Alabama law requires claimants seeking unemployment benefts to exhaust their administrative remedies before suing over those benefts in state court. Petitioners, the claimants here, failed to complete that process before they sued under Rev. Stat. § 1979, 42 U. S. C. § 1983. The Alabama Supreme Court accordingly held that it lacked jurisdiction over the suit. That holding was plainly permissible. As a matter of frst principles, States have unfettered discretion over whether to provide a forum for § 1983 claims in their courts. And, Alabama's exhaustion rule does not transgress the limitations that our precedents have recogPage Proof Pending Publication nized. The Court concludes otherwise by endorsing an as- applied theory of futility that is both forfeited and meritless, moving our jurisprudence even further off course. I respectfully dissent.
I
This case is straightforward under frst principles. Our federal system gives States “plenary authority to decide whether their local courts will have subject-matter jurisdiction over federal causes of action.” Haywood v. Drown, 556 U. S. 729, 743 (2009) (Thomas, J., dissenting). The Constitution allows States to hear federal claims in their courts, but it does “not impose a duty on state courts to do so.” Id., at 747. Thus, “[o]nce a State exercises its sovereign prerogative to deprive its courts of subject-matter jurisdiction over a federal cause of action, it is the end of the matter as far as the Constitution is concerned.” Id., at 749.
The only potential constraint that the Constitution places on a State's jurisdictional discretion is the possibility that a federal statute may preempt state law. The Supremacy Clause makes the “Constitution, and the Laws of the United States which shall be made in Pursuance thereof . . . the supreme Law of the Land.” Art. VI, cl. 2. Accordingly, “[f]ederal law must prevail when Congress validly enacts a statute that expressly supersedes state law, or when the state law conficts with a federal statute.” Haywood, 556 U. S., at 764 (Thomas, J., dissenting) (citations omitted). This preemption rule raises the “diffcult question” whether Congress can “require state courts to entertain a federal cause of action.” Ibid., n. 8.
We need not answer that question here because § 1983 does not raise any preemption issue. By its text, the provision does not “command” States to provide a forum for § 1983 plaintiffs. Id., at 765. Instead, it merely “addresses who may sue and be sued for violations of federal law.” Ibid.; see § 1983 (deeming “liable” state offcials who deny “any citizen of the United States or other person within the jurisdiction thereof . . . any rights, privileges, or immunities secured Page Proof Pending Publication by the Constitution and laws”). Nor does Alabama's exhaustion bar, which regulates state-court litigation, create any implicit confict with § 1983. Plaintiffs who do not exhaust state remedies are always free to bring their claims in a federal forum. Id., at 766; see also Felder v. Casey, 487 U. S. 131, 160 (1988) (O'Connor, J., dissenting) (“Every [§ 1983] plaintiff has the option of proceeding in federal court, and the [state] statute has not the slightest effect on that right”). Preemption analysis requires nothing further. This Court's precedents err to the extent they recognize a broader form of confict preemption for “state-court procedural rules that are perceived to `burde[n] the exercise of the federal right' in state court.” Haywood, 556 U. S., at 766 (Thomas, J., dissenting) (quoting Felder, 487 U. S., at 141). This form of confict preemption targets state-law rules that constitute an obstacle to the “goals” embodied in federal law. Id., at 138. But, only federal law itself can support preemption under the Supremacy Clause. Extratextual speculation about Congress's purposes cannot. See Wyeth v. Le vine, 555 U. S. 555, 603–604 (2009) (Thomas, J., concurring in judgment).
Our precedents also err in establishing the requirement at issue here—that state jurisdictional rules be “neutral,” even in the absence of a directly conficting federal law. See in fra, at 183. The Supremacy Clause does not of its own force “constrai[n] the States' authority to defne the subject-matter jurisdiction of their own courts.” Haywood, 556 U. S., at 750 (Thomas, J., dissenting). Rather, in making the Constitution and federal law supreme, “it provides only a rule of decision that the state court must follow if it adjudicates the claim.” Id., at 751. I would therefore disregard our further limitation as “demonstrably erroneous.” See Gamble v. United States, 587 U. S. 678, 717–718 (2019) (Thomas, J., concurring).* *Petitioners' suit implicates other precedents that may not withstand scrutiny. I doubt that petitioners have a true due process interest in “mere Government benefts and entitlements.” Axon Enterprise, Inc. v. Page Proof Pending Publication Taken together, this case should begin and end with Ala- bama's plenary authority to decide which federal matters its state courts will have subject-matter jurisdiction to hear. Alabama exercised that authority to create an exhaustion requirement, and we should respect its decision.
II
This Court should affrm even under existing precedents. Alabama's exhaustion requirement does not run afoul of the limitations that this Court has identifed on a State's authority to restrict federal causes of action from proceeding in state court. Petitioners misread our precedents in arguing otherwise, and the majority's theory likewise cannot pass muster.
A
Although this Court has held that there are limits on a State's discretion in regulating state-court jurisdiction over federal causes of action, our precedents emphasize that state authority predominates.
“The general rule `bottomed deeply in belief in the importance of state control of state FTC, 598 U. S. 175, 201, n. 3 (2023) (Thomas, J., concurring). Tellingly, the Court's original expansion of the Due Process Clause into this context came without meaningful legal analysis. The Court simply highlighted the social importance of “entitlements,” which had come to make up “[m]uch of the existing wealth in this country,” and which only the poor had been theretofore unable to effectively enforce. See Goldberg v. Kelly, 397 U. S. 254, 262, and n. 8 (1970) (citing C. Reich, Individual Rights and Social Welfare: The Emerging Legal Issues, 74 Yale L. J. 1245, 1255 (1965); C. Reich, The New Property, 73 Yale L. J. 733 (1964)). As Justice Black recognized at the time, it “strains credulity” as a textual matter “to say that the government's promise of charity to an individual is property” protected by the Fourteenth Amendment. 397 U. S., at 275 (dissenting opinion). Moreover, further examination may be required as to whether § 1983 can provide petitioners a cause of action in any event. Cf. T. Lindley, Anachronistic Readings of Section 1983, 75 Ala. L. Rev. 897, 900–901 (2024) (contending that, as originally understood, § 1983 did not provide a freestanding cause of action).
Page Proof Pending Publication judicial procedure, is that federal law takes the state courts as it fnds them.' ” Howlett v. Rose, 496 U. S. 356, 372 (1990) (quoting H. Hart, The Relations Between State and Federal Law, 54 Colum. L. Rev. 489, 508 (1954)). Each State thus has “great latitude to establish the structure and jurisdiction of [its] own courts.” Howlett, 496 U. S., at 372. This latitude allows States to decide which federal claims their courts can hear. Ibid. As relevant here, our precedents establish that States must exercise this jurisdictional latitude only through “neutral” rules that do not embody any “policy disagreement” with federal law. Haywood, 556 U. S., at 735–737. Based on this principle, we have identifed two narrow exceptions to a State's ordinary discretion. First, a State may not refuse to hear a federal claim “solely because [it] is brought under a federal law.” McKnett v. St. Louis & San Francisco R. Co., 292 U. S. 230, 233–234 (1934). Second, a State may not deprive its courts of jurisdiction over a “disfavored” federal claim, even if it simultaneously denies jurisdiction to an “identical state claim,” where doing so would “undermine federal law.” Haywood, 556 U. S., at 737–739.
For good reason, no one suggests that the frst exception applies. Alabama's exhaustion requirement by its terms does “not discriminate against rights arising under federal laws.” See McKnett, 292 U. S., at 234. Instead, it imposes a generally applicable exhaustion process “for the making of determinations with respect to claims for unemployment compensation benefts.” Ala. Code § 25–4–96 (2016). State and federal claims regarding unemployment benefts are equally subject to this process, including as to “procedural” challenges like the one here. Johnson v. Washington, 387 So. 3d 138, 143 (Ala. 2023).
The second exception does not apply either. Alabama's exhaustion requirement is nothing like the statute in Hay wood that this Court viewed as “disfavor[ing]” federal law. Page Proof Pending Publication 556 U. S., at 738. That statute deprived New York courts of jurisdiction over “damages suits fled by prisoners against state correction offcers,” based on the State's belief that they were “by and large frivolous and vexatious.” Id., at 733 (discussing N. Y. Correc. Law Ann. § 24 (West 1987)). This Court deemed New York's rule “effectively an immunity statute cloaked in jurisdictional garb,” which protected correction offcers from a subset of disfavored § 1983 claims even as New York courts continued to hear most § 1983 actions.
556 U. S., at 741–742.
According to the Hay wood majority, that policy-driven denial could not be squared with the supremacy of § 1983's countervailing policy. Id., at 740.
Alabama's decision to create an exhaustion requirement for all unemployment-benefts-related claims does not embody any comparable policy judgment. Rather, this requirement, which has existed since 1939, is an ordinary jurisdictional rule refecting the Alabama Department of Labor's comparative “competence over the subject matter” of unemployment benefts. Howlett, 496 U. S., at 381; see 1939 Ala. Acts no. 497, pp. 737–741. The exhaustion process serves all the useful functions that this Court has recognized: It allows the agency with subject-matter expertise to retain primary responsibility over the area; it avoids unnecessary litigation; and it creates a record in case judicial review is necessary. McCarthy v. Madigan, 503 U. S. 140, 145 (1992). In short, Alabama's exhaustion requirement is a procedural step that “promotes judicial effciency,” ibid., in contrast to the statute in Haywood, which created a de facto “immunity” shielding a class of claims from judicial review, 556 U. S., at 742. We have no authority to interfere with Alabama's choice.
B
Petitioners try to evade Alabama's exhaustion requirement by arguing for a different exception. On their view, our decisions in Patsy v. Board of Regents of Fla., 457 U. S. Page Proof Pending Publication 496 (1982), and Felder establish that States are categorically precluded from imposing exhaustion requirements in the § 1983 context.
But, petitioners badly misread both decisions.
Patsy addressed whether federal courts can impose an exhaustion requirement for § 1983 cases in the absence of a congressional directive to do so. See 457 U. S., at 501. The Court held that they cannot, reasoning that federal courts may create exhaustion requirements only where doing so is consistent with congressional intent, because “Congress is vested with the power to prescribe the basic procedural scheme under which claims may be heard in federal courts.” Id., at 501–502, 516. That analysis has no relevance to the question here: whether States have authority “to establish the structure and jurisdiction of their own courts.” How lett, 496 U. S., at 372.
Felder too is inapposite. That decision held that § 1983 preempted a Wisconsin notice-of-claim statute that effectively altered the scope of § 1983 liability on the merits. See 487 U. S., at 153. That is, the statute subjected state-court plaintiffs to a dismissal with prejudice if they did not frst submit their claims against the State or its offcers to the government for an advance merits determination. Id., at 136–137, and n. 2; see Haywood, 556 U. S., at 773–774, n. 11 (Thomas, J., dissenting). Failure to exhaust under that statute operated as a state-created merits defense to § 1983 liability. But, the impermissibility of such a merits defense says nothing about a State's discretion to create true jurisdictional rules, which speak only to the judiciary's “ `power' ” to “ `proceed at all.' ” Steel Co. v. Citizens for Better Envi ronment, 523 U. S. 83, 94 (1998) (quoting Ex parte McCardle, 7 Wall. 506, 514 (1869)).
Felder would remain inapposite even if it had involved a purportedly jurisdictional rule compelling dismissal without prejudice. In that event, the Wisconsin statute would simply have raised the problem that this Court later confronted Page Proof Pending Publication in Haywood, where the State singled out a “disfavored” category of claims for second-class treatment. 556 U. S., at 738. The notice-of-claim statute in Felder existed to “further the State's interest in minimizing liability and the expenses associated with it.” 487 U. S., at 143. And, although Felder noted that the statute “impose[d] an exhaustion requirement,” it treated that fact as one of multiple “interrelated” factors that caused the Wisconsin statute to “burden” § 1983 claimants. Id., at 141, 146. The exhaustion requirement was not an independently fatal problem, so Felder's language on exhaustion should not be overread. See Cohens v. Vir ginia, 6 Wheat. 264, 399 (1821) (“[G]eneral expressions, in every opinion, are to be taken in connection with the case in which those expressions are used. If they go beyond the case, they may be respected, but ought not to control the judgment in a subsequent suit when the very point is presented for decision”).
C
The majority rules for petitioners on narrower grounds, but its holding is equally unpersuasive. The majority does not dispute that, as a general matter, Alabama is entitled to apply its exhaustion requirement to § 1983 claims. See ante, at 174, n. 2. It instead holds that, under Haywood, Alabama's discretion cannot extend to the specifc claims here, which challenge delays in the exhaustion process itself. Ante, at 173–174. On the majority's view, maintaining the exhaustion requirement for such claims would mean that petitioners will never be able to advance to state court, leaving the State essentially “immun[e]” from challenges to its exhaustion process. Ante, at 175–176. This theory of futility is both forfeited and meritless.
“[T]his Court has almost unfailingly refused to consider any federal-law challenge to a state-court decision unless the federal claim `was either addressed by or properly presented Page Proof Pending Publication to the state court that rendered the decision we have been asked to review.' ” Howell v. Mississippi, 543 U. S. 440, 443 (2005) (per curiam) (quoting Adams v. Robertson, 520 U. S. 83, 86 (1997) (per curiam)). In fact, the Court's historical practice has generally been to treat this preservation requirement as jurisdictional, although our more recent cases have expressed uncertainty on this issue. Howell, 543 U. S., at 445–446. In view of petitioners' preservation obligation, we should reject as forfeited their newfound theory of futility, which was neither presented nor addressed below.
Until seeking certiorari, petitioners litigated this case as a facial challenge, arguing solely that § 1983 “categorically” preempted States from applying exhaustion requirements in the § 1983 context. Reply Brief for Appellant in Johnson v. Washington, No. SC–2022–0897 (Ala. Sup. Ct.), p. 16. The Alabama Supreme Court accordingly understood that this facial challenge was petitioners' “only” argument for federal preemption. 387 So. 3d, at 143–144. Petitioners belatedly contend that they also raised a futility-based argument, but the briefng they cite merely addressed how the futility of waiting for exhaustion affected the proper timing of their facial challenge. See Reply Brief 24, n. 3 (citing Reply Brief for Appellant in No. SC–2022–0897, at 16–17).
Because petitioners raised only a facial challenge below, they cannot press an as-applied challenge here. “[F]acial” and “as-applied” claims are distinct and must be individually preserved. See United States v. Stevens, 559 U. S. 460, 473, n. 3 (2010); see also, e.g., Moody v. NetChoice, LLC, 603 U. S. 707, 723 (2024) (“NetChoice chose to litigate these cases as facial challenges, and that decision comes at a cost”). So, petitioners cannot now argue that Alabama's exhaustion requirement is impermissible in the specific circumstance where exhaustion would be futile.
There is no reason to treat this case as the “very rare exceptio[n]” in which petitioners' forfeiture might be overlooked. Adams, 520 U. S., at 86 (internal quotation marks Page Proof Pending Publication Page Proof Pending Publication omitted). The majority, which ignores that petitioners needed to raise their as-applied objection specifcally, certainly provides no justifcation. See ante, at 178–179. Instead, its analysis only highlights why we should not decide petitioners' as-applied challenge in the frst instance. The majority's futility theory depends on the assumption that petitioners will never have their day in court if we leave Alabama's exhaustion requirement intact. See ante, at 175– 176. But, petitioners' failure to raise their as-applied claim below means that we have no way of knowing whether this assumption is true. It may be the case that the exhaustion requirement here contains an implicit futility exception. Cf. Graysville v. Glenn, 46 So. 3d 925, 929 (Ala. 2010) (identifying futility as a “recognized exceptio[n]” to the “exhaustionof-administrative-remedies doctrine” generally). Or, it may be the case that petitioners may obtain mandamus relief, as the dissent below suggested and the State underscored. See 387 So. 3d, at 146 (Cook, J., dissenting); Tr. of Oral Arg. 54–56. As a federal court assessing petitioners' objection in the frst instance, we have no way to assess the viability of these or any other mechanisms.
The majority's attempts to disregard this uncertainty are unpersuasive. The majority concludes that the uncertainty should count against the State, and expresses doubt about the availability of mandamus based on the Alabama Supreme Court's failure to address that form of relief. Ante, at 177– 178. But, that court had no reason to opine on the alternative pathways available to petitioners, given that petitioners failed to raise an as-applied challenge. We should not reward petitioners for their own mistake. Likewise, the majority's assertion that mandamus would be irrelevant even if it were available is puzzling. Ante, at 178. If petitioners can secure completion of the exhaustion process through mandamus, then by defnition they will not be in a “catch22” that “prevents [them] from obtaining a merits resolution of their § 1983 claims in state court.” Ante, at 176.
In any event, petitioners' as-applied challenge fails on the merits. Unlike the New York statute in Haywood, Ala- bama's exhaustion requirement is not “ `an immunity statute cloaked in jurisdictional garb.' ” Contra, ante, at 177 (quoting Haywood, 556 U. S., at 742).
Properly understood, Haywood directs our focus to the challenged statute's purpose. The Court there viewed the New York statute as an immunity statute because it was “designed to shield” correction offcers from damages claims brought by prisoners, “[b]ased on the belief” that these claims tended to be “frivolous and vexatious.” Id., at 741– 742. In other words, States cannot implicitly reject the supremacy of federal law by basing a jurisdictional limitation— even one that also applies to state claims—on “policy disagreement” with federal law. Id., at 737–738.
A focus on statutory purpose makes clear that Alabama's exhaustion requirement raises no Haywood problem. There is no credible argument that Alabama adopted its exhaustion requirement in order to defeat challenges to the exhaustion process itself. Alabama created its exhaustion scheme in 1939, decades before the understanding that public benefts give rise to a due process interest emerged. See supra, at 181–182, n. And, the Alabama exhaustion process is by all accounts an ordinary exhaustion requirement common among public-benefts schemes, which in the mine-run case serves to facilitate the adjudication of benefts determinations on the merits. There is no reason to think that Alabama intended to cause mischief in the rare context of a § 1983 challenge to its procedures.
At most, this case presents a circumstance in which Ala- bama's “neutral jurisdictional rule” has the effect of defeating a federal claim. See Haywood, 556 U. S., at 735. But, again, our precedents disallow a State's jurisdictional rule only if it is in fact not “neutral”—that is, if it is “based on a policy disagreement,” and so is intended to “shut the courtPage Proof Pending Publication Page Proof Pending Publication house door to federal claims that it considers at odds with its local policy.” Id., at 737–738, 740.
The majority's contrary conclusion misunderstands Hay wood. Ignoring that decision's purpose-focused language, the majority asserts that it disallows any state rule that “operates as an `immunity statute' . . . by wholly barring a `particular species' of § 1983 suits in state court.” Ante, at 178 (quoting 556 U. S., at 739, 742). But, in context, that quoted language only reinforces the majority's error. Those lines in Haywood reiterate that what mattered there was New York's illicit purpose: A State may not “dee[m]” “a particular species of suits . . . inappropriate for its trial courts.” Id., at 739–740. Nor may a State effectively create an “immunity statute” “[b]ased on the belief that [certain claims] are frivolous and vexatious.” Id., at 742. Nothing in Haywood suggests that a state rule could be impermissible just because it has the incidental effect of disallowing certain federal claims. The majority also does not grapple with the possible ripple effects of its reading of Haywood. It professes only that its opinion is “narrow” and does nothing more than “resolv[e] this dispute.” Ante, at 179. But, the majority's protestations do not make it so.
A constraint based on incidental effects is notably more amorphous than our prior focus on statutory purpose. After all, to the extent the Supremacy Clause bars States from enacting nominally jurisdictional rules that “registe[r their] dissent” from federal policy, States may craft their laws with an eye toward avoiding confict. Haywood, 556 U. S., at 737– 738. But, the same is not true for incidental effects. No statute can be perfectly drafted to anticipate every application that ultimately arises, so it is inevitable that exhaustion requirements will occasionally slow or defeat claims that we might think, as a policy matter, ought to go forward. That happenstance is not a reason for suspicion, just as we do not malign the many federal statutes with similarly categorical exhaustion requirements. See, e. g., Booth v. Churner, 532 U. S. 731, 733–734 (2001) (applying the Prison Litigation Reform Act's exhaustion requirement even where the exhaustion process could not provide the prisoner's requested relief). Here too, the Court should not encroach on Alabama's “latitude to establish the structure and jurisdiction of [its] own courts.” Howlett, 496 U. S., at 372.
III
The Court's decision is irreconcilable with both frst principles and precedent. I respectfully dissent.
Page Proof Pending Publication Page Proof Pending Publication Reporter’s Note The attached opinion has been revised to refect the usual publication and citation style of the United States Reports. The revised pagination makes available the offcial United States Reports citation in advance of publication. The syllabus has been prepared by the Reporter of Decisions for the convenience of the reader and constitutes no part of the opinion of the Court. A list of counsel who argued or fled briefs in this case, and who were members of the bar of this Court at the time this case was argued, has been inserted following the syllabus. Other revisions may include adjustments to formatting, captions, citation form, and any errant punctuation. The following additional edits were made: None