| Syllabus | Opinion [ Scalia ] | Concurrence [ Breyer ] | Concurrence [ Ginsburg ] | Dissent [ Stevens ] |
|---|---|---|---|---|
| HTML version PDF version | HTML version PDF version | HTML version PDF version | HTML version PDF version | HTML version PDF version |
VERMONT AGENCY OF NATURAL RESOURCES,
PETITIONER v. UNITED STATES
ex rel. STEVENS
ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT
[May 22, 2000]
Justice Stevens, with whom Justice Souter joins, dissenting.
In 1986, Congress amended the False Claims Act (FCA or Act) to create a new procedure known as a “civil investigative demand,” which allows the Attorney General to obtain documentary evidence “for the purpose of ascertaining whether any person is or has been engaged in” a violation of the Act–including a violation of 31 U.S.C. § 3729. The 1986 amendments also declare that a “person” who could engage in a violation of §3729–thereby triggering the civil investigative demand provision–includes “any State or political subdivision of a State.” See §6(a), 100 Stat. 3168 (codified at 31 U.S.C. § 3733(l)(1)(A), (2), (4)). In my view, this statutory text makes it perfectly clear that Congress intended the term “person” in §3729 to include States. This understanding is supported by the legislative history of the 1986 amendments, and is fully consistent with this Court’s construction of federal statutes in cases decided before those amendments were enacted.
Since the FCA was amended in 1986, however, the Court has decided a series of cases that cloak the States with an increasingly protective mantle of “sovereign immunity” from liability for violating federal laws. It is through the lens of those post-1986 cases that the Court has chosen to construe the statute at issue in this case. To explain my disagreement with the Court, I shall comment on pre-1986 cases, the legislative history of the 1986 amendments, and the statutory text of the FCA–all of which support the view that Congress understood States to be included within the meaning of the word “person” in §3729. I shall then briefly explain why the State’s constitutional defenses fail, even under the Court’s post-1986 construction of the doctrine of sovereign immunity.
I
Cases decided before 1986 uniformly support the proposition that the broad language used in the False Claims Act means what it says. Although general statutory references to “persons” are not normally construed to apply to the enacting sovereign, United States v. Mine Workers, 330 U.S. 258, 275 (1947), when Congress uses that word in federal statutes enforceable by the Federal Government or by a federal agency, it applies to States and state agencies as well as to private individuals and corporations. Thus, for example, the word “person” in the Sherman Act does not include the sovereign that enacted the statute (the Federal Government), United States v. Cooper Corp., 312 U.S. 600 (1941), but it does include the States, Georgia v. Evans, 316 U.S. 159 (1942). Similarly, States are subject to regulation as a “person” within the meaning of the Shipping Act of 1916, California v. United States, 320 U.S. 577 (1944), and as a “common carrier” within the meaning of the Safety Appliance Act, United States v. California, 297 U.S. 175 (1936). In the latter case, the State of California “invoke[d] the canon of construction that a sovereign is presumptively not intended to be bound” by a statute unless the act expressly declares that to be the case. Id., at 186. We rejected the applicability of that canon, stating:
“We can perceive no reason for extending it so as to exempt a business carried on by a state from the otherwise applicable provisions of an act of Congress, all-embracing in scope and national in its purpose, which is as capable of being obstructed by state as by individual action. Language and objectives so plain are not to be thwarted by resort to a rule of construction whose purpose is but to resolve doubts, and whose application in the circumstances would be highly artificial.” Id., at 186—187.1
The False Claims Act is also all-embracing in scope, national in its purpose, and as capable of being violated by state as by individual action.2 It was enacted during the Civil War, shortly after a congressional committee had decried the “fraud and peculation” by state officials in connection with the procurement of military supplies and Government contracts–specifically mentioning the purchases of supplies by the States of Illinois, Indiana, New York, and Ohio. See H. R. Rep. No. 2, 37th Cong., 2d Sess., pt. ii—a, pp. xxxviii—xxxix (1862). Although the FCA was not enacted until the following year, the Court of Appeals for the Second Circuit correctly observed that “it is difficult to suppose that when Congress considered the bills leading to the 1863 Act a year later it either meant to exclude the States from the ‘persons’ who were to be liable for the presentation of false claims to the federal government or had forgotten the results of this extensive investigation.” 162 F.3d 195, 206 (1998). That observation is faithful to the broad construction of the Act that this Court consistently endorsed in cases decided before 1986 (and hardly requires any “suspension of disbelief” as the majority supposes, ante, at 16, n. 12).
Thus, in United States v. Neifert&nbhyph;White Co., 390 U.S. 228, 232 (1968), after noting that the Act was passed as a result of investigations of the fraudulent use of federal funds during the Civil War, we inferred “that the Act was intended to reach all types of fraud, without qualification, that might result in financial loss to the Government.” See also Rainwater v. United States, 356 U.S. 590, 592 (1958) (“It seems quite clear that the objective of Congress [in the FCA] was broadly to protect the funds and property of the Government from fraudulent claims”); H. R. Rep. No. 99—660, p. 18 (1986) (“[T]he False Claims Act is used as … the primary vehicle by the Government for recouping losses suffered through fraud”). Indeed, the fact that Congress has authorized qui tam actions by private individuals to supplement the remedies available to the Federal Government provides additional evidence of its intent to reach all types of fraud that cause financial loss to the Federal Government. Finally, the breadth of the “claims” to which the FCA applies
The legislative history of the 1986 amendments discloses that both federal and state officials understood that States were “persons” within the meaning of the statute. Thus, in a section of the 1986 Senate Report describing the history of the Act, the committee unequivocally stated that the Act reaches all parties who may submit false claims and that “[t]he term ‘person’ is used in its broad sense to include partnerships, associations, and corporations . . . as well as States and political subdivisions thereof.” S. Rep. No. 99—345, pp. 8—9.5
Indeed, a few federal courts had accepted jurisdiction in qui tam cases brought by the States–thus indicating their view that States were included among the “persons” who may bring qui tam actions as relators under §3730(b)(1). See United States ex rel. Woodard v. Country View Care Center, Inc., 797 F.2d 888 (CA10 1986); United States ex rel. Wisconsin v. Dean, 729 F.2d 1100 (CA7 1984); see also United States ex rel. Hartigan v. Palumbo Bros., Inc., 797 F. Supp. 624 (ND Ill. 1992). Not only do these cases express the view of those federal judges who thought a State could be a “person” under §3730(b)(1), but the cases also demonstrate that the States considered themselves to be statutory “persons.” In fact, in the Dean case, the United States filed a statement with the court explicitly stating its view that “[t]he State is a proper relator.” 729 F.2d, at 1103, n. 2. And when the Seventh Circuit in that case dismissed Wisconsin’s qui tam claim on grounds unrelated to the definition of the word “person,” the National Association of Attorneys General adopted a resolution urging Congress to make it easier for States to be relators.6 When Congress amended the FCA in 1986–and enacted the word “person” in §3729 at issue here–it had all of this information before it, i.e., that federal judges had accepted States as relators (and hence as “persons”); that the States considered themselves to be statutory “persons” and wanted greater freedom to be “persons” who could sue under the Act; and that the United States had taken a like position. See S. Rep. No. 99—345, at 12—13.
In sum, it is quite clear that when the 1986 amendments were adopted, there was a general understanding that States and state agencies were “persons” within the meaning of the Act.
II
The text of the 1986 amendments confirms the pre-existing understanding. The most significant part of the amendments is the enactment of a new §3733 granting authority to the Attorney General to issue a civil investigative demand (CID) before commencing a civil proceeding on behalf of the United States. A series of interwoven definitions in §3733 unambiguously demonstrates that a State is a “person” who can violate §3729.
Section 3733 authorizes the Attorney General to issue a CID when she is conducting a “false claims law investigatio[n].” §3733(a). A “false claims law investigation” is defined as an investigation conducted “for the purpose of ascertaining whether any person is or has been engaged in any violation of a false claims law.” §3733(l)(2) (emphasis added). And a “false claims law” includes §3729–the provision at issue in this case. §3733(l)(1)(A). Quite plainly, these provisions contemplate that any “person” may be engaged in a violation of §3729. Finally, a “person” is defined to include “any State or political subdivision of a State.” §3733(l)(4). Hence, the CID provisions clearly state that a “person” who may be “engaged in any violation of a false claims law,” including §3729, includes a “State or a political subdivision of a State.”7 These CID provisions thus unmistakably express Congress’ understanding that a State may be a “person” who can violate §3729.
Elsewhere in the False Claims Act the term “person” includes States as well. For example, §3730 of the Act–both before and after the 1986 amendments–uses the word “person” twice. First, subsection (a) of §3730 directs the Attorney General to investigate violations of §3729, and provides that if she “finds that a person has violated or is violating” that section, she may bring a civil action “under this section against the person.” (Emphases added.) Second, subsection (b) of §3730 also uses the word “person,” though for a different purpose; in that subsection the word is used to describe the plaintiffs who may bring qui tam actions on behalf of themselves and the United
States.
Quite clearly, a State is a “person” against whom the Attorney General may proceed under §3730(a).8 And as I noted earlier, see supra, at 6, before 1986 States were considered “persons” who could bring a qui tam action as a relator under §3730(b)–and the Court offers nothing to question that understanding. See ante, at 21, n. 18. Moreover, when a qui tam relator brings an action on behalf of the United States, he or she is, in effect, authorized to act as an assignee of the Federal Government’s claim. See ante, at 6. Given that understanding, combined with the fact that §3730(a) does not make any distinction between possible defendants against whom the Attorney General may bring an action, the most normal inference to draw is that qui tam actions may be brought by relators against the same category of “persons” that may be sued by the Attorney General.
To recapitulate, it is undisputed that (under the CID provision) a State is a “person” who may violate §3729; that a State is a “person” who may be named as a defendant in an action brought by the Attorney General; and that a State is a “person” who may bring a qui tam action on behalf of the United States. It therefore seems most natural to read the adjacent uses of the term “person” in §§3729, 3730(a), 3730(b), and 3733 to cover the same category of defendants. See United States v. Cooper Corp., 312 U.S., at 606 (“It is hardly credible that Congress used the term ‘person’ in different senses in the same sentence”). And it seems even more natural to read the single word “person” (describing who may commit a violation under §3729) to have one consistent meaning regardless of whether the action against that violator is brought under §3730(a) or under §3730(b). See Ratzlaf v. United States, 510 U.S. 135, 143 (1994) (“A term appearing in several places in a statutory text is generally read the same way each time it appears. We have even stronger cause to construe a single formulation … the same way each time it is called into play”). Absent powerful arguments to the contrary, it should follow that a State may be named as a defendant in an action brought by an assignee of the United States. Rather than pointing to any such powerful arguments, however, the Court comes to a contrary conclusion on the basis of an inapplicable presumption and rather strained inferences drawn from three different statutory provisions.
The Court’s principal argument relies on “our longstanding interpretive presumption that ‘person’ does not include the sovereign.” Ante, at 13. As discussed earlier, that “presumption” does not quite do the heavy lifting the Court would like it to do. What’s more, the doctrinal origins of that “presumption” meant only that the enacting sovereign was not normally thought to be a statutory “person.” See, e.g., United States v. California, 297 U.S., at 186 (“[T]he canon of construction that a sovereign is presumptively not intended to be bound by its own statute unless named in it … has its historical basis in the English doctrine that the Crown is unaffected by acts of Parliament not specifically directed against it. The presumption is an aid to consistent construction of statutes of the enacting sovereign when their purpose is in doubt” (emphasis added)); see also United States v. Mine Workers, 330 U.S., at 275; United States v. Fox, 94 U.S. 315, 321 (1877); Will v. Michigan Dept. of State Police, 491 U.S. 58, 73 (1989) (Brennan, J., dissenting). The reason for presuming that an enacting sovereign does not intend to authorize litigation against itself simply does not apply to federal statutes that apply equally to state agencies and private entities. Finally, the “affirmative showing” the Court would require to demonstrate that the word “person” includes States, ante, at 14, is plainly found in the statutory text discussed above.
The Court’s first textual argument is based on the fact that the definition of the term “person” included in §3733’s CID provision expressly includes States. “The presence of such a definitional provision in §3733,” the Court argues, “together with the absence of such a provision from the definitional provisions contained in §3729 … suggests that States are not ‘persons’ for purposes of qui tam liability under §3729.” Ante, at 17. Leaving aside the fact that §3733’s definition actually cuts in the opposite direction, see supra, at 7—8, this argument might carry some weight if the definitional provisions in §3729 included some definition of “person” but simply neglected to mention States. But the definitional provisions in §3729 do not include any definition of “person” at all. The negative inference drawn by the Court, if taken seriously, would therefore prove too much. The definition of “person” in §3733 includes not only States, but also “any natural person, partnership, corporation, association, or other legal entity.” §3733(l)(4). If the premise of the Court’s argument were correct–that the inclusion of certain items as a “person” in §3733 implies their exclusion as a “person” in §3729–then there would be absolutely no one left to be a “person” under §3729.9 It is far more reasonable to assume that Congress simply saw no need to add a definition of “person” in §3729 because (as both the legislative history, see supra, at 3—7, and the definitions in the CID provisions demonstrate) the meaning of the term “person” was already well understood. Congress likely thought it unnecessary to include a definition in §3729 itself.
The Court also relies on the definition of “person” in a separate, but similar, statute, the Program Fraud Civil Remedies Act of 1986 (PFCRA). Ante, at 19—20. The definition of “person” found in that law includes “any individual, partnership, corporation, association, or private organization.” 31 U.S.C. § 3801(a)(6). It is first worth pointing out the obvious: Although the PFCRA sits next to the False Claims Act in the United States Code, they are separate statutes. It is therefore not altogether clear why the former has much bearing on the latter.10 Regardless, the Court’s whole argument about the PFCRA rests entirely on the premise that its definition of “person” does not include States. That premise, in turn, relies upon the fact that §3801(a)(6) in the PFCRA defines a “person” to include “any individual, partnership, corporation, association, or private organization,” but does not mention States. We have, however, interpreted similar definitions of “person,” which included corporations, partnerships, and associations, to include States as well, even though States were not expressly mentioned in the statutory definition. See California v. United States, 320 U.S., at 585; Georgia v. Evans, 316 U.S., at 160. (I draw no definitive conclusions as to whether States are subject to suit under the PFCRA; I only mean to suggest that the Court’s premise is not as obvious as it presumes it to be.) In any event, the ultimate relevant question is whether the text and legislative history of the False Claims Act make it clear that §3729’s use of the word “person” includes States. Because they do, nothing in any other piece of legislation narrows the meaning of that term.
Finally, the Court relies on the fact that the current version of the FCA includes a treble damages remedy that is “essentially punitive in nature.” Ante, at 18. Citing Newport v. Fact Concerts, Inc., 453 U.S. 247, 262—263 (1981), the Court invokes the “presumption against imposition of punitive damages on governmental entities.” Ante, at 18. But as Newport explains, “courts vie[w] punitive damages [against governmental bodies] as contrary to sound public policy, because such awards would burden the very taxpayers and citizens for whose benefit the wrongdoer was being chastised.” 453 U.S., at 263. That rationale is inapplicable here. The taxpaying “citizens for whose benefit” the False Claims Act is designed are the citizens of the United States, not the citizens of any individual State that might violate the Act. It is true, of course, that the taxpayers of a State that violates the FCA will ultimately bear the burden of paying the treble damages. It is not the coffers of the State (and hence state taxpayers), however, that the FCA is designed to protect, but the coffers of the National Government (and hence the federal taxpayers). Accordingly, a treble damages remedy against a State does not “burden the very taxpayers” the statute was designed to protect.11
III
Each of the constitutional issues identified in the Court’s opinion requires only a brief comment. The historical evidence summarized by the Court, ante, at 7—10, is obviously sufficient to demonstrate that qui tam actions are “cases” or “controversies” within the meaning of Article III. That evidence, together with the evidence that private prosecutions were commonplace in the 19th century, see Steel Co. v. Citizens for Better Environment, 523 U.S. 83, 127—128, and nn. 24—25 (1998) (Stevens, J., concurring in judgment), is also sufficient to resolve the Article II question that the Court has introduced sua sponte, ante, at 11, n. 8.
As for the State’s “Eleventh Amendment” sovereign immunity defense, I adhere to the view that Seminole Tribe of Fla. v. Florida, 517 U.S. 44 (1996), was wrongly decided. See Kimel v. Florida Bd. of Regents, 528 U.S. __, __ (2000) (Stevens, J., dissenting) (slip op., at 6—7); Seminole Tribe, 517 U.S., at 100—185 (Souter, J., dissenting). Accordingly, Congress’ clear intention to subject States to qui tam actions is also sufficient to abrogate any common-law defense of sovereign immunity. Moreover, even if one accepts Seminole Tribe as controlling, the State’s immunity claim would still fail. Given the facts that (1) respondent is, in effect, suing as an assignee of the United States, ante, at 6; (2) the Eleventh Amendment does not provide the States with a defense to claims asserted by the United States, see, e.g., United States v. Mississippi, 380 U.S. 128, 140 (1965) (“[N]othing in [the Eleventh Amendment] or any other provision of the Constitution prevents or has ever been seriously supposed to prevent a State’s being sued by the United States”); and (3) the Attorney General retains significant control over a relator’s action, see 162 F.3d, at 199—201 (case below), the Court of Appeals correctly affirmed the District Court’s order denying petitioner’s motion to dismiss. Compare New Hampshire v. Louisiana, 108 U.S. 76 (1883), with South Dakota v. North Carolina, 192 U. S 286 (1904).12 I would, accordingly, affirm the judgment of the Court of Appeals.
Notes
1. The difference between the post-1986 lens through which the Court views sovereign immunity issues, on the one hand, and the actual intent of Congress in statutes like the one before us today, on the other hand, is well illustrated by the congressional rejection of the holdings in Hoffman v. Connecticut Dept. of Income Maintenance, 492 U.S. 96 (1989), and United States v. Nordic Village, Inc., 503 U.S. 30 (1992). In those cases, the Court refused to find the necessary unequivocal waiver of sovereign immunity against both the States and the Federal Government in §106(c) of the Bankruptcy Code. Congress, however, thought differently: “In enacting section 106(c), Congress intended … to make the States subject to a money judgment. But the Supreme Court in Hoffman v. Connecticut Department of Income Maintenance, 492 U.S. 96 (1989), held [otherwise.] In using such a narrow construction, the Court . . . did not find in the text of the statute an ‘unmistakenly clear’ intent of Congress to waive sovereign immunity … . The Court applied this reasoning in United States v. Nordic Village, Inc.” See 140 Cong. Rec. 27693 (1994). Congress therefore overruled both of those decisions by enacting the current version of 11 U.S.C. § 106.
2. It is thus at the opposite pole from the statute construed in Wilson v. Omaha Tribe, 442 U.S. 653 (1979), which held that the term “white person” did not include the State of Iowa because “it is apparent that in adopting §22 Congress had in mind only disputes arising in Indian country, disputes that would not arise in or involve any of the States.” Id., at 668.
3. Title 31 U.S.C. § 3729(c) reads: “For purposes of this section, ‘claim’ includes any request or demand, whether under contract or otherwise, for money or property which is made to a contractor, grantee, or other recipient if the United States Government provides any portion of the money or property which is requested or demanded, or if the Government will reimburse such contractor, grantee, or other recipient for any portion of the money or property which is requested or demanded.”
4. When Congress amended the FCA in 1986, it noted that “[e]vidence of fraud in Government programs and procurement is on a steady rise.” H. R. Rep. No. 99—660, at 18. And at that time, federal grants to state and local governments had totaled over $108 billion. See U.S. Dept. of Commerce National Data Book and Guide to Sources, Statistical Abstract of the United States 301 (108th ed. 1988) (compiling data from 1986). It is therefore difficult to believe, as the Court contends, that Congress intended “to cover all types of fraud, [but not] all types of fraudsters,” ante, at 15, n. 10, a conclusion that would exclude from coverage such a large share of potential fraud.
5. Petitioner argues that the Senate Report’s statement was simply inaccurate, because the three cases to which the Report cited for support did not interpret the meaning of the word “person” in the False Claims Act. Brief for Petitioner 25—26. The cases stand for the proposition that the statutory term “person” may include States and local governments–exactly the proposition I have discussed above. See supra, at 2—3. Petitioner’s observation that none of the cases cited is directly on point only indicates that the Senate’s understanding was based on an analogy rather than on controlling precedent. Petitioner further argues that the text of the FCA as it was originally enacted in 1863 could not have included States as “persons,” and therefore the Senate’s understanding of the pre-1986 Act was erroneous. See also ante, at 14—15. Assuming for argument’s sake that the Senate incorrectly ascertained what Congress meant in 1863, petitioner’s argument is beside the point. The term “person” in §3729(a) that we are interpreting today was enacted by the 1986 Congress, not by the 1863 Congress. See 100 Stat. 3153 (deleting entirely the previously existing introductory clause in §3729, including the phrase “[a] person not a member of an armed force of the United States” and replacing it with the new phrase “[a]ny person”). Therefore, even if the 1986 Congress were mistaken about what a previous Legislature had meant by the word “person,” it clearly expressed its own view that when the 1986 Congress itself enacted the word “person” (and not merely the word “any” as the Court insists, ante, at 16, n. 12), it meant the reference to include States. There is not the least bit of contradiction (as the Court suggests, ibid.) in one Congress informing itself of the general understanding of a statutory term it enacts based on its own (perhaps erroneous) understanding of what a past Congress thought the term meant.
6. Congress adopted the suggestion of the Attorneys General in §3730(e)(4)(A).
7. Because this concatenation of definitions expressly references and incorporates §3729, it is no answer that the definitions listed in §3733 apply, by their terms, “[f]or the purposes of” §3733.
8. Justice Ginsburg, who joins in the Court’s judgment, is careful to point out that the Court does not disagree with this reading of §3730(a). Ante, at 2.
9. Not so, the Court says, because natural persons and other entities, unlike States, are presumed to be included within the term “person.” Ante, at 17—18, n. 14. In other words, this supposedly independent textual argument does nothing on its own without relying entirely on the presumption already discussed. See supra, at 9—10; ante, at 13—15. The negative inference adds nothing on its own.
10. Indeed, reliance on the PFCRA seems to contradict the Court’s central premise–that in 1863 the word “person” did not include States and that scattered intervening amendments have done nothing to change that. Ante, at 14—16. If that were so, the relevant meaning of the word “person” would be the meaning adopted by the 1863 Congress, not the 1986 Congress. And on that premise, why should it matter what a different Congress, in a different century, did in a separate statute? Of course, as described earlier, see n. 5, supra, I believe it is the 1986 Congress’ understanding of the word “person” that controls, because it is that word as enacted by the 1986 Congress that we are interpreting in this case. But on the Court’s premise, it is the 1863 Congress’ understanding that controls and the PFCRA should be irrelevant.
11. It is also worth mentioning that treble damages may be reduced to double damages if the court makes the requisite findings under §§3729(a)(7)(A)—(C).
12. The State argues that this is essentially an “end run” around the Eleventh Amendment. Brief for Petitioner 33. It is not at all clear to me, though, why a qui tam action would be considered an “end run” around that Amendment, yet precisely the same form of action is not an “end run” around Articles II and III.