed’s state-law motion for postconviction DNA testing. Reed petitioned this Court for certiorari, arguing that the CCA’s interpretation and application of the relevant state law violated his federal due process rights. After we denied his petition, Reed repackaged it as a complaint in Federal District Court, naming respondent (the Bastrop County District Attorney) as a placeholder defendant. Like his earlier certiorari petition, Reed’s complaint assails the CCA’s state-law reasoning as inconsistent with due process, and it seeks a declaration that the CCA’s interpretation and application of state law was unconstitutional.

Reed’s action should be dismissed for lack of subject-matter jurisdiction. Federal district courts lack appellate jurisdiction to review state-court judgments, and Reed’s action presents no original Article III case or controversy between him and the district attorney. Because the Court erroneously holds that the District Court had jurisdiction over Reed’s action, I respectfully dissent.

I

A

On April 23, 1996, 19-year-old Stacey Stites failed to report for her 3:30 a.m. shift at the H.E.B. grocery store in Bastrop, Texas. The truck Stites drove to work was found abandoned in the Bastrop High School parking lot a couple of hours later. That afternoon, a passerby discovered Stites’ body in a ditch by a country road, her clothing disturbed in a manner suggesting sexual violence. Medical examiners determined that Stites had been strangled to death with her own belt, which was found in two pieces—one near the truck, the other near Stites’ body. There was semen in Stites’ vagina and rectum and saliva on her breasts. The police concluded that Stites had been raped and murdered.

Despite a wide-ranging investigation, the police were initially unable to find a DNA match for the bodily fluids recovered from Stites’ corpse. Then, about six months after Stites’ death, Reed was arrested for kidnaping and attempting to rape and murder another young woman near the route Stites typically took to work and around the same time of night when Stites had gone missing. Reed lived near the high school and was often seen walking the surrounding area at night. Intrigued, the police checked Reed’s DNA profile, which Texas had on file from an earlier sexual-assault case against him. A series of tests established a conclusive, one-in-the-world-population match between Reed and the fluids recovered from Stites’ corpse.

When first questioned, Reed insisted that he did not know Stites at all, unaware that the police had DNA evidence disproving that claim. By the time of his trial, he had changed his story: He and Stites were having a consensual affair, and someone else—perhaps her jealous fiancé—had committed the murder. The jury rejected that post hoc narrative and found Reed guilty. In the separate penalty phase, Reed’s kidnaping victim testified about how Reed had abducted, threatened, and attempted to rape her before she was fortuitously able to escape. Four other women—and one underage girl—also testified that Reed had brutally beaten and raped them in the past. Reed was sentenced to death.

The CCA affirmed Reed’s conviction and sentence in 2000. In the 23 years since, he has kept up a constant stream of postconviction filings asserting his innocence. Every few years, Reed’s lawyers have produced a new theory and a new purportedly exculpatory affidavit. With the patience of Job, the Texas courts have duly considered them all. On one such occasion, the CCA noted “the complete lack of a cohesive theory of innocence” across Reed’s unending series of attempts to relitigate his guilt. Ex parte Reed, 271 S.W.3d 698, 746 (2008).[1]

B

In Texas, a convict has two distinct avenues to obtain postconviction DNA testing of evidence—one executive and discretionary, the other judicial and legal. As for the first, the convict can attempt to reach an agreement with the district attorney, who has broad discretion to order or allow DNA testing. See Tr. of Oral Arg. 39–40. In the case of the second, the convict can file a motion under Tex. Code Crim. Proc. Ann., Arts. 64.01 through 64.05 (Vernon 2018) (Chapter 64), which requires “the convicting court” to “order testing” if the movant establishes certain requirements. Ex parte Gutierrez, 337 S.W.3d 883, 889–890 (Tex. Crim. App. 2011).

In 2014, on the same day that the trial court held a hearing to set Reed’s execution date, Reed filed a Chapter 64 motion for DNA testing of a large number of items. The district attorney consented to test some of the items outside of the Chapter 64 framework, but he otherwise opposed Reed’s request. The trial court denied the motion, finding that Reed had not established two necessary elements for Chapter 64 testing: (1) that he “would not have been convicted if exculpatory results had been obtained through DNA testing,” Art. 64.03(a)(2)(A); and (2) that his Chapter 64 motion was “not made to unreasonably delay the execution of sentence or administration of justice.” Art. 64.03(a)(2)(B). Reed appealed, and the CCA remanded for the trial court to address the other elements of the Chapter 64 rubric. After making supplemental findings, the trial court again denied Reed’s motion, and Reed again appealed.

In April 2017, the CCA issued an opinion affirming the trial court. First, the CCA held that the record supported the trial court’s finding that many of the items had not “been subjected to a chain of custody sufficient to establish that [they had] not been substituted, tampered with, replaced, or altered in any material respect.” Art. 64.03(a)(1)(A)(ii); see Reed v. State, 541 S.W.3d 759, 769–770. Second, it held that Reed had not shown “a reasonable likelihood” that many of the items “contain[ed] biological material suitable for DNA testing.” Art. 64.03(a)(1)(B); see 541 S. W. 3d, at 772. Third, addressing only the items that survived the previous two holdings, the CCA held that Reed had not established that exculpatory results from DNA testing of those items would have prevented his conviction. See id., at 773–777. Finally, the CCA held that Reed had failed to establish that his Chapter 64 motion was not made for purposes of delay. See id., at 777–780. The CCA noted that “Chapter 64 had existed with only slight variations for over thirteen years at the time Reed filed his motion,” and that Reed’s motion was suspiciously filed “on the same day the judge heard the State’s motion to set an execution date.” Id., at 779.

Reed moved for rehearing, arguing that the CCA had misapplied the Chapter 64 elements and asserting, in broad terms, that those errors violated his due process rights. See App. to Pet. for Cert. in Reed v. Texas, O. T. 2017, No. 17–1093, pp. 263a–272a. The CCA denied rehearing by summary order in October 2017.

Reed then timely petitioned this Court for a writ of certiorari to review the CCA’s judgment. His petition contended that the CCA’s judgment “violate[d his] due process rights” because it was based on “arbitrary and fundamentally unfair interpretation[s]” of Chapter 64’s chain-of-custody and unreasonable-delay elements. Pet. for Cert. in No. 17–1093, pp. i–ii. We denied certiorari. See Reed v. Texas, 585 U. S. ___ (2018).

C

In August 2019, Reed sued the district attorney under Rev. Stat. §1979, 42 U. S. C. §1983 in the U. S. District Court for the Western District of Texas. As relevant here, Reed’s complaint alleges that he successfully “proved each of the statutory requirements of [Chapter] 64” in the state-court proceedings, App. 31, ¶52, but that “the CCA’s adoption of non-statutory criteria to preclude . . . Reed from testing key trial evidence to prove his innocence violate[d] fundamental notions of fairness and denie[d] him due process of law,” id., at 14, ¶2. Reed proceeds to allege “several ways” in which “[t]he CCA’s interpretation and application of [Chapter] 64 violate[d] fundamental fairness,” id., at 41, ¶79, with particular focus on the CCA’s allegedly arbitrary constructions of the chain-of-custody, unreasonable-delay, and exculpatory-results elements, see id., at 41–42, ¶¶79–81; 43–45, ¶¶84–87. For relief, “Reed seeks a declaration that [Chapter] 64, as interpreted, construed and applied by the Texas courts to deny his motion for DNA testing, violates his rights under” the Constitution. Id., at 14, ¶3; see also id., at 49 (prayer for relief ).

The district attorney moved to dismiss Reed’s complaint for lack of subject-matter jurisdiction and for failure to state a claim. See Fed. Rule Civ. Proc. 12(b)(1) and (b)(6). The District Court held that it had jurisdiction but dismissed Reed’s complaint on the merits, concluding that Reed had alleged only “that he disagree[d] with the state court’s construction of Texas law” and that none of the issues in the complaint “r[ose] to the level of a procedural due-process violation.” 2019 WL 12073901, *7 (WD Tex., Nov. 15, 2019). The Fifth Circuit affirmed on the alternative ground that Reed’s claim was untimely: Applying Texas’ 2-year statute of limitations for personal-injury claims, it reasoned that Reed’s due process claim accrued when the trial court first denied his Chapter 64 motion, rendering his complaint several years too late. 995 F.3d 425, 431 (2021).

II

Two intertwined principles of federal jurisdiction—Article III standing and the Rooker-Feldman doctrine[2]—mandate a finding that the District Court lacked jurisdiction over this action. The majority gives short shrift to these principles, and its holding that Reed’s claim was timely serve only to underscore its antecedent jurisdictional errors.

A

The Constitution limits the federal courts’ jurisdiction to “Cases” and “Controversies,” Art. III, §2, cl. 1, constraining judicial power to “the determination of real, earnest and vital controvers[ies] between” contending litigants. Chicago & Grand Trunk R. Co. v. Wellman, 143 U.S. 339, 345 (1892). “[A]n essential and unchanging part of [this] case-or-controversy requirement” is the doctrine of Article III standing. Lujan v. Defenders of Wildlife, 504 U.S. 555, 560 (1992). Under that doctrine, any party requesting relief from a federal court must assert “an injury” that is “concrete, particularized, and actual or imminent,” and he must show that his injury is both “fairly traceable to the challenged action” and “redressable by a favorable ruling.” Monsanto Co. v. Geertson Seed Farms, 561 U.S. 139, 149 (2010); see also Town of Chester v. Laroe Estates, Inc., 581 U.S. 433, 438–439 (2017). Absent that showing, the court has no jurisdiction and thus no “power to adjudicate the case.” Steel Co. v. Citizens for Better Environment, 523 U.S. 83, 89 (1998) (emphasis deleted).

Jurisdiction, moreover, comes in two types—original and appellate—and the application of the Article III standing elements is interwoven with that constitutionally grounded distinction. See Art. III, §2, cl. 2. In an original case or controversy, the plaintiff traces his injury “to the defendant’s allegedly unlawful conduct,” Allen v. Wright, 468 U.S. 737, 751 (1984), and, correspondingly, seeks a remedy that runs against the defendant and determines that defendant’s duties or liabilities (e.g., a judgment for money damages or an injunction). On the other hand, “[t]he criterion which distinguishes appellate from original jurisdiction, is that it revises and corrects the decisions of another tribunal.” Ex parte Bollman, 4 Cranch 75, 86 (1807); see also Marbury v. Madison, 1 Cranch 137, 175–176 (1803). As such, a case or controversy is appellate in nature when the relief-seeking party’s injury is traceable to the allegedly erroneous action of another court and requires a remedy correcting that judicial action (e.g., reversal or vacatur of the challenged judgment).[3] See, e.g., Food Marketing Institute v. Argus Leader Media, 588 U. S. ___, ___–___ (2019) (slip op., at 4–5); Monsanto, 561 U. S., at 150–153. Thus, whenever a party seeks relief from a federal court, the elements that bring his claim within Article III in the first place—the nature and source of his injury and the remedy needed to redress it—also dictate whether his claim invokes original or appellate jurisdiction.

The conceptual distinction between original and appellate jurisdiction also animates the Rooker-Feldman doctrine—which, despite its name, is not so much a “doctrine” as a basic fact of federal statutory law. This Court has discretionary appellate jurisdiction to review certain state-court judgments by certiorari. 28 U. S. C. §1257(a). But no other federal court has appellate jurisdiction over state-court judgments, and, in particular, “[t]he jurisdiction possessed by the District Courts is strictly original.” Rooker v. Fidelity Trust Co., 263 U.S. 413, 416 (1923); see, e.g., 28 U. S. C. §1331 (“The district courts shall have original jurisdiction of all civil actions arising under [federal law]”). Thus, if the losing party in a state judicial proceeding “claim[s] that the state judgment itself violates [his] federal rights”—a claim that calls for an exercise of appellate jurisdiction—his only remedy in the federal system is certiorari in this Court. Johnson v. De Grandy, 512 U.S. 997, 1006 (1994). He may not “see[k] what in substance would be appellate review of the state judgment” under the guise of an original action in federal district court. Id., at 1005–1006; see also Exxon Mobil Corp. v. Saudi Basic Industries Corp., 544 U.S. 280, 284–285 (2005); District of Columbia Court of Appeals v. Feldman, 460 U.S. 462, 482–488, and nn. 15, 16 (1983).

Yet, that is precisely what Reed has done here. While his complaint purports to bring an original action against the district attorney, in reality, it seeks appellate review to redress an alleged injury inflicted by the CCA’s adverse decision “in [his] particular cas[e].” Id., at 487, n. 18. The gravamen of Reed’s claim—made clear again and again throughout his complaint—is that the CCA violated his due process rights through its reasoning in his case. See, e.g., App. 14, ¶2; 31–33, ¶¶53–57; 38, ¶69; 39–40, ¶¶71, 74; 41, ¶79; 42–43, ¶¶83–84; 44–45, ¶¶86–87. All of those alleged injuries are traceable to the CCA, not the district attorney. And, redressing them would require an exercise of appellate jurisdiction over the CCA—jurisdiction that the District Court does not have. Confirming the point, Reed’s complaint does not ask the District Court to control the district attorney’s actions at all. Instead, the only relief it requests is “[a] declaration that the CCA’s interpretation and application of [Chapter] 64 . . . is unconstitutional.” Id., at 49. The complaint transparently seeks nothing more than the District Court’s “review and rejection” of the CCA’s judgment. Exxon Mobil, 544 U. S., at 284. As such, it founders upon the Rooker-Feldman doctrine as well as the Article III traceability and redressability requirements with which that doctrine is intertwined.

Any doubt that Reed seeks de facto appellate review should be dispelled by one undisputed fact: Every due process violation that Reed alleges could have been considered on direct review of the CCA’s judgment in this Court. After all, determining whether state-court judgments applied unconstitutional constructions of state law is a classic use of this Court’s appellate jurisdiction under §1257(a). See, e.g., Rogers v. Tennessee, 532 U.S. 451, 453 (2001); Bouie v. City of Columbia, 378 U.S. 347, 349 (1964); Brinkerhoff-Faris Trust & Sav. Co. v. Hill, 281 U.S. 673, 678 (1930). That is why Reed originally petitioned this Court for certiorari to review the CCA’s judgment. And it is why he agreed at oral argument that we could have granted that petition. See Tr. of Oral Arg. 32–33.

Yet, even after repackaging his failed certiorari petition as an original §1983 complaint, Reed not only concedes but affirmatively argues that his claim is analogous to the due process arguments presented in Rogers, Bouie, and Brinkerhoff-Faris. See Brief for Petitioner 33–34. That he is correct on that front should be fatal to his complaint. Like the petitioners in those cases, Reed contends that the rules of decision applied against him in a state-court proceeding violated his due process rights. Because those contentions would have been appropriate subjects for this Court’s appellate review, it follows that Reed cannot press the same due process challenges and seek the same relief in an original action in the District Court. See Feldman, 460 U. S., at 482–486, and n. 15; accord, id., at, 489 (Stevens, J., dissenting).

In holding otherwise, the majority improperly separates the Rooker-Feldman and Article III inquiries and applies a different theory of Reed’s claim to each. But, Reed’s claim must satisfy two conditions at once: It must implicate an Article III case or controversy between the parties to this action, and that case or controversy must fall within the District Court’s “strictly original” jurisdiction. Rooker, 263 U. S., at 416. The majority articulates no theory of how Reed’s claim can satisfy both conditions. That is because there is no such theory.

A useful way to view this is to work backwards from the majority’s Rooker-Feldman holding. The majority accepts Reed’s representation that he “does ‘not challenge the adverse’ state-court decisions themselves,” but only “ ‘targets as unconstitutional the Texas statute [Chapter 64] they authoritatively construed.’ ” Ante, at 4 (quoting Skinner v. Switzer, 562 U.S. 521, 532 (2011)). But this workaround to Rooker-Feldman raises a glaring Article III problem: As this Court has repeatedly explained, a federal court may not entertain a free-floating challenge to a statute unmoored from a concrete case or controversy. See, e.g., California v. Texas, 593 U. S. ___, ___–___ (2021) (slip op., at 7–9); Valley Forge Christian College v. Americans United for Separation of Church and State, Inc., 454 U.S. 464, 471–472 (1982); Massachusetts v. Mellon, 262 U.S. 447, 488 (1923); Muskrat v. United States, 219 U.S. 346, 360–362 (1911). Unless Reed merely seeks an advisory opinion, his due process challenge to Chapter 64 must seek relief from some concrete enforcement or application of that law that affects him. More specifically, Reed must be challenging either (1) some conduct of the district attorney constituting enforcement of Chapter 64 against him or (2) the CCA’s application of Chapter 64 as a rule of decision in his case.[4] If it is the former, Reed’s suit is original; if it is the latter, it requires an exercise of appellate jurisdiction.

So, which is it? As already indicated, the correct answer is the latter: Fundamentally, Reed’s complaint—like his certiorari petition before it—contests how “the Texas courts” “interpreted, construed[,] and applied” Chapter 64 “to deny his motion for DNA testing,” App. 14, ¶3, which is why the only relief he requests is an abstract “declaration that the CCA’s interpretation and application of [Chapter] 64 . . . is unconstitutional.” Id., at 49. The idea that his claim “does not challenge the adverse state-court decisions,” ante, at 4 (internal quotation marks omitted), cannot survive even a cursory examination of his complaint. See supra, at 9–10.

Nor would the other possibility make any sense. Reed cannot be seeking relief from the district attorney’s enforcement of Chapter 64, because the district attorney has not enforced that law against Reed at all. The sum total of the district attorney’s relevant conduct is as follows. First, he declined to order Reed’s desired testing in his executive discretion, independent of Chapter 64. Next, when Reed asked the Texas courts to grant testing under Chapter 64, the district attorney opposed his motion. Finally, after Reed’s motion proved unsuccessful, the district attorney continued to decline to order Reed’s desired testing. To say that this conduct amounts to enforcing Chapter 64 makes as much sense as saying that a party to a discovery dispute, who defeats a motion to compel, in effect, “enforces” the Federal Rules of Civil Procedure by continuing not to turn over the demanded documents. Again, any due process injury that Chapter 64 has caused Reed is traceable to the CCA’s judicial application of that law in his case, not to any executive acts or omissions of the district attorney.

The majority permits Reed to evade that problem by framing his Article III injury as the mere lack of access to his desired evidence, independent of any alleged due process denial. See ante, at 3. But, if framing Reed’s injury that way helps with traceability, it only worsens his redressability problem. Suppose that the District Court accepted Reed’s due process arguments and issued his requested relief: an abstract declaration that the interpretation of Chapter 64 that the CCA applied in his case is unconstitutional. How, exactly, would that redress Reed’s injury of not having the evidence tested? The CCA’s Chapter 64 judgment would remain untouched; Reed would have obtained an opinion disapproving its reasoning, but without any appellate “revis[ion] and correct[ion]” to disturb its finality. Bollman, 4 Cranch, at 86. Nor would a declaration that the CCA’s construction of Chapter 64 was unconstitutional imply anything about the district attorney’s duties or liabilities.

The majority asserts that such a declaration would cause “ ‘a significant increase in the likelihood’ ” that the district attorney would grant Reed’s desired testing. Ante, at 3 (quoting Utah v. Evans, 536 U.S. 452, 464 (2002)). But the district attorney has made clear that he does not understand Reed’s requested relief to “require any change in conduct” from him and that it is not “likely to bring about such change.” Brief for Respondent 38–39. If the majority thinks the district attorney is wrong about that, it would only be fair to explain exactly what change in conduct would be legally required of him if Reed prevailed on his due process claim. The majority fails to do so.[5]

Instead, it offers a number of vague pronouncements, all of which wilt under scrutiny. Consider the claim that Reed’s victory in this action would “eliminate the [district attorney’s] justification for denying DNA testing.” Ante, at 3. If this means that Reed’s requested relief would entitle him to testing under Chapter 64, it is wrong because the CCA’s unreversed judgment would stand as a final, binding determination of Reed’s Chapter 64 rights even if the District Court were to declare that, in its opinion, the CCA had applied that law unconstitutionally in Reed’s case. Alternatively, if the majority means that the success of Reed’s due process claim would require the district attorney to permit testing in his independent executive discretion, it is also wrong because Reed is not challenging the district attorney’s denial of discretionary testing as unlawful—only the CCA’s “interpretation and application of [Chapter] 64.” App. 49.

The majority also misses the mark when it asserts that it is “substantially likely that the [district attorney] would abide by [Reed’s requested] court order.” Ante, at 3 (internal quotation marks omitted). Again, the only “court order” Reed seeks is a declaration disapproving the legal underpinnings of the CCA’s judgment. Such an “order” would have no bearing on the district attorney’s future conduct; in a literal sense, there would be nothing for him to “abide by.”

Finally, the majority says that the District Court “ ‘would have ordered a change in a legal status’ ” were it to grant the declaration Reed seeks. Ibid. (quoting Utah, 536 U. S., at 464). The intended meaning of this statement is completely obscure. The “status” that the majority has in mind cannot be that of Chapter 64 itself. See California, 593 U. S., at ___ (slip op., at 8) (explaining that judicial remedies “ ‘operate with respect to specific parties,’ ” not “ ‘on legal rules in the abstract’ ”); see also Mellon, 262 U. S., at 488 (explaining that courts “have no power per se to review and annul [statutes] on the ground that they are unconstitutional,” only “the negative power to disregard an unconstitutional enactment” when “declaring the law applicable to [a justiciable] controversy”). Nor can the majority mean that the District Court could change the “status” of the CCA’s judgment. In reality, the only way that the District Court could possibly help Reed obtain DNA testing is by directly controlling the district attorney’s actions. But, again, Reed’s complaint neither requests nor sets forth a basis for any such relief.[6]

In sum, there is no getting around the essential problem with Reed’s due process claim: To the extent he is not merely seeking an advisory opinion, he is complaining about a court-inflicted injury, and redressing that injury would require an exercise of appellate jurisdiction that the District Court does not possess. In substance, his complaint in this action is a mere reprise of his prior certiorari petition, camouflaged as an original action against the district attorney. Thus, I would vacate the Fifth Circuit’s judgment and remand this case to the District Court with instructions to dismiss the complaint for lack of subject-matter jurisdiction.

B

The majority next holds that Reed’s §1983 due process claim was timely because it did not accrue until the CCA denied rehearing. The little reasoning the majority offers for this conclusion helpfully accentuates its antecedent jurisdictional errors.

First, the majority points out that a procedural due process claim is not necessarily “ ‘complete when the deprivation occurs,’ ” but “only when ‘the State fails to provide due process.’ ” Ante, at 5 (quoting Zinermon v. Burch, 494 U.S. 113, 126 (1990)). Yet, “the general rule” is that due process itself “requir[es] predeprivation notice and hearing,” so the truism for which the majority quotes Zinermon matters only in those “extraordinary situations” in which “[w]e tolerate” postdeprivation process as sufficient. United States v. James Daniel Good Real Property, 510 U.S. 43, 53 (1993) (internal quotation marks omitted); see Zinermon, 494 U. S., at 127–130. The majority proceeds to show, however, that it does not regard this case as a postdeprivation case at all, for it says that the State “deprived Reed of his asserted liberty interest in DNA testing” at the very moment when “the State’s alleged failure to provide Reed with a fundamentally fair process was complete.” Ante, at 5. Given this understanding of Reed’s claim, the “[i]mportan[t]” proposition with which the majority begins its analysis is doctrinally irrelevant. Ibid.

After that red herring, the majority engages in an obvious equivocation, conflating the Chapter 64 “process” that Reed challenges as “fundamentally unfair” with the Texas courts’ generally applicable decisional procedures. Ibid. But of course, those procedures are not what Reed challenges. Instead (and, again, exactly like the arguments in his prior certiorari petition), his due process claim “ ‘targets as unconstitutional’ ” the substantive requirements of Chapter 64 as construed. Ante, at 4. His claim plainly would be no different if the CCA did not entertain rehearing motions.

Still, the majority’s confused accrual reasoning is useful for the added light that it shines on Reed’s jurisdictional problems. As the majority says, a procedural due process claim has two elements: (1) a deprivation and (2) inadequate process. The majority then acknowledges that the state courts effectuated Reed’s deprivation, and it treats the state courts’ ordinary decisional mechanics as the allegedly inadequate process. But, after both elements of Reed’s claim are thus laid at the feet of the state courts, what role is left for the nominal defendant here, the district attorney? What part did he play in violating Reed’s procedural due process rights, and what makes him a proper defendant to Reed’s §1983 claim?

The majority has no answer. At bottom, its approval of Reed’s claim is intelligible only upon the supposition that the district attorney may be sued as a mere stand-in for the State as a whole, such that Reed can urge against him the due process violations that the State allegedly committed through its courts. That is a profound mistake. True, the district attorney and the CCA are both state actors. But, States act in different ways through their different entities and officers, and the nature of a challenged state action determines what federal-court remedy may be available.

As this Court has explained, “[a] State acts by its legislative, its executive, or its judicial authorities,” and “in no other way.” Ex parte Virginia, 100 U.S. 339, 347 (1880). The Due Process Clause applies to action through any of these agencies, ibid., but not every alleged due process violation may be asserted in an original §1983 action. By itself, a State’s legislative enactment of an unconstitutional law does not give rise to a justiciable case or controversy. See California, 593 U. S., at ___–___ (slip op., at 7–9); Muskrat, 219 U. S., at 361. Next, when a State allegedly violates due process through executive action, the aggrieved party may bring an original action for appropriate relief against the relevant executive officer. See Mellon, 262 U. S., at 488; Ex parte Young, 209 U.S. 123 (1908).[7] And, when a State allegedly violates due process through its judicial actions—be it through the denial of a fundamentally fair judicial procedure or through the application of a rule of decision that itself violates due process—the remedy that Congress has provided is appellate “review of the [challenged] judgmen[t] in this Court.” Feldman, 460 U. S., at 482; see, e.g., Rogers, 532 U. S., at 453; Bouie, 378 U. S., at 349; Brinkerhoff-Faris, 281 U. S., at 678; cf. Stop the Beach Renourishment, Inc. v. Florida Dept. of Environmental Protection, 560 U.S. 702 (2010) (reviewing judicial-taking claim on certiorari to the challenged state-court judgment). But, if that remedy proves unsuccessful—as it did for Reed—the aggrieved party cannot simply substitute an executive officer as a defendant, charge the state court’s errors to that officer, and seek redress for a court-inflicted injury in a purported original action.

Properly understood, therefore, Article III, the Rooker-Feldman doctrine, and procedural due process principles work in harmony. The majority’s piecemeal analysis replaces this natural coherence with chaos. It dilutes Article III’s traceability and redressability requirements to the point of irrelevance. It creates a system in which the same state-court actions simultaneously give rise to identical original and appellate claims for relief. See this Court’s Rule 13.3 (“[T]he time to file [a] petition for a writ of certiorari . . . runs from the date of the denial of rehearing” by the lower court). It allows Reed to convert his failed certiorari petition into a §1983 complaint. And, in doing so, it authorizes a proceeding in which the District Court can do nothing except opine on the constitutional merits of a state-court adjudication.

*  *  *

If there is a mitigating factor to today’s decision, it is that the §1983 action that the Court misguidedly allows to proceed is no barrier to the prompt execution of Reed’s lawful sentence. See Hill v. McDonough, 547 U.S. 573, 583–584 (2006). Indeed, Reed conceded at oral argument “that you do not get a stay of execution just because you brought [a Chapter] 64 proceeding or just because you’re in [§]1983 proceedings . . . challenging the adequacy of the procedures available to you from the state.” Tr. of Oral Arg. 68. Texas is free to take him at his word. But, because the majority undermines vital principles of federal jurisdiction and destabilizes the orderly working of our judicial system, I respectfully dissent.

SUPREME COURT OF THE UNITED STATES

_________________

No. 21–442

_________________

RODNEY REED, PETITIONER v. BRYAN GOERTZ

on writ of certiorari to the united states court of appeals for the fifth circuit

[April 19, 2023]

Justice Alito, with whom Justice Gorsuch joins, dissenting.

This case involves a suit brought by petitioner Rodney Reed under Rev. Stat. §1979, 42 U. S. C. §1983, against Bryan Goertz, the District Attorney of Bastrop County, Texas. Reed claims that Goertz violated his due process rights when, based on the Texas Court of Criminal Appeals’ interpretation of the Texas statute that allows post-trial DNA testing under specified circumstances, Article 64 of the Texas Code of Criminal Procedure, Goertz continued to deny Reed’s request for DNA testing of certain items found near the scene of the murder for which he was convicted 25 years ago.

As the Court notes and the parties agree, the statute of limitations for Reed’s claim is two years. Ante, at 4; Brief for Petitioner 17; Brief for Respondent 17. Reed filed his complaint on August 8, 2019, and the lower courts held that this was too late. The question before us is when the 2-year statute of limitations began to run, that is in legal parlance, when Reed’s claim “accrued.” As the parties agree, the general rule is that a claim accrues when the plaintiff has “a complete and present cause of action,” Wallace v. Kato, 549 U.S. 384, 388 (2007) (internal quotation marks omitted). Reed contends that his claim did not accrue until the Texas Court of Criminal Appeals (CCA) denied his petition for rehearing on October 4, 2017, and thus refused to retract the interpretation of Article 64 that the court had unanimously adopted on April 12, 2017. Goertz, on the other hand, argues that Reed’s claim accrued no later than the date of the CCA’s April 12 decision, and because that date preceded the federal lawsuit by more than two years, Goertz maintains that we should affirm the Fifth Circuit’s decision that Reed’s complaint was filed too late.

As I will explain, there is room for debate about exactly when Reed’s DNA testing claim accrued, but in my view, the notion that this did not take place until rehearing was denied is clearly wrong.

I

Before getting to the nub of this case, I briefly explain why Reed’s claim might have accrued even earlier than April 12, 2017. First, it can be argued that Reed’s claim against Goertz accrued on or before July 2014, when Goertz initially refused Reed’s testing request.[1] The general rule is that a plaintiff ’s §1983 claim against a state official for violating a constitutional right accrues when the alleged violation takes place. See Wallace, 549 U. S., at 384 (§1983 claim “normally commence[s] to run” from when wrong occurs). And the Court does not disclaim the possibility that a plaintiff could file a §1983 claim as soon as a state prosecutor denies a DNA testing request. See ante, at 6, and n. 1.

Another possibility is that the particular claim Reed now asserts did not accrue until the state trial court held that Goertz had properly denied Reed’s testing request. Reed does not claim that the bare text of Article 64 is unconstitutional. (Had he done so, he could hardly argue, as he does now, that his claim did not accrue until the end of the appellate process.) Instead, he stresses that his claim concerns the state courts’ construction of that statute and in particular, their holding that the evidence for which testing is sought must not be contaminated. Brief for Petitioner 3, 15, 29. In response to this argument, Goertz contends that every allegedly unconstitutional aspect of the judicial interpretation of the statute was adopted by the time the state trial court issued amended findings of fact and conclusions of law in 2016, and Goertz therefore takes the position that Reed’s unconstitutional-construction claim accrued at that time. Brief for Respondent 18–19.

II

A

For present purposes, it is not necessary to decide whether Reed’s claim accrued on either of these two dates. We need only decide whether accrual was put off until the CCA denied rehearing, and it is clear to me that this delayed accrual date is wrong.[2] As noted, the claim that Reed asserts is not based on the bare text of Article 64, but on what he claims is an erroneous interpretation of that provision by the Texas courts. He thus submits that his claim accrued when the “authoritative construction of Article 64” that he challenges was pronounced by the CCA. Brief for Petitioner 17.

I will assume for the sake of argument that Reed’s claim accrued when the CCA issued its “authoritative construction of Article 64,” but I cannot agree with Reed’s argument—which the Court conspicuously declines to defend—that the CCA’s interpretation did not become “authoritative” until rehearing was denied.

Reed cites no authority for the proposition that the filing of a petition for rehearing typically suspends the authoritative force of an appellate court’s decision, and in fact, it appears that the opposite is true—as this Court’s “GVR” practice illustrates. On or shortly after the day when we hand down a decision, we often “GVR” cases in which petitions raising similar issues are pending before us. (That is, we grant the petition, vacate the decision below, and remand the case for reconsideration in light of the decision we have handed down.) On June 30, 2022, for example, we did this in no fewer than 33 cases.[3] We do not wait to see if a petition for rehearing will be filed; nor do we hold off until a mandate is issued or a certified copy of the judgment is prepared. See this Court’s Rules 45.2 and 45.3. If our decisions did not become authoritative and binding as soon as they are issued, this practice would be impermissible.

There is no reason why decisions of the CCA should be viewed any differently. On the contrary, it appears that the CCA has followed a practice similar to our GVR practice. See Oliver v. State, 872 S.W.2d 713, 716 (Tex. Crim. App. 1994) (vacating judgment and remanding for reconsideration in light of decision on same day). And neither Reed nor the Court has cited any contrary Texas authority. Accordingly, Reed’s “authoritative construction” argument became complete, at the latest, when the CCA adopted that construction on April 12, 2017, two years and 11 months before Reed filed his §1983 complaint.

B

Unlike Reed, the Court does not contend that the CCA’s interpretation lacked “authoritative” status until rehearing was denied. Instead, the Court merely proclaims that the State, acting through Goertz, did not deny Reed due process of law until “the state litigation ended.” Ante, at 5.[4] I certainly see the logic in this view: until the process afforded by a State has been exhausted, it may be said that the State has not definitively denied the process that the Constitution is alleged to demand. This logic leads to the conclusion that a prisoner like Reed should exhaust state remedies—something that would generally be required if the proper vehicle for contesting the denial of a DNA testing claim were a petition for a writ of habeas corpus. See 28 U. S. C. §2254(b)(1). But the Court rejected that proposition in District Attorney’s Office for Third Judicial Dist. v. Osborne, 557 U.S. 52 (2009), and it is well-established that a §1983 plaintiff need not exhaust state remedies. Patsy v. Board of Regents of Fla., 457 U.S. 496, 500–501 (1982); Edwards v. Balisok, 520 U.S. 641, 649 (1997). Not only is this the general rule, but the Osborne Court found that the rule applies in cases involving constitutional challenges to the denial of requested DNA testing. 557 U. S., at 71. Thus, the Court’s reasoning collides with precedent.

On top of this, the Court’s reasoning, if taken to its logical conclusion, points to a result that neither Reed nor the Court is willing to embrace: namely, that a due process challenge to the denial of a request for DNA testing is not ripe until state remedies have been exhausted. (Reed squarely rejects that conclusion, Brief for Petitioner 48; Tr. of Oral Arg. 12–13, and the Court reserves judgment. Ante, at 6, n. 1.) But that is where the Court’s reasoning is likely to lead.

Reed tries to circumvent this problem by distinguishing between a claim that challenges the literal terms of a state law and one that challenges the law as authoritatively interpreted by the State’s highest court. Brief for Petitioner 30, 48. On this view, only claims of the latter type would have to proceed through the entire state court appellate process before a §1983 challenge could be brought. But this categorization of DNA-testing claims is problematic. When a State’s high court interprets a state law, it generally settles what the law always meant, and therefore it is hard to see the difference between a claim that the text of a state statute is unconstitutional and a claim that the text is unconstitutional as interpreted by the State’s highest court. In the case of a state law like Article 64, which permits DNA testing under limited circumstances, the court may interpret the statute to impose requirements that are not expressly spelled out in the statutory text. (That is what happened here.) Or the state high court may interpret requirements in the text more leniently than a literal reading of the text would demand. In either event, the statute means what the state high court says it means, and if accrual in the first of these situations does not take place until the end of appellate review, it is hard to see why the same should not be true in the second as well.

In light of these problems, it is not surprising that the Court declines to say anything about whether prisoners who wish to challenge a state DNA testing law may sue as soon as their testing requests are denied. The Court says only that it “need not address th[e] hypothetical scenario” of a plaintiff who declines “full appellate review,” ante, at 6, n. 1, but what does that mean? Does it mean that such a plaintiff must exhaust state remedies at the trial level but need not appeal? Does it mean that such a plaintiff must pursue some (but not “full”) appellate review? Litigants and the lower courts are left to guess. Instead of clarifying the law, the Court’s decision may sow confusion.

C

Much of Reed’s argumentation is not aimed at the argument that his claim accrued when the CCA issued its contested interpretation of Article 64. Instead, Reed directs his attack on the earlier possible accrual dates discussed in Part I of this opinion and in particular the Fifth Circuit’s holding that a claim like Reed’s accrues when testing is denied at the trial level. He says that this rule is unfair because he “isn’t Nostradamus,” lacks “supernatural foresight,” and therefore could not have predicted at the time of the trial court decision whether the CCA would ultimately agree. Brief for Petitioner 26, 32. He argues that his rule promotes federalism (because it encourages resort to state court litigation before turning to the federal courts), judicial economy (because it tends to avoid contemporaneous litigation in both state and federal court), comity (because it allows state courts to adopt interpretations of their statutes that avoid federal constitutional problems), and practical reality (because a prisoner bringing an authoritative-construction claim cannot know in advance how a State’s high court will interpret the relevant statute). Id., at 36–39. The Court makes related arguments. Ante, at 5–6.

Whatever merit these arguments might have in relation to the accrual date adopted by the Fifth Circuit, they ring hollow as applied to the choice between the date when a state high court issues a decision interpreting the state testing statute and the date when that court refuses to rehear and overturn that interpretation. One need not have “supernatural foresight” in order to predict that rehearing is unlikely to be granted. And it is hard to see how requiring a §1983 plaintiff to sue within two years after a state high court decision is issued is unfair or does any damage to federalism, comity, or judicial economy.

Reed has provided no explanation why he could not have filed his §1983 action within two years after the CCA’s decision. Instead, he waited until an execution date was set. While that event may have “concentrate[d] his mind wonderfully,” that is not an excuse for the basic mistake of missing a statute of limitations.[5]

*  *  *

For these reasons, I would affirm the judgment below, and I therefore respectfully dissent.