COUNTY OF MAUI, HAWAII v. HAWAII WILDLIFE FUND, 590 U.S. ___ (2020)


Issues: ,

140 S.Ct. 1462 (2020)

COUNTY OF MAUI, HAWAII, Petitioner
v.
HAWAII WILDLIFE FUND, et al.

No. 18-260.
Supreme Court of United States.

Argued November 6, 2019.
Decided April 23, 2020.
ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT.

Documents

County of Maui, Moana M. Lutey, Richelle M. Thomson, Maui, HI, Hunton Andrews Kurth LLP, Elbert Lin, Michael R. Shebelskie, Richmond, VA, Colleen P. Doyle, Diana Pfeffer Martin, Los Angeles, CA, for Petitioner.

Scott L. Nelson, Public Citizen, Litigation Group, Washington, DC, Amanda C. Leiter, American University, Washington College of Law, Washington, DC, David L. Henkin, Earthjustice, Honolulu, HI, Janette K. Brimmer, Earthjustice, Seattle, WA, for Respondents.

*1465

Syllabus[*]

The Clean Water Act forbids “any addition” of any pollutant from “any point source” to “navigable waters” without an appropriate permit from the Environmental Protection Agency (EPA). §§ 301(a), 502(12), 86 Stat. 844, 886. The Act defines “pollutant” broadly, § 502(6); defines a “point source” as “`any discernible, confined and discrete conveyance … from which pollutants are or may be discharged,'” including, e.g., any “`container,'”

*1466

“`pipe, ditch, channel, tunnel, conduit,'” or “`well,'” § 502(14); and defines the term “discharge of a pollutant” as “`any addition of any pollutant to navigable waters [including navigable streams, rivers, the ocean, or coastal waters] from any point source,'” § 502(12). It then uses those terms in making “unlawful” “`the discharge of any pollutant by any person'” without an appropriate permit. § 301.

Petitioner County of Maui’s wastewater reclamation facility collects sewage from the surrounding area, partially treats it, and each day pumps around 4 million gallons of treated water into the ground through four wells. This effluent then travels about a half mile, through groundwater, to the Pacific Ocean. Respondent environmental groups brought a citizens’ Clean Water Act suit, alleging that Maui was “discharg[ing]” a “pollutant” to “navigable waters” without the required permit. The District Court found that the discharge from Maui’s wells into the nearby groundwater was “functionally one into navigable water,” 24 F.Supp.3d 980, 998, and granted summary judgment to the environmental groups. The Ninth Circuit affirmed, stating that a permit is required when “pollutants are fairly traceable from the point source to a navigable water.” 886 F.3d 737, 749.

Held: The statutory provisions at issue require a permit when there is a direct discharge from a point source into navigable waters or when there is the functional equivalent of a direct discharge. Pp. 1469-1478.

(a) Statutory context limits the reach of the phrase “from any point source” to a range of circumstances narrower than that which the Ninth Circuit’s “fairly traceable” interpretation suggests. At the same time, it is significantly broader than the total exclusion of all discharges through groundwater, as urged by Maui and by the Solicitor General as amicus curiae. Pp. 1469-1470.

(b) The Ninth Circuit’s “fairly traceable” limitation could allow EPA to assert permitting authority over the release of pollutants that reach navigable waters many years after their release. But Congress did not intend to provide EPA with such broad authority. First, to interpret “from” so broadly might require a permit in unexpected circumstances, such as, e.g., the 100-year migration of pollutants through 250 miles of groundwater to a river. Second, the statute’s structure indicates that, as to groundwater pollution and nonpoint source pollution, Congress left substantial responsibility and autonomy to the States and did not give EPA authority that could seriously interfere with this state responsibility. Third, the Act’s legislative history strongly supports the conclusion that the permitting provision does not extend so far. Finally, longstanding regulatory practice shows that EPA has successfully applied the permitting provision to pollution discharges from point sources that reached navigable waters through groundwater using a narrower interpretation than that of the Ninth Circuit. Pp. 1470-1473.

(c) Maui, the Government, and the two dissents argue for interpretations that, in light of the statute’s language, structure, and purposes, are also too extreme. Pp. 1473-1476.

(1) Maui and the Solicitor General argue that the statute’s permitting requirement does not apply if a pollutant, having emerged from a “point source,” must travel through any amount of groundwater before reaching navigable waters. That narrow interpretation would risk serious interference with EPA’s ability to regulate point source discharges, and Congress would not have intended to create such a large and obvious loophole in one of the

*1467

Clean Water Act’s key regulatory innovations. P. 1474.

(2) Reading “from” in the phrase “from any point source” together with “conveyance” in the point source definition “any … conveyance,” Maui argues that the meaning of “from any point source” is not about where the pollution originated, but about how it got there. Thus, Maui claims, a permit is required only if a point source ultimately delivers the pollutant to navigable waters. By contrast, if a pollutant travels through groundwater, then the groundwater is the conveyance and no permit is required. But Maui’s definition of “from” as connoting a means does not fit in context. Coupling “from” with “to” is strong evidence that Congress was referring to a destination (“navigable waters”) and an origin (“any point source”). That Maui’s reading would create a serious loophole in the permitting regime also indicates that it is unreasonable. Pp. 1473-1474.

(3) The Solicitor General argues that the proper interpretation of the statute is the one reflected in EPA’s recent Interpretive Statement, namely, that “all releases of pollutants to groundwater” are excluded from the scope of the permitting program, “even where pollutants are conveyed to jurisdictional surface waters via groundwater.” 84 Fed. Reg. 16810, 16811. That reading, which would open a loophole allowing easy evasion of the statutory provision’s basic purposes, is neither persuasive nor reasonable. EPA is correct that Congress did not require a permit for all discharges to groundwater, and it did authorize study and funding related to groundwater pollution. But the most that the study and funding provisions show is that Congress thought that the problem of pollution in groundwater would primarily be addressed by the States or perhaps by other federal statutes. EPA’s new interpretation is also difficult to reconcile with the statute’s reference to “any addition” of a pollutant to navigable waters; with the statute’s inclusion of “wells” in the “point source” definition, since wells would ordinarily discharge pollutants through groundwater; and with statutory provisions that allow EPA to delegate its permitting authority to a State only if the State, inter alia, provides “`adequate authority'” to “`control the disposal of pollutants into wells,'” § 402(b). Pp. 1473-1475.

(4) Perhaps, as the dissents suggest, the statute’s language could be narrowed by reading the statute to refer only to the pollutant’s immediate origin, but there is no linguistic basis for this limitation. Pp. 1474-1476.

(d) The statute’s words reflect Congress’ basic aim to provide federal regulation of identifiable sources of pollutants entering navigable waters without undermining the States’ longstanding regulatory authority over land and groundwater. The reading of the statute that best captures Congress’ meaning, reflected in the statute’s words, structure, and purposes, is that a permit is required when there is a discharge from a point source directly into navigable waters or when there is the functional equivalent of a direct discharge. Many factors may be relevant to determining whether a particular discharge is the functional equivalent of one directly into navigable waters. Time and distance will be the most important factors in most cases, but other relevant factors may include, e.g., the nature of the material through which the pollutant travels and the extent to which the pollutant is diluted or chemically changed as it travels. Courts will provide additional guidance through decisions in individual cases. The underlying statutory objectives can also provide guidance, and EPA can provide administrative guidance. Although this interpretation

*1468

does not present as clear a line as the other interpretations proffered, the EPA has applied the permitting provision to some discharges through groundwater for over 30 years, with no evidence of inadministrability or an unmanageable expansion in the statute’s scope. Pp. 1475-1478.

886 F.3d 737, vacated and remanded.

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

For U.S. Supreme Court briefs, see: 2019 WL 2068597 (Pet.Brief) 2019 WL 2085683 (Pet.Brief) 2019 WL 3230945 (Resp.Brief) 2019 WL 3801429 (Reply.Brief)

Justice BREYER delivered the opinion of the Court.

The Clean Water Act forbids the “addition” of any pollutant from a “point source” to “navigable waters” without the appropriate permit from the Environmental Protection Agency (EPA). Federal Water Pollution Control Act, §§ 301(a), 502(12)(A), as amended by the Federal Water Pollution Control Act Amendments of 1972 (Clean Water Act) § 2, 86 Stat. 844, 886, 33 U.S.C. §§ 1311(a), 1362(12)(A). The question presented here is whether the Act “requires a permit when pollutants originate from a point source but are conveyed to navigable waters by a nonpoint source,” here, “groundwater.” Pet. for Cert. i. Suppose, for example, that a sewage treatment plant discharges polluted water into the ground where it mixes with groundwater, which, in turn, flows into a navigable river, or perhaps the ocean. Must the plant’s owner seek an EPA permit before emitting the pollutant? We conclude that the statutory provisions at issue require a permit if the addition of the pollutants through groundwater is the functional equivalent of a direct discharge from the point source into navigable waters.

I

A

Congress’ purpose as reflected in the language of the Clean Water Act is to “`restore and maintain the … integrity of the Nation’s waters,'” § 101(a), 86 Stat. 816. Prior to the Act, Federal and State Governments regulated water pollution in large part by setting water quality standards. See EPA v. California ex rel. State Water Resources Control Bd., 426 U.S. 200, 202-203, 96 S.Ct. 2022, 48 L.Ed.2d 578 (1976). The Act restructures federal regulation by insisting that a person wishing to discharge any pollution into navigable waters first obtain EPA’s permission to do so. See id., at 203-205, 96 S.Ct. 2022; Milwaukee v. Illinois, 451 U.S. 304, 310-311, 101 S.Ct. 1784, 68 L.Ed.2d 114 (1981).

*1469

The Act’s provisions use specific definitional language to achieve this result. First, the Act defines “pollutant” broadly, including in its definition, for example, any solid waste, incinerator residue, “`heat,'” “`discarded equipment,'” or sand (among many other things). § 502(6), 86 Stat. 886. Second, the Act defines a “point source” as “`any discernible, confined and discrete conveyance … from which pollutants are or may be discharged,'” including, for example, any “`container,'” “`pipe, ditch, channel, tunnel, conduit,'” or “`well.'” § 502(14), id., at 887. Third, it defines the term “discharge of a pollutant” as “`any addition of any pollutant to navigable waters [including navigable streams, rivers, the ocean, or coastal waters] from any point source.'” § 502(12), id., at 886.

The Act then sets forth a statutory provision that, using these terms, broadly states that (with certain exceptions) “`the discharge of any pollutant by any person'” without an appropriate permit “`shall be unlawful.'” § 301, id., at 844. The question here, as we have said, is whether, or how, this statutory language applies to a pollutant that reaches navigable waters only after it leaves a “point source” and then travels through groundwater before reaching navigable waters. In such an instance, has there been a “discharge of a pollutant,” that is, has there been “any addition of any pollutant to navigable waters from any point source?”

B

The petitioner, the County of Maui, operates a wastewater reclamation facility on the island of Maui, Hawaii. The facility collects sewage from the surrounding area, partially treats it, and pumps the treated water through four wells hundreds of feet underground. This effluent, amounting to about 4 million gallons each day, then travels a further half mile or so, through groundwater, to the ocean.

In 2012, several environmental groups, the respondents here, brought this citizens’ Clean Water Act lawsuit against Maui. See § 505(a), id., at 888. They claimed that Maui was “discharg[ing]” a “pollutant” to “navigable waters,” namely, the Pacific Ocean, without the permit required by the Clean Water Act. The District Court, relying in part upon a detailed study of the discharges, found that a considerable amount of effluent from the wells ended up in the ocean (a navigable water). It wrote that, because the “path to the ocean is clearly ascertainable,” the discharge from Maui’s wells into the nearby groundwater was “functionally one into navigable water.” 24 F.Supp.3d 980, 998 (Haw. 2014). And it granted summary judgment in favor of the environmental groups. See id., at 1005.

The Ninth Circuit affirmed the District Court, but it described the relevant statutory standard somewhat differently. The appeals court wrote that a permit is required when “the pollutants are fairly traceable from the point source to a navigable water such that the discharge is the functional equivalent of a discharge into the navigable water.” 886 F.3d 737, 749 (2018) (emphasis added). The court left “for another day the task of determining when, if ever, the connection between a point source and a navigable water is too tenuous to support liability….” Ibid.

Maui petitioned for certiorari. In light of the differences in the standards adopted by the different Courts of Appeals, we granted the petition. Compare, e.g., 886 F.3d at 749 (“fairly traceable”), with Upstate Forever v. Kinder Morgan Energy Partners, L. P., 887 F.3d 637, 651 (C.A.4 2018) (“direct hydrological connection”), and Kentucky Waterways Alliance v. Kentucky Util. Co., 905 F.3d 925, 932-938

*1470

(C.A.6 2018) (discharges through groundwater are excluded from the Act’s permitting requirements).

II

The linguistic question here concerns the statutory word “from.” Is pollution that reaches navigable waters only through groundwater pollution that is “from” a point source, as the statute uses the word? The word “from” is broad in scope, but context often imposes limitations. “Finland,” for example, is often not the right kind of answer to the question, “Where have you come from?” even if long ago you were born there.

The parties here disagree dramatically about the scope of the word “from” in the present context. The environmental groups, the respondents, basically adopt the Ninth Circuit’s view—that the permitting requirement applies so long as the pollutant is “fairly traceable” to a point source even if it traveled long and far (through groundwater) before it reached navigable waters. They add that the release from the point source must be “a proximate cause of the addition of pollutants to navigable waters.” Brief for Respondents 20.

Maui, on the other hand, argues that the statute creates a “bright-line test.” Brief for Petitioner 27-28. A point source or series of point sources must be “the means of delivering pollutants to navigable waters.” Id., at 28. They add that, if “at least one nonpoint source (e.g., unconfined rainwater runoff or groundwater)” lies “between the point source and the navigable water,” then the permit requirement “does not apply.” Id., at 54. A pollutant is “from” a point source only if a point source is the last “conveyance” that conducted the pollutant to navigable waters.

The Solicitor General, as amicus curiae, supports Maui, at least in respect to groundwater. Reiterating the position taken in a recent EPA “Interpretive Statement,” see 84 Fed. Reg. 16810 (2019), he argues that, given the Act’s structure and history, “a release of pollutants to groundwater is not subject to” the Act’s permitting requirement “even if the pollutants subsequently migrate to jurisdictional surface waters,” such as the ocean. Brief for United States as Amici Curiae 12 (capitalization omitted).

We agree that statutory context limits the reach of the statutory phrase “from any point source” to a range of circumstances narrower than that which the Ninth Circuit’s interpretation suggests. At the same time, it is significantly broader than the total exclusion of all discharges through groundwater described by Maui and the Solicitor General.

III

Virtually all water, polluted or not, eventually makes its way to navigable water. This is just as true for groundwater. See generally 2 Van Nostrand’s Scientific Encyclopedia 2600 (10th ed. 2008) (defining “Hydrology”). Given the power of modern science, the Ninth Circuit’s limitation, “fairly traceable,” may well allow EPA to assert permitting authority over the release of pollutants that reach navigable waters many years after their release (say, from a well or pipe or compost heap) and in highly diluted forms. See, e.g., Brief for Aquatic Scientists et al. as Amici Curiae 13-28.

The respondents suggest that the standard can be narrowed by adding a “proximate cause” requirement. That is, to fall within the permitting provision, the discharge from a point source must “proximately cause” the pollutants’ eventual addition to navigable waters. But the term “proximate cause” derives from general

*1471

tort law, and it takes on its specific content based primarily on “policy” considerations. See CSX Transp., Inc. v. McBride, 564 U.S. 685, 701, 131 S.Ct. 2630, 180 L.Ed.2d 637 (2011) (plurality opinion). In the context of water pollution, we do not see how it significantly narrows the statute beyond the words “fairly traceable” themselves.

Our view is that Congress did not intend the point source-permitting requirement to provide EPA with such broad authority as the Ninth Circuit’s narrow focus on traceability would allow. First, to interpret the word “from” in this literal way would require a permit in surprising, even bizarre, circumstances, such as for pollutants carried to navigable waters on a bird’s feathers, or, to mention more mundane instances, the 100-year migration of pollutants through 250 miles of groundwater to a river.

Second, and perhaps most important, the structure of the statute indicates that, as to groundwater pollution and non-point source pollution, Congress intended to leave substantial responsibility and autonomy to the States. See, e.g., § 101(b), 86 Stat. 816 (stating Congress’ purpose in this regard). Much water pollution does not come from a readily identifiable source. See 3 Van Nostrand’s Scientific Encyclopedia, at 5801 (defining “Water Pollution”). Rainwater, for example, can carry pollutants (say, as might otherwise collect on a roadway); it can pollute groundwater, and pollution collected by unchanneled rainwater runoff is not ordinarily considered point source pollution. Over many decades, and with federal encouragement, the States have developed methods of regulating nonpoint source pollution through water quality standards, and otherwise. See, e.g., Nonpoint Source Program, Annual Report (California) 6 (2016-2017) (discussing state timberland management programs to address addition of sediment-pollutants to navigable waters); id., at 10-11 (discussing regulations of vineyards to control water pollution); id. at 17-19 (discussing livestock grazing management, including utilization ratios and time restrictions); Nonpoint Source Management Program, Annual Report (Maine) 8-10 (2018) (discussing installation of livestock fencing and planting of vegetation to reduce nonpoint source pollution); Oklahoma’s Nonpoint Source Management Program, Annual Report 5, 14 (2017) (discussing program to encourage voluntary no-till farming to reduce sediment pollution).

The Act envisions EPA’s role in managing nonpoint source pollution and groundwater pollution as limited to studying the issue, sharing information with and collecting information from the States, and issuing monetary grants. See §§ 105, 208, 86 Stat. 825, 839; see also Water Quality Act of 1987, § 316, 101 Stat. 52 (establishing Nonpoint Source Management Programs). Although the Act grants EPA specific authority to regulate certain point source pollution (it can also delegate some of this authority to the States acting under EPA supervision, see § 402(b), 86 Stat. 880), these permitting provisions refer to “point sources” and “navigable waters,” and say nothing at all about nonpoint source regulation or groundwater regulation. We must doubt that Congress intended to give EPA the authority to apply the word “from” in a way that could interfere as seriously with States’ traditional regulatory authority—authority the Act preserves and promotes—as the Ninth Circuit’s “fairly traceable” test would.

Third, those who look to legislative history to help interpret a statute will find that this Act’s history strongly supports our conclusion that the permitting provision does not extend so far. Fifty years ago, when Congress was considering the bills that became the Clean Water Act,

*1472

William Ruckelshaus, the first EPA Administrator, asked Congress to grant EPA authority over “ground waters” to “assure that we have control over the water table… so we can … maintai[n] a control over all the sources of pollution, be they discharged directly into any stream or through the ground water table.” Water Pollution Control Legislation—1971 (Proposed Amendments to Existing Legislation): Hearings before the House Committee on Public Works, 92d Cong., 1st Sess., 230 (1971). Representative Les Aspin similarly pointed out that there were “conspicuou[s]” references to groundwater in all sections of the bill except the permitting section at issue here. Water Pollution Control Legislation—1971: Hearings before the House Committee on Public Works on H. R. 11896 and H. R. 11895, 92d Cong., 1st Sess., 727 (1972). The Senate Committee on Public Works “recognize[d] the essential link between ground and surface waters.” S. Rep. No. 92-414, p. 73 (1971).

But Congress did not accept these requests for general EPA authority over groundwater. It rejected Representative Aspin’s amendment that would have extended the permitting provision to groundwater. Instead, Congress provided a set of more specific groundwater-related measures such as those requiring States to maintain “affirmative controls over the injection or placement in wells” of “any pollutants that may affect ground water.” Ibid. These specific state-related programs were, in the words of the Senate Public Works Committee, “designed to protect ground waters and eliminate the use of deep well disposal as an uncontrolled alternative to toxic and pollution control.” Ibid. The upshot is that Congress was fully aware of the need to address groundwater pollution, but it satisfied that need through a variety of state-specific controls. Congress left general groundwater regulatory authority to the States; its failure to include groundwater in the general EPA permitting provision was deliberate.

Finally, longstanding regulatory practice undermines the Ninth Circuit’s broad interpretation of the statute. EPA itself for many years has applied the permitting provision to pollution discharges from point sources that reached navigable waters only after traveling through groundwater. See, e.g., United States Steel Corp. v. Train, 556 F.2d 822, 832 (C.A.7 1977) (permit for “deep waste-injection well” on the shore of navigable waters). But, in doing so, EPA followed a narrower interpretation than that of the Ninth Circuit. See, e.g., In re Bethlehem Steel Corp., 2 E. A. D. 715, 718 (EAB 1989) (Act’s permitting requirement applies only to injection wells “that inject into ground water with a physically and temporally direct hydrologic connection to surface water”). EPA has opposed applying the Act’s permitting requirements to discharges that reach groundwater only after lengthy periods. See McClellan Ecological Seepage Situation (MESS) v. Cheney, 763 F.Supp. 431, 437 (E.D. Cal. 1989) (United States argued that permitting provisions do not apply when it would take “literally dozens, and perhaps hundreds, of years for any pollutants” to reach navigable waters); Greater Yellowstone Coalition v. Larson, 641 F.Supp.2d 1120, 1139 (D. Idaho 2009) (same in respect to instances where it would take “between 60 and 420 years” for pollutants to travel “one to four miles” through groundwater before reaching navigable waters). Indeed, in this very case (prior to its recent Interpretive Statement, see infra, at 1474-1475), EPA asked the Ninth Circuit to apply a more limited “direct hydrological connection” test. See Brief for United States as Amicus Curiae in No. 15-17447 (CA9), pp. 13-20. The Ninth Circuit did not accept this suggestion.

*1473

We do not defer here to EPA’s interpretation of the statute embodied in this practice. Indeed, EPA itself has changed its mind about the meaning of the statutory provision. See infra, at 1474-1475. But this history, by showing that a comparatively narrow view of the statute is administratively workable, offers some additional support for the view that Congress did not intend as broad a delegation of regulatory authority as the Ninth Circuit test would allow.

As we have said, the specific meaning of the word “from” necessarily draws its meaning from context. The apparent breadth of the Ninth Circuit’s “fairly traceable” approach is inconsistent with the context we have just described.

IV

A

Maui and the Solicitor General argue that the statute’s permitting requirement does not apply if a pollutant, having emerged from a “point source,” must travel through any amount of groundwater before reaching navigable waters. That interpretation is too narrow, for it would risk serious interference with EPA’s ability to regulate ordinary point source discharges.

Consider a pipe that spews pollution directly into coastal waters. There is an “addition of” a “pollutant to navigable waters from [a] point source.” Hence, a permit is required. But Maui and the Government read the permitting requirement not to apply if there is any amount of groundwater between the end of the pipe and the edge of the navigable water. See Tr. of Oral Arg. 5-6, 24-25. If that is the correct interpretation of the statute, then why could not the pipe’s owner, seeking to avoid the permit requirement, simply move the pipe back, perhaps only a few yards, so that the pollution must travel through at least some groundwater before reaching the sea? Cf. Brief for State of Maryland et al. as Amici Curiae 9, n. 4. We do not see how Congress could have intended to create such a large and obvious loophole in one of the key regulatory innovations of the Clean Water Act. Cf. California ex rel. State Water Resources Control Bd., 426 U.S. at 202-204, 96 S.Ct. 2022 (basic purpose of Clean Water Act is to regulate pollution at its source); The Emily, 9 Wheat. 381, 390, 6 L.Ed. 116 (1824) (rejecting an interpretation that would facilitate “evasion of the law”).

B

Maui argues that the statute’s language requires its reading. That language requires a permit for a “discharge.” A “discharge” is “any addition” of a pollutant to navigable waters “from any point source.” And a “point source” is “any discernible, confined and discrete conveyance” (such as a pipe, ditch, well, etc.). Reading “from” and “conveyance” together, Maui argues that the statutory meaning of “from any point source” is not about where the pollution originated, but about how it got there. Under what Maui calls the means-of-delivery test, a permit is required only if a point source itself ultimately delivers the pollutant to navigable waters. Under this view, if the pollutant must travel through groundwater to reach navigable waters, then it is the groundwater, not the pipe, that is the conveyance.

Congress sometimes adopts less common meanings of common words, but this esoteric definition of “from,” as connoting a means, does not remotely fit in this context. The statute couples the word “from” with the word “to”—strong evidence that Congress was referring to a destination (“navigable waters”) and an origin (“any point source”). Further underscoring that Congress intended this every

*1474

day meaning is that the object of “from” is a “point source“—a source, again, connoting an origin. That Maui’s proffered interpretation would also create a serious loophole in the permitting regime also indicates it is an unreasonable one.

C

The Solicitor General agrees that, as a general matter, the permitting requirement applies to at least some additions of pollutants to navigable waters that come indirectly from point sources. See Brief for United States as Amicus Curiae 33-35. But the Solicitor General argues that the proper interpretation of the statute is the one reflected in EPA’s recent Interpretive Statement. After receiving more than 50,000 comments from the public, and after the Ninth Circuit released its opinion in this case, EPA wrote that “the best, if not the only, reading” of the statutory provisions is that “all releases of pollutants to groundwater” are excluded from the scope of the permitting program, “even where pollutants are conveyed to jurisdictional surface waters via groundwater.” 84 Fed. Reg. 16810, 16811.

Neither the Solicitor General nor any party has asked us to give what the Court has referred to as Chevron deference to EPA’s interpretation of the statute. See Chevron U.S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 844, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). Even so, we often pay particular attention to an agency’s views in light of the agency’s expertise in a given area, its knowledge gained through practical experience, and its familiarity with the interpretive demands of administrative need. See United States v. Mead Corp., 533 U.S. 218, 234-235, 121 S.Ct. 2164, 150 L.Ed.2d 292 (2001); Skidmore v. Swift & Co., 323 U.S. 134, 139-140, 65 S.Ct. 161, 89 L.Ed. 124 (1944). But here, as we have explained, to follow EPA’s reading would open a loophole allowing easy evasion of the statutory provision’s basic purposes. Such an interpretation is neither persuasive nor reasonable.

EPA correctly points out that Congress did not require a permit for all discharges to groundwater; rather, Congress authorized study and funding related to groundwater pollution. See Brief for United States as Amicus Curiae 15-19. But there is quite a gap between “not all” and “none.” The statutory text itself alludes to no exception for discharges through groundwater. These separate provisions for study and funding that EPA points to would be a “surprisingly indirect route” to convey “an important and easily expressed message”—that the permit requirement simply does not apply if the pollutants travel through groundwater. Landgraf v. USI Film Products, 511 U.S. 244, 262, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994). In truth, the most these provisions show is that Congress thought that the problem of groundwater pollution, as distinct from navigable water pollution, would primarily be addressed by the States or perhaps by other federal statutes.

EPA’s new interpretation is also difficult to reconcile with the statute’s reference to “any addition” of a pollutant to navigable waters. Cf. Milwaukee, 451 U.S. at 318, 101 S.Ct. 1784 (“Every point source discharge is prohibited unless covered by a permit” (footnote omitted)). It is difficult to reconcile EPA’s interpretation with the statute’s inclusion of “wells” in the definition of “point source,” for wells most ordinarily would discharge pollutants through groundwater. And it is difficult to reconcile EPA’s interpretation with the statutory provisions that allow EPA to delegate its permitting authority to a State only if the State (among other things) provides “`adequate

*1475

authority'” to “`control the disposal of pollutants into wells.'” § 402(b), 86 Stat. 881. What need would there be for such a proviso if the federal permitting program the State replaces did not include such discharges (from wells through groundwater) in the first place?

In short, EPA’s oblique argument about the statute’s references to groundwater cannot overcome the statute’s structure, its purposes, or the text of the provisions that actually govern.

D

Perhaps, as the two dissents suggest, the language could be narrowed to similar effect by reading the statute to refer only to the pollutant’s immediate origin. See post, at 1479-1480 (opinion of THOMAS, J.); post, at 1486-1487 (opinion of ALITO, J.). But there is no linguistic basis here to so limit the statute in that way. Again, whether that is the correct reading turns on context. Justice THOMAS insists that in the case of a discharge through groundwater, the pollutants are added “from the groundwater.” Post, at 1480. Indeed, but that does not mean they are not also “from the point source.” Ibid. When John comes to the hotel, John might have come from the train station, from Baltimore, from Europe, from any two of those three places, or from all three. A sign that asks all persons who arrive from Baltimore to speak to the desk clerk includes those who took a taxi from the train station. There is nothing unnatural about such a construction. As the plurality correctly noted in Rapanos v. United States, 547 U.S. 715, 126 S.Ct. 2208, 165 L.Ed.2d 159 (2006), the statute here does not say “directly” from or “immediately” from. Id., at 743, 126 S.Ct. 2208 (opinion of Scalia, J.). Indeed, the expansive language of the provision— any addition from any point source— strongly suggests its scope is not so limited.

Justice ALITO appears to believe that there are only two possible ways to read “from”: as referring either to the immediate source, or else to the original source. Post, at 1484-1485, 1486-1487. Because he agrees that the statute cannot reasonably be read always to reach the original source, he concludes the statute must refer only to the immediate origin. But as the foregoing example illustrates, context may indicate that “from” includes an intermediate stop—Baltimore, not Europe or the train station.

Justice THOMAS relies on the word “addition,” but we fail to see how that word limits the statute to discharges directly to navigable waters. Ordinary language abounds in counter examples: A recipe might instruct to “add the drippings from the meat to the gravy”; that instruction does not become incomprehensible, or even peculiar, simply because the drippings will have first collected in a pan or on a cutting board. And while it would be an unusual phrasing (as statutory phrasings often are), we do not see how the recipe’s meaning would transform if it instead said to “add the drippings to the gravy from the meat.” To take another example: If Timmy is told to “add water to the bath from the well” he will know just what it means—even though he will have to use a bucket to complete the task.

And although Justice THOMAS resists the inevitable implications of his reading of the statute, post, at 1481-1482, that reading would create the same loopholes as those offered by the petitioner and the Government, and more. It would necessarily exclude a pipe that drains onto the beach next to navigable waters, even if the pollutants then flow to those waters. It also seems to exclude a pipe that hangs out over the water and adds pollutants to

*1476

the air, through which the pollutants fall to navigable waters. The absurdity of such an interpretation is obvious enough.

We therefore reject this reading as well: Like Maui’s and the Government’s, it is inconsistent with the statutory text and simultaneously creates a massive loophole in the permitting scheme that Congress established.

E

For the reasons set forth in Part III and in this Part, we conclude that, in light of the statute’s language, structure, and purposes, the interpretations offered by the parties, the Government, and the dissents are too extreme.

V

Over the years, courts and EPA have tried to find general language that will reflect a middle ground between these extremes. The statute’s words reflect Congress’ basic aim to provide federal regulation of identifiable sources of pollutants entering navigable waters without undermining the States’ longstanding regulatory authority over land and groundwater. We hold that the statute requires a permit when there is a direct discharge from a point source into navigable waters or when there is the functional equivalent of a direct discharge. We think this phrase best captures, in broad terms, those circumstances in which Congress intended to require a federal permit. That is, an addition falls within the statutory requirement that it be “from any point source” when a point source directly deposits pollutants into navigable waters, or when the discharge reaches the same result through roughly similar means.

Time and distance are obviously important. Where a pipe ends a few feet from navigable waters and the pipe emits pollutants that travel those few feet through groundwater (or over the beach), the permitting requirement clearly applies. If the pipe ends 50 miles from navigable waters and the pipe emits pollutants that travel with groundwater, mix with much other material, and end up in navigable waters only many years later, the permitting requirements likely do not apply.

The object in a given scenario will be to advance, in a manner consistent with the statute’s language, the statutory purposes that Congress sought to achieve. As we have said (repeatedly), the word “from” seeks a “point source” origin, and context imposes natural limits as to when a point source can properly be considered the origin of pollution that travels through groundwater. That context includes the need, reflected in the statute, to preserve state regulation of groundwater and other nonpoint sources of pollution. Whether pollutants that arrive at navigable waters after traveling through groundwater are “from” a point source depends upon how similar to (or different from) the particular discharge is to a direct discharge.

The difficulty with this approach, we recognize, is that it does not, on its own, clearly explain how to deal with middle instances. But there are too many potentially relevant factors applicable to factually different cases for this Court now to use more specific language. Consider, for example, just some of the factors that may prove relevant (depending upon the circumstances of a particular case): (1) transit time, (2) distance traveled, (3) the nature of the material through which the pollutant travels, (4) the extent to which the pollutant is diluted or chemically changed as it travels, (5) the amount of pollutant entering the navigable waters relative to the amount of the pollutant that leaves the point source, (6) the manner by or area in which the pollutant enters the

*1477

navigable waters, (7) the degree to which the pollution (at that point) has maintained its specific identity. Time and distance will be the most important factors in most cases, but not necessarily every case.

At the same time, courts can provide guidance through decisions in individual cases. The Circuits have tried to do so, often using general language somewhat similar to the language we have used. And the traditional common-law method, making decisions that provide examples that in turn lead to ever more refined principles, is sometimes useful, even in an era of statutes.

The underlying statutory objectives also provide guidance. Decisions should not create serious risks either of undermining state regulation of groundwater or of creating loopholes that undermine the statute’s basic federal regulatory objectives.

EPA, too, can provide administrative guidance (within statutory boundaries) in numerous ways, including through, for example, grants of individual permits, promulgation of general permits, or the development of general rules. Indeed, over the years, EPA and the States have often considered the Act’s application to discharges through groundwater.

Both Maui and the Government object that to subject discharges to navigable waters through groundwater to the statute’s permitting requirements, as our interpretation will sometimes do, would vastly expand the scope of the statute, perhaps requiring permits for each of the 650,000 wells like petitioner’s or for each of the over 20 million septic systems used in many Americans’ homes. Brief for Petitioner 44-48; Brief for United States as Amicus Curiae 24-25. Cf. Utility Air Regulatory Group v. EPA, 573 U.S. 302, 324, 134 S.Ct. 2427, 189 L.Ed.2d 372 (2014).

But EPA has applied the permitting provision to some (but not to all) discharges through groundwater for over 30 years. See supra, at 1472-1473. In that time we have seen no evidence of unmanageable expansion. EPA and the States also have tools to mitigate those harms, should they arise, by (for example) developing general permits for recurring situations or by issuing permits based on best practices where appropriate. See, e.g., 40 CFR § 122.44(k) (2019). Judges, too, can mitigate any hardship or injustice when they apply the statute’s penalty provision. That provision vests courts with broad discretion to set a penalty that takes account of many factors, including “any good-faith efforts to comply” with the Act, the “seriousness of the violation,” the “economic impact of the penalty on the violator,” and “such other matters as justice may require.” See 33 U.S.C. § 1319(d). We expect that district judges will exercise their discretion mindful, as we are, of the complexities inherent to the context of indirect discharges through groundwater, so as to calibrate the Act’s penalties when, for example, a party could reasonably have thought that a permit was not required.

In sum, we recognize that a more absolute position, such as the means-of-delivery test or that of the Government or that of the Ninth Circuit, may be easier to administer. But, as we have said, those positions have consequences that are inconsistent with major congressional objectives, as revealed by the statute’s language, structure, and purposes. We consequently understand the permitting requirement, § 301, as applicable to a discharge (from a point source) of pollutants that reach navigable waters after traveling through groundwater if that discharge is the functional equivalent of a direct discharge from the point source into navigable waters.

*1478

VI

Because the Ninth Circuit applied a different standard, we vacate its judgment and remand the case for further proceedings consistent with this opinion.

It is so ordered.

Justice KAVANAUGH, concurring.

I join the Court’s opinion in full. I write separately to emphasize three points.

First, the Court’s interpretation of the Clean Water Act regarding pollution “from” point sources adheres to the interpretation set forth in Justice Scalia’s plurality opinion in Rapanos v. United States, 547 U.S. 715, 126 S.Ct. 2208, 165 L.Ed.2d 159 (2006). The Clean Water Act requires a permit for “any addition of any pollutant to navigable waters from any point source.” 33 U.S.C. § 1362(12)(A); see §§ 1311(a), 1342(a). The key word is “from.” The question in this case is whether the County of Maui needs a permit for its Lahaina Wastewater Reclamation Facility. No one disputes that pollutants originated at Maui’s wastewater facility (a point source), and no one disputes that the pollutants ended up in the Pacific Ocean (a navigable water). Maui contends, however, that it does not need a permit. Maui says that the pollutants did not come “from” the Lahaina facility because the pollutants traveled through groundwater before reaching the ocean.

Justice Scalia’s plurality opinion in Rapanos explained why Maui’s interpretation of the Clean Water Act is incorrect. In that case, Justice Scalia stated that polluters could not “evade the permitting requirement of § 1342(a) simply by discharging their pollutants into noncovered intermittent watercourses that lie upstream of covered waters.” 547 U.S. at 742-743, 126 S.Ct. 2208. Justice Scalia reasoned that the Clean Water Act does not merely “forbid the `addition of any pollutant directly to navigable waters from any point source,’ but rather the `addition of any pollutant to navigable waters.’ Thus, from the time of the CWA’s enactment, lower courts have held that the discharge into intermittent channels of any pollutant that naturally washes downstream likely violates § 1311(a), even if the pollutants discharged from a point source do not emit `directly into’ covered waters, but pass `through conveyances’ in between.” Id., at 743, 126 S.Ct. 2208 (citations omitted).

In other words, under Justice Scalia’s interpretation in Rapanos, the fact that the pollutants from Maui’s wastewater facility reach the ocean via an indirect route does not itself exempt Maui’s facility from the Clean Water Act’s permitting requirement for point sources. The Court today adheres to Justice Scalia’s analysis in Rapanos on that issue.

Second, as Justice Scalia’s opinion in Rapanos pointed out and as the Court’s opinion today explains, the statute does not establish a bright-line test regarding when a pollutant may be considered to have come “from” a point source. The source of the vagueness is Congress’ statutory text, not the Court’s opinion. The Court’s opinion seeks to translate the vague statutory text into more concrete guidance.

Third, JUSTICE THOMAS’ dissent states that “the Court does not commit” to “which factors are the most important” in determining whether pollutants that enter navigable waters come “from” a point source. Post, at 1481-1482. That critique is not accurate, as I read the Court’s opinion. The Court identifies relevant factors to consider and emphasizes that “[t]ime and distance are obviously important.” Ante, at 1476. And the Court expressly adds that “[t]ime and distance will be the most important factors in most cases, but

*1479

not necessarily every case.” Ante, at 1477. Although the statutory text does not supply a bright-line test, the Court’s emphasis on time and distance will help guide application of the statutory standard going forward.

With those additional comments, I join the Court’s opinion in full.

Justice THOMAS, with whom Justice GORSUCH joins, dissenting.

The Clean Water Act (CWA) requires a federal permit for “the discharge of any pollutant by any person.” 33 U.S.C. § 1311(a); see § 1342. The CWA defines a “discharge” as “any addition of any pollutant to navigable waters from any point source.” § 1362(12).[1] Based on the statutory text and structure, I would hold that a permit is required only when a point source discharges pollutants directly into navigable waters. The Court adopts this interpretation in part, concluding that a permit is required for “a direct discharge.” Ante, at 1475-1476. But the Court then departs from the statutory text by requiring a permit for “the functional equivalent of a direct discharge,” ibid., which it defines through an open-ended inquiry into congressional intent and practical considerations. Because I would adhere to the text, I respectfully dissent.

I

A

In interpreting the statutory definition of “discharge,” the Court focuses on the word “from,” but the most helpful word is “addition.” That word, together with “to” and “from,” limits the meaning of “discharge” to the augmentation of navigable waters.

Dictionary definitions of “addition” denote an augmentation or increase. Webster’s Third New International Dictionary defines “addition” as “the act or process of adding: the joining or uniting of one thing to another.” Webster’s Third New International Dictionary 24 (1961); see also ibid. (listing “increase” and “augmentation” as synonyms for “addition”). Other dictionary definitions from around the time of the statute’s enactment are in accord. See, e.g., American Heritage Dictionary 14, 15 (1981) (defining “addition” as “[t]he act or process of adding” and defining “add” as “[t]o join or unite so as to increase in size, quantity, or scope”); see also Webster’s New International Dictionary 29, 30 (2d ed. 1957) (defining “addition” as the “[a]ct, process, or instance of adding,” and defining “add” as to “join or unite, as one thing to another, or as several particulars, so as to increase the number, augment the quantity, enlarge the magnitude, or so as to form into one aggregate”).

The inclusion of the term “addition” in the CWA indicates that the statute excludes anything other than a direct discharge. When a point source releases pollutants to groundwater, one would naturally say that the groundwater has been augmented with pollutants from the point source. If the pollutants eventually reach navigable waters, one would not naturally say that the navigable waters have been augmented with pollutants from the point

*1480

source. The augmentation instead occurs with pollutants from the groundwater.

The prepositions “from” and “to” reinforce this reading. When pollutants are released from a point source to another point source or groundwater, they are added to the second from the first. If the pollutants are later released to navigable waters, they are added to the navigable waters from the second point source or the groundwater. One would not naturally say that the pollutants are added to the navigable waters from the original point source.

Interpreting “discharge” to mean a direct discharge makes sense of other parts of the definition as well. It respects the statutory definition of a point source as a “conveyance,” see § 1362(14), because a point source that releases pollutants directly into navigable waters is a means of conveyance. And it makes sense of the word “any” before “point source,” because that term clarifies that any kind of point source may require a permit.

The structure of the CWA confirms this interpretation. It authorizes the Environmental Protection Agency (EPA) to regulate discharges from point sources, including through the permitting process, but it reserves to the States the primary responsibility for regulating other sources of pollution, including groundwater. With respect to these sources, the EPA merely collects information, coordinates with the States, and provides funding. See 33 U.S.C. §§ 1252(a), 1254(a)(5), 1282(b)(2), 1288, 1314(a), 1329; ante, at 1470-1472. In the CWA, Congress expressly stated its “policy … to recognize, preserve, and protect the primary responsibilities and rights of States to prevent, reduce, and eliminate pollution.” § 1251(b). Thus, construing the EPA’s power to regulate point sources to allow the agency to regulate nonpoint sources and groundwater is in serious tension with Congress’ design.

My reading is also consistent with our decision in South Fla. Water Management Dist. v. Miccosukee Tribe, 541 U.S. 95, 124 S.Ct. 1537, 158 L.Ed.2d 264 (2004). The petitioner in that case argued that no permit was required when a point source was not the original source of the pollutant but instead conveyed the pollutant from further up a chain of sources. Id., at 104, 124 S.Ct. 1537. We rejected that argument because “a point source need not be the original source of the pollutant; it need only convey the pollutant to `navigable waters.'” Id., at 105, 124 S.Ct. 1537. Although that case did not involve the exact question presented here, the direct-discharge interpretation comports well with that previous decision.

B

The Court’s main textual argument reads the word “from” in isolation. But as the Court recognizes, “the word `from’ necessarily draws its meaning from context.” Ante, at 1473. The Court’s example using “arrive” instead of “addition” is thus unpersuasive, ante, at 1474-1475, because “from” takes different meanings with different verbs. The Court’s culinary example also misses the mark, ante, at 1475, because if the drippings from the meat collect in the pan before the chef adds them to the gravy, the drippings are added to the gravy from the pan, not from the meat. This point becomes clear if we reorder the majority’s recipe to match the statute; the chef has not added the drippings to the gravy from the meat. The Court’s bathwater example, ante, at 1475, suffers from the same problem; if the well water is put in a bucket before it is put in the bathtub, it is added to the bathtub from the bucket. Only by reading the phrase in its entirety can we interpret the definition of “discharge.”

*1481

See Deal v. United States, 508 U.S. 129, 132, 113 S.Ct. 1993, 124 L.Ed.2d 44 (1993).

The Court also asserts that a narrower reading than the one it adopts would create a “massive loophole” in the statute. Ante, at 1476. Far from creating a loophole, my reading is the most logical because it is consonant with the scope of Congress’ power. The CWA presumably was passed as an exercise of Congress’ authority “to regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes.” U.S. Const., Art. I, § 8, cl. 3. My interpretation ties the statute more closely to navigable waters, on the theory that they are at least a channel of these kinds of commerce.

Further, the Court’s interpretation creates practical problems of its own. As the Court acknowledges, its opinion gives almost no guidance, save for a list of seven factors. But the Court does not commit to whether those factors are the only relevant ones, whether those factors are always relevant, or which factors are the most important. See ante, at 1475-1477. It ultimately does little to explain how functionally equivalent an indirect discharge must be to require a permit.[2]

The Court suggests that the EPA could clarify matters through “administrative guidance,” ante, at 1477, but so far the EPA has provided only limited advice and recently shifted its position, see 84 Fed. Reg. 16810 (2019); ante, at 1473-1474. In any event, the sort of “`general rules'” that the Court hopes the EPA will promulgate are constitutionally suspect. See Department of Transportation v. Association of American Railroads, 575 U.S. 43, 67-87, 135 S.Ct. 1225, 191 L.Ed.2d 153 (2015) (THOMAS, J., concurring in judgment).

Despite giving minimal guidance as to how this case should be decided on remand, the majority speculates about whether a permit would be required in other factual circumstances. It poses the examples of a pipe that releases pollutants over navigable waters and a pipe that releases pollutants onto land near navigable waters. As an initial matter, I am not as sure as the majority that a “pollutant,” as defined by the CWA, may be added to the air.[3] Even if the majority is correct that a permit is not required in these hypothetical cases, drawing the line at discharges to water is not so absurd as to undermine the most natural reading of the statute. In any event, it is unnecessary to decide these hypothetical cases today.

*1482

Finally, the Court speculates as to “those circumstances in which Congress intended to require a federal permit.” Ante, at 1476. But we are not a superlegislature (or super-EPA) tasked with making good policy—assuming that is even what the Court accomplishes today. “Our job is to follow the text even if doing so will supposedly undercut a basic objective of the statute.” Baker Botts L. L. P. v. ASARCO LLC, 576 U.S. 121, 135, 135 S.Ct. 2158, 192 L.Ed.2d 208 (2015) (internal quotation marks omitted).

II

I do agree with the Court on several points. First, the interpretation adopted by respondents and the Ninth Circuit is unsupportable. That interpretation—which would require permits for discharges that are “`fairly traceable'” to, and proximately caused by, a point source—is atextual and unsettles the CWA’s careful balance between federal regulation of point-source pollution and state regulation of nonpoint-source pollution. Ante, at 1470-1473.

Second, I agree that the interpretation adopted by petitioner and Justice ALITO reads the word “any” unnaturally, ante, at 1473-1474, although the majority appears to deploy that argument itself in another part of the opinion, ante, at 1475. Petitioner’s and Justice ALITO’s interpretation also gives insufficient weight to the meaning of “addition,” see supra, at 1479-1480.

Third, I agree that the EPA’s interpretation is not entitled to deference for at least two reasons: No party requests it, and the EPA’s reading is not the best one. Ante, at 1474-1475. I add only that deference under Chevron U.S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984), likely conflicts with the Vesting Clauses of the Constitution. See Baldwin v. United States, 589 U.S. ___, ___-___, 140 S.Ct. 690, 690-694, ___ L.Ed.2d ___ (2020) (THOMAS, J., dissenting from denial of certiorari); Michigan v. EPA, 576 U.S. 743, 761-764, 135 S.Ct. 2699, 192 L.Ed.2d 674 (2015) (THOMAS, J., concurring); see also Perez v. Mortgage Bankers Assn., 575 U.S. 92, 115-126, 135 S.Ct. 1199, 191 L.Ed.2d 186 (2015) (THOMAS, J., concurring in judgment).

Finally, I agree with the Court’s implicit conclusion that Rapanos v. United States, 547 U.S. 715, 126 S.Ct. 2208, 165 L.Ed.2d 159 (2006), does not resolve this case. That plurality opinion, which I joined, observed that lower courts have required a permit when pollutants pass through a chain of point sources. Id., at 743-744, 126 S.Ct. 2208. But we expressly said in Rapanos that “we [did] not decide this issue.” Id., at 743, 126 S.Ct. 2208. We are not bound by dictum in a plurality opinion or by the lower court opinions it cited.

III

The best reading of the statute is that a “discharge” is the release of pollutants directly from a point source to navigable waters. The application of this interpretation to the undisputed facts of this case makes a remand unnecessary. Petitioner operates a wastewater treatment facility and injects treated wastewater into four underground injection control wells. All parties agree that the wastewater enters groundwater from the wells and does not directly enter navigable waters. Based on these undisputed facts, there is no “discharge,” so I would reverse the judgment of the Ninth Circuit. I respectfully dissent.

Justice ALITO, dissenting.

If the Court is going to devise its own legal rules, instead of interpreting those enacted by Congress, it might at least adopt rules that can be applied with a modicum of consistency. Here, however,

*1483

the Court makes up a rule that provides no clear guidance and invites arbitrary and inconsistent application.

The text of the Clean Water Act generally requires a permit when a discharge “from” a “point source” (such as a pipe) “add[s]” a pollutant “to” navigable waters (such as the Pacific Ocean). 33 U.S.C. § 1362(12). There are two ways to read this text. A pollutant that reaches the ocean could be understood to have been added “from” a pipe if the pipe originally discharged the pollutant and the pollutant eventually made its way to the ocean by flowing over or under the surface of the ground. Or a pollutant that reaches the ocean could be understood to have come “from” a pipe if the pollutant is discharged from the pipe directly into the ocean.

There is no comprehensible alternative to these two interpretations, but the Court refuses to accept either. Both alternatives, it believes, lead to unacceptable results, and it therefore tries to find a middle way. It holds that a permit is required “when there is a direct discharge from a point source into navigable waters or when there is the functional equivalent of a direct discharge.” Ante, at 1476. This is not a plausible interpretation of the statutory text and, to make matters worse, the Court’s test has no clear meaning.

Just what is the “functional equivalent” of a “direct discharge”? The Court provides no real answer. All it will say is that the distance a pollutant travels and the time this trip entails are the most important factors, but at least five other factors may have a bearing on the question, and even this list is not exhaustive. Ante, at 1476-1477. Entities like water treatment authorities that need to know whether they must get a permit are left to guess how this nebulous standard will be applied. Regulators are given the discretion, at least in the first instance, to make of this standard what they will. And the lower courts? The Court’s advice, in essence, is: “That’s your problem. Muddle through as best you can.”

I

Petitioner, the County of Maui (County), built the Lahaina Wastewater Reclamation Facility in the 1970s. Excerpts of Record 304. The facility receives sewage and then discharges treated wastewater into wells (essentially long pipes) that extend 200 feet or more below ground level. Id., at 694-695. Some of this discharge enters an aquifer below the facility. Id., at 696.

In all the years of its operation, the facility has never had a National Pollution Discharge Elimination System (NPDES) permit for discharges from the wells, a fact that has been well known to both the EPA and the Hawaii Department of Health (HDOH). The EPA helped to finance the construction of the facility with a Clean Water Act grant. Id., at 141. In 1973, before breaking ground on the facility, the County prepared an environmental impact report and shared it with the EPA and the HDOH. Id., at 140, 342. The report predicted that effluent injected into groundwater from the wells would “eventually reach the ocean some distance from the shore.” Id., at 342. Both the EPA and the HDOH received and submitted comments on the report without any mention of a need for permitting discharges from the wells. Id., at 140. Six years later, the HDOH issued an NPDES permit to the facility—but not for the wells. (The permit covered separate discharges to the Honokowai Stream.) Id., at 141, 223-224. And in a May 1985 NPDES Compliance Monitoring Report, the EPA concluded that the County was operating in compliance with the permit, because all effluent was entering the injection wells—and was thus destined for groundwater rather than for navigable waters or for use in irrigation. Id.,

*1484

at 141, 222. In 1994, HDOH again informed the EPA that “all experts agree that the wastewater does enter the ocean.” Id., at 369. And again—nothing from the federal authorities.

Thus, despite nearly five decades of notice that effluent from the facility would make, or was making, its way via groundwater to the ocean, neither the EPA nor the HDOH required NPDES permitting for the Lahaina wells. App. to Pet. for Cert. 138, 143. Indeed, none of the more than 6,600 underground injection wells in Hawaii currently has an NPDES permit.[1]

In 2012, however, as the Court recounts, respondents filed a citizen suit claiming that the Lahaina facility was violating the Clean Water Act by discharging pollutants into the ocean without a permit. The District Court granted summary judgment against the County on the issue of liability because pollutants “can be directly traced from the injection wells to the ocean.” 24 F.Supp.3d 980, 998 (Haw. 2014) (emphasis deleted).

The parties then entered into a conditional settlement

that would take effect if the County were unsuccessful on appeal. Under that agreement, the County must: make good-faith efforts to obtain and comply with an NPDES permit; pay $100,000 in civil penalties; spend $2.5 million on a “supplemental environmental project” in the western part of the island of Maui; and pay nearly $1 million for respondents’ attorney’s fees and other costs of litigation.[2]

On appeal, the Ninth Circuit affirmed on the ground that pollutants that eventually reached the ocean were “fairly traceable” to the wells. 886 F.3d 737, 749 (2018). We granted review and must now decide whether the Court of Appeals erred in holding that the discharge of effluent from the wells into groundwater requires a permit.

II

The Clean Water Act generally makes it unlawful to “discharge” a “pollutant”[3] without a permit. 33 U.S.C. § 1311(a). The Act defines the “discharge of a pollutant” as “any addition of any pollutant to navigable waters[[4]] from any point source.” § 1362(12). And a “point source” is broadly defined as “any discernible, confined and discrete conveyance … from which pollutants are or may be discharged.” § 1362(14). The Act includes a non-exhaustive list of conveyances that fall within this

*1485

definition, and included on that list are such things as “pipe[s],” “ditch[es],” “channel[s],” and “well[s].” Ibid.

Putting all these statutory terms together, the rule can be stated as follows: A permit is required when a pollutant is “add[ed]” to navigable waters “from” a “point source.” In this case, the parties and the EPA agree that most of the elements of this rule are met. Specifically, they agree that: The effluent emitted by the wells is a “pollutant”; this effluent reaches navigable waters (the Pacific Ocean); and the wells are “point source[s].” The disputed question is whether the emission of effluent from those wells qualifies as a “discharge,” that is, the addition of a pollutant “from” a point source. § 1362(12) (emphasis added).

There are two possible interpretations of this phrase. The first is that pollutants are added to navigable waters from a point source whenever they originally came from the point source. The second is that pollutants are added to navigable waters only if they were discharged from a point source directly into navigable waters.

Dissatisfied with those options, the Court tries to find a third, but its interpretation is very hard to fit into the statutory text. Under the Court’s interpretation, it appears that a pollutant that leaves a point source and heads toward navigable waters via some non-point source (such as by flowing over the ground or by means of groundwater) is “from” the point source for some portion of its journey, but once it has travelled a certain distance or once a certain amount of time has elapsed, it is no longer “from” the point source and is instead “from” a non-point source.

This is an implausible reading of the statute. The Court has many inventive examples of the different meanings that can be conveyed by the simple statement that A comes from B, but one of the Court’s examples—the traveler who flies from Europe to Baltimore—illustrates the problem. If we apply the Court’s interpretation of § 1362 to this traveler’s journey, he would be “from” Europe for the first part of the flight, but at some point he might cease to be “from” Europe and would then be from someplace else, maybe Greenland or geographical coordinates in the middle of the Atlantic. This is a very strange notion, and therefore, I think the statutory text compels us to choose between the two alternatives set out above.

The Court rejects both of these because it thinks they lead to unacceptably extreme results. “Originally from” would impose liability even if pollutants discharged into ground water had to travel 250 miles over the course of 100 years before reaching navigable waters. See ante, at 1470-1471. And “`immediately'” or “`directly’ from,” the Court thinks, would mean that a polluter could evade the permit requirement by discharging pollutants from a pipe located just a few feet from navigable waters. Ante, at 1475-1476.

To escape these possibilities, the Court devises its own test: A permit is required, the Court holds, “when there is a direct discharge from a point source into navigable waters or when there is the functional equivalent of a direct discharge.” Ante, at 1476 (emphasis in original). The Clean Water Act, however, says nothing about “the functional equivalent” of a direct discharge. That is the Court’s own concoction, and the Court provides no clear explanation of its meaning.

The term “functional equivalent” may have a quasi-technical ring, but what does it mean? “Equivalent” means “equal” in some respect, and “functional” signifies a relationship to a function. The function of a direct discharge from a point source into navigable waters is to convey the entirety

*1486

of the discharge into navigable waters without any delay. Therefore, the “functional equivalent” of a direct discharge of a pollutant into navigable waters would seem to be a discharge that is equal to a direct discharge in these respects.

If that is what the Court meant by “the functional equivalent of a direct discharge,” the test would apply at best to only a small set of situations not involving a direct discharge. The Court’s example of a pipe that emits pollutants a few feet from the ocean would presumably qualify on de minimis grounds, but if the pipe were moved back any significant distance, the discharge would not be exactly equal to a direct discharge. There would be some lag from the time of the discharge to the time when the pollutant reaches navigable waters; some of the pollutant might not reach that destination; and the pollutant might have changed somewhat in composition by the time it reached the navigable waters.

For these reasons, the Court’s reference to “the functional equivalent of a direct discharge,” if taken literally, would be of little importance, but the Court’s understanding of this concept is very different from the literal meaning of the phrase. As used by the Court, “the functional equivalent of a direct discharge” means a discharge that is sufficiently similar to a direct discharge to warrant a permit in light of the Clean Water Act’s “language, structure, and purposes.” See ante, at 1477-1478. But what, in concrete terms, does this mean? How similar is sufficiently similar?

The Court provides this guidance. It explains that time and distance are the most important factors, ante, at 1476-1477, but it does not set any time or distance limits except to observe that a permit is needed where the discharge is a few feet away from navigable waters and that a permit is not required where the discharge is far away and it takes “many years” for the pollutants to complete the journey. Ante, at 1475-1477. Beyond this, the Court provides a list (and a non-exhaustive one at that!) of five other factors that may be relevant: “the nature of the material through which the pollutant travels,” “the extent to which the pollutant is diluted or chemically changed as it travels,” “the amount of pollutant entering the navigable waters relative to the amount of the pollutant that leaves the point source,” “the manner by or area in which the pollutant enters the navigable waters,” and “the degree to which the pollution (at that point) has maintained its specific identity.” Ante, at 1476-1477.

The Court admits that its rule “does not, on its own, clearly explain how to deal with middle instances,” ibid., but that admission does not go far enough. How the rule applies to “middle instances” is anybody’s guess. Except in extreme cases, dischargers will be able to argue that the Court’s multifactor test does not require a permit. Opponents will be able to make the opposite argument. Regulators will be able to justify whatever result they prefer in a particular case. And judges will be left at sea.

III

A

Instead of concocting our own rule, I would interpret the words of the statute, and in my view, the better of the two possible interpretations is that a permit is required when a pollutant is discharged directly from a point source to navigable waters. This interpretation is consistent with the statutory language and better fits the overall scheme of the Clean Water Act. And properly understood, it does not have the sort of extreme consequences that the Court finds unacceptable.

*1487

Take the Court’s example of a pipe that discharges pollutants a short distance from the ocean. Ante, at 1473. This pipe qualifies as a point source. 33 U.S.C. § 1362(14). If its discharge goes directly into another point source and that point source discharges directly into navigable waters, there is a direct discharge into navigable waters, and a permit is needed. See Rapanos v. United States, 547 U.S. 715, 743-744, 126 S.Ct. 2208, 165 L.Ed.2d 159 (2006) (plurality opinion).[5]

That a permit is required in this situation is important because the Clean Water Act’s definition of a “point source” is very broad, and as a result, many discharges onto the surface of land are likely to be covered. As noted, “point source[s]” include “ditch[es]” and “channel[s],” as well as “any discernible, confined and discrete conveyance … from which pollutants … may be discharged.” § 1362(14). Therefore if water discharged on the surface of the land finds or creates a passage leading to navigable waters, a permit may be required if the course that the discharge takes is (1) a “conveyance” that is (2) “discernible” and (3) “confined.”

*1488

Those three requirements are rather easily satisfied. When a liquid flows over the surface of land to navigable waters, the surface is a conveyance, i.e., a “means of carrying or transporting something” from one place to another. Webster’s Third New International Dictionary 499 (1971) (Webster’s Third); Random House Dictionary of the English Language 320 (1967) (Random House).[6] This conveyance would be “discernible,” i.e., capable of being seen. Webster’s Third 644; Random House 409. And it would be “confined,” i.e., held within bounds, see Webster’s Third 476; Random House 308, if the topography of the land in question imposes some boundaries on its flow.

If the term “point source” is read in this way, it would have a broad reach and would cover many of the cases that trouble the Court. Moreover—and I find this point particularly important—even if this interpretation fails to capture every case that seems to call out for regulation, that would not mean that these cases would necessarily remain unchecked. The States have the authority to regulate the discharge of pollutants by non-point sources. See 33 U.S.C. §§ 1285(j), 1314(f), 1329(i), 1329(b)(1), (h). They are entrusted with a vital role under the Clean Water Act, and there is no reason to believe that they would tolerate cases of abuse.

The interpretation I have outlined is not only consistent with the statutory language; it is strongly supported by the Clean Water Act’s regulatory scheme for at least two reasons. First, it respects Congress’ decision to treat point-source pollution differently from non-point-source pollution, including pollution conveyed by groundwater. See 84 Fed. Reg. 16832.[7] The Court itself recognizes this:

“[T]he structure of the statute indicates that, as to groundwater pollution and non[-]point source pollution, Congress intended to leave substantial responsibility and autonomy to the States.” Ante, at 1471.

“Over many decades, and with federal encouragement, the States have developed methods of regulating nonpoint source pollution through water quality standards, and otherwise.” Ibid.

“The Act envisions EPA’s role in managing non[-]point source pollution and groundwater pollution as limited to studying the issue, sharing information with and collecting information from the States, and issuing monetary grants.” Ante, at 1471.

Point sources are readily identifiable and therefore more susceptible to uniform nationwide regulation. Non-point source pollution, on the other hand, often presents

*1489

more complicated issues that are better suited to individualized local solutions. See Shanty Town Assocs. L. P. v. EPA, 843 F.2d 782, 791 (C.A.4 1988) (“[T]he control of non[-]point source pollution was so dependent on such site-specific factors as topography, soil structure, rainfall, vegetation, and land use that its uniform federal regulation was virtually impossible”); Natural Resources Defense Council v. EPA, 915 F.2d 1314, 1316 (C.A.9 1990) (“The Act focused on point source polluters presumably because they could be identified and regulated more easily than non[-]point source polluters”); Brief for State of West Virginia et al. as Amici Curiae 14-18.

Second, this bright-line rule is consistent with the Act’s remedial scheme. The Clean Water Act imposes a regime of strict liability, §§ 1311, 1342, 1344, backed by criminal penalties and steep civil fines, § 1319. Thus, “the consequences to landowners even for inadvertent violations can be crushing.” Army Corps of Engineers v. Hawkes Co., 578 U.S. ___, ___, 136 S.Ct. 1807, 1816, 195 L.Ed.2d 77 (2016) (Kennedy, J., concurring). The Act authorizes as much as $54,833 in fines per day (or more than $20 million per year), 40 CFR § 19.4; 84 Fed. Reg. 2059, and contains a 5-year statute of limitations, 28 U.S.C. § 2462. And the availability of citizen suits only exacerbates the danger to ordinary landowners. Even when the EPA and the relevant state agency conclude that a permit is not needed, there is always the possibility that a citizen suit will result in a very costly judgment. The interpretation set out above, by providing a relatively straight-forward rule, provides a measure of fair notice and promotes good-faith compliance.

B

The alternative way in which the statutory language could be interpreted—reading “from” to mean “originally from”— would lead to extreme results, as the Court recognizes. And while state regulation could fill any unwarranted gaps left by the interpretation I have outlined, there would be no apparent remedy for the overreach that would result from interpreting “from” to mean “originally from.”

The extreme consequences of that interpretation are shown most dramatically by its potential application to ordinary homeowners with septic tanks, a problem that the EPA highlighted in a recent Interpretive Statement. See Interpretive Statement on Application of the Clean Water Act NPDES Program to Releases of Pollutants From a Point Source to Groundwater, 84 Fed. Reg. 16824 (2019). Septic systems —used by 26 million American homes—generally operate by “discharging liquid effluent into perforated pipes buried in a leach field, chambers, or other special units designed to slowly release the effluent into the soil.” Id., at 16812. That effluent then percolates through the soil and “can in certain circumstances ultimately enter groundwater.” Ibid.[8] Congress most certainly did not intend that ordinary homeowners with septic systems obtain NPDES permits—or that they face severe penalties for failing to do so. That, however, is where this alternative interpretation would lead.

*1490

And the same is true for the test adopted by the Ninth Circuit. The Ninth Circuit held that a permit is required if a pollutant that reaches navigable waters is “fairly traceable,” but there is no real difference between “fairly traceable” and “originally from.” Unless a pollutant is “traceable” to a point source, how could that point source be required to get a permit? And the addition of the qualifier “fairly” does not seem to add anything. What would it mean for a pollutant to be “unfairly traceable” to a point source? Traceable only as a result of a method that is scientifically unsound? In that situation, why would a court consider the pollutant to be traceable to the source in question at all? So if a pollutant can be reliably determined to have originally come from a point source, a permit would appear to be required under the Ninth Circuit’s test.

Respondents, instead of defending the Ninth Circuit’s interpretation, argue that a discharge from a point source must be the “proximate cause” of a pollutant’s reaching navigable waters. Brief for Respondents 12. But as the Court concludes, ante, at 1470-1471, there is no basis for transplanting this concept from the law of torts into the Clean Water Act, and it is unclear what it would mean in that context.

For these reasons, of the two possible interpretations of the statutory terms, the better is the interpretation that reads “from” to mean “directly from.”

C

Even if the Court were to find § 1362(12) ambiguous, applicable clear statement rules foreclose the “functional equivalent” standard and favor the test just described. The Court has required a clear statement of congressional intent when an administrative agency seeks to interpret a statute in a way that entails “a significant impingement of the States’ traditional and primary power over land and water use,” Solid Waste Agency of Northern Cook Cty. v. Army Corps of Engineers, 531 U.S. 159, 174, 121 S.Ct. 675, 148 L.Ed.2d 576 (2001) (SWANCC), and when it adopts a new and expanded interpretation of a statute, Utility Air Regulatory Group v. EPA, 573 U.S. 302, 134 S.Ct. 2427, 189 L.Ed.2d 372 (2014) (UARG). The same rules should apply here where what is at issue is a new theory propounded by private plaintiffs.

First, the Court’s “functional equivalent” test unquestionably impinges on the States’ traditional authority. In SWANCC, the Court struck down the Army Corps of Engineers’ “Migratory Bird Rule” as inconsistent with the Clean Water Act because the rule effectively displaced state authority over land and water use. In this case, the federalism interest is even stronger because the Clean Water Act itself assigns non-point-source-pollution regulation to the States and explicitly recognizes and protects the state role in environmental protection. 33 U.S.C. § 1251(b). The “functional equivalent” standard expands federal point source regulation at the expense of state non-point source regulation. And as a practical matter, States would be saddled with the costs of increased NPDES permitting (because States generally award permits in place of the EPA), while exercising diminished control over non-point source pollution within their territory. See Brief for State of West Virginia et al. as Amici Curiae 27-34.

Second, the Court’s test offends the clear-statement rule recognized in UARG by expanding the authority of the EPA. Congress must speak clearly if it “wishes to assign to an agency decisions of vast `economic and political significance.'” 573 U.S. at 324, 134 S.Ct. 2427 (quoting FDA v. Brown & Williamson Tobacco Corp.,

*1491

529 U.S. 120, 160, 120 S.Ct. 1291, 146 L.Ed.2d 121 (2000)). In UARG, the EPA had promulgated greenhouse-gas emission standards for stationary sources that “constitute[d] an `unprecedented expansion of EPA authority that would have a profound effect on virtually every sector of the economy and touch every household in the land.'” 573 U.S. at 310-311, 134 S.Ct. 2427 (quoting 73 Fed. Reg. 44355 (2008)). The statutory scheme, designed for large stationary sources like factories, would have been extended to smaller sources like hotels and retail establishments. The number of permits (and associated expenses) would have skyrocketed.

Here, as the EPA explained in a recent Interpretive Statement, the Fourth and Ninth Circuit “discharge” tests—which I struggle to distinguish from the “functional equivalent” formulation—broaden the Act’s coverage to “potentially swee[p] into the scope of the statute commonplace and ubiquitous activities such as releases from homeowners’ backyard septic systems.” 84 Fed. Reg. 16823.

IV

The Court does little to justify its newfound standard, other than to point to certain past EPA enforcement actions, see ante, at 1472-1473, 1477, but the EPA’s position on the regulation of groundwater has been anything but consistent. It is true, as the Court recounts, that the EPA has required NPDES permits for the discharge of some pollutants that migrate through groundwater before reaching navigable waters. See ante, at 1472-1473. But the EPA has contradicted itself on this important question multiple times. See Brief for Edison Electric Institute et al. as Amici Curiae 21-32 (reviewing EPA NPDES interpretations and permitting practices).

In the Act’s earliest years, the EPA deputy general counsel wrote in a formal memorandum that “[d]ischarges into ground waters” do not require NPDES permits. Memorandum to EPA Region IX Regional Counsel 2-3 (Dec. 13, 1973).[9] More recently, the EPA recognized “conflicting legal precedents” on this question. Compare NPDES Permit Regulation and Effluent Limitation Guidelines and Standards for Concentrated Animal Feeding Operations (CAFOs), 68 Fed. Reg. 7216 (2003), with 66 Fed. Reg. 3018 (2001).

Similarly, in its 2019 Interpretive Statement, the EPA acknowledged its “[l]ack of consistent and comprehensive direction” on this issue. See 84 Fed. Reg. 16820; see also Brief for Edison Electric Institute et al. as Amici Curiae 21-32 (recounting EPA historical approach to NPDES permitting). But it added that “the best, if not the only, reading of the [Act] is that all releases to groundwater are excluded from the scope of the NPDES program, even where pollutants are conveyed to jurisdictional surface waters via groundwater.” 84 Fed. Reg. 16814.

In short, the EPA’s inconsistent position on the groundwater issue does not provide a sufficient basis for the Court’s new “functional equivalent” test.

* * *

The Court adopts a nebulous standard, enumerates a non-exhaustive list of potentially relevant factors, and washes its hands of the problem. We should not require regulated parties to “feel their way on a case-by-case basis” where the costs of uncertainty are so great. Rapanos, 547 U.S. at 758, 126 S.Ct. 2208 (ROBERTS,

*1492

C.J., concurring). The Court’s decision invites “arbitrary and inconsistent decisionmaking.” UARG, 573 U.S. at 350, 134 S.Ct. 2427 (ALITO, J., concurring in part and dissenting in part). And “[t]hat is not what the Clean [Water] Act contemplates.” Ibid.

I would reverse the judgment below and instruct the lower courts to apply the test set out above. I therefore respectfully dissent.

Footnotes

[*] 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, 26 S.Ct. 282, 50 L.Ed. 499.

[1] The CWA defines “navigable waters” as “the waters of the United States, including the territorial seas.” § 1362(7). It defines a “point source” as “any discernible, confined and discrete conveyance, including but not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock, concentrated animal feeding operation, or vessel or other floating craft, from which pollutants are or may be discharged,” excluding “agricultural stormwater discharges and return flows from irrigated agriculture.” § 1362(14).

[2] Justice KAVANAUGH believes that the Court’s opinion provides enough guidance when it states that “[t]ime and distance will be the most important factors in most cases, but not necessarily every case,” ante, at 1477 (majority opinion) (emphasis added). See ante, at 1478 (concurring opinion). His hope for guidance appears misplaced. For all we know, these factors may not be the most important in 49 percent of cases. The majority’s nonexhaustive seven-factor test “may aid in identifying relevant facts for analysis, but— like most multifactor tests—it leaves courts adrift once those facts have been identified.” Dietz v. Bouldin, 579 U.S. ___, ___, 136 S.Ct. 1885, 1898, 195 L.Ed.2d 161 (2016) (THOMAS, J., dissenting); see also Scalia, The Rule of Law as a Law of Rules, 56 U. Chi. L. Rev. 1175, 1186-1187 (1989) (noting that “when balancing is the mode of analysis, not much general guidance may be drawn from the opinion” and arguing that “totality of the circumstances tests and balancing modes of analysis” should “be avoided where possible”).

[3] The CWA defines a “pollutant” as “dredged spoil, solid waste, incinerator residue, sewage, garbage, sewage sludge, munitions, chemical wastes, biological materials, radioactive materials, heat, wrecked or discarded equipment, rock, sand, cellar dirt and industrial, municipal, and agricultural waste discharged into water,” with certain exceptions. § 1362(6).

[1] EPA, FY 2018 State Underground Injection Control Inventory, https://www.epa.gov/uic/uic-injection-well-inventory; EPA, Hawaii NPDES Permits: Draft and Final NPDES Permits, https://www.epa.gov/npdes-permits/hawaii-npdes-permits.

[2] Settlement Agreement and Order re: Remedies in No. 1:12cv198, Doc. 259 (Haw.); Stipulated Settlement Agreement Regarding Award of Plaintiffs’ Costs of Litigation, ibid., Doc. 267 (Haw.).

[3] The Act defines a “pollutant” as:

“dredged spoil, solid waste, incinerator residue, sewage, garbage, sewage sludge, munitions, chemical wastes, biological materials, radioactive materials, heat, wrecked or discarded equipment, rock, sand, cellar dirt and industrial, municipal, and agricultural waste discharged into water….” § 1362(6).

 

[4] The Act defines “navigable waters” as “waters of the United States, including the territorial seas.” § 1362(7). The term “navigable waters” has a well-known meaning, but the broader term “waters of the United States” is not defined by the Clean Water Act and has presented a difficult issue for this Court. See Rapanos v. United States, 547 U.S. 715, 126 S.Ct. 2208, 165 L.Ed.2d 159 (2006). The EPA’s definition of “waters of the United States” expressly excludes groundwater. see 40 CFR § 122.2 (2019); 84 Fed. Reg. 4190 (2019), and no party in this case disputes that interpretation.

[5] Justice THOMAS describes his preferred holding in similar terms: “[A] permit is required only when a point source discharges pollutants directly into navigable waters.” Ante, at 1479 (dissenting opinion). But I take Justice THOMAS’s opinion to foreclose liability in one situation where I believe a permit would be required: a discharge from multiple, linked point sources. In my view, a permit is required in that instance because a pollutant would ultimately be added to navigable waters directly from a point source.

Justice Scalia’s opinion in Rapanos, 547 U.S. at 743-744, 126 S.Ct. 2208 (plurality opinion), supports this conclusion. Rapanos addressed the meaning of the term “waters of the United States,” and Justice Scalia’s opinion concluded that this term does not apply to “[w]etlands with only an intermittent, physically remote hydrologic connection to [such waters].” Id., at 742, 126 S.Ct. 2208. At one point in his opinion, Justice Scalia responded to the argument that this interpretation would allow polluters to evade the permit requirement “simply by discharging their pollutants into noncovered intermittent watercourses that lie upstream of covered waters.” Id., at 743, 126 S.Ct. 2208. Arguing that this was not likely to occur, he identified two lines of lower court authority that would prevent such evasion, but he did not endorse either. Ibid. One of these lines was based on exactly the interpretation set out in this opinion, namely, that “such upstream, intermittently flowing channels themselves constitute `point sources'” under the Act’s broad definition of that term. Ibid. The other line, as described in Justice Scalia’s opinion, “held that the discharge into intermittent channels of any pollutant that naturally washes downstream likely [requires a permit] even if the pollutants discharged from a point source do not emit `directly into’ covered waters, but pass `through conveyances’ in between.” Ibid. (emphasis in original). To the extent these lower court cases are understood as holding that a permit is required whenever a pollutant “naturally” reaches waters of the United States, their reasoning would conflict with the Court’s rejection of the theory that a permit is required whenever a pollutant that originated from a point source ultimately reaches covered waters. But as Justice Scalia noted, in the two cases he cited, the pollutants were discharged from point sources into “conveyances” that, in turn, brought the pollutants to covered waters. Ibid. And the conveyances in both cases, a sewer system and tunnel, ibid., could easily fall within the broad definition of a point source.

In short, at least one and perhaps both of the lines of lower court cases to which Justice Scalia referred are fully consistent with the interpretation set out in this opinion. The same is true of his statement, discussed by Justice KAVANAUGH, ante, at 1478-1479 (concurring opinion), that the Clean Water Act “does not forbid the `addition of any pollutant directly to navigable waters from any point source.'” 547 U.S. at 743, 126 S.Ct. 2208. As noted, Justice Scalia’s opinion is open to the possibility that a permit is required if point source A discharges into point source B, and point source B then discharges into covered waters. Thus, his opinion apparently regards that situation as involving an indirect discharge. I would describe that discharge as direct because point source B discharges directly into covered waters, but the difference is purely semantic.

 

[6] As we have said, the Act’s point-source “definition makes plain that a point source need not be the original source of the pollutant; it need only convey the pollutant to `navigable waters,’ which are, in turn, defined as `the waters of the United States.'” South Fla. Water Management Dist. v. Miccosukee Tribe, 541 U.S. 95, 105, 124 S.Ct. 1537, 158 L.Ed.2d 264 (2004). The label is a bit of a misnomer: Although labeled “point sources,” “[t]ellingly, the examples … listed by the Act include pipes, ditches, tunnels, and conduits, objects that do not themselves generate pollutants but merely transport them.” Ibid. (citing § 1362(14)).

[7] The Act contains a number of references to groundwater (a non-point source) outside the NPDES context. The Act textually distinguishes groundwater from surface water and navigable waters, § 1252(a), provides funding for state regulation of groundwater pollution, and suggests that groundwater is a non-point source. See § 1329(h)(5)(D) (authorizing EPA to prioritize grants to States that have implemented or proposed “carry[ing] out groundwater quality protection activities which [EPA] determines are part of a comprehensive non[-]point source pollution control program”).

[8] According to the EPA, numerous other conveyances that deposit pollutants into groundwater could now require NPDES permits. “Activities listed by commentators included aquifer recharge, leaks from sewage collection systems, … treatment systems such as constructed wetlands, spills and accidental releases, manure management, and coal ash impoundment seepage.” 84 Fed. Reg. 16812. The County and amici also assert that respondents’ theory would require permits for green infrastructure, water reuse, and groundwater discharge. See, e.g., Brief for National Association of Clean Water Agencies et al. as Amici Curiae 20-26.

[9] This early understanding, as the Court describes, is consistent with the legislative history, which shows that Congress intentionally left regulation of groundwater pollution to the States. See ante, at 1471-1472.