Medina v. Planned Parenthood South Atlantic, 606 U.S. 357

See case details

MEDINA, DIRECTOR, SOUTH CAROLINA DEPARTMENT OF HEALTH AND HUMAN SERVICES v. PLANNED PARENTHOOD SOUTH ATLANTIC ET AL.

No. 23–1275

SUPREME COURT OF THE UNITED STATES

June 26, 2025

606 U. S. ____ (2025)

CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

Argued April 2, 2025

Syllabus

NOTE: Where it is feasible, a syllabus (headnote) will be released, as is being done in connection with this case, at the time the opinion is issued. The syllabus constitutes no part of the opinion of the Court but has been prepared by the Reporter of Decisions for the convenience of the reader. See United States v. Detroit Timber & Lumber Co., 200 U. S. 321, 337.

SUPREME COURT OF THE UNITED STATES

Syllabus

MEDINA, DIRECTOR, SOUTH CAROLINA DEPARTMENT OF HEALTH AND HUMAN SERVICES v. PLANNED PARENTHOOD SOUTH ATLANTIC ET AL.

CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

No. 23–1275. Argued April 2, 2025—Decided June 26, 2025

Congress created Medicaid in 1965 to subsidize state healthcare for families and individuals “whose income and resources are insufficient to meet the costs of necessary medical services.” §1396–1. Medicaid offers States “a bargain“: federal funds in exchange for compliance with congressionally imposed conditions. To participate in Medicaid, States must submit a “plan for medical assistance” satisfying over 80 conditions in §1396a(a). If a State fails “to comply substantially” with any condition, the Secretary of Health and Human Services may withhold federal funding. §1396c. This case involves the any-qualified-provider provision in §1396a(a)(23)(A), which requires States to ensure that “any individual eligible for medical assistance . . . may obtain” it “from any [provider] qualified to perform the service . . . who undertakes to provide” it. The provision does not define “qualified,” leaving that to States’ traditional authority over health and safety matters. The question is whether individual Medicaid beneficiaries may sue state officials under 42 U. S. C. §1983 for failing to comply with the any-qualified-provider provision.

Planned Parenthood South Atlantic operates two clinics in South Carolina, offering a wide range of services to Medicaid and non-Medicaid patients. It also performs abortions. Citing state law prohibiting public funds for abortion, South Carolina in July 2018 determined that Planned Parenthood could no longer participate in the State‘s Medicaid program. At the same time, the State took steps that, it said, would help ensure that other providers would continue offering necessary medical care and family planning services. Planned Parenthood and patient Julie Edwards sued, claiming the exclusion of Planned Parenthood violated the any-qualified-provider provision. Edwards alleged she preferred Planned Parenthood for gynecological care but needed Medicaid coverage. They brought a §1983 class action “to vindicate rights secured by the federal Medicaid statutes.”

Section 1983 allows private parties to sue state actors who violate their “rights” under the federal “Constitution and laws.” But federal statutes do not automatically confer §1983-enforceable “rights.” This is especially true of spending-power statutes like Medicaid, where “the typical remedy” for violations is federal funding termination, not private suits. Gonzaga Univ. v. Doe, 536 U. S. 273, 280.

The district court granted summary judgment for plaintiffs and enjoined the exclusion. The Fourth Circuit affirmed. This Court then granted certiorari, vacated, and remanded in light of Health and Hospital Corporation of Marion Cty. v. Talevski, 599 U. S. 166, which addressed whether another spending-power statute created §1983-enforceable rights. On remand, the Fourth Circuit reaffirmed.

Held: Section 1396a(a)(23)(A) does not clearly and unambiguously confer individual rights enforceable under §1983. Pp. 5–24.

(a) Congress sometimes allows private enforcement through §1983, which authorizes suits against state actors who deprive individuals of federal “rights, privileges, or immunities.” But statutes create individual rights only in “atypical case[s].” Talevski, 599 U. S., at 183. Section 1983 provides causes of action for deprivation of ” ‘rights,’ ” not mere ” ‘benefits’ or ‘interests.’ ” Gonzaga, 536 U. S., at 283.

To prove an enforceable right, plaintiffs must show the statute “clear[ly] and unambiguous[ly]” uses “rights-creating terms” with “an unmistakable focus” on individuals. Id., at 284, 290. This is a “strin-gent” and “demanding” test. Talevski, 599 U. S., at 180, 186. Even qualifying statutes may be unenforceable if Congress provided alternative remedies.

These rules vindicate separation of powers. Courts once assumed authority to provide whatever remedies seemed necessary for statutory purposes. But statutes do not pursue single purposes “at all costs,” American Express Co. v. Italian Colors Restaurant, 570 U. S. 228, 234, and Congress may not wish to authorize private suits, Hernandez v. Mesa, 589 U. S. 93, 100. Deciding whether to permit private enforcement poses delicate policy questions involving competing costs and benefits—decisions for elected representatives, not judges. Pp. 6–7.

(b) Spending-power statutes are especially unlikely to confer enforceable rights. Unlike Commerce Clause or other regulatory powers, Congress‘s spending authority rests on the “Taxing Clause” (Art. I, §8, cl. 1), which does not expressly authorize regulating conduct or issuing direct orders to States.

Early courts described federal grants as contracts, not commands. Federal-state agreements resemble treaties “between two sovereignties.” Neil, Moore & Co. v. Ohio, 3 How. 720, 742. Treaties may benefit citizens but generally do not confer individually enforceable rights against sovereigns, instead depending on the contracting governments for enforcement. Thus, “Congress alone has the power to enforce” grant conditions. Emigrant Co. v. County of Adams, 100 U. S. 61, 69. Pp. 8–10.

(c) In Pennhurst State School and Hospital v. Halderman, 451 U. S. 1, the Court established that spending-power legislation is “much in the nature of a contract.” Id., at 17. The “typical remedy for state noncompliance” is federal funding termination. Id., at 28. Private enforcement requires showing States “voluntarily and knowingly” consented to private suits, meaning Congress must “clearly” and “unambiguously” alert States that private enforcement was a funding condition. Id., at 17.

Gonzaga held that spending-power legislation cannot support §1983 suits unless Congress “speaks with a clear voice, and manifests an unambiguous intent to confer individual rights.” 536 U. S., at 280. Only “unmistakable” notice suffices. Id., at 286–287, and n. 5

Talevski reaffirmed that Gonzaga “sets forth [the] established method.” 599 U. S., at 183. Statutory provisions must “unambiguously confer individual federal rights“—a “demanding bar” cleared only in “atypical” cases. Id., at 180, 183–184. The statutes there qualified because they “expressly” used clear “rights-creating language.” Id., at 184, 186 (internal quotation marks omitted).

Earlier cases like Wilder v. Virginia Hospital Assn., 496 U. S. 498, Wright v. Roanoke Redevelopment and Housing Authority, 479 U. S. 418, and Blessing v. Freestone, 520 U. S. 329, suggested less demanding standards, but Gonzaga “reject[ed]” any approach permitting “anything short of an unambiguously conferred right.” 536 U. S., at 283. Lower courts should not rely on these repudiated precedents. Pp. 10–15.

(d) Section 1396a(a)(23)(A) lacks the required clear rights-creating language. Since Pennhurst, only three sets of spending-power statutes have been found to confer §1983 rights: those in Wright, Wilder, and Talevski. Given this Court‘s repudiation of Wright and Wilder‘s reasoning, Talevski provides the only reliable measure.

Talevski addressed Federal Nursing Home Reform Act provisions requiring facilities to “protect and promote” residents “right to be free from” restraints and provisions titled “[t]ransfer and discharge rights” in a subsection called “[r]equirements relating to residents’ rights.” §1396r(c) (emphasis added).

The any-qualified-provider provision looks nothing like these. Section 1396a(a)(23)(A) states that Medicaid plans must “provide that . . . any individual eligible for medical assistance . . . may obtain such assistance from any . . . qualified” provider. This language addresses state duties and may benefit providers and patients, but lacks FNHRA‘s clear “rights-creating language,” Talevski, 599 U. S., at 186 (internal quotation marks omitted).

Congress knows how to create clear rights, as FNHRA shows by giving nursing-home residents “the right to choose a personal attending physician.” §1396r(c)(1)(A)(i) (emphasis added). But that is not the law here.

The provision‘s exceptions confirm this reading. States may exclude providers “convicted of a felony” and “determin[e]” which convictions qualify. §1396a(a)(23)(B). This makes sense if the provision addresses state duties to the federal government, but creates problems if it also confers individual rights—Congress would grant rights in one breath while letting States control their scope in the next.

The statutory context supports this conclusion. The Medicaid Act requires only “substantia[l]” compliance, §1396c, suggesting focus on “‘aggregate‘” compliance with federal obligations rather than rights “‘of any particular person.‘” Gonzaga, 536 U. S., at 288. The provision appears as paragraph 23 of 87 plan requirements directed to the Secretary, without discernible organizational principle. If §1396a(a)(23)(A) created individual rights, many similar Medicaid provisions would too, making rights-creating provisions the rule rather than “atypical” exceptions. Pp. 15–19.

(e) Four counterarguments are offered. First, the claim that Congress modeled §1396a(a)(23)(A) on a Medicare provision titled ” ‘Free choice by patient guaranteed.’ ” 79 Stat. 291, 42 U. S. C. §1395a. But no court has addressed whether that Medicare provision creates §1983 rights. Moreover, while the Medicare provision “guarantee[s]” patient “free choice,” the Medicaid provision never uses “guarantee” or “free choice“—Congress omitted the very language claimed to create rights. Second, the appeal to legislative history suggesting Congress intended individual rights. But statutory interpretation focuses on what Congress enacted, not speculated intentions. For spending-power statutes, “the key is not what a majority of the Members . . . intend but what the States are clearly told.” Arlington Central School Dist. Bd. of Ed. v. Murphy, 548 U. S. 291, 304. Third, the proposal to remodel the established test by arguing that “individual-centric, mandatory language” is necessarily “rights-creating” without requiring the “ex-plicit rights-creating terms” this Court has long required. This standard lacks foundation in precedent and obliterates the distinction between mere benefits and enforceable rights. It would make rights-creating provisions the rule rather than “atypical” exceptions and leave States guessing about their obligations. Fourth, the policy argument that only §1983 litigation can effectively enforce the provision, claiming the federal government lacks capacity or appetite for funding cutoffs. This Court has rejected the notion that funding cutoffs are “too massive” to be realistic relief. Armstrong v. Exceptional Child Center, Inc., 575 U. S. 320, 331. Alternative enforcement exists—States have administrative processes for provider challenges, reviewable by state courts. If existing remedies prove insufficient, Congress can create new ones. But balancing enforcement costs and benefits is a policy question for Congress, not courts. Pp. 19–24.

95 F. 4th 152, reversed and remanded.

GORSUCH, J., delivered the opinion of the Court, in which ROBERTS, C. J., and THOMAS, ALITO, KAVANAUGH, and BARRETT, JJ., joined. THOMAS, J., filed a concurring opinion. JACKSON, J., filed a dissenting opinion, in which SOTOMAYOR and KAGAN, JJ., joined.

JUSTICE GORSUCH delivered the opinion of the Court.

Medicaid offers States “a bargain.” Armstrong v. Exceptional Child Center, Inc., 575 U. S. 320, 323 (2015). In return for federal funds, States agree “to spend them in accordance with congressionally imposed conditions.” Ibid. Should a State fail to comply substantially with those conditions, the Secretary of Health and Human Services can withhold some or all of its federal Medicaid funding. This case poses the question whether, in addition to that remedy, individual Medicaid beneficiaries may sue state officials for failing to comply with one funding condition spelled out in 42 U. S. C. §1396a(a)(23)(A).

I

Congress created Medicaid in 1965 to subsidize state efforts to provide healthcare to families and individuals “‘whose income and resources are insufficient to meet the costs of necessary medical services.‘” Armstrong, 575 U. S., at 323 (quoting §1396–1). Today, all 50 States participate in Medicaid. Congressional Research Service, Medicaid: An Overview 1 (2025) (CRS). In order to do so, a State must submit to the Secretary a “plan for medical assistance.” §1396a(a); see also §1396–1. To win the Secretary‘s approval, that plan must satisfy more than 80 separate conditions Congress has set out in §1396a(a). Once the Secretary approves a plan, federal funds begin flowing to help the State execute it. Of course, States must contribute their own money, too. See §1396d(b). Historically, the federal government has provided on average about 57% of the funds required to implement Medicaid, and States have supplied the balance. CRS 21.

This case concerns one of the conditions state plans must meet. Located in §1396a(a)(23)(A), Medicaid‘s any-qualified-provider provision, as it is sometimes called, requires States to ensure that “any individual eligible for medical assistance . . . may obtain” it “from any [provider] qualified to perform the service . . . who undertakes to provide” it. The provision does not define the term “qualified,” perhaps because States have traditionally exercised primary responsibility over “matters of health and safety,” including the regulation of the practice of medicine. De Buono v. NYSA–ILA Medical and Clinical Services Fund, 520 U. S. 806, 814 (1997); see also Linder v. United States, 268 U. S. 5, 18 (1925); 42 CFR §431.51(c)(2) (2024). But everyone acknowledges that, if a State fails “to comply substantially” with this (or any) congressionally specified condition, the Secretary may withhold some or all of the State‘s federal funding until he is “satisfied that there will no longer be any such failure to comply.” §1396c.

The parties’ dispute concerns whether, in addition to that remedy, the law recognizes another. The dispute arose this way. Planned Parenthood South Atlantic operates two clinics in South Carolina, one in each of the State‘s two most populous cities. Planned Parenthood South Atlantic v. Kerr, 95 F. 4th 152, 156–157 (CA4 2024). At both locations, the group offers “a wide range” of services to Medicaid and non-Medicaid patients. Ibid. It also performs abortions. Ibid. Citing a state law prohibiting the use of its own public funds for abortion, South Carolina announced in July 2018 that Planned Parenthood could no longer participate in the State‘s Medicaid program. App. to Pet. for Cert. 157a–162a. At the same time, the State took steps that, it said, would help ensure that a “variety of other nongovernmental entities and governmental agencies” would continue to provide “access to necessary medical care and important women‘s health and family planning services.” Id., at 158a. According to the State, it has “140 [other] federally qualified health clinics and pregnancy centers, not counting the numerous private health providers who accept Medicaid.” Brief for Petitioner 9.

In response to the State‘s announcement, Planned Parenthood and one of its patients, Julie Edwards, sued the director of the State‘s Department of Health and Human Services. They argued that South Carolina‘s exclusion of Planned Parenthood from its Medicaid program violated the any-qualified-provider provision. Specifically, Ms. Edwards alleged that, while she regularly visits other medical care providers, she has had especially positive experiences with Planned Parenthood and would like “to shift all [her] gynecological and reproductive health care there.” App. 32, 33. But none of that will be possible, she continued, unless Medicaid covers those services. Ibid. Based on these allegations, Ms. Edwards and Planned Parenthood brought a putative class action “pursuant to 42 U. S. C. §1983 to vindicate rights secured by the federal Medicaid statutes.” Id., at 1.

First enacted as part of the Civil Rights Act of 1871, §1983 allows private parties to sue state actors who violate their “rights” under “the Constitution and laws” of the United States. But federal statutes do not confer “rights” enforceable under §1983 “as a matter of course.” Health and Hospital Corporation of Marion Cty. v. Talevski, 599 U. S. 166, 183 (2023). That is particularly true of statutes, like Medicaid, enacted pursuant to Congress‘s spending power. The spending power allows Congress to offer funds to States that agree to certain conditions. See, e.g., South Dakota v. Dole, 483 U. S. 203, 207–208 (1987). But when a State violates those conditions, “‘the typical remedy‘” is not a private enforcement suit “‘but rather action by the Federal Government to terminate funds to the State.‘” Gonzaga Univ. v. Doe, 536 U. S. 273, 280 (2002) (quoting Pennhurst State School and Hospital v. Halderman, 451 U. S. 1, 28 (1981)).

Appreciating all this, the plaintiffs argued that their case implicated an exception to the usual rule. The any-qualified-provider provision, they said, is among those rare federal spending-power statutes that confer individual rights enforceable under §1983. And, they submitted, South Carolina violated Ms. Edwards‘s rights under that provision when it denied her the opportunity to select Planned Parenthood as her healthcare provider. Agreeing with the plaintiffs’ assessment, the district court granted summary judgment to them and entered a permanent injunction preventing the State from excluding Planned Parenthood from its Medicaid program. Planned Parenthood South Atlantic v. Baker, 487 F. Supp. 3d 443, 448 (SC 2020).

On appeal, the Fourth Circuit affirmed the district court‘s decision. Planned Parenthood South Atlantic v. Kerr, 27 F. 4th 945 (2022). Writing separately, Judge Richardson expressed “confusion and uncertainty” about this Court‘s directions addressing when spending-power legislation creates enforceable rights under §1983. Id., at 959 (opinion concurring in judgment). And he voiced “hop[e]” that we might provide “clarity . . . soon.” Ibid.

Seeking review of the Fourth Circuit‘s decision, the State filed a petition for certiorari in this Court. In light of our intervening decision in Talevski, 599 U. S. 166, in which we addressed whether another spending-power statute created §1983-enforceable rights, we granted the State‘s petition, vacated the decision of the court of appeals, and remanded the case for further proceedings. Kerr v. Planned Parenthood South Atlantic, 599 U. S. ___ (2023).

On remand, the court of appeals reaffirmed its earlier decision. 95 F. 4th, at 153. And, once more, Judge Richardson wrote separately. Even after Talevski, he said, lower courts “continue[d] to lack the guidance” they need from this Court to determine when a federal spending-power statute creates a right that private parties can enforce under §1983. 95 F. 4th, at 170 (opinion concurring in judgment). Other circuit judges have expressed similar concerns. See, e.g., Saint Anthony Hospital v. Whitehorn, 132 F. 4th 962, 971 (CA7 2025) (en banc); id., at 982 (Hamilton, J., dissenting); New York State Citizens’ Coalition for Children v. Poole, 935 F. 3d 56, 60 (CA2 2019) (Livingston, J., dissenting from denial of rehearing en banc).

In response to the Fourth Circuit‘s latest decision, the State filed another petition for certiorari. In it, South Carolina noted that other lower courts have disagreed with the Fourth Circuit regarding whether §1396a(a)(23)(A) confers an individually enforceable right. Cf. Planned Parenthood of Greater Tex. Family Planning & Preventative Health Servs., Inc. v. Kauffman, 981 F. 3d 347, 350 (CA5 2020) (en banc); Does v. Gillespie, 867 F. 3d 1034, 1037 (CA8 2017). We agreed to hear the case. 604 U. S. ___ (2024).

II

To resolve the circuits’ disagreement and address our lower court colleagues’ calls for clarification, we begin by outlining how to determine whether a statute confers an individually enforceable right under §1983.

A

The Constitution charges the Executive Branch with enforcing federal law. Art. II, §3. But sometimes Congress also allows private parties to enforce the law through civil litigation. In §1983, Congress did just that, authorizing individuals to sue anyone who, under color of state law, deprives them of “rights, privileges, or immunities secured by the Constitution and laws” of the United States.

Historically, individuals brought §1983 suits to vindicate rights protected by the Constitution. But, in 1980, this Court recognized that §1983 also authorizes private parties to pursue violations of their federal statutory rights. Maine v. Thiboutot, 448 U. S. 1. Still, this Court has emphasized, statutes create individual rights only in “atypical case[s].” Talevski, 599 U. S., at 183. Routinely, of course, federal legislation seeks to benefit one group or another. (Why pass legislation otherwise?) But §1983 provides a cause of action “only for the deprivation of ‘rights, privileges, or immunities,‘” not “‘benefits’ or ‘interests.‘” Gonzaga, 536 U. S., at 283.

To prove that a statute secures an enforceable right, privilege, or immunity, and does not just provide a benefit or protect an interest, a plaintiff must show that the law in question “clear[ly] and unambiguous[ly]” uses “rights-creating terms.” Id., at 284, 290. In addition, the statute must display “‘an unmistakable focus‘” on individuals like the plaintiff. Id., at 284 (emphasis deleted); accord, Talevski, 599 U. S., at 183. We have described this as a “strin-gent” and “demanding” test. Id., at 180, 186; accord, post, at 9 (JACKSON, J., dissenting) (describing Gonzaga as setting forth “a restrictive test“). And even for the rare statute that satisfies it, this Court has said, a §1983 action still may not be available if Congress has displaced §1983‘s general cause of action with a more specific remedy. Rancho Palos Verdes v. Abrams, 544 U. S. 113, 120 (2005).

These rules seek to “vindicat[e] the separation of powers.” Talevski, 599 U. S., at 183. To be sure, there was a time in the mid-20th century when “the Court assumed it to be a proper judicial function to provide” whatever “remedies” it deemed “necessary to make effective a statute‘s purpose.” Ziglar v. Abbasi, 582 U. S. 120, 131–132 (2017) (internal quotation marks omitted). But, as this Court has since come to appreciate, no statute pursues any single “pur-pos[e] at all costs.” American Express Co. v. Italian Colors Restaurant, 570 U. S. 228, 234 (2013) (internal quotation marks omitted). And, often enough, Congress may “not wish to pursue [a] provision‘s purpose to the extent of authorizing private suits.” Hernandez v. Mesa, 589 U. S. 93, 100 (2020). After all, the decision whether to let private plaintiffs enforce a new statutory right poses delicate questions of public policy. New rights for some mean new duties for others. And private enforcement actions, meritorious or not, can force governments to direct money away from public services and spend it instead on litigation. See ibid. The job of resolving how best to weigh those competing costs and benefits belongs to the people‘s elected representatives, not unelected judges charged with applying the law as they find it. See Alexander v. Sandoval, 532 U. S. 275, 286 (2001); Gonzaga, 536 U. S., at 285.¹

B

Though it is rare enough for any statute to confer an enforceable right, spending-power statutes like Medicaid are especially unlikely to do so. The reasons why take a little unpacking.

When Congress passes a law, say, regulating commerce between the States or outlawing piracy, it can point for authority to the Commerce Clause, U. S. Const., Art. I, §8, cl. 3, or the Piracies Clause, cl. 10. In enumerated areas like those, the Constitution vests Congress with the power to regulate conduct. But when Congress distributes money, its authority rests on a different footing.

The Constitution has no “Spending Clause,” strictly speaking. Instead, we usually trace Congress‘s spending power to Article I, section eight, clause one, which gives Congress the “Power To lay and collect Taxes, Duties, Imposts and Excises, to pay the Debts and provide for the common Defence and general Welfare of the United States.” Unlike other enumerated powers, this provision does not expressly endow Congress with the power to regulate conduct. Nor does it include “the power to issue direct orders to the governments of the States.” Murphy v. National Collegiate Athletic Assn., 584 U. S. 453, 471 (2018).

As the Court observed in United States v. Butler, the meaning of Article I‘s “general welfare” language provoked fierce debate right from the start. 297 U. S. 1, 65–67 (1936). At one extreme, Gouverneur Morris thought it authorized Congress to tax, spend, and regulate broadly in pursuit of the “general Welfare.” D. Schwartz, Mr. Madison‘s War on the General Welfare Clause, 56 U. C. D. L. Rev. 887, 915 (2022). Alexander Hamilton took a more modest view. He thought the language gave Congress the power to raise and “appropriate money” for “objects” of “General” (as opposed to “local“) importance. Report on the Subject of Manufactures (Dec. 5, 1791), in 10 Papers of Alexander Hamilton 230, 303–304 (H. Syrett ed. 1966) (emphasis deleted). But he denied that those powers included as well “a power to do whatever else should appear to Congress conducive to the General Welfare.” Ibid. James Madison advanced a narrower position still. As he saw it, the language authorized Congress to spend money only in support of its other enumerated powers. A. LaCroix, The Interbellum Constitution: Federalism in the Long Founding Moment, 67 Stan. L. Rev. 397, 407 (2015) (LaCroix).

Over time, Hamilton‘s view gained ground. So, for example, as Justice Story saw it, Congress may raise and “appropriat[e] . . . money” to advance the “general welfare.” 3 J. Story, Commentaries on the Constitution of the United States §1269, p. 150 (1833). But nothing in Article I, section eight, clause one endows Congress with a power to regulate, for if it did, the “enumeration of specific powers” elsewhere in Article I would be rendered largely pointless, and the Nation would trade a limited federal government for “an unlimited” one. 2 id., §§904, 906, pp. 367, 369; see also Butler, 297 U. S., at 66 (Justice Story‘s “reading . . . is the correct one“); J. Monroe, Message From the President of the United States 32–33 (1822); E. Corwin, The Spending Power of Congress—Apropos the Maternity Act, 36 Harv. L. Rev. 548, 564–566 (1923).

Consistent with this understanding, early courts described federal grants not as commands but as contracts. Consider, for example, how this Court approached a dispute concerning the first major federal highway. The Cumberland Road once supplied a vital link between the East Coast and the old Northwest. LaCroix 420. Starting in the 1830s, the federal government gradually transferred control of the road to several States. J. Young, A Political and Constitutional Study of the Cumberland Road 78–98 (1902). One transfer to Ohio came with a condition: The State could not charge tolls on wagons carrying federal property. Id., at 96–98. When a disagreement arose about the scope of that toll exemption, this Court looked to “the expectations of the parties,” a familiar feature of contract law, to resolve it. Neil, Moore & Co. v. Ohio, 3 How. 720, 741 (1845). In doing so, the Court emphasized that it was enforcing requirements “well known” to the parties when the “compact was made.” Ibid.; see also McGee v. Mathis, 4 Wall. 143, 155 (1866) (“It is not doubted that the grant by the United States to the State upon conditions, and the acceptance of the grant by the State, constituted a contract“).

At the same time, the Court recognized that agreements between state and federal governments are not exactly the same as contracts “between individuals.” Searight v. Stokes, 3 How. 151, 167 (1845). In many respects, the Court suggested, federal-state agreements are really more like treaties “between two sovereignties.” See Neil, Moore & Co., 3 How., at 742. And, while treaties may seek to benefit the citizens of the compacting nations, they generally do not confer individually enforceable rights against a sovereign, but “depen[d] for the enforcement of [their] provisions on . . . the governments which are parties to” them. Head Money Cases, 112 U. S. 580, 598 (1884).² Adapting this logic to the context of federal grants, the Court concluded that, as a rule, “Congress alone has the power to enforce” the conditions it attaches to its grants. Emigrant Co. v. County of Adams, 100 U. S. 61, 69 (1879); see also Mills County v. Railroad Cos., 107 U. S. 557, 566 (1883).

C

For much of the Nation‘s history, the Court had little occasion to employ these ideas. Congress rarely granted money to States and, when it did, those grants rarely came with many conditions. See D. Currie, The Constitution in Congress: Democrats and Whigs, 1829–1861, pp. 42–45 (2005). But that began to change during the New Deal. And when disputes about those grant conditions arose, this Court returned to the old contract and treaty analogies to ensure that spending-power legislation did not pass the “point at which pressure turns into compulsion.” Steward Machine Co. v. Davis, 301 U. S. 548, 590 (1937); see also Massachusetts v. Mellon, 262 U. S. 447, 480 (1923); Butler, 297 U. S., at 73–75; Oklahoma v. Civil Serv. Comm‘n, 330 U. S. 127, 143–144 (1947). The same analogies guided the Court, too, after federal grants exploded in the 1960s, generating “an unprecedented” wave of litigation in which private parties sought to challenge state compliance with federal grant conditions. E. Tomlinson & J. Mashaw, The Enforcement of Federal Standards in Grant-In-Aid Programs: Suggestions for Beneficiary Involvement, 58 Va. L. Rev. 600, 630 (1972).³

Take Pennhurst. There, private plaintiffs sought to sue the Commonwealth of Pennsylvania for failing to fulfill the terms of a federal healthcare grant. 451 U. S., at 6. In assessing whether the suit could proceed, the Court began by observing that “legislation enacted pursuant to the spending power is much in the nature of a contract: in return for federal funds, the States agree to comply with federally imposed conditions.” Id., at 17. And the “typical remedy for state noncompliance” with a federal grant‘s conditions is an “action by the Federal Government to terminate funds to the State.” Id., at 28. Given these principles, the Court reasoned, whether a private party may sue to enforce the terms of a federal grant depends on “whether the State voluntarily and knowingly” consented to answer private claims as part of its bargain with the federal government. Id., at 17. And to satisfy this standard, the Court held, a plaintiff must show, at a minimum, that Congress alerted the State in advance, “clear[ly]” and “unambiguously,” that responding to private enforcement suits was a condition of its offer. Ibid.⁴

In Gonzaga, the Court restated these principles and explored how they interact with §1983. Spending-power legislation, the Court explained, cannot provide the basis for a §1983 enforcement suit unless Congress “speaks with a clear voice, and manifests an unambiguous intent to confer individual rights.” 536 U. S., at 280 (alteration and internal quotation marks omitted). Only that kind of “unmistakable” notice, the Court said, suffices to alert grantees that they might be subject “to private suits . . . whenever they fail to comply with a federal funding condition.” Id., at 286–287, and n. 5 (internal quotation marks omitted). And, the Court concluded, because the statute at issue before it did not clearly and unambiguously confer a “right to support a cause of action under §1983,” the plaintiff‘s suit could not proceed. Id., at 283, 290.⁵

Just two Terms ago, we reaffirmed these points. In Talevski, the Court faced another private §1983 suit alleging that recipients of federal funding had violated grant conditions. To decide whether the plaintiffs could proceed, we turned to Gonzaga, recognizing that it “sets forth our established method” for analyzing suits like that. Talevski, 599 U. S., at 183. In doing so, we reiterated that the relevant “[s]tatutory provisions must unambiguously confer individual federal rights” before a §1983 claim might proceed. Id., at 180. That standard, we emphasized, is a “demanding bar” and a “significant hurdle” that will be cleared only in the “atypical case.” Id., at 180, 183–184. And, applying that test, we found the statutes in question satisfied it precisely because they “expressly” employed the sort of clear and unambiguous “rights-creating language” Gonzaga demands. 599 U. S., at 184, 186 (internal quotation marks omitted).

Admittedly, this Court briefly experimented with a different approach, and that fact has given rise to some confusion in the lower courts. For a time, as we have seen, the Court sometimes took an expansive view of its power to imply private causes of action to enforce federal laws. See Part II–A, supra. Moved by the same spirit, the Court sometimes took a broad view of its authority to confer new rights under spending-power statutes that did not expressly provide them. In Wilder v. Virginia Hospital Association, for example, the Court suggested that spending-power legislation can give rise to an enforceable right under §1983 so long as the legislation is “intended to benefit the putative plaintiff” and the plaintiff‘s interest in the statute is not “too vague and amorphous.” 496 U. S. 498, 509 (1990) (alteration and internal quotation marks omitted); see Wright v. Roanoke Redevelopment and Housing Authority, 479 U. S. 418, 423–424, 431–432 (1987). Building on those same ideas in Blessing v. Freestone, 520 U. S. 329, 340–341 (1997). Some lower court judges, including in this case, still consult Wilder, Wright, and Blessing when asking whether a spending-power statute creates an enforceable individual right. See, e.g., 95 F. 4th, at 163–165; id., at 170 (Richardson, J., concurring in judgment).

They should not. Gonzaga “reject[ed]” any reading of our prior cases that would “permit anything short of an unambiguously conferred right to support a cause of action brought under §1983.” 536 U. S., at 283. Armstrong “repu-diate[d]” any other approach. 575 U. S., at 330, n. And Talevski reaffirmed that ”Gonzaga sets forth our established method” for determining whether a spending-power statute confers individual rights. 599 U. S., at 183.

All of these warnings came for now-familiar reasons. Because spending-power legislation is “in the nature of a contract,” a grantee must “voluntarily and knowingly” consent to answer private §1983 enforcement suits before they may proceed. Pennhurst, 451 U. S., at 17; see id., at 28. And that consent cannot be fairly inferred if the federal spending-power statute fails to provide “clear and unambig-uous” notice that it creates a personally enforceable right. Gonzaga, 536 U. S., at 290. To the extent lower courts feel obliged, or permitted, to consider the contrary reasoning of Wilder, Wright, or Blessing, they should resist the impulse.

III

With these principles in hand, we turn to the question whether the plaintiffs before us may maintain a §1983 suit to enforce Medicaid‘s any-qualified-provider provision. To succeed, they must show, at a minimum, that §1396a(a)(23)(A) does not just seek to benefit them or serve their interests but “clear[ly] and unambiguous[ly]” gives them individual federal rights. Gonzaga, 536 U. S., at 290.⁶

Since Pennhurst, this Court has identified only three sets of spending-power statutes that confer enforceable rights under §1983—those at issue in Wright, Wilder, and Talev-ski. But given this Court‘s longstanding repudiation of Wright and Wilder‘s reasoning, the statutes at issue in Talevski supply the only reliable yardstick against which to measure whether spending-power legislation confers a privately enforceable right.

Talevski addressed two provisions of the Federal Nursing Home Reform Act (FNHRA). See 599 U. S., at 181–182. The first obliged nursing-home facilities to “protect and promote” residents’ “right to be free from” unnecessary “physical or chemical restraints.” 42 U. S. C. §1396r(c)(1)(A)(ii) (emphasis added). The second appeared in a subparagraph titled “[t]ransfer and discharge rights.” §1396r(c)(2)(A) (emphasis added). And both provisions sat in a subsection called “[r]equirements relating to residents’ rights.” §1396r(c) (emphasis added).

The any-qualified-provider provision before us looks

nothing like those FNHRA provisions. Section 1396a(a)(23)(A) indicates that state Medicaid plans must “provide that . . . any individual eligible for medical assistance (including drugs) may obtain such assistance from any institution, agency, community pharmacy, or person, qualified to perform the service or services required . . . who undertakes to provide him such services.” Doubtless, this language speaks to what a State must do to participate in Medicaid, and a State that fails to fulfill its duty might lose federal funding. Doubtless, too, this provision seeks to benefit both providers and patients. But missing from §1396a(a)(23)(A) is anything like FNHRA‘s clear and unambiguous “rights-creating language.” Talevski, 599 U. S., at 186 (internal quotation marks omitted).

To be sure, Congress could have taken a different approach when drafting §1396a(a)(23)(A). In fact, FNHRA offers an example almost perfectly on point. One of its provisions gives nursing-home residents the right to choose their own attending physicians. Here is the provision in context:

“(c) Requirements relating to residents’ rights

“(1) General rights

“(A) Specified rights

“A nursing facility must protect and promote the rights of each resident, including each of the following rights:

“(i) Free choice

“The right to choose a personal attending physician . . . .” §1396r(c) (emphasis added).

As this language shows, Congress knows how to give a grantee clear and unambiguous notice that, if it accepts federal funds, it may face private suits asserting an individual right to choose a medical provider. Tellingly, too, Congress adopted this FNHRA provision in legislation that also amended §1396a(a)(23). Yet Congress‘s work in the two provisions could not have been more different. See 101 Stat. 1330-152; Talevski, 599 U. S., at 181, n. 10. Someday, Congress might choose to revise §1396a(a)(23) to resemble FNHRA. But that is not the law we have. Cf. Feliciano v. Department of Transportation, 605 U. S. ___ (2025) (slip op., at 6).

The remainder of §1396a(a)(23) only serves to confirm our conclusion. After announcing that state Medicaid plans must allow individuals to obtain care from any qualified provider, the provision proceeds to carve out various exceptions to that rule. So, for example, the statute allows States to exclude from their Medicaid programs certain providers “convicted of a felony“—and, what is more, to “determin[e]” which felony convictions qualify for that exclusion. §1396a(a)(23)(B). All that makes perfect sense if §1396a(a)(23)(A) speaks only to a State‘s duties to the federal government. But it is an arrangement a good deal harder to understand if §1396a(a)(23)(A) also confers an individually enforceable right, for that would mean Congress sought to convey a right against the States in one breath but let States control its scope in the next.

Expanding our view beyond §1396a(a)(23) to the surrounding statutory context yields similar clues. To continue receiving federal funding, the Medicaid Act says, a State need only “comply substantially” with the any-qualified-provider mandate. §1396c. And, as this Court recognized in Gonzaga, that focus on “aggregate” compliance suggests that a statute addresses a State‘s obligations to the federal government, not the rights “‘of any particular person.‘” 536 U. S., at 288. Sometimes, we appreciate, a provision may overcome this weighty statutory evidence. In Talevski, after all, the Court found two FNHRA provisions to confer individual rights even though that statute also speaks of “substantial compliance.” See Brief for Respondents 35-36. But, at risk of repetition, the provisions at issue there employed explicit and unmistakable “‘rights- creating language,‘” 599 U. S., at 186, and §1396a(a)(23)(A) does not.

Notable, too, is where Congress placed the any-qualified-provider provision. It appears in a subsection titled “Contents.” §1396a(a). That subsection outlines scores of things a state plan must include to qualify for federal funding. Ibid. Those requirements do not appear in any discernible order, and the any-qualified-provider provision does not crop up until paragraph 23 of 87. All of §1396a(a)‘s requirements are directed to the Secretary of Health and Human Services, who must “approve any plan” that meets them. §1396a(b); see Armstrong, 575 U. S., at 331-332 (plurality opinion). None of this may suffice to prove that the any-qualified-provider provision is unenforceable under §1983. See §1320a-2. But it does show, once more, that the statute before us stands in stark contrast to the ones we faced in Talevski, where Congress set its rights-creating provisions apart from others and, in doing so, helped alert grantees that accepting federal funds comes with a duty to answer private suits.

Observe, as well, what it would mean if §1396a(a)(23)(A) did create an individually enforceable right. Many other Medicaid plan requirements would likely do the same. And instead of remaining “atypical” exceptions, as our cases have said they are, rights-creating provisions might more nearly become the rule. Talevski, 599 U. S., at 183.

Take one example. See Brief for United States as Amicus Curiae 27-29 (offering others). Section 1396a(a)(32) follows several paragraphs down from the any-qualified-provider provision. It requires state Medicaid plans to “provide,” with certain exceptions, “that no payment under the plan for any care or service provided to an individual shall be made to anyone other than such individual or the person or institution providing such care or service.” As the plaintiffs acknowledge, this provision “uses language with some similarities to” §1396a(a)(23)(A). Brief for Respondents 38-39. Both speak in mandatory terms (“must provide“; “shall“). Both discuss “individual[s].” Neither mentions “rights.” Yet, while the plaintiffs insist that paragraph (23)(A) clearly and unambiguously creates an individual right, they suggest that a court could reasonably “determine” that paragraph (32) “does not.” Ibid. (citing Polk v. Yee, 36 F. 4th 939, 945-946 (CA9 2022)). Rather than try to square that circle, we think the better course is the one our precedents suggest: Neither paragraph uses clear and unambiguous rights-creating language, so neither supports a private suit under §1983.

IV

Seeking to persuade us otherwise, the plaintiffs and dissent offer four principal counterarguments.

First, the plaintiffs and dissent appeal to legislative history. The hearings and committee reports leading to §1396a(a)(23)(A)‘s adoption, they say, reveal that Congress meant for the statute to secure an individual right. See Brief for Respondents 30-32; see also post, at 3, 13 (JACKSON, J., dissenting). But that does not move the needle. When it comes to interpreting the law, speculation about what Congress may have intended matters far less than what Congress actually enacted. See Epic Systems Corp. v. Lewis, 584 U. S. 497, 523 (2018) (“[L]egislative history is not the law“). And that goes double for spending-power statutes, where “the key is not what a majority of the Members of both Houses intend but what the States are clearly told.” Arlington Central School Dist. Bd. of Ed. v. Murphy, 548 U. S. 291, 304 (2006).⁷

Second, the plaintiffs and dissent contend, Congress modeled §1396a(a)(23)(A) on a Medicare provision titled “Free choice by patient guaranteed.” 79 Stat. 291, 42 U. S. C. § 1395a. It reads: “Any individual entitled to insurance benefits under this subchapter may obtain health services from any person qualified to participate under this subchapter if such person undertakes to provide him such services.” §1395a(a). And because that Medicare provision “confer[s] an individual right,” the plaintiffs and dissent reason, its Medicaid offshoot must as well. Brief for Respondents 34; see post, at 13 (JACKSON, J., dissenting).

This argument stumbles out of the gate. Its premise—that §1395a(a) confers an enforceable right—is questionable. As the plaintiffs admit, “[n]o court has addressed whether a Medicare beneficiary can enforce this provision under Section 1983.” Brief for Respondents 34, n. 7. Even overlooking that deficiency, another quickly emerges. While the title of §1395a(a) “guarantee[s]” a patient‘s “free choice” of provider—and while the plaintiffs and dissent insist this language can create a right—the any-qualified-provider provision never uses “guarantee” or its equivalent. So if the comparison between the Medicaid and Medicare provisions reveals anything, it is that Congress did not include in §1396a(a)(23)(A) the language from §1395a that the plaintiffs and dissent think most likely to confer enforceable rights.

Third, instead of grappling meaningfully with the test our precedents provide, the dissent proposes to rewrite it. In the dissent‘s view, a statute confers a privately enforceable right whenever it uses “compulsory” and “individual-centric terminology,” as long as it also evokes “language classically associated with establishing rights.” Post, at 12 (opinion of JACKSON, J.). When it comes to that last requirement, the dissent reasons this way: Congress enacted §1396a(a)(23)(A) under the title “free choice by individuals eligible for medical assistance,” 81 Stat. 903 (capitalization omitted); the phrase “free choice” calls to the dissent‘s mind a phrase from the First Amendment (“free exercise” of religion); that Amendment declares rights; so §1396a(a)(23)(A) likely must as well. Post, at 12 (opinion of JACKSON, J.).

Our precedents do not authorize anything like the dissent‘s approach—and for good reasons. To start, while a title may underscore that the statutory text creates a right, “[i]t has long been established that the title of an Act cannot enlarge or confer powers” by itself. Pennhurst, 451 U. S., at 19, n. 14 (internal quotation marks omitted); see A. Scalia & B. Garner, Reading Law: The Interpretation of Legal Texts 221-224 (2012). That must be especially so where, as here, Congress chose not to enact into the U. S. Code the very title on which the dissent relies. See 81 Stat. 903 (enacting the title of a different section, but not “free choice by individuals eligible for medical assistance,” into the U. S. Code (capitalization omitted)).

Even beyond that, the dissent‘s test would risk obliterating the longstanding line between mere benefits and enforceable rights. See supra, at 6, 13, 15. If, as the dissent says, §1396a(a)(23)(A) creates an enforceable right because it contains “compulsory” and “individual-centric terminology” plus an iffy analogy to the Bill of Rights, then many other provisions (in Medicaid and elsewhere) previously thought to confer only benefits would suddenly create rights instead. Supra, at 18-19. All despite Talevski‘s insight just two Terms ago that, while many statutes supply benefits, only “atypical” statutes confer enforceable rights under §1983. 599 U. S., at 183. To be sure, the dissent assures us that other Medicaid provisions are distinguishable from this one. Post, at 19 (opinion of JACKSON, J.). How? Not based on their text (which the dissent never addresses) but, it seems, based on an unspoken judicial intuition that the provision before us is just more important than others. So, on top of all its other flaws, the dissent‘s approach would leave States guessing about the terms of their deals with the federal government and invite courts to revive their long-abandoned approach of usurping Congress‘s role in creating rights and remedies. Supra, at 7.⁸

Fourth and finally, the plaintiffs and dissent advance a policy argument. Only §1983 litigation, they submit, can give the any-qualified-provider provision the teeth it needs. Yes, they acknowledge, the federal government can audit States’ compliance with §1396a(a)(23)(A) and withhold some or all Medicaid funds from noncompliant States. Brief for Respondents 44. But, the plaintiffs and dissent insist, the federal government has neither the capacity nor the appetite for taking that “drastic step.” Ibid.; see Tr. of Oral Arg. 110; see also post, at 2-3 (JACKSON, J., dissenting).

This argument suffers from a number of problems. For one, this Court has specifically rejected the notion that “the cut-off of funding” is “too massive” a remedy “to be a realistic source of relief” for violations of §1396a(a) provisions. Armstrong, 575 U. S., at 331. To the contrary, this Court has called funding cutoffs “the typical remedy” when a grant recipient violates the terms of spending-power legislation. Pennhurst, 451 U. S., at 17.

For another, funding cutoffs may not be the only way to enforce §1396a(a)(23)(A). Like other States, South Carolina has an administrative process that lets providers challenge their exclusion from the State‘s Medicaid program. See Brief for United States as Amicus Curiae 30; Gillespie, 867 F. 3d, at 1038. That process can culminate with state judicial review—and, if necessary, with a petition for certiorari to this Court. See S. C. Code Ann. §1-23-380 (Cum. Supp. 2024). Indeed, Planned Parenthood itself pursued just such an administrative claim at one point. See App. 61-63.

For another thing still, if existing remedies prove insufficient, Congress can create new ones. So, for example, it might do as it did in FNHRA and revise §1396a(a)(23)(A) to provide States with clear and unambiguous notice of an individually enforceable right. Of course, as we have observed, a decision like that comes with tradeoffs. At their best, individual suits under §1983 can vindicate plaintiffs’ rights while pushing States to fulfill their obligations. But private enforcement does not always benefit the public, not least because it requires States to divert money and attention away from social services and toward litigation. And balancing those costs and benefits poses a question of public policy that, under our system of government, only Congress may answer. See Sandoval, 532 U. S., at 286; Gonzaga, 536 U. S., at 285-286.⁹

*

Section 1983 permits private plaintiffs to sue for violations of federal spending-power statutes only in “atypical” situations, Talevski, 599 U. S., at 183, where the provision in question “clear[ly]” and “unambiguous[ly]” confers an individual “right,” Gonzaga, 536 U. S., at 290. Section 1396a(a)(23)(A) is not such a statute. Because the Fourth Circuit concluded otherwise, its judgment is reversed and the case is remanded for further proceedings consistent with this opinion.

It is so ordered.

THOMAS, J., concurring

SUPREME COURT OF THE UNITED STATES

No. 23-1275

EUNICE MEDINA, DIRECTOR, SOUTH CAROLINA DEPARTMENT OF HEALTH AND HUMAN SERVICES, PETITIONER v. PLANNED PARENTHOOD SOUTH ATLANTIC, ET AL.

ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

[June 26, 2025]

JUSTICE THOMAS, concurring.

Individual plaintiffs may invoke Rev. Stat. §1979, 42 U. S. C. §1983, to sue state or local officials who have deprived them of “any rights, privileges, or immunities secured by the Constitution and laws.” In other words, §1983 provides a mechanism for plaintiffs to enforce constitutional or statutory provisions that confer personally held federal rights. The Court correctly holds today that §1396a(a)(23)(A) of the Medicaid Act is not such a provision. Its decision properly limits plaintiffs’ ability to bring §1983 suits premised on conditional spending legislation, and I join in full. I write separately because it behooves us to reexamine more broadly this Court‘s §1983 jurisprudence, which bears little resemblance to the statute as originally understood. In appropriate cases, we should reassess §1983‘s bounds, including its application in the spending context and our understanding of the “rights” enforceable under §1983.

I

The history of §1983 makes clear that the statute has ex- ceeded its original limits. Section 1983 originated as a narrow, Reconstruction era statute.

A

Congress enacted §1983 as §1 of the Civil Rights Act of 1871, 17 Stat. 13. The 1871 Act was designed “to enforce the Provisions of the Fourteenth Amendment,” ibid., “in response to an ongoing pattern of violence and intimidation” against former slaves, W. Baude, J. Goldsmith, J. Manning, J. Pfander, & A. Tyler, Hart and Wechsler‘s The Federal Courts and the Federal System 1279 (8th ed. 2025) (Hart & Wechsler). In its original form, §1983 provided a means by which private plaintiffs could obtain redress from state and local officials for certain constitutional violations:

”Be it enacted . . . That any person who, under color of any law, statute, ordinance, regulation, custom, or usage of any State, shall subject, or cause to be subjected, any person within the jurisdiction of the United States to the deprivation of any rights, privileges, or immunities secured by the Constitution of the United States, shall . . . be liable to the party injured in any action at law, suit in equity, or other proper proceeding for redress . . . .” 17 Stat. 13.

In 1874, Congress extended §1983‘s reach to some statutory violations, amending the language on “rights, privileges, or immunities” to encompass “rights . . . secured by the Constitution and laws.” Rev. Stat. §1979 (emphasis added). Congress made this change as part of a general 1874 revision that aimed to “simplify, organize, and consolidate all federal statutes” into a single volume. Chapman v. Houston Welfare Rights Organization, 441 U. S. 600, 624 (1979) (Powell, J., concurring). In undertaking this revision, Congress “did not intend . . . to alter the content of federal statutory law,” id., at 625, but only to “reproduc[e]” the “existing laws,” with “such additions . . . as shall give to these provisions their intended effect,” H. R. Misc. Doc. No. 31, 40th Cong., 3d Sess., 2 (1869).

Section 1983 has remained virtually unchanged, with only relatively minor revisions. The current provision allows injured parties to sue state and local officials for “the deprivation of any rights, privileges, or immunities secured by the Constitution and laws” that they cause “under color of” state law.

B

Although the text of §1983 has remained largely constant, the judicial understanding of its scope is an entirely different matter. At the time of its enactment, ”§1983 was the least controversial provision in the 1871 Act, attracting little attention or debate.” Crawford-El v. Britton, 93 F. 3d 813, 829-830 (CADC 1996) (en banc) (Silberman, J., concurring). The “provision spawned relatively few cases for many decades.” Hart & Wechsler 1280. By one count, “there were only 21 cases decided under §1983 in its first 50 years.” Crawford-El, 93 F. 3d, at 830 (Silberman, J., concurring).

When courts did face §1983 cases, they construed the statute narrowly. This Court early on deemed §1983‘s protection of “rights, privileges, or immunities” to “refer to civil rights only.” Holt v. Indiana Mfg. Co., 176 U. S. 68, 72 (1900). The Court “never was precise about what these civil rights were,” but case law generally focused on “the rights that Congress had delineated in the Civil Rights Act of 1866,” which “mandated racial equality respecting a citizen‘s ability to sue and be a party in state court, to testify, to make contracts, and to buy, sell, and inherit property.” M. Collins, “Economic Rights,” Implied Constitutional Actions, and the Scope of Section 1983, 77 Geo. L. J. 1493, 1500-1501 (1989) (Collins); see 14 Stat. 27, as amended, 42 U. S. C. §§1981–1982. Courts later coalesced around then-Justice Stone‘s view that the relevant rights were “one[s] of personal liberty,” such as free speech and assembly, but not “property rights.” Hague v. Committee for Industrial Organization, 307 U. S. 496, 527, 531 (1939); see Maine v. Thiboutot, 448 U. S. 1, 27–28, and nn. 17–18 (1980) (Powell, J., dissenting) (discussing then-Justice Stone‘s “prevailing view“). The courts also adopted “a restrictive reading of the statute‘s reference to rights ‘secured by the Constitution and laws,‘” construing that phrase to “exclud[e] rights that did not . . . take their origin in or derive ‘directly’ from the Constitution or federal law.” Collins 1502-1503, and nn. 59-60.

This Court‘s §1983 jurisprudence took a sharp turn when the Court decided Monroe v. Pape, 365 U. S. 167 (1961). Prior to Monroe, §1983 was understood to impose liability only for actions “taken by officials pursuant to state law.” Id., at 184; see Crawford-El v. Britton, 523 U. S. 574, 611 (1998) (Scalia, J., dissenting). But, Monroe held that an official acts “under color of law” and becomes subject to the statute so long as he “is clothed with the authority of state law,” regardless of whether the State has authorized his actions. 365 U. S., at 184, 187 (internal quotation marks omitted). As a result, individuals can now bring §1983 actions for “violations committed without the authority of any” state law or “indeed even . . . violations committed in stark violation of state civil or criminal law.” Crawford-El, 523 U. S., at 611 (Scalia, J., dissenting). Monroe thus “breathed new life” into §1983. E. Zagrans, “Under Color Of” What Law: A Reconstructed Model of Section 1983 Liability, 71 Va. L. Rev. 499, 500–501 (1985).

The Court continued to broaden §1983 in the years that followed. In Lynch v. Household Finance Corp., 405 U. S. 538 (1972), it rejected then-Justice Stone‘s exclusion of “property’ rights” from the scope of §1983. Id., at 542. Then, in 1980, the Court recognized for the first time in Thiboutot that §1983 could reach statutory violations in addition to constitutional ones. See 448 U. S., at 4–5; ante, at 6.

The upshot of these decisions was that §1983 can reach “any and all violations” of rights secured by the Constitution or federal law. Health and Hospital Corporation of Marion Cty. v. Talevski, 599 U. S. 166, 225 (2023) (THOMAS, J., dissenting); see Dennis v. Higgins, 498 U. S. 439, 445 (1991); Thiboutot, 448 U. S., at 4-5. Moreover, under the Court‘s modern standard, a right is “secured by” the Constitution or federal law as long as it “unambiguously confer[s] individual rights upon a class of beneficiaries,” and Congress did not manifest any contrary intent to make §1983 unavailable. Talevski, 599 U. S., at 183, 186 (internal quotation marks omitted). The consequence is that litigants can now invoke §1983 to challenge myriad “state actions that have little or nothing to do with” civil rights. Thiboutot, 448 U. S., at 25 (Powell, J., dissenting).¹

This jurisprudential shift has transformed §1983 litigation. In 1961, the year the Court issued Monroe, federal courts heard just 296 civil rights actions. Crawford-El, 93 F. 3d, at 830 (Silberman, J., concurring). Post-Monroe, courts have faced a “deluge” of §1983 filings numbering in the tens of thousands each year. R. Aldisert, Judicial Expansion of Federal Jurisdiction: A Federal Judge‘s Thoughts on Section 1983, Comity and the Federal Caseload, 1973 Law & Social Order 557, 563; see Federal Judicial Center, M. Schwartz, Section 1983 Litigation 4 (3d ed. 2014) (“Each year, the federal courts face dockets filled with huge numbers of §1983 cases“).² Section 1983 has become

“easily the most important statute authorizing suits against state officials for violations of the Constitution and [federal] laws.” Hart & Wechsler 1280. Notwithstanding its origins as an “extraordinary remedy passed during Reconstruction to protect basic civil rights against oppressive state action,” §1983 now serves as “simply one more weapon in the litigant‘s arsenal.” Dennis, 498 U. S., at 465 (Kennedy, J., dissenting).

II

The “scant resemblance” between §1983 today and §1983 as it was traditionally understood creates good reason to doubt our modern understanding. Crawford-El, 523 U. S., at 611 (Scalia, J., dissenting). After all, a statute‘s meaning turns on what its words “conveyed to reasonable people at the time they were written.” A. Scalia & B. Garner, Reading Law 16 (2012). To ensure that we are not “elevat[ing] demonstrably erroneous decisions” over “duly enacted federal law,” we should in appropriate cases revisit the proper bounds of §1983. Gamble v. United States, 587 U. S. 678, 711 (2019) (THOMAS, J., concurring). Although the potential problems are numerous, this case implicates two in particular: the extension of §1983 into the spending-power context, and an ahistorically modern understanding of the “rights” protected by §1983.³

A

As I explained at length in Talevski, this Court has erred in extending §1983 into the spending-power context. See 599 U. S., at 196-230 (dissenting opinion). Section 1983 provides a means to redress the deprivation of “rights, privileges, or immunities secured by the Constitution and laws.” But, legislation enacted under Congress‘s spending power cannot “secure” rights as required by §1983.

This conclusion flows from a proper understanding of spending legislation. An exercise of Congress‘s power to spend “is no more than a disposition of funds.” Id., at 196. That description holds even when Congress imposes conditions on the receipt of federal funds: Conditional spending legislation amounts to a “contractual offer,” whose conditions “have no effect . . . unless and until they are freely accepted by the” recipient. Id., at 196, 201. It thus is “‘only the agreement—and not the statute—[that] makes the terms obligatory on the funds recipient.‘” Id., at 204.

In other words, conditional spending legislation does not itself “secure any rights.” Id., at 201. It cannot “make certain” or “guarantee” the obligations imposed by the spending conditions. J. Worcester, A Dictionary of the English Language 1299 (1860); accord, Webster‘s New International Dictionary 1911 (1909). Accordingly, any third parties who benefit from those obligations cannot derive an enforceable federal right from the legislation: “[S]uch third-party rights . . . are ‘secured’ (if at all) . . . only by the contract between the recipient and the United States.” Talev- ski, 599 U. S., at 205 (THOMAS, J., dissenting) (some internal quotation marks omitted).

Were it otherwise, conditional spending legislation would be unconstitutional. When the would-be recipient of federal funds is a State, treating spending conditions as imposing mandatory obligations “would contradict the bedrock constitutional prohibition against federal commandeering of the States.” Id., at 196. That prohibition protects state sovereignty by barring Congress from “conscript[ing] state governments as its agents” or “requir[ing] the States to govern according to [its] instructions.” New York v. United States, 505 U. S. 144, 162, 178 (1992). Moreover, the historical record makes clear that Congress‘s “spending power is the power to spend only” and does not “carry with it any independent regulatory authority.” Talevski, 599 U. S., at 206, 224 (THOMAS, J., dissenting).

I therefore continue to think that the Talevski majority erred “[i]n holding that spending conditions . . . can directly impose obligations on the States with the force of federal law.” Id., at 229; see id., at 177–180 (majority opinion). When “fairly possible,” we ordinarily read statutes “to avoid . . . the conclusion that [they are] unconstitutional.” United States v. Jin Fuey Moy, 241 U. S. 394, 401 (1916). Yet, Talevski chose an implausible reading of §1983 that created constitutional infirmity—and substantial infirmity, at that, given the frequency with which modern spending legislation imposes spending conditions. See 599 U. S., at 202 (THOMAS, J., dissenting).

This case does not present an occasion to remedy our error because the petitioner did not ask us to revisit our precedents. But, in a case where the issue is properly presented, I would make clear that spending conditions—which are by definition conditional—cannot “secure” rights.

B

Separately, I question whether our current understanding of §1983 is overbroad with respect to the range of “rights, privileges, or immunities” covered by that statute. Given the degree to which the judicial conception of “rights” evolved over the 20th century, I doubt that §1983, as originally understood, protects the full range of “rights” that courts now construe it to cover.

Our cases have glossed over the threshold question of what constitutes a “right” under §1983. As to constitutional rights, the Court has simply assumed that the term “rights” has the same meaning in §1983 as elsewhere. Accordingly, the Court has allowed §1983 to evolve “into an all-purpose constitutional litigation statute,” with its reach growing in proportion to the Court‘s recognition of novel constitutional “rights” in other contexts, without consideration of whether §1983‘s original meaning can be so flexible. Collins 1537; see Dennis, 498 U. S., at 445 (“[W]e have rejected attempts to limit the types of constitutional rights that are encompassed [under §1983]“). As to statutory rights, the Court has essentially collapsed the question whether a “right” exists into the broader inquiry whether there is a “righ[t] . . . secured by the Constitution and laws,” as §1983 requires. Our current test asks whether a law “clearly and unambiguously uses rights-creating terms” and displays “an unmistakable focus on individuals like the plaintiff.” Ante, at 6 (internal quotation marks and alterations omitted). But, the test does not consider the meaning of the term “rights” standing alone.

Applying these inquiries, the Court has recognized a wide variety of constitutional and statutory “rights” enforceable under §1983. Particularly given how broadly the Court has construed “the due process or cruel and unusual punishment clauses, almost any common law tort committed by a state officer” now can be “converted into a constitutional vi- olation and thereby made the basis of a section 1983 action.” Developments in the Law: Section 1983 and Federalism, 90 Harv. L. Rev. 1133, 1173 (1977).⁴ And, the Court has found a variety of “rights” conferred through statutes far removed from §1983‘s Reconstruction era roots, such as laws concerning federal entitlement programs. See, e.g., Thiboutot, 448 U. S., at 4-6 (Social Security); Talevski, 599 U. S., at 184-186 (Medicaid).

We should revisit the threshold question of what constitutes a “right” under §1983. Because we interpret statutes at the time of their enactment, see Loper Bright Enterprises v. Raimondo, 603 U. S. 369, 400 (2024), the answer to that question turns on how ordinary readers would have understood the phrase “rights, privileges, or immunities” in 1871. And, it seems more than likely that contemporaneous readers would have understood those terms more narrowly than our current §1983 doctrine does. For example, such readers presumably would have read §1983 in light of its Reconstruction era context, especially given that the provision‘s “‘rights, privileges or immunities’ language suggestively echoed the fourteenth amendment‘s ‘privileges or immunities’ clause.” Collins 1505; cf. Talevski, 599 U. S., at 225, n. 12 (THOMAS, J., dissenting) (questioning whether statu-

tory §1983 actions should be “confined to laws enacted under Congress’ Reconstruction Amendments enforcement powers“).

Even assuming that courts should give the term “rights” in §1983 the broadest meaning it could have received in 1871, that meaning almost certainly was narrower than our understanding today. Case law from the period surrounding §1983 emphasized a distinction between rights and mere government benefits. For example, in cases concerning military pensions, this Court made clear that pensions were simply “bounties of the government,” to which “[n]o pensioner has a vested legal right.” United States v. Teller, 107 U. S. 64, 68 (1883); accord, e.g., Frisbie v. United States, 157 U. S. 160, 166 (1895). Likewise, while serving on the Massachusetts Supreme Judicial Court, Justice Holmes famously summarized the once-prevailing understanding of government employees’ free speech rights when he declared that “[t]he petitioner may have a constitutional right to talk politics, but he has no constitutional right to be a police-man” that is, he has no right to public employment. McAuliffe v. Mayor and Bd. of Aldermen of New Bedford, 155 Mass. 216, 220, 29 N. E. 517 (1892). That view remained “unchallenged dogma” for “most of th[e 20th] century.” Connick v. Myers, 461 U. S. 138, 143 (1983).

Only in the 1960s and 1970s did the Court replace its traditional distinction between rights and benefits with a dramatically expanded conception of “rights.” Most notably, in Goldberg v. Kelly, 397 U. S. 254 (1970), the Court held that welfare benefits, previously thought of as gratuities, are in fact property for purposes of the Fourteenth Amendment‘s Due Process Clause. Id., at 261-262.⁵ Goldberg and other

contemporaneous cases formed a “due process revolution” that extended the Due Process Clause to cover traditionally unprotected categories such as “a government job or benefits.” R. Pierce, The Due Process Counterrevolution of the 1990s? 96 Colum. L. Rev. 1973, 1974, 1977–1980 (1996).

The modern §1983 framework developed during the same period as this rights “revolution,” and the Court‘s shift in cases like Goldberg inevitably influenced the Court‘s understanding of “rights” in the §1983 context. Plaintiffs now routinely bring §1983 claims alleging constitutional violations that would have been unimaginable in 1871. Compare, e.g., Escoe v. Zerbst, 295 U. S. 490, 492 (1935) (deeming parole an “act of grace” not protected by the Due Process Clause), with Wilkinson v. Dotson, 544 U.S. 74, 76–77 (2005) (allowing prisoners’ challenges to state parole procedures to proceed under §1983). And, much of our case law on statutory §1983 actions stems from plaintiffs’ efforts to enforce so-called rights conferred through entitlement programs. See, e.g., Thiboutot, 448 U. S., at 2-3. In light of the distinctly modern nature of our §1983 jurisprudence, I doubt that we have correctly interpreted the term “rights” for purposes of §1983.⁶

* * *

The Court properly applies our precedents to resolve the question presented. As it makes clear, even under current doctrine, courts should not too readily recognize a statutory right as enforceable under §1983. Ante, at 12-14. But, given the remarkable gap between the original understanding of §1983 and its current role, a more fundamental reexamination of our §1983 jurisprudence is in order.

JUSTICE JACKSON, with whom JUSTICE SOTOMAYOR and JUSTICE KAGAN join, dissenting.

The Civil Rights Act of 1871 was an exercise in grand ambition. It had to be. In the wake of the Civil War, the American South was consumed by a wave of terrorist violence designed to disenfranchise and intimidate the country‘s newly freed citizens and their allies. The threat was existential—not just for the newly liberated, but for democracy itself—and required bold intervention. It was precisely because the goals of the 1871 Act were so ambitious that those most committed to the structures it targeted, including many in South Carolina, opposed the measure so vehemently.

A century and a half later, the project of stymying one of the country‘s great civil rights laws continues. In this latest chapter, South Carolina urges our Court to adopt a narrow and ahistorical reading of the 1871 Act‘s first section, which is codified today at 42 U. S. C. §1983. That venerable provision permits any citizen to obtain redress in federal court for “the deprivation of any rights, privileges, or immunities secured by the Constitution and laws” of the United States. South Carolina asks us to hollow out that provision so that the State can evade liability for violating the rights of its Medicaid recipients to choose their own doctors. The Court abides South Carolina‘s request. I would not. For that reason, I respectfully dissent.

I

This case concerns South Carolina‘s obligations under the Medicaid Act. Signed into law by President Lyndon B. Johnson in 1965, the Medicaid Act establishes “a cooperative federal-state program that provides medical care to needy individuals.” Douglas v. Independent Living Center of Southern Cal., Inc., 565 U. S. 606, 610 (2012). “Like other Spending Clause legislation, Medicaid offers the States a bargain: Congress provides federal funds in exchange for the States’ agreement to spend them in accordance with congressionally imposed conditions.” Armstrong v. Exceptional Child Center, Inc., 575 U. S. 320, 323 (2015).

Any State that wishes to receive federal funds under the program must submit a proposed Medicaid plan to the Department of Health and Human Services (HHS). 42 U. S. C. §1396–1. If HHS approves the plan, the State will receive the funding. States enjoy relatively wide discretion in crafting their Medicaid plans. They have significant control, for instance, over who is eligible to receive Medicaid benefits and which types of services are covered. E.g., §§1396a(a)(10)(A)(ii), (70), (86).

Still, the Medicaid Act imposes certain plan requirements on States as a condition of receiving federal funding. If a State “fail[s] to comply substantially” with those conditions, HHS may withhold further funding from that State. §1396c; see also 42 CFR §430.12(c) (2023). In practice, however, HHS rarely invokes its authority to withhold funding because doing so would inevitably harm the program‘s beneficiaries.¹

One of the conditions that the Medicaid Act imposes on participating States is the requirement that Medicaid recipients be able to choose their own healthcare providers without government interference. The statute explicitly requires that every State‘s Medicaid plan must “provide that ... any individual eligible for medical assistance (including drugs) may obtain such assistance from any institution, agency, community pharmacy, or person, qualified to perform the service or services required.” §1396a(a)(23)(A). Congress enacted that provision, known as the “free-choice-of-provider provision,” in order to prevent States from steering Medicaid recipients to the States’ preferred healthcare providers. See H. R. Rep. No. 544, 90th Cong., 1st Sess., 122 (1967).

The dispute in this case arises from South Carolina‘s failure to comply with that provision. In 2018, the State‘s Governor issued an executive order deeming all “abortion clinics” unqualified to provide healthcare services and directing the State‘s Department of Health and Human Services to terminate them from the State‘s Medicaid program. App. to Pet. for Cert. 157a-160a. That executive order would have forced two clinics operated by Planned Parenthood South Atlantic (PPSAT)—one in Charleston and one in Columbia—to stop serving any patients who rely on Medicaid.

One of those patients is respondent Julie Edwards. Before she became a PPSAT patient, Edwards had struggled to find a healthcare provider capable of meeting her needs as a diabetic whose condition heightened the risks associated with pregnancy. At PPSAT, she found doctors who were able to provide her with the services she needed, as well as a respectful and judgment-free environment to receive care.

Edwards filed this lawsuit against state health officials under §1983 seeking to enjoin PPSAT‘s termination from the Medicaid program. She asserted that the termination decision violated her rights under the free-choice-of-provider provision to obtain care from her doctors of choice.

The District Court entered summary judgment in Edwards‘s favor and enjoined the State from terminating PPSAT from its Medicaid program. Planned Parenthood v. Baker, 487 F. Supp. 3d 443, 448 (SC 2020). The Fourth Circuit affirmed. Planned Parenthood v. Kerr, 27 F. 4th 945, 959 (2022). In a careful opinion authored by Judge Wilkinson, the panel held that the free-choice-of-provider provision conferred an individual right on Medicaid recipients to select their own healthcare providers and that, as such, that right was enforceable under §1983. Rejecting South Carolina‘s arguments to the contrary, the court concluded that the “statutory text ... unmistakably evinces Congress‘s intention to confer on Medicaid beneficiaries a right to the free choice of their provider.” Id., at 956.

South Carolina petitioned for certiorari. While its petition was pending, this Court decided Health and Hospital Corporation of Marion County v. Talevski, 599 U. S. 166 (2023), which considered whether a different provision of the Medicaid Act conferred rights enforceable under §1983. We therefore granted South Carolina‘s petition, vacated the judgment below, and remanded the case for the Fourth Circuit to reconsider the parties’ arguments in light of our decision in Talevski. 599 U. S. ____ (2023).

On remand, the Fourth Circuit once again determined that the free-choice-of-provider provision establishes an individual right that can be enforced under §1983. Planned Parenthood South Atlantic v. Kerr, 95 F. 4th 152, 154 (2024). The panel, in another thoughtful opinion by Judge Wilkinson, “conclude[d] that Talevski did not change the law to an extent that would call our previous determinations into question.” Id., at 159. It therefore affirmed the District Court‘s order granting summary judgment to Edwards and enjoining the State from terminating PPSAT from its Medicaid program. Id., at 170.

II

Two years ago, in Health and Hospital Corporation of Marion County v. Talevski, 599 U. S. 166, this Court outlined the test for determining whether a federal statute is privately enforceable under §1983. The majority accepts that the touchstone for that inquiry is whether the law in question “unambiguously confer[s] individual federal rights.” Id., at 180 (emphasis deleted); see ante, at 13. But the opinion it hands down today suggests that, as a practical matter, the character of the law—and, in particular, whether it was enacted under Congress‘s spending power—is all but dispositive of the required rights determination. That view distorts the unambiguous-conferral test beyond recognition and strains our precedential holding that §1983‘s unqualified use of the word “laws” means exactly what it says. As I explain below, under a faithful application of our unambiguous-conferral test, the Medicaid Act‘s free-choice-of-provider provision readily creates an enforceable right.

A

The Civil Rights Act of 1871 was designed to bolster the protections of the Civil War Amendments and earlier Reconstruction statutes, which had failed to “preven[t] postbellum state actors from continuing to deprive American citizens of federally protected rights.” Talevski, 599 U. S., at 176. White supremacist violence was spreading across the South, aided at times by state and local officials, and the mayhem posed a fundamental threat to both public safety and the rule of law. E. Foner, Reconstruction: America‘s Unfinished Revolution 1863–1877, pp. 442–444 (1988). The 1871 Act aimed to combat that threat in various ways. One of them, embedded in the Act‘s very first section, was to “ope[n] the federal courts to private citizens, offering a uniquely federal remedy against incursions under the claimed authority of state law upon rights secured by the Constitution and laws of the Nation.” Mitchum v. Foster, 407 U. S. 225, 239 (1972).

The text of that provision, now codified at 42 U. S. C. §1983, is straightforward. It authorizes private individuals to sue state or local officials who deprive them of “any rights, privileges, or immunities secured by the Constitution and laws” of the United States. Mindful of the statute‘s ambitious goals, the Court has traditionally “given full effect to its broad language, recognizing that §1983 ‘provide[s] a remedy, to be broadly construed, against all forms of official violation of federally protected rights.‘” Dennis v. Higgins, 498 U. S. 439, 445 (1991).

Thus, in Maine v. Thiboutot, 448 U. S. 1, 4 (1980), we expressly rejected a State‘s contention that the phrase “and laws” refers only to civil rights laws enacted under Congress‘s Fourteenth Amendment powers. As we explained, the statute‘s “plain language“—and, in particular, the fact that “Congress attached no modifiers to the phrase“—makes clear that the word “laws” “means what it says” and is not “limited to some subset of laws.” Ibid.

At the same time, our cases also recognize that §1983 “speaks in terms of ‘rights, privileges, or immunities,’ not violations of federal law” more generally. Golden State Transit Corp. v. Los Angeles, 493 U. S. 103, 106 (1989). Accordingly, we have held that a plaintiff may not prevail under §1983 merely by identifying a violation of any federal statute; rather, she must identify a violation of a statute that creates “rights, privileges, or immunities.” Ibid.

The test we apply for determining whether a statute creates such “rights, privileges, or immunities” has gradually grown more restrictive over the years. During the 1980s and 1990s, the Court adhered to Thiboutot‘s plain-language approach to §1983 and thus freely recognized individual rights in federal “laws,” absent clear congressional intent to the contrary. E.g., Livadas v. Bradshaw, 512 U. S. 107, 132-134 (1994) (holding that employees could use §1983 to enforce a provision of the National Labor Relations Act); Wright v. Roanoke Redevelopment and Housing Authority, 479 U. S. 418, 430–432 (1987) (holding that public-housing tenants could use §1983 to enforce a provision of the Housing Act of 1937 capping their rental payments).

In Wilder v. Virginia Hospital Assn., 496 U. S. 498, 509–510 (1990), for instance, we held that healthcare providers could use §1983 to enforce a provision of the Medicaid Act that required States to reimburse them at “reasonable and adequate” rates. We determined that the provision was enforceable because it left “little doubt that health care providers [were] the intended beneficiaries,” and it was “cast in mandatory rather than precatory terms.” Id., at 510, 512. We also rejected the defendant‘s argument that the provision‘s “reasonable and adequate” mandate was “too ‘vague and amorphous’ to be judicially enforceable,” observing that the statute provided an “objective benchmark” for States to judge those criteria. Id., at 519.

A few years after Wilder, Congress endorsed our holistic approach to evaluating whether statutes create rights that are enforceable under §1983. In 1994, it passed a statute confirming that a provision may create enforceable rights even if the provision is framed as a directive to States as part of a federal spending program. §555(a), 108 Stat. 4057. Congress enacted that statute in direct response to our decision in Suter v. Artist M., where we had held that a provision of the Adoption Assistance and Child Welfare Act was not enforceable under §1983. 503 U. S. 347, 363 (1992). Our decision in that case had relied, in part, on the fact that the provision at issue appeared in a section of the statute that required States to submit specific plans to HHS as a condition of receiving federal funds. Id., at 358. Rejecting that line of reasoning, Congress adopted what has come to be called the “Suter fix.” The statute it enacted provides that a “provision is not to be deemed unenforceable because of its inclusion in a section of this chapter [of the U. S. Code] requiring a State plan or specifying the required contents of a State plan.” §1320a-2.² The statute explicitly “overturn[ed]” any suggestion in Suter that state-plan requirements cannot be enforced under §1983—an interpretive approach that, in Congress‘s view, had “not [been] applied in prior Supreme Court decisions respecting [§1983] enforceability.” Ibid.

The Court decided Blessing v. Freestone, 520 U. S. 329 (1997), three years after Congress adopted the Suter fix. With no mention of §1320a–2, Blessing summarized how the Court had previously approached determining whether a federal law is privately enforceable under §1983. After surveying our past cases on the subject, we identified three key factors that bore on “whether a particular statutory provision gives rise to a federal right.” 520 U. S., at 340. Those three factors were: (1) whether “Congress ... intended that the provision in question benefit the plaintiff“; (2) whether “the right assertedly protected by the statute” is “so ‘vague or amorphous’ that its enforcement would strain judicial competence“; and (3) whether the statute “unambiguously impose[s] a binding obligation on the States.” Id., at 340-341.

Although the Blessing factors aimed merely to synthesize our past decisions, they also struck a balance between §1983‘s broad remedial goals and our historical concern that States receive fair notice of their statutory obligations under federal law. That balance began to shift dramatically in the years following Blessing.

B

In Gonzaga University v. Doe, 536 U. S. 273 (2002), the Court adopted a restrictive test for determining whether a federal statute creates rights enforceable under §1983. There, we held that a university student could not use §1983 to enforce a provision of the Family Educational Rights and Privacy Act (FERPA)—a statute that directed the Secretary of Education to withhold federal funds from schools that had failed to maintain the confidentiality of their students’ educational records. We suggested that Blessing had led to “confusion” among some lower courts about how to determine whether a statute confers rights that are enforceable under §1983. Id., at 282-283. Citing a need for greater clarity, Gonzaga stated: “We now reject the notion that our cases permit anything short of an unambiguously conferred right to support a cause of action brought under §1983.” Id., at 283.

To justify that stricter standard, the Court relied heavily on the fact that Congress had enacted FERPA under its spending powers. We noted that, in “legislation enacted pursuant to the spending power, the typical remedy for state noncompliance with federally imposed conditions is ... action by the Federal Government to terminate funds.” Id., at 280 (quoting Pennhurst State School and Hospital v. Halderman, 451 U. S. 1, 28 (1981)). For that reason, we explained, recipients of federal funds must have clear notice that their failure to comply with a particular funding condition might “subjec[t] them to private suits for money damages” under §1983. 536 U. S., at 286–287, n. 5. We thus concluded that “if Congress wishes to create new rights enforceable under §1983, it must do so in clear and unambiguous terms.” Id., at 290. And we held that FERPA flunked that test because its confidentiality mandate—which was framed principally as a directive to the Secretary of Education—“lack[ed] the sort of ‘rights-creating’ language critical to showing the requisite congressional intent to create new rights.” Id., at 287.

But while Gonzaga made the test for evaluating the enforceability of statutory rights under §1983 more stringent, it did not close the door on §1983 enforcement altogether. Just two years ago, in Talevski, we applied Gonzaga‘s analytical framework and held that a pair of Medicaid provisions created individual rights. 599 U. S., at 183. There, we determined that plaintiffs could use §1983 to enforce two provisions of the Federal Nursing Home Reform Act, or FNHRA—one imposing certain predischarge-notice requirements on nursing facilities and the other barring those facilities from using unnecessary chemical restraints on their residents. Id., at 171.

Talevski‘s analysis began by restating “the Gonzaga test.” Id., at 183 (citing Gonzaga, 536 U. S., at 284, 287). As we recounted, that test asks whether “the provision in question is ‘phrased in terms of the persons benefited’ and contains ‘rights-creating,’ individual-centric language with an ‘unmistakable focus on the benefited class.‘” 599 U. S., at 183 (quoting Gonzaga, 536 U. S., at 284, 287). Although we recognized that this test was “stringent,” we held that the two FNHRA provisions at issue satisfied it. 599 U. S., at 186. We cited the fact that both provisions appeared in a list of “[r]equirements ‘relating to residents’ rights.‘” Id., at 184. And we outlined how the text of each provision “unambiguously confer[red] rights upon the residents of nursing-home facilities“: The unnecessary-restraint provision required nursing homes to “protect and promote ... [t]he right to be free from ... any physical or chemical restraints” not needed for treatment, while the predischarge-notice provision referred to “transfer and discharge rights” and stated that nursing homes “must not transfer or discharge [a] resident” without notice. Id., at 184-185.

Perhaps most importantly, our opinion in Talevski also squarely rejected the defendant‘s argument that ”§1983 contains an implicit carveout for laws that Congress enacts via its spending power.” Id., at 171. The defendant, an Indiana hospital system, had argued that “Spending Clause statutes do not give rise to privately enforceable rights under Section 1983” because such statutes operate like contracts, which “were not ‘generally’ enforceable by third-party beneficiaries at common law.” Id., at 178 (quoting defendant‘s brief). In rejecting that attempt to dilute §1983‘s power, we affirmed once again that “‘[l]aws’ means ‘laws,’ no less today than in the 1870s.” Id., at 172. Our decision thus preserved §1983‘s central remedial aims, even as it faithfully applied Gonzaga‘s “demanding” test for whether statutes “unambiguously confer individual federal rights.” 599 U. S., at 180.³

C

Medicaid‘s free-choice-of-provider provision easily satisfies the unambiguous-conferral test. To start, the text of the provision is plainly “phrased in terms of the persons benefited“—namely, Medicaid recipients. Id., at 183. The provision states that every Medicaid plan “must ... provide that ... any individual eligible for medical assistance (including drugs) may obtain such assistance from any institution, agency, community pharmacy, or person, qualified to perform the service or services required.” §1396a(a)(23)(A) (emphasis added). This “individual-centric” formulation reflects an “unmistakable focus on the benefited class.” Talevski, 599 U. S., at 183 (quoting Gonzaga, 536 U. S., at 284).

Congress also used rights-creating language in the heading of the provision when it enacted the original session law. The provision was entitled: “FREE CHOICE BY INDIVIDUALS ELIGIBLE FOR MEDICAL ASSISTANCE,” 81 Stat. 903 (emphasis added).⁴ This phrasing indisputably invokes language classically associated with establishing rights. E.g., U. S. Const., Amdt. 1 (protecting the “free exercise” of religion); Faretta v. California, 422 U. S. 806, 833-834 (1975) (“[W]hatever else may be said of those who wrote the Bill of Rights, surely there can be no doubt that they understood the inestimable worth of free choice“). And Congress reinforced its rights-creating intent by making the provision mandatory—it specifically inserted the word “must” into the statute—to make clear that the obligation imposed on the States was binding. If Congress did not want to protect Medicaid recipients’ freedom to choose their own providers, it would have likely avoided using a combination of classically compulsory language and explicit individual-centric terminology. As the Fourth Circuit rightly put it, it is “difficult to imagine a clearer or more affirmative directive.” Planned Parenthood v. Baker, 941 F. 3d 687, 694 (2019).

The provision‘s history confirms what the text makes evident: that Congress intended the provision to be binding. Congress enacted the free-choice-of-provider provision in 1967—just two years after the original Medicaid Act—in direct response to efforts by some jurisdictions to steer Medicaid beneficiaries to specific providers. See H. R. Rep. No. 544, at 122. To prevent States from interfering with Medicaid recipients’ freedom to choose their own providers, Congress adopted nearly identical language from a provision of the Medicare Act that—in both purpose and effect—had guaranteed that right to Medicare beneficiaries. §1395a(a). In other words, Congress made a deliberate choice to protect Medicaid recipients’ ability to choose their own providers by employing statutory language that it knew, based on its Medicare experience, would achieve that end. Congress‘s intent could not have been clearer.

That clarity is perhaps why, in the only other case where we have had occasion to construe the free-choice-of-provider provision, we repeatedly used the word “right” to describe the protection it confers. In O‘Bannon v. Town Court Nursing Center, 447 U. S. 773 (1980), a group of elderly Medicaid recipients sought to leverage the provision to assert “a constitutional right to a hearing” before Medicaid officials could strip their nursing home of funding. Id., at 775. In rejecting the recipients’ understanding of the provision, we explained what the provision does protect. As we put it, ”§1396a(a)(23) ... gives recipients the right to choose among a range of qualified providers, without government interference.” Id., at 785 (first emphasis added; citation omitted). We used the word “right” again in the next sentence to elaborate on that description: “By implication,” we said, the provision “also confers an absolute right to be free from government interference with the choice to remain in a home that continues to be qualified.” Ibid. (emphasis added).⁵

Although O‘Bannon was not a case about §1983 enforceability, our description of the free-choice-of-provider provision confirms that the most natural and obvious way to read the provision‘s individual-centric, mandatory language is as “rights-creating.”

III

The majority‘s effort to resist the natural and obvious rights-creating reading of the Medicaid Act‘s free-choice-of-provider provision is, ultimately, unpersuasive. The Court holds that the provision does not confer any individual rights on Medicaid recipients, but reaches that conclusion by applying a version of the unambiguous-conferral test that we did not endorse in Talevski or Gonzaga. In doing so, the Court adopts an approach to §1983 that not only undermines the statute‘s core function but also stretches our doctrine beyond anything that can be justified as a matter of text, precedent, or first principles.

A

The approach that the Court follows today differs conspicuously from the approach we developed in Gonzaga and reaffirmed in Talevski. To see how, start by observing that the majority chooses not to frame its analysis around the question that guided our thinking in those cases: namely, whether “the provision in question is ‘phrased in terms of the persons benefited’ and contains ‘rights-creating,’ individual-centric language with an ‘unmistakable focus on the benefited class.‘” Talevski, 599 U. S., at 183 (quoting Gonzaga, 536 U. S., at 284, 287). Instead, the Court builds its analysis around the simplistic premise that Medicaid‘s free-choice-of-provider provision “looks nothing like th[e] FNHRA provisions” we upheld in Talevski. Ante, at 15–16.

That approach warps our reasoning in Talevski. Nowhere in our opinion did we single out FNHRA as the sole or definitive model for conferring individual rights. To the contrary, the reason we went out of our way to reaffirm “the Gonzaga test” was to remove any doubts about “our established method for ascertaining unambiguous conferral.” 599 U. S., at 183. Talevski was merely an application of that methodology to the statutory provision at issue in that case.

Yet, now, the majority disregards the established method and, in its place, looks to FNHRA itself as “the only reliable yardstick against which to measure whether spending-power legislation confers a privately enforceable right.” Ante, at 15. In short, the majority construes our requirement that Congress “manifes[t] an ‘unambiguous’ intent to confer individual rights,” Gonzaga, 536 U. S., at 280, as a requirement that Congress manifest an unambiguous intent to imitate FNHRA.

The majority‘s hyperfocus on FNHRA also widens the gap between our Gonzaga test and the text of §1983 itself. As noted, §1983 protects against deprivations of “any rights, privileges, or immunities secured by the ... laws” of the United States—not just the specific rights secured by FNHRA. (Emphasis added.) It is therefore strange to treat FNHRA as the “only reliable yardstick,” ante, at 15, for assessing whether a statute unambiguously creates enforceable rights per Gonzaga. Cf. Dennis, 498 U. S., at 445 (“[W]e have rejected attempts to limit the types of constitutional rights that are encompassed within the phrase ‘rights, privileges, or immunities‘“). Put simply, the fact that FNHRA happens to be the subject of one of the few cases this Court has opted to review concerning §1983 enforceability does not lend it talismanic status.

The majority‘s FNHRA-or-bust approach makes even less sense when framed against the Court‘s concerns about ensuring that States have fair notice of their statutory obligations. As the majority recognizes, the whole reason we require clear rights-creating language in spending statutes is because “[o]nly that kind of ‘unmistakable’ notice ... suffices to alert grantees” that they might be sued under §1983. Ante, at 12. But focusing myopically on a given statute‘s resemblance to FNHRA does little to advance the goal of providing fair notice to federal grantees. That is because, as we have often recognized, Congress “need not use magic words in order to speak clearly.” Henderson v. Shinseki, 562 U. S. 428, 436 (2011).

Indeed, if actual notice were the touchstone, this would be an easy case: By the time South Carolina chose to terminate PPSAT as a Medicaid provider in 2018, the State had ample reason to know that it could be sued under §1983—even beyond the clarity of the free-choice-of-provider provision‘s text. By that point, the Federal Government had long taken the position that the free-choice-of-provider provision was privately enforceable via §1983.⁶

Our decision in O‘Bannon had also explicitly described the provision as “giv[ing] recipients the right to choose” their providers “without government interference.” 447 U. S., at 785 (emphasis added). And Congress itself had reaffirmed, via the Suter fix, that the Medicaid Act‘s “State plan requirements” could create enforceable rights. §1320a-2. Our Wilder decision had long since held that a similarly structured provision of the Medicaid Act—codified in the same section as the free-choice-of-provider provision—was enforceable under §1983. 496 U. S., at 524. With all that information, South Carolina could not reasonably claim surprise that its decision to restrict Medicaid recipients’ access to particular healthcare providers might trigger a §1983 suit under the free-choice-of-provider provision.⁷

In any event, the majority‘s FNHRA-centric approach to fair notice fails on its own terms. The free-choice-of-provider provision mirrors the FNHRA provisions from Talevski in all respects that matter: both employ individual-centric language that focuses on the relevant beneficiaries and combine it with mandatory language directed at the relevant grant recipients. The provision also employs rights- creating language: As explained above, Congress explicitly used the words “free choice” in the provision‘s original heading—words that plainly reflect rights-creating intent. See Part II-C, supra. The fact that the provision does not specifically use the word “right” is not dispositive. We have never required Congress to use specific verbiage to establish individual rights. And forcing Congress to use the specific word “right” would make little sense in this context anyway in light of §1983‘s more capacious phrase “rights, privileges, or immunities.” Nor does it matter that FNHRA contains its own free-choice provision protecting the “‘right to choose a personal attending physician.‘” Ante, at 16 (quoting §1396r(c)(1)(A)(i)). If anything, that Congress chose to use “Free choice” in the heading of both provisions reflects its understanding that the two provisions would have the same rights-protecting effect.⁸

Congress ultimately has wide discretion to use whatever language it wishes to create individual rights. We require only that it do so unambiguously. As the court below aptly put it, it is not our role “to limit Congress to a thin thesaurus of our own design.” 95 F. 4th, at 166.

B

In typical parade-of-horribles-like fashion, the majority also expresses the concern that, if the Court were to hold that the free-choice-of-provider provision confers an individual right, it would mean that “[m]any other Medicaid plan requirements would likely do the same.” Ante, at 18. But case law from the lower courts demonstrates that this fear is unfounded. Those courts have recognized only a tiny handful of the nearly 90 provisions contained in the Medicaid Act‘s list of state-plan requirements as actually conferring individual rights. See Brief for National Health Law Program et al. as Amici Curiae 18–24 (highlighting the small number of provisions in §1396a(a) that lower courts have found to confer individual rights and noting the near unanimity of the Circuits as to each provision‘s enforceability). And the lower courts have consistently refused to recognize individual rights in the Medicaid Act‘s various other state-plan provisions. Ibid.

Meanwhile, the vast majority of the provisions on the Medicaid Act‘s list of state-plan requirements have never generated any §1983 litigation whatsoever. There is thus little reason to think that a decision holding that the free-choice-of-provider provision confers individual rights would unleash a sudden torrent of §1983 suits under the Act‘s other state-plan provisions. Indeed, recent history confirms as much: Prior to 2017, every Circuit to consider the question had held that the free-choice-of-provider provision confers an individual right enforceable under §1983. See n. 7, supra. But those decisions did not spawn a bevy of §1983 suits seeking to enforce other state-plan provisions.

Nor were the floodgates opened by this Court‘s decisions in Wilder, Blessing, or any other cases that predate the restrictive test for §1983 enforceability that this Court adopted in Gonzaga. As the majority readily acknowledges, prior to Gonzaga, the Court “experimented with a different approach.” Ante, at 13. Indeed, in Gonzaga itself, the Court rationalized its newly restrictive approach to §1983 enforceability by indicating that some lower courts had become too permissive in recognizing enforceable statutory rights. 536 U. S., at 283. Yet, even during the pre-Gonzaga period, there is no evidence that lower courts treated the Medicaid Act—which spans multiple volumes of the U. S. Code—as a wellspring of generally enforceable rights. Rather, the state of affairs before our tightening of the test reinforces that the free-choice-of-provider provision is, in fact, the “atypical” spending statute that creates individual rights, Talevski, 599 U. S., at 183—contrary to the majority‘s assertions otherwise. Ante, at 18, 21.

C

Finally, JUSTICE THOMAS‘s concurrence calls for a “fundamental reexamination of our §1983 jurisprudence” based on his view that the “history of §1983 makes clear that the statute has exceeded its original limits.” Ante, at 1-2, 13. Because his opinion is not tethered to the specific facts or arguments presented in this case, an extensive response is not necessary here. But it is worth pausing briefly to think about whether the historical account he offers reflects the level of depth, nuance, or context needed to support the wholesale reappraisal he is envisioning.

Take his observation that courts decided relatively few cases under §1983 during its first several decades. Ante, at 3. Like other §1983 skeptics, JUSTICE THOMAS seems to view the paucity of early §1983 lawsuits as evidence that the statute was originally understood to do very little. But other explanations come to mind, too—such as the fact that filing civil rights lawsuits during the Jim Crow era could be quite perilous, especially for the people whom the statute was originally meant to benefit. Many would-be plaintiffs had reason to fear that filing a lawsuit would lead to physical or economic reprisals.⁹ Add to that the difficulty of finding a lawyer, prevailing before often-hostile juries, and (if successful) enforcing a judgment, and it is not hard to imagine that the dearth of §1983 lawsuits in the wake of Reconstruction might have myriad alternative explanations.¹⁰

JUSTICE THOMAS also suggests that the word “rights,” as used in §1983, was originally understood more narrowly than it is today. Ante, at 8-12. But his support for that claim is limited to a handful of late-19th-century cases, mostly about government pensions and employment. If “a statute‘s meaning turns on what its words ‘conveyed to reasonable people at the time they were written,‘” ante, at 6, a broader—and more inclusive—survey of historical sources would seem to be in order.¹¹

All of which is to say: more caution (and more research) may be warranted before our longstanding precedents in this area can be seriously scrutinized or attacked—especially in cases where no party has made such a claim or presented any such argument.

* * *

Congress enacted the Medicaid Act‘s free-choice-of-provider provision to ensure that Medicaid recipients have the right to choose their own doctors. The Court‘s decision to foreclose Medicaid recipients from using §1983 to enforce that provision thwarts Congress‘s will twice over: once, in dulling the tool Congress created for enforcing all federal rights, and again in vitiating one of those rights altogether.

The Court‘s decision today is not the first to so weaken the landmark civil rights protections that Congress enacted during the Reconstruction Era. See, e.g., Civil Rights Cases, 109 U. S. 3 (1883); United States v. Cruikshank, 92 U. S. 542 (1876); Blyew v. United States, 13 Wall. 581 (1872). That means we do have a sense of what comes next: as with those past rulings, today‘s decision is likely to result in tangible harm to real people. At a minimum, it will deprive Medicaid recipients in South Carolina of their only meaningful way of enforcing a right that Congress has expressly granted to them. And, more concretely, it will strip those South Carolinians—and countless other Medicaid recipients around the country—of a deeply personal freedom: the “ability to decide who treats us at our most vulnerable.” Kerr, 95 F. 4th, at 169. The Court today disregards Congress‘s express desire to prevent that very outcome.

Notes

1

Cannon v. University of Chicago, 441 U. S. 677, 691 (1979), and its aftermath illustrate the shift in this Court‘s approach. In Cannon, the Court inferred new private causes of action from the terms of Title VI of the Civil Rights Act of 1964 and Title IX of the Education Amendments of 1972. Though Congress later “ratified Cannon‘s holding,” Sandoval, 532 U. S., at 280, the Court has retreated from Cannon‘s reasoning, which “exemplified” an “expansive rights-creating approach” that later decisions “abandoned,” Franklin v. Gwinnett County Public Schools, 503 U. S. 60, 77 (1992) (Scalia, J., concurring in judgment); see also Stoneridge Investment Partners, LLC v. Scientific-Atlanta, Inc., 552 U. S. 148, 164–165 (2008) (quoting Justice Powell‘s Cannon dissent). So while this Court has said it remains bound by Cannon‘s “holdin[g],” it has emphasized that the decision‘s “language” no longer controls. Sandoval, 532 U. S., at 282. To be sure, our §1983 jurisprudence is not without guardrails. As today‘s decision emphasizes, few federal laws truly “secure” individual rights: Our cases in the spending-power context make clear that federal laws unambiguously confer such rights “only in ‘atypical case[s],‘” where a statutory provision meets a “‘stringent’ and ‘demanding’ test.” Ante, at 6 (quoting Health and Hospital Corporation of Marion Cty. v. Talevski, 599 U. S. 166, 180, 183, 186 (2023)). But, limits like this one do not alter the overall thrust of our §1983 case law. Kaiser Family Foundation, Focus on Health Reform: A Guide to the Supreme Court‘s Decision on the ACA‘s Medicaid Expansion 1 (Aug. 2012), https://www.kff.org/affordable-care-act/issue-brief/a-guide-to-the-supreme-courts-decision/.

2

Much the same holds true today. While treaties may benefit individuals or groups, this Court has said, “the background presumption” is that treaties “‘do not create private rights or provide for a private cause of action.‘” Medellin v. Texas, 552 U. S. 491, 506, n. 3 (2008) (quoting 2 Restatement (Third) of Foreign Relations Law of the United States §907, Comment a, p. 395 (1986); emphasis added). In the 12-month period before September 30, 2024, federal district courts docketed over 65,000 new civil rights actions. See U. S. Courts, U. S. District Courts—Civil Cases Commenced, by Nature of Suit, During the 12-Month Periods Ending Sept. 30, 2020 through 2024 (Table C-2A), https://uscourts.gov/sites/default/files/2025-01/jb_c2a_0930.2024.pdf. Although the data is not granular enough to determine the precise number of §1983 cases within this total, §1983 cases undoubtedly make up a sizable fraction. Cf. C. Whitman, Constitutional Torts, 79 Mich. L. Rev. 5, 6, n. 9 (1980) (“In practice, virtually all civil rights cases filed against states in federal court include a §1983 claim“). The law at issue in this case—Medicaid‘s free-choice-of-provider provision, 42 U. S. C. §1396a(a)(23)—is codified in the same chapter of Title 42 as the Suter fix.

3

Between 1940 and 2023, federal outlays to state and local governments increased by more than 50 times in constant dollars. Office of Management and Budget, Historical Tables, Budget of the United States Government, Summary Comparison of Total Outlays for Grants to State and Local Governments: 1940–2029 (2024) (Table 12.1), https://www.govinfo.gov/app/details/BUDGET-2025-TAB/BUDGET-2025-TAB-13-1. I have elsewhere identified other questionable aspects of our §1983 jurisprudence. For example, there is reason to doubt the broad reading of §1983‘s “under color of” language in Monroe v. Pape, 365 U. S. 167 (1961). See Baxter v. Bracey, 590 U. S. ___, ___, n. 2 (2020) (THOMAS, J., dissenting from denial of certiorari) (slip op., at 5-6, n. 2). Given §1983‘s Reconstruction era context, it also is questionable whether statutory §1983 actions can be based on laws besides those “enacted under Congress’ Reconstruction Amendments enforcement powers.” Talevski, 599 U. S., at 225, n. 12 (THOMAS, J., dissenting). And, reexamination may be warranted as to whether §1983 even supplies a freestanding cause of action. See Williams v. Reed, 604 U. S. ___, ___ n. (2025) (THOMAS, J., dissenting) (slip op., at 4, n.). Talevski also recognized that “[e]ven if a statutory provision unambiguously secures rights, a defendant ‘may defeat t[he] presumption by demonstrating that Congress did not intend’ that §1983 be available to enforce those rights.” 599 U. S., at 186. South Carolina has not invoked that proposition here as a basis for arguing that Medicaid‘s free-choice-of-provider provision is not enforceable under §1983.

4

Beyond the rule that Congress must clearly and unambiguously alert States to conditions associated with federal funding, our cases have articulated other limits on spending-power legislation. First, as previously observed, “the exercise of the spending power must be in pursuit of ‘the general welfare,’ ” rather than private or merely local interests. South Dakota v. Dole, 483 U. S. 203, 207 (1987); see supra, at 8–9. Second, grant conditions must relate “to the federal interest in particular national projects or programs.” Dole, 483 U. S., at 207 (internal quotation marks omitted). Third, “other constitutional provisions may provide an independent bar to the conditional grant of federal funds.” Id., at 208. Finally, spending-power conditions are legitimate only if the State‘s acceptance of them is in fact voluntary. National Federation of Independent Business v. Sebelius, 567 U. S. 519, 581–582 (2012) (opinion of ROBERTS, C. J.); see also id., at 676 (Scalia, Kennedy, THOMAS, and ALITO, JJ., dissenting). In one case, the Court even deemed enforceable under §1983 the negative Commerce Clause “‘right’ to engage in interstate trade free from restrictive state regulation.” Dennis v. Higgins, 498 U. S. 439, 448-451 (1991). Setting aside that “[t]he negative Commerce Clause has no basis in the text of the Constitution,” Camps Newfound/Owatonna, Inc. v. Town of Harrison, 520 U. S. 564, 610 (1997) (THOMAS, J., dissenting), the holding in Dennis relied on reasoning from our statutory §1983 cases that we have since repudiated. Compare 498 U. S., at 448-449, with ante, at 13-14. As the Court emphasizes today, our more recent statutory §1983 case law makes clear that statutes must include “unambiguous rights-creating language” to be enforceable through §1983. Ante, at 19. In an appropriate case, we should at minimum extend similar scrutiny to the range of constitutional rights enforceable through §1983. The majority seeks to downplay the title Congress assigned to the free-choice-of-provider provision by noting that a title “by itself” cannot confer rights. Ante, at 21. But the majority does not appear to dispute that statutory titles offer at least some insight into Congress‘s intent, as evidenced by the majority‘s own reliance on statutory titles elsewhere in its opinion. See ante, at 15 (highlighting the title of one of FNHRA‘s subprovisions); ante, at 18 (citing the title of §1396a(a)). In any event, as the rest of the discussion above illustrates, Congress‘s decision to use the “free choice” language in its session-law heading is not the only evidence of its rights-creating intent with respect to the provision at issue here.

5

Gonzaga involved federal funds granted to a private university, not a State. But our spending-power cases have applied similar principles to state and private recipients of federal aid. See, e.g., Cummings v. Premier Rehab Keller, 596 U. S. 212, 219–220 (2022). Whether a State or private recipient is involved, after all, §1983 actions to enforce federal statutes present a question sounding in the separation of powers, given that it is for Congress, not the courts, to confer “rights upon a class of beneficiaries” sufficient to support a cause of action. See Gonzaga, 536 U. S., at 285; Part II–A, supra. And grants to private parties can risk altering the Constitution‘s balance of federal-state authority, too, by expanding federal regulation beyond Congress‘s enumerated powers and into areas traditionally reserved for the States. See Gonzaga, 536 U. S., at 286, and n. 5; cf. Gregory v. Ashcroft, 501 U. S. 452, 460–461 (1991). As I have previously explained, Goldberg rests on tenuous grounds. Forgoing “meaningful legal analysis,” the decision “simply highlighted the social importance of ‘entitlements’ in modern America.” Williams, 604 U. S., at ___, n. (THOMAS, J., dissenting opinion) (slip op., at 3-4, n.) (quoting Goldberg, 397 U. S., at 262, and n. 8); see also Gutierrez v. Saenz, 606 U. S. ___ (2025). In their certiorari-stage briefs, the parties disputed whether and to what extent O‘Bannon bears on the scope of the free-choice-of-provider provision. We declined to grant certiorari on that question. 604 U. S. ____ (2024) (limiting our grant of certiorari to only the first question presented in the petition). Undeterred by that choice, the majority proceeds to address the question we took off the table: It suggests that O‘Bannon is inapposite because our opinion in that case purportedly rejected the particular right that respondent has asserted here. Ante, at 22, n. 8. But the question of how broadly to construe the rights conferred by the free-choice-of-provider provision is distinct from the question of whether the provision creates rights in the first place. And as to that latter question—the sole question presented in this case—O‘Bannon‘s repeated use of the word “right” to describe the provision‘s protections underscores how the provision‘s text is naturally read to create rights. What is more, the majority has quoted the O‘Bannon passage completely out of context; when read in its entirety, the quoted passage has little bearing on this case. The full sentence states that the free-choice-of-provider provision “clearly does not confer a right on a recipient to enter an unqualified home and demand a hearing to certify it, nor does it confer a right on a recipient to continue to receive benefits for care in a home that has been decertified.” 447 U. S., at 785. This language does not come close to suggesting that the free-choice-of-provider provision does not confer a right to choose one‘s provider (i.e., the right respondent has asserted here), as the majority suggests.

6

As we have seen, the plaintiffs must also show that the provision in question displays “an unmistakable focus” on individuals like them. Gonzaga, 536 U. S., at 284 (emphasis deleted; internal quotation marks omitted). And even then, a §1983 action may not be available if Congress has displaced that general cause of action with a more specific remedy. See supra, at 6. To resolve this case, however, we need not reach those questions. The dissent questions whether sufficient “research” supports my current conclusions. Post, at 20-21 (opinion of JACKSON, J.). But, my point is precisely that further examination is warranted. Insofar as the dissent highlights the existence of other “historical sources” beyond the scope of this concurring opinion, the “broader” historical record at which the dissent gestures only reinforces the need to consider the relationship (or lack thereof) between our current §1983 jurisprudence and §1983‘s original meaning. Post, at 21. See Brief for United States as Amicus Curiae 4, n. 1 (citing amicus briefs filed by the Government, across multiple administrations, in cases dating back to 2005).

7

If anything, the legislative history of the any-qualified-provider provision illustrates the pitfalls of trying to equate an unenacted legislative record with the law. On the plaintiffs’ telling, Congress first enacted the any-qualified-provider provision “to prevent [the] second-class treatment” of Medicaid patients, as exemplified by Puerto Rico‘s policy of requiring them “to be treated only at designated government facilities.” Brief for Respondents 30 (citing Hearing on H. R. 5710 before the House Committee on Ways and Means, 90th Cong., 1st Sess., pt. 4, pp. 2273, 2301 (1967)). And yet §1396a(a)(23), as it stands today, expressly excludes Puerto Rican beneficiaries from its protections. See §1396a(a)(23)(B) (“[T]his paragraph shall not apply in the case of Puerto Rico, the Virgin Islands, and Guam“). The Court‘s repudiation of Wilder today does not alter any of those historical facts. Indeed, prior to this Court‘s attempt to disavow Wilder in a footnote in Armstrong v. Exceptional Child Center, Inc., 575 U. S. 320, 330, n. (2015), it was widely accepted—not just by the Government, but by every Circuit to consider the question—that the free-choice-of-provider provision conferred privately enforceable rights. See Planned Parenthood Ariz. Inc. v. Betlach, 727 F. 3d 960, 963 (CA9 2013); Planned Parenthood of Ind., Inc. v. Commissioner of Ind. State Dept. of Health, 699 F. 3d 962, 974 (CA7 2012); Harris v. Olszewski, 442 F. 3d 456, 461 (CA6 2006). South Carolina accepted Medicaid funding for years with knowledge of those facts. Only recently—in its brief in this case—has the Federal Government changed its longstanding position. That the Court has now succeeded in injecting ambiguity where none previously existed underscores the extent to which the Court‘s practical concerns about fair notice to grantees seem to have been displaced by a general aversion to recognizing individual rights.

8

We agree with the dissent that we did not grant certiorari to resolve “whether and to what extent O‘Bannon [v. Town Court Nursing Center, 447 U. S. 773 (1980)] bears on the scope of” §1396a(a)(23)(A). Post, at 14, n. 5. But since the dissent relies heavily on that decision, post, at 13-14, we should make plain that we read it as consistent with all we have said. O‘Bannon held only that residents of a nursing facility had no right under the Due Process Clause to a hearing before a State ended that facility‘s participation in its Medicaid program. 447 U. S., at 775, 790. To the extent O‘Bannon addressed any right, then, it was an asserted property right under the Due Process Clause, not a clear and unambiguous statutory right under §1983. Id., at 779. Notably, too, O‘Bannon expressly recognized that 42 U. S. C. §1396a(a)(23) “does not confer a right on a recipient . . . to continue to receive benefits for care [from a provider] that has been decertified.” 447 U. S., at 785. And that is precisely the right the plaintiffs assert here. Separately, the dissent suggests that amicus briefs the government filed in other cases might suffice to supply States with notice of a condition attached to federal funding. Post, at 16-17, and n. 6 (opinion of JACKSON, J.). But, as this case attests, the government‘s views can shift from administration to administration. And our decisions have never suggested that anything less than clear statutory language can supply States with the unambiguous notice required. Instead, given the separation of powers and federalism concerns we have outlined, our decisions have always “insist[ed] that Congress speak with a clear voice.” Pennhurst State School and Hospital v. Halderman, 451 U. S. 1, 17 (1981) (emphasis added). Compare §1396r(c)(1)(A)(i) (“Free choice” (boldface deleted)) with §227, 81 Stat. 903 (“Free choice by individuals eligible for medical assistance” (some capitalization omitted)). Notably, the same omnibus legislation that included the FNHRA free-choice provision also included an amendment to the original free-choice-of-provider provision that appeared under the heading “Freedom of choice,” further reinforcing the view that the provision is rights creating. §4113(c), 101 Stat. 1330-152 (some capitalization omitted).

9

In the end, the dissent resorts to the extravagant charge that our decision represents the “latest chapter” in a “project of stymying . . . civil rights.” Post, at 1 (opinion of JACKSON, J.); see also post, at 22. As we have explained at length, our decision simply applies the same test this Court applied in Gonzaga and again in Talevski (with the support of today‘s dissenters). And in doing so, we reach the unsurprising conclusion that it generally belongs to the federal government to supervise compliance with its own spending programs. As the dissenters themselves put it in Talevski, spending-power legislation creates privately enforceable rights only in “atypical case[s].” 596 U. S., at 183. Our decision merely recognizes that this case is not an atypical one. See, e.g., M. Klarman, From Jim Crow to Civil Rights: The Supreme Court and the Struggle for Racial Equality 49 (2004) (Klarman) (explaining that, “[b]y the 1890s, southern black challenges to segregation would have invited physical retaliation and perhaps even lynching“); Equal Justice Initiative, Bob Hudson Lynched and Wife Beaten in Weakley County, Tennessee (last visited June 15, 2025), https://calendar.eji.org/racial-injustice/oct/9 (recounting the lynching of a Black man whose wife had filed a civil suit against a White man).

10

See Klarman 48-49 (describing the dearth of lawyers willing to litigate civil rights cases, the lack of sympathy among southern juries, and the unlikelihood that local authorities would be willing to enforce judgments obtained by certain civil rights plaintiffs).

11

E.g., Colored People‘s Convention of the State of South Carolina (1865, Charleston, SC), Colored Conventions Project Digital Records (last visited June 15, 2025), https://omeka.coloredconventions.org/items/show/570 (“‘Right’ is defined to be the just claim, ownership, or lawful title which a person has to anything“).

Case Details

Case Name: Medina v. Planned Parenthood South Atlantic

Court Name: Supreme Court of the United States

Date Published: Jun 26, 2025

Citations: 606 U.S. 357; 23-1275

Docket Number: 23-1275

Court Abbreviation: SCOTUS