Petitioners Michael and Chantell Sackett purchased property near Priest Lake, Idaho, and began backfilling the lot with dirt to prepare for building a home. The Environmental Protection Agency informed the Sacketts that their property contained wetlands and that their backfilling violated the Clean Water Act, which prohibits discharging pollutants into “the waters of the United States.” 33 U. S. C. §1362(7). The EPA ordered the Sacketts to restore the site, threatening penalties of over $40,000 per day. The EPA classified the wetlands on the Sacketts’ lot as “waters of the United States” because they were near a ditch that fed into a creek, which fed into Priest Lake, a navigable, intrastate lake. The Sacketts sued, alleging that their property was not “waters of the United States.” The District Court entered summary judgment for the EPA. The Ninth Circuit affirmed, holding that the CWA covers wetlands with an ecologically significant nexus to traditional navigable waters and that the Sacketts’ wetlands satisfy that standard.

Held: The CWA’s use of “waters” in §1362(7) refers only to “geographic[al] features that are described in ordinary parlance as ‘streams, oceans, rivers, and lakes’ ” and to adjacent wetlands that are “indistinguishable” from those bodies of water due to a continuous surface connection. Rapanos v. United States, 547 U. S. 715, 755, 742, 739 (plurality opinion). To assert jurisdiction over an adjacent wetland under the CWA, a party must establish “first, that the adjacent [body of water constitutes] . . . ‘water[s] of the United States’ (i.e., a relatively permanent body of water connected to traditional interstate navigable waters); and second, that the wetland has a continuous surface connection with that water, making it difficult to determine where the ‘water’ ends and the ‘wetland’ begins.” Ibid. Pp. 6–28.

 (a) The uncertain meaning of “the waters of the United States” has been a persistent problem, sparking decades of agency action and litigation. Resolving the CWA’s applicability to wetlands requires a review of the history surrounding the interpretation of that phrase. Pp. 6–14.

  (1) During the period relevant to this case, the two federal agencies charged with enforcement of the CWA—the EPA and the Army Corps of Engineers—similarly defined “the waters of the United States” broadly to encompass “[a]ll . . . waters” that “could affect interstate or foreign commerce.” 40 CFR §230.3(s)(3). The agencies likewise gave an expansive interpretation of wetlands adjacent to those waters, defining “adjacent” to mean “bordering, contiguous, or neighboring.” §203.3(b). In United States v. Riverside Bayview Homes, Inc., 474 U. S. 121, the Court confronted the Corps’ assertion of authority under the CWA over wetlands that “actually abut[ted] on a navigable waterway.” Id., at 135. Although concerned that the wetlands fell outside “traditional notions of ‘waters,’ ” the Court deferred to the Corps, reasoning that “the transition from water to solid ground is not necessarily or even typically an abrupt one.” Id., 132–133. Following Riverside Bayview, the agencies issued the “migratory bird rule,” extending CWA jurisdiction to any waters or wetlands that “are or would be used as [a] habitat” by migratory birds or endangered species. 53 Fed. Reg. 20765. The Court rejected the rule after the Corps sought to apply it to several isolated ponds located wholly within the State of Illinois, holding that the CWA does not “exten[d] to ponds that are not adjacent to open water.” Solid Waste Agency of Northern Cook Cty. v. Army Corps of Engineers, 531 U. S. 159, 168 (SWANCC) (emphasis deleted). The agencies responded by instructing their field agents to determine the scope of the CWA’s jurisdiction on a case-by-case basis. Within a few years, the agencies had “interpreted their jurisdiction over ‘the waters of the United States’ to cover 270-to-300 million acres” of wetlands and “virtually any parcel of land containing a channel or conduit . . . through which rainwater or drainage may occasionally or intermittently flow.” Rapanos, 547 U. S., at 722 (plurality opinion).

 Against that backdrop, the Court in Rapanos vacated a lower court decision that had held that the CWA covered wetlands near ditches and drains that emptied into navigable waters several miles away. As to the rationale for vacating, however, no position in Rapanos commanded a majority of the Court. Four Justices concluded that the CWA’s coverage was limited to certain relatively permanent bodies of water connected to traditional interstate navigable waters and to wetlands that are “as a practical matter indistinguishable” from those waters. Id., at 755 (emphasis deleted). Justice Kennedy, concurring only in the judgment, wrote that CWA jurisdiction over adjacent wetlands requires a “significant nexus” between the wetland and its adjacent navigable waters, which exists when “the wetlands, either alone or in combination with similarly situated lands in the region, significantly affect the chemical, physical, and biological integrity” of those waters. Id., at 779–780. Following Rapanos, field agents brought nearly all waters and wetlands under the risk of CWA jurisdiction by engaging in fact-intensive “significant-nexus” determinations that turned on a lengthy list of hydrological and ecological factors.

 Under the agencies’ current rule, traditional navigable waters, interstate waters, and the territorial seas, as well as their tributaries and adjacent wetlands, are waters of the United States. See 88 Fed. Reg. 3143. So too are any “[i]ntrastate lakes and ponds, streams, or wetlands” that either have a continuous surface connection to categorically included waters or have a significant nexus to interstate or traditional navigable waters. Id., at 3006, 3143. Finding a significant nexus continues to require consideration of a list of open-ended factors. Ibid. Finally, the current rule returns to the agencies’ longstanding definition of “adjacent.” Ibid. Pp. 6–12.

  (2) Landowners who even negligently discharge pollutants into navigable waters without a permit potentially face severe criminal and civil penalties under the Act. As things currently stand, the agencies maintain that the significant-nexus test is sufficient to establish jurisdiction over “adjacent” wetlands. By the EPA’s own admission, nearly all waters and wetlands are potentially susceptible to regulation under this test, putting a staggering array of landowners at risk of criminal prosecution for such mundane activities as moving dirt. Pp. 12–14.

 (b) Next, the Court considers the extent of the CWA’s geographical reach. Pp. 14–22.

  (1) To make sense of Congress’s choice to define “navigable waters” as “the waters of the United States,” the Court concludes that the CWA’s use of “waters” encompasses “only those relatively permanent, standing or continuously flowing bodies of water ‘forming geographic[al] features’ that are described in ordinary parlance as ‘streams, oceans, rivers, and lakes.’ ” Rapanos, 547 U. S., at 739 (plurality opinion). This reading follows from the CWA’s deliberate use of the plural “waters,” which refers to those bodies of water listed above, and also helps to align the meaning of “the waters of the United States” with the defined term “navigable waters.” More broadly, this reading accords with how Congress has employed the term “waters” elsewhere in the CWA—see, e.g., 33 U. S. C. §§1267(i)(2)(D), 1268(a)(3)(I)—and in other laws—see, e.g., 16 U. S. C. §§745, 4701(a)(7). This Court has understood CWA’s use of “waters” in the same way. See, e.g., Riverside Bayview, 474 U. S., at 133; SWANCC, 531 U. S., at 168–169, 172.

 The EPA’s insistence that “water” is “naturally read to encompass wetlands” because the “presence of water is ‘universally regarded as the most basic feature of wetlands’ ” proves too much. Brief for Respondents 19. It is also tough to square with SWANCC’s exclusion of isolated ponds or Riverside Bayview’s extensive focus on the adjacency of wetlands to covered waters. Finally, it is difficult to see how the States’ “responsibilities and rights” in regulating water resources would remain “primary” if the EPA had such broad jurisdiction. §1251(b). Pp. 14–18.

  (2) Statutory context shows that some wetlands nevertheless qualify as “waters of the United States.” Specifically, §1344(g)(1), which authorizes States to conduct certain permitting programs, specifies that discharges may be permitted into any waters of the United States, except for traditional navigable waters, “including wetlands adjacent thereto,” suggesting that at least some wetlands must qualify as “waters of the United States.” But §1344(g)(1) cannot define what wetlands the CWA regulates because it is not the operative provision that defines the Act’s reach. Instead, the reference to adjacent wetlands in §1344(g)(1) must be harmonized with “the waters of the United States,” which is the operative term that defines the CWA’s reach. Because the “adjacent” wetlands in §1344(g)(1) are “includ[ed]” within “waters of the United States,” these wetlands must qualify as “waters of the United States” in their own right, i.e., be indistinguishably part of a body of water that itself constitutes “waters” under the CWA. To hold otherwise would require implausibly concluding that Congress tucked an important expansion to the reach of the CWA into convoluted language in a relatively obscure provision concerning state permitting programs. Understanding the CWA to apply to wetlands that are distinguishable from otherwise covered “waters of the United States” would substantially broaden §1362(7) to define “navigable waters” as “waters of the United States and adjacent wetlands.” But §1344(g)(1)’s use of the term “including” makes clear that it does not purport to do any such thing. It merely reflects Congress’s assumption that certain “adjacent” wetlands are part of the “waters of the United States.”

 To determine when a wetland is part of adjacent “waters of the United States,” the Court agrees with the Rapanos plurality that the use of “waters” in §1362(7) may be fairly read to include only wetlands that are “indistinguishable from waters of the United States.” This occurs only when wetlands have “a continuous surface connection to bodies that are ‘waters of the United States’ in their own right, so that there is no clear demarcation between ‘waters’ and wetlands.” 547 U. S., at 742.

 In sum, the CWA extends to only wetlands that are “as a practical matter indistinguishable from waters of the United States.” This requires the party asserting jurisdiction to establish “first, that the adjacent [body of water constitutes] . . . ‘water[s] of the United States’ (i.e., a relatively permanent body of water connected to traditional interstate navigable waters); and second, that the wetland has a continuous surface connection with that water, making it difficult to determine where the ‘water’ ends and the ‘wetland’ begins.” Rapanos, 547 U. S., at 755, 742. Pp. 18–22.

 (c) The EPA asks the Court to defer to its most recent rule providing that “adjacent wetlands are covered by the [CWA] if they ‘possess a significant nexus to’ traditional navigable waters” and that wetlands are “adjacent” when they are “neighboring” to covered waters. Brief for Respondents 32, 20. For multiple reasons, the EPA’s position lacks merit. Pp. 22–27.

  (1) The EPA’s interpretation is inconsistent with the CWA’s text and structure and clashes with “background principles of construction” that apply to the interpretation of the relevant provisions. Bond v. United States, 572 U. S. 844, 857. First, “exceedingly clear language” is required if Congress wishes to alter the federal/state balance or the Government’s power over private property. United States Forest Service v. Cowpasture River Preservation Assn., 590 U. S. ___, ___. The Court has thus required a clear statement from Congress when determining the scope of “the waters of the United States.” Second, the EPA’s interpretation gives rise to serious vagueness concerns in light of the CWA’s criminal penalties, thus implicating the due process requirement that penal statutes be defined “ ‘with sufficient definiteness that ordinary people can understand what conduct is prohibited.’ ” McDonnell v. United States, 579 U. S. 550, 576. Where penal statutes could sweep broadly enough to render criminal a host of what might otherwise be considered ordinary activities, the Court has been wary about going beyond what “Congress certainly intended the statute to cover.” Skilling v. United States, 561 U. S. 358, 404. Under these two principles, the judicial task when interpreting “the waters of the United States” is to ascertain whether clear congressional authorization exists for the EPA’s claimed power. Pp. 22–25.

  (2) The EPA claims that Congress ratified the EPA’s regulatory definition of “adjacent” when it amended the CWA to include the reference to “adjacent” wetlands in §1344(g)(1). This argument fails for at least three reasons. First, the text of §§1362(7) and 1344(g) shows that “adjacent” cannot include wetlands that are merely nearby covered waters. Second, EPA’s argument cannot be reconciled with this Court’s repeated recognition that §1344(g)(1) “ ‘does not conclusively determine the construction to be placed on . . . the relevant definition of “navigable waters.” ’ ” SWANCC, 531 U. S., at 171. Third, the EPA falls short of establishing the sort of “overwhelming evidence of acquiescence” necessary to support its argument in the face of Congress’s failure to amend §1362(7). Finally, the EPA’s various policy arguments about the ecological consequences of a narrower definition of “adjacent” are rejected. Pp. 25–27.

8 F. 4th 1075, reversed and remanded.

 Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Thomas, Gorsuch, and Barrett, JJ., joined. Thomas, J., filed a concurring opinion, in which Gorsuch, J., joined. Kagan, J., filed an opinion concurring in the judgment, in which Sotomayor and Jackson, JJ., joined. Kavanaugh, J., filed an opinion concurring in the judgment, in which Sotomayor, Kagan, and Jackson, JJ., joined.

Geraldine Tyler owned a condominium in Hennepin County, Minnesota, that accumulated about $15,000 in unpaid real estate taxes along with interest and penalties. The County seized the condo and sold it for $40,000, keeping the $25,000 excess over Tyler’s tax debt for itself. Minn. Stat. §§281.18, 282.07, 282.08. Tyler filed suit, alleging that the County had unconstitutionally retained the excess value of her home above her tax debt in violation of the Takings Clause of the Fifth Amendment and the Excessive Fines Clause of the Eighth Amendment. The District Court dismissed the suit for failure to state a claim, and the Eighth Circuit affirmed.

Held: Tyler plausibly alleges that Hennepin County’s retention of the excess value of her home above her tax debt violated the Takings Clause. Pp. 3–14.

  (a) Tyler’s claim that the County illegally appropriated the $25,000 surplus constitutes a classic pocketbook injury sufficient to give her standing. TransUnion LLC v. Ramirez, 594 U. S. ___, ___. Even if there are debts on her home, as the County claims, Tyler still plausibly alleges a financial harm, for the County has kept $25,000 that she could have used to reduce her personal liability for those debts. Pp. 3–4.

  (b) Tyler has stated a claim under the Takings Clause, which provides that “private property [shall not] be taken for public use, without just compensation.” Whether remaining value from a tax sale is property protected under the Takings Clause depends on state law, “traditional property law principles,” historical practice, and the Court’s precedents. Phillips v. Washington Legal Foundation, 524 U. S. 156, 165–168. Though state law is an important source of property rights, it cannot be the only one because otherwise a State could “sidestep the Takings Clause by disavowing traditional property interests” in assets it wishes to appropriate. Id., at 167. History and precedent dictate that, while the County had the power to sell Tyler’s home to recover the unpaid property taxes, it could not use the tax debt to confiscate more property than was due. Doing so effected a “classic taking in which the government directly appropriates private property for its own use.” Tahoe-Sierra Preservation Council, Inc. v. Tahoe Regional Planning Agency, 535 U. S. 302, 324 (internal quotation marks omitted).

  The principle that a government may not take from a taxpayer more than she owes is rooted in English law and can trace its origins at least as far back as the Magna Carta. From the founding, the new Government of the United States could seize and sell only “so much of [a] tract of land . . . as may be necessary to satisfy the taxes due thereon.” Act of July 14, 1798, §13, 1 Stat. 601. Ten States adopted similar statutes around the same time, and the consensus that a government could not take more property than it was owed held true through the ratification of the Fourteenth Amendment. Today, most States and the Federal Government require excess value to be returned to the taxpayer whose property is sold to satisfy outstanding tax debt.

  The Court’s precedents have long recognized the principle that a taxpayer is entitled to the surplus in excess of the debt owed. See United States v. Taylor, 104 U. S. 216; United States v. Lawton, 110 U. S. 146. Nelson v. City of New York, 352 U. S. 103, did not change that. The ordinance challenged there did not “absolutely preclud[e] an owner from obtaining the surplus proceeds of a judicial sale,” but instead simply defined the process through which the owner could claim the surplus. Id., at 110. Minnesota’s scheme, in comparison, provides no opportunity for the taxpayer to recover the excess value from the State.

  Significantly, Minnesota law itself recognizes in many other contexts that a property owner is entitled to the surplus in excess of her debt. If a bank forecloses on a mortgaged property, state law entitles the homeowner to the surplus from the sale. And in collecting past due taxes on income or personal property, Minnesota protects the taxpayer’s right to surplus. Minnesota may not extinguish a property interest that it recognizes everywhere else to avoid paying just compensation when the State does the taking. Phillips, 524 U. S., at 167. Pp. 4–12.

  (c) The Court rejects the County’s argument that Tyler has no property interest in the surplus because she constructively abandoned her home by failing to pay her taxes. Abandonment requires the “surrender or relinquishment or disclaimer of” all rights in the property, Rowe v. Minneapolis, 51 N. W. 907, 908. Minnesota’s forfeiture law is not concerned about the taxpayer’s use or abandonment of the property, only her failure to pay taxes. The County cannot frame that failure as abandonment to avoid the demands of the Takings Clause. Pp. 12–14.

26 F. 4th 789, reversed.

 Roberts, C. J., delivered the opinion for a unanimous Court. Gorsuch, J., filed a concurring opinion, in which Jackson, J., joined.

Respondent Kevin Younger claims that during his pretrial detention in a Maryland state prison, petitioner Neil Dupree, then a correctional officer lieutenant, ordered three prison guards to attack him. Younger sued Dupree for damages under 42 U. S. C. §1983, alleging excessive use of force. Prior to trial, Dupree moved for summary judgment under Federal Rule of Civil Procedure 56(a), arguing that Younger had failed to exhaust administrative remedies as required by law. Rule 56 requires a district court to enter judgment on a claim or defense if there is “no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” The District Court denied the motion, finding no dispute that the Maryland prison system had internally investigated Younger’s assault, and concluding that this inquiry satisfied Younger’s exhaustion obligation. At trial, Dupree did not present evidence relating to his exhaustion defense. The jury found Dupree and four codefendants liable and awarded Younger $700,000 in damages. Dupree did not file a post-trial motion under Rule 50(b), which allows a disappointed party to file a renewed motion for judgment as a matter of law. He appealed a single issue to the Fourth Circuit: the District Court’s rejection of his exhaustion defense. The Fourth Circuit—bound by its precedent which holds that any claim or defense rejected at summary judgment is not preserved for appellate review unless it was renewed in a post-trial motion—dismissed the appeal.

Held: A post-trial motion under Rule 50 is not required to preserve for appellate review a purely legal issue resolved at summary judgment. In Ortiz v. Jordan, the Court held that an order denying summary judgment on sufficiency-of-the-evidence grounds is not appealable after trial. 562 U. S. 180, 184. Because the factual record developed at trial “supersedes the record existing at the time of the summary-judgment motion,” ibid., it follows that a party must raise a sufficiency claim in a post-trial motion in order to preserve it for appeal, id., at 191–192. That motion allows the district court to take first crack at the question that the appellate court will ultimately face: Was there sufficient evidence in the trial record to support the jury’s verdict?

  The same is not true for pure questions of law resolved in an order denying summary judgment. These conclusions are not “supersede[d]” by later developments in the litigation, id., at 184, and so such rulings merge into the final judgment, at which point they are reviewable on appeal, Quackenbush v. Allstate Ins. Co., 517 U. S. 706, 712. The reviewing court does not benefit from having a district court reexamine a purely legal pretrial ruling after trial, because nothing at trial will have given the district court any reason to question its prior analysis.

  Younger’s counterarguments are unpersuasive. Ortiz does not hold, as Younger contends, that any order denying summary judgment—whether decided on legal or factual grounds—is unreviewable under 28 U. S. C. §1291. While an interlocutory order denying summary judgment is typically not immediately appealable, §1291 does not insulate interlocutory orders from appellate scrutiny, but rather delays their review until final judgment. And while Younger insists there should be no two-track system of summary judgment, in which factual and legal claims follow different routes, nothing in Rule 56 supports his argument for uniformity. On the contrary, fitting the preservation rule to the rationale (factual or legal) underlying the summary-judgment order is consistent with the text of Rule 56. It also makes sense: Factual development at trial will not change the district court’s pretrial answer to a purely legal question, so a post-trial motion requirement would amount to an empty exercise. Finally, while Younger predicts that a separate preservation rule for legal issues will prove unworkable because the line between factual and legal questions can be “vexing” for courts and litigants, Pullman-Standard v. Swint, 456 U. S. 273, 288, experience demonstrates that Younger overstates the need for a bright-line rule. “Courts of appeals have long found it possible to separate factual from legal matters.” Teva Pharmaceuticals USA, Inc. v. Sandoz, Inc., 574 U. S. 318, 328. Here, the Court does not decide whether the issue Dupree raised on appeal is purely legal, and remands for the Fourth Circuit to evaluate that question in the first instance. Pp. 4–9.

Vacated and remanded.

 Barrett, J., delivered the opinion for a unanimous Court.