holding that the supplemental jurisdiction statute permits the exercise of diversity jurisdiction over additional plaintiffs who fail to satisfy the minimum amount-in-controversy requirement, as long as the other elements of diversity jurisdiction are present and at least one named plaintiff does satisfy the amount-in-controversy requirementSummary of this case from Auto-Owners Ins. Co. v. Stevens & Ricci Inc.
Together with No. 04-79, del Rosario Ortega et al. v. Star-Kist Foods, Inc., on certiorari to the United States Court of Appeals for the First Circuit.
In No. 04-70, Exxon dealers filed a class action against Exxon Corporation, invoking the Federal District Court's 28 U. S. C. § 1332(a) diversity jurisdiction. After the dealers won a jury verdict, the court certified the case for interlocutory review on the question whether it had properly exercised § 1367 supplemental jurisdiction over the claims of class members who had not met § 1332(a)'s minimum amount-in-controversy requirement. The Eleventh Circuit upheld this extension of supplemental jurisdiction. In No. 04-79, a girl and her family sought damages from Star-Kist Foods, Inc., in a diversity action. The District Court granted Star-Kist summary judgment, finding that none of the plaintiffs had met the amount-in-controversy requirement. The First Circuit ruled that the girl, but not her family, had alleged the requisite amount, and then held that supplemental jurisdiction over the family's claims was improper because original jurisdiction is lacking in a diversity case if one plaintiff fails to satisfy the amount-in-controversy requirement.
Held: Where the other elements of jurisdiction are present and at least one named plaintiff in the action satisfies § 1332(a)'s amount-in-controversy requirement, § 1367 authorizes supplemental jurisdiction over the claims of other plaintiffs in the same Article III case or controversy, even if those claims are for less than the requisite amount. Pp. 552-572.
(a) Although district courts may not exercise jurisdiction absent a statutory basis, once a court has original jurisdiction over some claims in an action, it may exercise supplemental jurisdiction over additional claims arising from the same case or controversy. See Mine Workers v. Gibbs, 383 U. S. 715. This expansive interpretation does not apply to § 1332's complete diversity requirement, for incomplete diversity destroys original jurisdiction with respect to all claims, leaving nothing to which supplemental claims can adhere. But other statutory prerequisites, including the federal-question and amount-in-controversy requirements, can be analyzed claim by claim. Before § 1367 was enacted, every plaintiff had to separately satisfy the amount-in-controversy requirement, Clark v. Paul Gray, Inc., 306 U.S. 583; Zahn v. International Paper Co., 414 U. S. 291, and the grant of original jurisdiction over claims involving particular parties did not itself confer supplemental jurisdiction over additional claims involving other parties, Finley v. United States, 490 U.S. 545, 556. Pp. 552-557.
(b) All parties here agree that § 1367 overturned Finley, but there is no warrant for assuming that is all it did. To determine § 1367's scope requires examination of the statute's text in light of context, structure, and related statutory provisions. Section 1367(a) is a broad grant of supplemental jurisdiction over other claims within the same case or controversy, as long as the action is one in which district courts would have original jurisdiction. Its last sentence makes clear that this grant extends to claims involving joinder or intervention of additional parties. The question here is whether a diversity case in which the claims of some, but not all, plaintiffs satisfy the amount-in-controversy requirement qualifies as a "civil action of which the district courts have original jurisdiction," § 1367(a). Pp. 557-558.
(c) The answer must be yes. When a well-pleaded complaint has at least one claim satisfying the amount-in-controversy requirement, and there are no other relevant jurisdictional defects, the district court, beyond all question, has original jurisdiction over that claim. A court with original jurisdiction over a single claim in the complaint has original jurisdiction over a "civil action" under § 1367(a), even if that action comprises fewer claims than were included in the complaint. Once a court has original jurisdiction over the action, it can then decide whether it has a constitutional and statutory basis for exercising supplemental jurisdiction over other claims in the action. Section 1367(b), which contains exceptions to § 1367(a)'s broad rule, does not withdraw supplemental jurisdiction over the claims of the additional parties here. In fact, its exceptions support this Court's conclusion. Pp. 559-560.
(d) The Court cannot accept the alternative view, or its supporting theories, that a district court lacks original jurisdiction over a civil action unless it has original jurisdiction over every claim in the complaint. The "indivisibility theory" — that all claims must stand or fall as a single, indivisible action — is inconsistent with the whole notion of supplemental jurisdiction and is belied by this Court's practice of allowing federal courts to cure jurisdictional defects by dismissing the offending parties instead of the entire action. And the statute's broad and general language does not permit the theory to apply in diversity cases when it does not apply in federal-question cases. The "contamination theory" — that inclusion of a claim or party falling outside the district court's original jurisdiction contaminates every other claim in the complaint — makes sense with respect to the complete diversity requirement because a non-diverse party's presence eliminates the justification for a federal forum. But it makes little sense with regard to the amount-in-controversy requirement, which is meant to ensure that a dispute is sufficiently important to warrant federal-court attention. It is fallacious to suppose, simply from the proposition that § 1332 imposes both requirements, that the contamination theory germane to the former also applies to the latter. This Court has already considered and rejected a virtually identical argument in the closely analogous removal-jurisdiction context. See Chicago v. International College of Surgeons, 522 U. S. 156. Pp. 560-566.
(e) In light of the statute's text and structure, § 1367's only plausible reading is that a court has original jurisdiction over a civil action comprising the claims for which there is no jurisdictional defect. Though a single nondiverse party can contaminate every other claim in a law-suit, contamination does not occur with respect to jurisdictional defects going only to the substantive importance of individual claims. Thus, § 1367(a)'s threshold requirement is satisfied in cases, such as these, where some but not all of the plaintiffs in a diversity action allege a sufficient amount in controversy. Section 1367 by its plain text over-ruled Clark and Zahn and authorized supplemental jurisdiction over all claims by diverse parties arising out of the same case or controversy, subject only to enumerated exceptions not applicable here. Pp. 566-567.
(f) Because § 1367 is not ambiguous, this Court need not examine other interpretative tools, including legislative history. Even were it appropriate to do so, the Court would not give the legislative history significant weight. Pp. 567-571.
(g) The Class Action Fairness Act has no impact on the analysis of these cases. Pp. 571-572.
KENNEDY, J., delivered the opinion of the Court, in which REHNQUIST, C. J., and SCALIA, SOUTER, and THOMAS, JJ., joined. STEVENS, J., filed a dissenting opinion, in which BREYER, J., joined, post, p. 572. GINSBURG, J., filed a dissenting opinion, in which STEVENS, O'CONNOR, and BREYER, JJ., joined, post, p. 577.
Carter G. Phillips argued the cause for petitioner in No. 04-70. With him on the briefs was Virginia A. Seitz. Donald B. Ayer argued the cause for petitioners in No. 04-79. With him on the briefs were Michael S. Fried, Christian G. Vergonis, Freddie Perez-Gonzalez, and Robert H. Klonoff. Eugene E. Stearns argued the cause for respondents in No. 04-70. With him on the briefs were Mark P. Dikeman, Mona E. Markus, Matthew W. Buttrick, and David C Pollack. Robert A. Long, Jr., argued the cause for respondent in No. 04-79. With him on the brief were Jeremy D. Kernodle, Scott T. Rickman, and David J. Herman.
A brief of amicus curiae urging reversal in No. 04-79 was filed for the United States by Acting Solicitor General Clement, Assistant Attorney General Keisler, Deputy Solicitor General Hungar, Deanne E. Maynard, Mark B. Stern, and Alisa B. Klein.
Briefs of amici curiae urging affirmance in No. 04-70 were filed for the United States by Mr. Clement, Acting Assistant Attorney General Schiffer, Mr. Hungar, Ms. Maynard, Mr. Stern, and Ms. Klein; and for the Product Liability Advisory Council, Inc., by Robert N. Weiner.
These consolidated cases present the question whether a federal court in a diversity action may exercise supplemental jurisdiction over additional plaintiffs whose claims do not satisfy the minimum amount-in-controversy requirement, provided the claims are part of the same case or controversy as the claims of plaintiffs who do allege a sufficient amount in controversy. Our decision turns on the correct interpretation of 28 U. S. C. § 1367. The question has divided the Courts of Appeals, and we granted certiorari to resolve the conflict. 543 U. S. 924 (2004).
We hold that, where the other elements of jurisdiction are present and at least one named plaintiff in the action satisfies the amount-in-controversy requirement, § 1367 does authorize supplemental jurisdiction over the claims of other plaintiffs in the same Article III case or controversy, even if those claims are for less than the jurisdictional amount specified in the statute setting forth the requirements for diversity jurisdiction. We affirm the judgment of the Court of Appeals for the Eleventh Circuit in No. 04-70, and we reverse the judgment of the Court of Appeals for the First Circuit in No. 04-79.
In 1991, about 10,000 Exxon dealers filed a class-action suit against the Exxon Corporation in the United States District Court for the Northern District of Florida. The dealers alleged an intentional and systematic scheme by Exxon under which they were overcharged for fuel purchased from Exxon. The plaintiffs invoked the District Court's § 1332(a) diversity jurisdiction. After a unanimous jury verdict in favor of the plaintiffs, the District Court certified the case for interlocutory review, asking whether it had properly exercised § 1367 supplemental jurisdiction over the claims of class members who did not meet the jurisdictional minimum amount in controversy.
The Court of Appeals for the Eleventh Circuit upheld the District Court's extension of supplemental jurisdiction to these class members. Allapattah Services, Inc. v. Exxon Corp., 333 F. 3d 1248 (2003). [W]e find," the court held, "that § 1367 clearly and unambiguously provides district courts with the authority in diversity class actions to exercise supplemental jurisdiction over the claims of class members who do not meet the minimum amount in controversy as long as the district court has original jurisdiction over the claims of at least one of the class representatives." Id., at 1256. This decision accords with the views of the Courts of Appeals for the Fourth, Sixth, and Seventh Circuits. See Rosmer v. Pfizer, Inc., 263 F. 3d 110 (CA4 2001); Olden v. LaFarge Corp., 383 F. 3d 495 (CA6 2004); Stromberg Metal Works, Inc. v. Press Mechanical, Inc., 77 F. 3d 928 (CA7 1996); In re Brand Name Prescription Drugs Antitrust Litigation, 123 F. 3d 599 (CA7 1997). The Courts of Appeals for the Fifth and Ninth Circuits, adopting a similar analysis of the statute, have held that in a diversity class action the unnamed class members need not meet the amount-in-controversy requirement, provided the named class members do. These decisions, however, are unclear on whether all the named plaintiffs must satisfy this requirement. In re Abbott Labs., 51 F. 3d 524 (CA5 1995); Gibson v. Chrysler Corp., 261 F. 3d 927 (CA9 2001).
In the other case now before us the Court of Appeals for the First Circuit took a different position on the meaning of § 1367(a). 370 F. 3d 124 (2004). In that case, a 9-year-old girl sued Star-Kist in a diversity action in the United States District Court for the District of Puerto Rico, seeking damages for unusually severe injuries she received when she sliced her finger on a tuna can. Her family joined in the suit, seeking damages for emotional distress and certain medical expenses. The District Court granted summary judgment to Star-Kist, finding that none of the plaintiffs met the minimum amount-in-controversy requirement. The Court of Appeals for the First Circuit, however, ruled that the injured girl, but not her family members, had made allegations of damages in the requisite amount.
The Court of Appeals then addressed whether, in light of the fact that one plaintiff met the requirements for original jurisdiction, supplemental jurisdiction over the remaining plaintiffs' claims was proper under § 1367. The court held that § 1367 authorizes supplemental jurisdiction only when the district court has original jurisdiction over the action, and that in a diversity case original jurisdiction is lacking if one plaintiff fails to satisfy the amount-in-controversy requirement. Although the Court of Appeals claimed to "express no view" on whether the result would be the same in a class action, id., at 143, n. 19, its analysis is inconsistent with that of the Court of Appeals for the Eleventh Circuit. The Court of Appeals for the First Circuit's view of § 1367 is, however, shared by the Courts of Appeals for the Third, Eighth, and Tenth Circuits, and the latter two Courts of Appeals have expressly applied this rule to class actions. See Meritcare, Inc. v. St. Paul Mercury Ins. Co., 166 F. 3d 214 (CA3 1999); Trimble v. Asarco, Inc., 232 F. 3d 946 (CA8 2000); Leonhardt v. Western Sugar Co., 160 F. 3d 631 (CA10 1998).
The district courts of the United States, as we have said many times, are "courts of limited jurisdiction. They possess only that power authorized by Constitution and statute," Kokkonen v. Guardian Life Ins. Co. of America, 511 U. S. 375, 377 (1994). In order to provide a federal forum for plaintiffs who seek to vindicate federal rights, Congress has conferred on the district courts original jurisdiction in federal-question cases — civil actions that arise under the Constitution, laws, or treaties of the United States. 28 U. S. C. § 1331. In order to provide a neutral forum for what have come to be known as diversity cases, Congress also has granted district courts original jurisdiction in civil actions between citizens of different States, between U. S. citizens and foreign citizens, or by foreign states against U. S. citizens. § 1332. To ensure that diversity jurisdiction does not flood the federal courts with minor disputes, § 1332(a) requires that the matter in controversy in a diversity case exceed a specified amount, currently $75,000.
Although the district courts may not exercise jurisdiction absent a statutory basis, it is well established — in certain classes of cases — that, once a court has original jurisdiction over some claims in the action, it may exercise supplemental jurisdiction over additional claims that are part of the same case or controversy. The leading modern case for this principle is Mine Workers v. Gibbs, 383 U.S. 715 (1966). In Gibbs, the plaintiff alleged the defendant's conduct violated both federal and state law. The District Court, Gibbs held, had original jurisdiction over the action based on the federal claims. Gibbs confirmed that the District Court had the additional power (though not the obligation) to exercise supplemental jurisdiction over related state claims that arose from the same Article III case or controversy. Id., at 725 ("The federal claim must have substance sufficient to confer subject matter jurisdiction on the court. . . . [A]ssuming substantiality of the federal issues, there is power in federal courts to hear the whole").
As we later noted, the decision allowing jurisdiction over pendent state claims in Gibbs did not mention, let alone come to grips with, the text of the jurisdictional statutes and the bedrock principle that federal courts have no jurisdiction without statutory authorization. Finley v. United States, 490 U. S. 545, 548 (1989). In Finley, we nonetheless reaffirmed and rationalized Gibbs and its progeny by inferring from it the interpretive principle that, in cases involving supplemental jurisdiction over additional claims between parties properly in federal court, the jurisdictional statutes should be read broadly, on the assumption that in this context Congress intended to authorize courts to exercise their full Article III power to dispose of an "`entire action before the court [which] comprises but one constitutional "case."'" 490 U. S., at 549 (quoting Gibbs, supra, at 725).
We have not, however, applied Gibbs' expansive interpretive approach to other aspects of the jurisdictional statutes. For instance, we have consistently interpreted § 1332 as requiring complete diversity: In a case with multiple plaintiffs and multiple defendants, the presence in the action of a single plaintiff from the same State as a single defendant deprives the district court of original diversity jurisdiction over the entire action. Strawbridge v. Curtiss, 3 Cranch 267 (1806); Owen Equipment Erection Co. v. Kroger, 437 U. S. 365, 375 (1978). The complete diversity requirement is not mandated by the Constitution, State Farm Fire Casualty Co. v. Tashire, 386 U. S. 523, 530-531 (1967), or by the plain text of § 1332(a). The Court, nonetheless, has adhered to the complete diversity rule in light of the purpose of the diversity requirement, which is to provide a federal forum for important disputes where state courts might favor, or be perceived as favoring, home-state litigants. The presence of parties from the same State on both sides of a case dispels this concern, eliminating a principal reason for conferring § 1332 jurisdiction over any of the claims in the action. See Wisconsin Dept. of Corrections v. Schacht, 524 U. S. 381, 389 (1998); Newman-Green, Inc. v. Alfonzo-Larrain, 490 U. S. 826, 829 (1989). The specific purpose of the complete diversity rule explains both why we have not adopted Gibbs' expansive interpretive approach to this aspect of the jurisdictional statute and why Gibbs does not undermine the complete diversity rule. In order for a federal court to invoke supplemental jurisdiction under Gibbs, it must first have original jurisdiction over at least one claim in the action. Incomplete diversity destroys original jurisdiction with respect to all claims, so there is nothing to which supplemental jurisdiction can adhere.
In contrast to the diversity requirement, most of the other statutory prerequisites for federal jurisdiction, including the federal-question and amount-in-controversy requirements, can be analyzed claim by claim. True, it does not follow by necessity from this that a district court has authority to exercise supplemental jurisdiction over all claims provided there is original jurisdiction over just one. Before the enactment of § 1367, the Court declined in contexts other than the pendent-claim instance to follow Gibbs' expansive approach to interpretation of the jurisdictional statutes. The Court took a more restrictive view of the proper interpretation of these statutes in so-called pendent-party cases involving supplemental jurisdiction over claims involving additional parties — plaintiffs or defendants — where the district courts would lack original jurisdiction over claims by each of the parties standing alone.
Thus, with respect to plaintiff-specific jurisdictional requirements, the Court held in Clark v. Paul Gray, Inc., 306 U. S. 583 (1939), that every plaintiff must separately satisfy the amount-in-controversy requirement. Though Clark was a federal-question case, at that time federal-question jurisdiction had an amount-in-controversy requirement analogous to the amount-in-controversy requirement for diversity cases. "Proper practice," Clark held, "requires that where each of several plaintiffs is bound to establish the jurisdictional amount with respect to his own claim, the suit should be dismissed as to those who fail to show that the requisite amount is involved." Id., at 590. The Court reaffirmed this rule, in the context of a class action brought invoking § 1332(a) diversity jurisdiction, in Zahn v. International Paper Co., 414 U. S. 291 (1973). It follows "inescapably" from Clark, the Court held in Zahn, that "any plaintiff without the jurisdictional amount must be dismissed from the case, even though others allege jurisdictionally sufficient claims." 414 U. S., at 300.
The Court took a similar approach with respect to supplemental jurisdiction over claims against additional defendants that fall outside the district courts' original jurisdiction. In Aldinger v. Howard, 427 U. S. 1 (1976), the plaintiff brought a Rev. Stat. § 1979, 42 U. S. C. § 1983, action against county officials in District Court pursuant to the statutory grant of jurisdiction in 28 U. S. C. § 1343(3) (1976 ed.). The plaintiff further alleged the court had supplemental jurisdiction over her related state-law claims against the county, even though the county was not suable under § 1983 and so was not subject to § 1343(3)'s original jurisdiction. The Court held that supplemental jurisdiction could not be exercised because Congress, in enacting § 1343(3), had declined (albeit implicitly) to extend federal jurisdiction over any party who could not be sued under the federal civil rights statutes. 427 U. S., at 16-19. "Before it can be concluded that [supplemental] jurisdiction [over additional parties] exists," Aldinger held, "a federal court must satisfy itself not only that Art[icle] III permits it, but that Congress in the statutes conferring jurisdiction has not expressly or by implication negated its existence." Id., at 18.
In Finley v. United States, 490 U. S. 545 (1989), we confronted a similar issue in a different statutory context. The plaintiff in Finley brought a Federal Tort Claims Act negligence suit against the Federal Aviation Administration in District Court, which had original jurisdiction under § 1346(b). The plaintiff tried to add related claims against other defendants, invoking the District Court's supplemental jurisdiction over so-called pendent parties. We held that the District Court lacked a sufficient statutory basis for exercising supplemental jurisdiction over these claims. Relying primarily on Zahn, Aldinger, and Kroger, we held in Finley that "a grant of jurisdiction over claims involving particular parties does not itself confer jurisdiction over additional claims by or against different parties." 490 U. S., at 556. While Finley did not "limit or impair" Gibbs' liberal approach to interpreting the jurisdictional statutes in the context of supplemental jurisdiction over additional claims involving the same parties, 490 U. S., at 556, Finley nevertheless declined to extend that interpretive assumption to claims involving additional parties. Finley held that in the context of parties, in contrast to claims, "we will not assume that the full constitutional power has been congressionally authorized, and will not read jurisdictional statutes broadly." Id., at 549.
As the jurisdictional statutes existed in 1989, then, here is how matters stood: First, the diversity requirement in § 1332(a) required complete diversity; absent complete diversity, the district court lacked original jurisdiction over all of the claims in the action. Strawbridge, 3 Cranch, at 267-268; Kroger, 437 U. S., at 373-374. Second, if the district court had original jurisdiction over at least one claim, the jurisdictional statutes implicitly authorized supplemental jurisdiction over all other claims between the same parties arising out of the same Article HI case or controversy. Gibbs, 383 U. S., at 725. Third, even when the district court had original jurisdiction over one or more claims between particular parties, the jurisdictional statutes did not authorize supplemental jurisdiction over additional claims involving other parties. Clark, 306 U.S., at 590; Zahn, supra, at 300-301; Finley, supra, at 556.
In Finley we emphasized that "[w]hatever we say regarding the scope of jurisdiction conferred by a particular statute can of course be changed by Congress." 490 U. S., at 556. In 1990, Congress accepted the invitation. It passed the Judicial Improvements Act, 104 Stat. 5089, which enacted § 1367, the provision which controls these cases.
Section 1367 provides, in relevant part:
"(a) Except as provided in subsections (b) and (c) or as expressly provided otherwise by Federal statute, in any civil action of which the district courts have original jurisdiction, the district courts shall have supplemental jurisdiction over all other claims that are so related to claims in the action within such original jurisdiction that they form part of the same case or controversy under Article III of the United States Constitution. Such supplemental jurisdiction shall include claims that involve the joinder or intervention of additional parties. "(b) In any civil action of which the district courts have original jurisdiction founded solely on section 1332 of this title, the district courts shall not have supplemental jurisdiction under subsection (a) over claims by plaintiffs against persons made parties under Rule 14, 19, 20, or 24 of the Federal Rules of Civil Procedure, or over claims by persons proposed to be joined as plaintiffs under Rule 19 of such rules, or seeking to intervene as plaintiffs under Rule 24 of such rules, when exercising supplemental jurisdiction over such claims would be inconsistent with the jurisdictional requirements of section 1332."
All parties to this litigation and all courts to consider the question agree that § 1367 overturned the result in Finley. There is no warrant, however, for assuming that § 1367 did no more than to overrule Finley and otherwise to codify the existing state of the law of supplemental jurisdiction. We must not give jurisdictional statutes a more expansive interpretation than their text warrants, 490 U. S., at 549, 556; but it is just as important not to adopt an artificial construction that is narrower than what the text provides. No sound canon of interpretation requires Congress to speak with extraordinary clarity in order to modify the rules of federal jurisdiction within appropriate constitutional bounds. Ordinary principles of statutory construction apply. In order to determine the scope of supplemental jurisdiction authorized by § 1367, then, we must examine the statute's text in light of context, structure, and related statutory provisions.
Section 1367(a) is a broad grant of supplemental jurisdiction over other claims within the same case or controversy, as long as the action is one in which the district courts would have original jurisdiction. The last sentence of § 1367(a) makes it clear that the grant of supplemental jurisdiction extends to claims involving joinder or intervention of additional parties. The single question before us, therefore, is whether a diversity case in which the claims of some plaintiffs satisfy the amount-in-controversy requirement, but the claims of other plaintiffs do not, presents a "civil action of which the district courts have original jurisdiction." If the answer is yes, § 1367(a) confers supplemental jurisdiction over all claims, including those that do not independently satisfy the amount-in-controversy requirement, if the claims are part of the same Article III case or controversy. If the answer is no, § 1367(a) is inapplicable and, in light of our holdings in Clark and Zahn, the district court has no statutory basis for exercising supplemental jurisdiction over the additional claims.
We now conclude the answer must be yes. When the well-pleaded complaint contains at least one claim that satisfies the amount-in-controversy requirement, and there are no other relevant jurisdictional defects, the district court, beyond all question, has original jurisdiction over that claim. The presence of other claims in the complaint, over which the district court may lack original jurisdiction, is of no moment. If the court has original jurisdiction over a single claim in the complaint, it has original jurisdiction over a "civil action" within the meaning of § 1367(a), even if the civil action over which it has jurisdiction comprises fewer claims than were included in the complaint. Once the court determines it has original jurisdiction over the civil action, it can turn to the question whether it has a constitutional and statutory basis for exercising supplemental jurisdiction over the other claims in the action.
Section 1367(a) commences with the direction that §§ 1367(b) and (c), or other relevant statutes, may provide specific exceptions, but otherwise § 1367(a) is a broad jurisdictional grant, with no distinction drawn between pendent-claim and pendent-party cases. In fact, the last sentence of § 1367(a) makes clear that the provision grants supplemental jurisdiction over claims involving joinder or intervention of additional parties. The terms of § 1367 do not acknowledge any distinction between pendent jurisdiction and the doctrine of so-called ancillary jurisdiction. Though the doctrines of pendent and ancillary jurisdiction developed separately as a historical matter, the Court has recognized that the doctrines are "two species of the same generic problem," Kroger, 437 U.S., at 370. Nothing in § 1367 indicates a congressional intent to recognize, preserve, or create some meaningful, substantive distinction between the jurisdictional categories we have historically labeled pendent and ancillary.
If § 1367(a) were the sum total of the relevant statutory language, our holding would rest on that language alone. The statute, of course, instructs us to examine § 1367(b) to determine if any of its exceptions apply, so we proceed to that section. While § 1367(b) qualifies the broad rule of § 1367(a), it does not withdraw supplemental jurisdiction over the claims of the additional parties at issue here. The specific exceptions to § 1367(a) contained in § 1367(b), moreover, provide additional support for our conclusion that § 1367(a) confers supplemental jurisdiction over these claims. Section 1367(b), which applies only to diversity cases, with-holds supplemental jurisdiction over the claims of plaintiffs proposed to be joined as indispensable parties under Federal Rule of Civil Procedure 19, or who seek to intervene pursuant to Rule 24. Nothing in the text of § 1367(b), however, withholds supplemental jurisdiction over the claims of plaintiffs permissively joined under Rule 20 (like the additional plaintiffs in No. 04-79) or certified as class-action members pursuant to Rule 23 (like the additional plaintiffs in No. 04-70). The natural, indeed the necessary, inference is that § 1367 confers supplemental jurisdiction over claims by Rule 20 and Rule 23 plaintiffs. This inference, at least with respect to Rule 20 plaintiffs, is strengthened by the fact that § 1367(b) explicitly excludes supplemental jurisdiction over claims against defendants joined under Rule 20.
We cannot accept the view, urged by some of the parties, commentators, and Courts of Appeals, that a district court lacks original jurisdiction over a civil action unless the court has original jurisdiction over every claim in the complaint. As we understand this position, it requires assuming either that all claims in the complaint must stand or fall as a single, indivisible "civil action" as a matter of definitional necessity — what we will refer to as the "indivisibility theory" — or else that the inclusion of a claim or party falling outside the district court's original jurisdiction somehow contaminates every other claim in the complaint, depriving the court of original jurisdiction over any of these claims — what we will refer to as the "contamination theory."
The indivisibility theory is easily dismissed, as it is inconsistent with the whole notion of supplemental jurisdiction. If a district court must have original jurisdiction over every claim in the complaint in order to have "original jurisdiction" over a "civil action," then in Gibbs there was no civil action of which the district court could assume original jurisdiction under § 1331, and so no basis for exercising supplemental jurisdiction over any of the claims. The indivisibility theory is further belied by our practice — in both federal-question and diversity cases — of allowing federal courts to cure jurisdictional defects by dismissing the offending parties rather than dismissing the entire action. Clark, for example, makes clear that claims that are jurisdictionally defective as to amount in controversy do not destroy original jurisdiction over other claims. 306 U. S., at 590 (dismissing parties who failed to meet the amount-in-controversy requirement but retaining jurisdiction over the remaining party). If the presence of jurisdictionally problematic claims in the complaint meant the district court was without original jurisdiction over the single, indivisible civil action before it, then the district court would have to dismiss the whole action rather than particular parties.
We also find it unconvincing to say that the definitional indivisibility theory applies in the context of diversity cases but not in the context of federal-question cases. The broad and general language of the statute does not permit this result. The contention is premised on the notion that the phrase "original jurisdiction of all civil actions" means different things in §§ 1331 and 1332. It is implausible, however, to say that the identical phrase means one thing (original jurisdiction in all actions where at least one claim in the complaint meets the following requirements) in § 1331 and something else (original jurisdiction in all actions where every claim in the complaint meets the following requirements) in § 1332.
The contamination theory, as we have noted, can make some sense in the special context of the complete diversity requirement because the presence of nondiverse parties on both sides of a lawsuit eliminates the justification for providing a federal forum. The theory, however, makes little sense with respect to the amount-in-controversy requirement, which is meant to ensure that a dispute is sufficiently important to warrant federal-court attention. The presence of a single nondiverse party may eliminate the fear of bias with respect to all claims, but the presence of a claim that falls short of the minimum amount in controversy does nothing to reduce the importance of the claims that do meet this requirement.
It is fallacious to suppose, simply from the proposition that § 1332 imposes both the diversity requirement and the amount-in-controversy requirement, that the contamination theory germane to the former is also relevant to the latter. There is no inherent logical connection between the amount-in-controversy requirement and § 1332 diversity jurisdiction. After all, federal-question jurisdiction once had an amount-in-controversy requirement as well. If such a requirement were revived under § 1331, it is clear beyond per-adventure that § 1367(a) provides supplemental jurisdiction over federal-question cases where some, but not all, of the federal-law claims involve a sufficient amount in controversy. In other words, § 1367(a) unambiguously overrules the holding and the result in Clark. If that is so, however, it would be quite extraordinary to say that § 1367 did not also over-rule Zahn, a case that was premised in substantial part on the holding in Clark.
In addition to the theoretical difficulties with the argument that a district court has original jurisdiction over a civil action only if it has original jurisdiction over each individual claim in the complaint, we have already considered and rejected a virtually identical argument in the closely analogous context of removal jurisdiction. In Chicago v. Intenational College of Surgeons, 522 U. S. 156 (1997), the plaintiff brought federal- and state-law claims in state court. The defendant removed to federal court. The plaintiff objected to removal, citing the text of the removal statute, § 1441(a). That statutory provision, which bears a striking similarity to the relevant portion of § 1367, authorizes removal of "any civil action . . . of which the district courts of the United States have original jurisdiction. . . ." The College of Surgeons plaintiff urged that, because its state-law claims were not within the District Court's original jurisdiction, § 1441(a) did not authorize removal. We disagreed. The federal-law claims, we held, "suffice to make the actions `civil actions' within the `original jurisdiction' of the district courts. . . . Nothing in the jurisdictional statutes suggests that the presence of related state law claims somehow alters the fact that [the plaintiff's] complaints, by virtue of their federal claims, were `civil actions' within the federal courts' `original jurisdiction.'" Id., at 166. Once the case was removed, the District Court had original jurisdiction over the federal-law claims and supplemental jurisdiction under § 1367(a) over the state-law claims. Id., at 165.
The dissent in College of Surgeons argued that because the plaintiff sought on-the-record review of a local administrative agency decision, the review it sought was outside the scope of the District Court's jurisdiction. Id., at 177 (opinion of GINSBURG, J.). We rejected both the suggestion that state-law claims involving administrative appeals are beyond the scope of § 1367 supplemental jurisdiction, id., at 168-172 (opinion of the Court), and the claim that the administrative review posture of the case deprived the District Court of original jurisdiction over the federal-law claims in the case, id., at 163-168. More importantly for present purposes, College of Surgeons stressed that a district court has original jurisdiction of a civil action for purposes of § 1441(a) as long as it has original jurisdiction over a subset of the claims constituting the action. Even the College of Surgeons dissent, which took issue with the Court's interpretation of § 1367, did not appear to contest this view of § 1441(a).
Although College of Surgeons involved additional claims between the same parties, its interpretation of § 1441(a) applies equally to cases involving additional parties whose claims fall short of the jurisdictional amount. If we were to adopt the contrary view that the presence of additional parties means there is no "civil action . . . of which the district courts . . . have original jurisdiction," those cases simply would not be removable. To our knowledge, no court has issued a reasoned opinion adopting this view of the removal statute. It is settled, of course, that absent complete diversity a case is not removable because the district court would lack original jurisdiction. Caterpillar Inc. v. Lewis, 519 U.S. 61, 73 (1996). This, however, is altogether consistent with our view of § 1441(a). A failure of complete diversity, unlike the failure of some claims to meet the requisite amount in controversy, contaminates every claim in the action.
We also reject the argument, similar to the attempted distinction of College of Surgeons discussed above, that while the presence of additional claims over which the district court lacks jurisdiction does not mean the civil action is out-side the purview of § 1367(a), the presence of additional parties does. The basis for this distinction is not altogether clear, and it is in considerable tension with statutory text. Section 1367(a) applies by its terms to any civil action of which the district courts have original jurisdiction, and the last sentence of § 1367(a) expressly contemplates that the court may have supplemental jurisdiction over additional parties. So it cannot be the case that the presence of those parties destroys the court's original jurisdiction, within the meaning of § 1367(a), over a civil action otherwise properly before it. Also, § 1367(b) expressly withholds supplemental jurisdiction in diversity cases over claims by plaintiffs joined as indispensable parties under Rule 19. If joinder of such parties were sufficient to deprive the district court of original jurisdiction over the civil action within the meaning of § 1367(a), this specific limitation on supplemental jurisdiction in § 1367(b) would be superfluous. The argument that the presence of additional parties removes the civil action from the scope of § 1367(a) also would mean that § 1367 left the Finley result undisturbed. Finley, after all, involved a Federal Tort Claims Act suit against a federal defendant and state-law claims against additional defendants not otherwise subject to federal jurisdiction. Yet all concede that one purpose of § 1367 was to change the result reached in Finley.
Finally, it is suggested that our interpretation of § 1367(a) creates an anomaly regarding the exceptions listed in § 1367(b): It is not immediately obvious why Congress would withhold supplemental jurisdiction over plaintiffs joined as parties "needed for just adjudication" under Rule 19 but would allow supplemental jurisdiction over plaintiffs permissively joined under Rule 20. The omission of Rule 20 plaintiffs from the list of exceptions in § 1367(b) may have been an "unintentional drafting gap," Meritcare, 166 F. 3d, at 221, and n. 6. If that is the case, it is up to Congress rather than the courts to fix it. The omission may seem odd, but it is not absurd. An alternative explanation for the different treatment of Rules 19 and 20 is that Congress was concerned that extending supplemental jurisdiction to Rule 19 plaintiffs would allow circumvention of the complete diversity rule: A nondiverse plaintiff might be omitted intentionally from the original action, but joined later under Rule 19 as a necessary party. See Stromberg Metal Works, 77 F 3d, at 932. The contamination theory described above, if applicable, means this ruse would fail, but Congress may have wanted to make assurance double sure. More generally, Congress may have concluded that federal jurisdiction is only appropriate if the district court would have original jurisdiction over the claims of all those plaintiffs who are so essential to the action that they could be joined under Rule 19.
To the extent that the omission of Rule 20 plaintiffs from the list of § 1367(b) exceptions is anomalous, moreover, it is no more anomalous than the inclusion of Rule 19 plaintiffs in that list would be if the alternative view of § 1367(a) were to prevail. If the district court lacks original jurisdiction over a civil diversity action where any plaintiff's claims fail to comply with all the requirements of § 1332, there is no need for a special § 1367(b) exception for Rule 19 plaintiffs who do not meet these requirements. Though the omission of Rule 20 plaintiffs from § 1367(b) presents something of a puzzle on our view of the statute, the inclusion of Rule 19 plaintiffs in this section is at least as difficult to explain under the alternative view.
And so we circle back to the original question. When the well-pleaded complaint in district court includes multiple claims, all part of the same case or controversy, and some, but not all, of the claims are within the court's original jurisdiction, does the court have before it "any civil action of which the district courts have original jurisdiction"? It does. Under § 1367, the court has original jurisdiction over the civil action comprising the claims for which there is no jurisdictional defect. No other reading of § 1367 is plausible in light of the text and structure of the jurisdictional statute. Though the special nature and purpose of the diversity requirement mean that a single nondiverse party can contaminate every other claim in the lawsuit, the contamination does not occur with respect to jurisdictional defects that go only to the substantive importance of individual claims.
It follows from this conclusion that the threshold requirement of § 1367(a) is satisfied in cases, like those now before us, where some, but not all, of the plaintiffs in a diversity action allege a sufficient amount in controversy. We hold that § 1367 by its plain text overruled Clark and Zahn and authorized supplemental jurisdiction over all claims by diverse parties arising out of the same Article III case or controversy, subject only to enumerated exceptions not applicable in the cases now before us.
The proponents of the alternative view of § 1367 insist that the statute is at least ambiguous and that we should look to other interpretive tools, including the legislative history of § 1367, which supposedly demonstrate Congress did not intend § 1367 to overrule Zahn. We can reject this argument at the very outset simply because § 1367 is not ambiguous. For the reasons elaborated above, interpreting § 1367 to foreclose supplemental jurisdiction over plaintiffs in diversity cases who do not meet the minimum amount in controversy is inconsistent with the text, read in light of other statutory provisions and our established jurisprudence. Even if we were to stipulate, however, that the reading these proponents urge upon us is textually plausible, the legislative history cited to support it would not alter our view as to the best interpretation of § 1367.
Those who urge that the legislative history refutes our interpretation rely primarily on the House Judiciary Committee Report on the Judicial Improvements Act. H. R. Rep. No. 101-734 (1990) (House Report or Report). This Report explained that § 1367 would "authorize jurisdiction in a case like Finley, as well as essentially restore the pre- Finley understandings of the authorization for and limits on other forms of supplemental jurisdiction." Id., at 28. The Report stated that § 1367(a) "generally authorizes the district court to exercise jurisdiction over a supplemental claim whenever it forms part of the same constitutional case or controversy as the claim or claims that provide the basis of the district court's original jurisdiction," and in so doing codifies Gibbs and fills the statutory gap recognized in Finley. House Report, at 28-29, and n. 15. The Report then remarked that § 1367(b) "is not intended to affect the jurisdictional requirements of [§ 1332] in diversity-only class actions, as those requirements were interpreted prior to Finley," citing, without further elaboration, Zahn and Supreme Tribe of Ben-Hur v. Cauble, 255 U. S. 356 (1921). House Report, at 29, and n. 17. The Report noted that the "net effect" of § 1367(b) was to implement the "principal rationale" of Kroger, House Report, at 29, and n. 16, effecting only "one small change" in pre- Finley practice with respect to diversity actions: § 1367(b) would exclude "Rule 23(a) plaintiff-intervenors to the same extent as those sought to be joined as plaintiffs under Rule 19." House Report, at 29. (It is evident that the report here meant to refer to Rule 24, not Rule 23.)
As we have repeatedly held, the authoritative statement is the statutory text, not the legislative history or any other extrinsic material. Extrinsic materials have a role in statutory interpretation only to the extent they shed a reliable light on the enacting Legislature's understanding of otherwise ambiguous terms. Not all extrinsic materials are reliable sources of insight into legislative understandings, however, and legislative history in particular is vulnerable to two serious criticisms. First, legislative history is itself often murky, ambiguous, and contradictory. Judicial investigation of legislative history has a tendency to become, to borrow Judge Leventhal's memorable phrase, an exercise in "`looking over a crowd and picking out your friends.'" See Wald, Some Observations on the Use of Legislative History in the 1981 Supreme Court Term, 68 Iowa L. Rev. 195, 214 (1983). Second, judicial reliance on legislative materials like committee reports, which are not themselves subject to the requirements of Article I, may give unrepresentative committee members — or, worse yet, unelected staffers and lobbyists — both the power and the incentive to attempt strategic manipulations of legislative history to secure results they were unable to achieve through the statutory text. We need not comment here on whether these problems are sufficiently prevalent to render legislative history inherently unreliable in all circumstances, a point on which Members of this Court have disagreed. It is clear, however, that in this instance both criticisms are right on the mark.
First of all, the legislative history of § 1367 is far murkier than selective quotation from the House Report would suggest. The text of § 1367 is based substantially on a draft proposal contained in a Federal Court Study Committee working paper, which was drafted by a Subcommittee chaired by Judge Posner. Report of the Subcommittee on the Role of the Federal Courts and Their Relationship to the States 567-568 (Mar. 12, 1990), reprinted in 1 Judicial Conference of the United States, Federal Courts Study Committee, Working Papers and Subcommittee Reports (July 1, 1990) (Subcommittee Working Paper). See also Judicial Conference of the United States, Report of the Federal Courts Study Committee 47-48 (Apr. 2, 1990) (Study Committee Report) (echoing, in brief summary form, the Subcommittee Working Paper proposal and noting that the Subcommittee Working Paper "contains additional material on this subject"); House Report, at 27 ("[Section 1367] implements a recommendation of the Federal Courts Study Committee found on pages 47 and 48 of its Report"). While the Sub-committee explained, in language echoed by the House Report, that its proposal "basically restores the law as it existed prior to Finley," Subcommittee Working Paper, at 561, it observed in a footnote that its proposal would overrule Zahn and that this would be a good idea, Subcommittee Working Paper, at 561, n. 33. Although the Federal Courts Study Committee did not expressly adopt the Subcommittee's specific reference to Zahn, it neither explicitly disagreed with the Subcommittee's conclusion that this was the best reading of the proposed text nor substantially modified the proposal to avoid this result. Study Committee Report, at 47-48. Therefore, even if the House Report could fairly be read to reflect an understanding that the text of § 1367 did not overrule Zahn, the Subcommittee Working Paper on which § 1367 was based reflected the opposite understanding. The House Report is no more authoritative than the Subcommittee Working Paper. The utility of either can extend no further than the light it sheds on how the enacting Legislature understood the statutory text. Trying to figure out how to square the Subcommittee Working Paper's understanding with the House Report's understanding, or which is more reflective of the understanding of the enacting legislators, is a hopeless task.
Second, the worst fears of critics who argue legislative history will be used to circumvent the Article I process were realized in this case. The telltale evidence is the statement, by three law professors who participated in drafting § 1367, see House Report, at 27, n. 13, that § 1367 "on its face" permits "supplemental jurisdiction over claims of class members that do not satisfy section 1332's jurisdictional amount requirement, which would overrule [Zahn]. [There is] a disclaimer of intent to accomplish this result in the legislative history. . . . It would have been better had the statute dealt explicitly with this problem, and the legislative history was an attempt to correct the oversight." Rowe, Burbank, Mengler, Compounding or Creating Confusion About Supplemental Jurisdiction? A Reply to Professor Freer, 40 Emory L. J. 943, 960, n. 90 (1991). The professors were frank to concede that if one refuses to consider the legislative history, one has no choice but to "conclude that section 1367 has wiped Zahn off the books." Ibid. So there exists an acknowledgment, by parties who have detailed, specific knowledge of the statute and the drafting process, both that the plain text of § 1367 overruled Zahn and that language to the contrary in the House Report was a post hoc attempt to alter that result. One need not subscribe to the wholesale condemnation of legislative history to refuse to give any effect to such a deliberate effort to amend a statute through a committee report.
In sum, even if we believed resort to legislative history were appropriate in these cases — a point we do not concede — we would not give significant weight to the House Report. The distinguished jurists who drafted the Subcommittee Working Paper, along with three of the participants in the drafting of § 1367, agree that this provision, on its face, overrules Zahn. This accords with the best reading of the statute's text, and nothing in the legislative history indicates directly and explicitly that Congress understood the phrase "civil action of which the district courts have original jurisdiction" to exclude cases in which some but not all of the diversity plaintiffs meet the amount-in-controversy requirement.
No credence, moreover, can be given to the claim that, if Congress understood § 1367 to overrule Zahn, the proposal would have been more controversial. We have little sense whether any Member of Congress would have been particularly upset by this result. This is not a case where one can plausibly say that concerned legislators might not have realized the possible effect of the text they were adopting. Certainly, any competent legislative aide who studied the matter would have flagged this issue if it were a matter of importance to his or her boss, especially in light of the Subcommittee Working Paper. There are any number of reasons why legislators did not spend more time arguing over § 1367, none of which are relevant to our interpretation of what the words of the statute mean.
Finally, we note that the Class Action Fairness Act (CAFA), Pub.L. 109-2, 119 Stat. 4, enacted this year, has no bearing on our analysis of these cases. Subject to certain limitations, the CAFA confers federal diversity jurisdiction over class actions where the aggregate amount in controversy exceeds $6 million. It abrogates the rule against aggregating claims, a rule this Court recognized in Ben-Hur and reaffirmed in Zahn. The CAFA, however, is not retroactive, and the views of the 2005 Congress are not relevant to our interpretation of a text enacted by Congress in 1990. The CAFA, moreover, does not moot the significance of our interpretation of § 1367, as many proposed exercises of supplemental jurisdiction, even in the class-action context, might not fall within the CAFA's ambit. The CAFA, then, has no impact, one way or the other, on our interpretation of § 1367.
* * *
The judgment of the Court of Appeals for the Eleventh Circuit is affirmed. The judgment of the Court of Appeals for the First Circuit is reversed, and the case is remanded for proceedings consistent with this opinion.
It is so ordered.
JUSTICE GINSBURG'S carefully reasoned opinion, post, at 577 (dissenting opinion), demonstrates the error in the Court's rather ambitious reading of this opaque jurisdictional statute. She also has demonstrated that "ambiguity" is a term that may have different meanings for different judges, for the Court has made the remarkable declaration that its reading of the statute is so obviously correct — and JUSTICE GINSBURG'S SO obviously wrong — that the text does not even qualify as "ambiguous." See ante, at 567. Because ambiguity is apparently in the eye of the beholder, I remain convinced that it is unwise to treat the ambiguity vel non of a statute as determinative of whether legislative history is consulted. Indeed, I believe that we as judges are more, rather than less, constrained when we make ourselves accountable to all reliable evidence of legislative intent. See Koons Buick Pontiac GMC, Inc. v. Nigh, 543 U. S. 50, 65-66, and n. 1 (2004) (STEVENS, J., concurring).
The legislative history of 28 U. S. C. § 1367 provides powerful confirmation of JUSTICE GINSBURG'S interpretation of that statute. It is helpful to consider in full the relevant portion of the House Report, which was also adopted by the Senate:
"This section would authorize jurisdiction in a case like Finley [v. United States, 490 U. S. 545 (1989)], as well as essentially restore the pre- Finley understandings of the authorization for and limits on other forms of supplemental jurisdiction. In federal question cases, it broadly authorizes the district courts to exercise supplemental jurisdiction over additional claims, including claims involving the joinder of additional parties. In diversity cases, the district courts may exercise supplemental jurisdiction, except when doing so would be in-consistent with the jurisdictional requirements of the diversity statute.
. . . .
"Subsection 114(b) [§ 1367(b)] prohibits a district court in a case over which it has jurisdiction founded solely on the general diversity provision, 28 U. S. C. § 1332, from exercising supplemental jurisdiction in specified circumstances. [Footnote 16: `The net effect of subsection (b) is to implement the principal rationale of Owen Equipment Erection Co. v. Kroger, 437 U. S. 365 (1978)'.] In diversity-only actions the district courts may not hear plaintiffs' supplemental claims when exercising supplemental jurisdiction would encourage plaintiffs to evade the jurisdictional requirement of 28 U. S. C. § 1332 by the simple expedient of naming initially only those defendants whose joinder satisfies section 1332s requirements and later adding claims not within original federal jurisdiction against other defendants who have intervened or been joined on a supplemental basis. In accord with case law, the subsection also prohibits the joinder or intervention of persons or plaintiffs if adding them is inconsistent with section 1332's requirements. The section is not intended to affect the jurisdictional requirements of 28 U. S. C. § 1332 in diversity-only class actions, as those requirements were interpreted prior to Finley. [Footnote 17: `See Supreme Tribe of Ben-Hur v. Cauble, 255 U. S. 356 (1921); Zahn v. International Paper Co., 414 U. S. 291 (1973)'.]
"Subsection (b) makes one small change in pre- Finley practice. Anomalously, under current practice, the same party might intervene as of right under Federal Rule of Civil Procedure 23(a) and take advantage of supplemental jurisdiction, but not come within supplemental jurisdiction if parties already in the action sought to effect the joinder under Rule 19. Subsection (b) would eliminate this anomaly, excluding Rule 23(a) plaintiff-intervenors to the same extent as those sought to be joined as plaintiffs under Rule 19." H. R. Rep. No. 101-734, pp. 28-29 (1990) (footnote omitted) (hereinafter House Report or Report).
The last quoted paragraph was intended to refer to Rule 24, not Rule 23. See ante, at 568.
Not only does the House Report specifically say that § 1367 was not intended to upset Zahn v. International Paper Co., 414 U. S. 291 (1973), but its entire explanation of the statute demonstrates that Congress had in mind a very specific and relatively modest task — undoing this Court's 5-to-4 decision in Finley v. United States, 490 U. S. 545 (1989). In addition to overturning that unfortunate and much-criticized decision, the statute, according to the Report, codifies and preserves "the pre- Finley understandings of the authorization for and limits on other forms of supplemental jurisdiction," House Report, at 28, with the exception of making "one small change in pre- Finley practice," id., at 29, which is not relevant here.
As I pointed out in my dissent in Finley, the majority's decision was "not faithful to our precedents," 490 U. S., at 558, and casually dismissed the accumulated wisdom of judges such as Henry Friendly, who had "special learning and expertise in matters of federal jurisdiction," id., at 565.
The sweeping purpose that the Court's decision imputes to Congress bears no resemblance to the House Report's description of the statute. But this does not seem to trouble the Court, for its decision today treats statutory interpretation as a pedantic exercise, divorced from any serious attempt at ascertaining congressional intent. Of course, there are situations in which we do not honor Congress' apparent intent unless that intent is made "clear" in the text of a statute — in this way, we can be certain that Congress considered the issue and intended a disfavored outcome, see, e. g., Landgraf v. USI Film Products, 511 U. S. 244 (1994) (requiring clear statement for retroactive civil legislation). But that principle provides no basis for discounting the House Report, given that our cases have never recognized a presumption in favor of expansive diversity jurisdiction.
The Court's reasons for ignoring this virtual billboard of congressional intent are unpersuasive. That a subcommittee of the Federal Courts Study Committee believed that an earlier, substantially similar version of the statute overruled Zahn, see ante, at 569, only highlights the fact that the statute is ambiguous. What is determinative is that the House Report explicitly rejected that broad reading of the statutory text. Such a report has special significance as an indicator of legislative intent. In Congress, committee reports are normally considered the authoritative explication of a statute's text and purposes, and busy legislators and their assistants rely on that explication in casting their votes. Cf. Garcia v. United States, 469 U. S. 70, 76 (1984) ("In surveying legislative history we have repeatedly stated that the authoritative source for finding the Legislature's intent lies in the Committee Reports on the bill, which `represen[t] the considered and collective understanding of those Congressmen involved in drafting and studying proposed legislation'" (quoting Zuber v. Allen, 396 U. S. 168, 186 (1969); brackets in original)).
The Court's second reason — its comment on the three law professors who participated in drafting § 1367, see ante, at 570 — is similarly off the mark. In the law review article that the Court refers to, the professors were merely saying that the text of the statute was susceptible to an overly broad (and simplistic) reading, and that clarification in the House Report was therefore appropriate. See Rowe, Burbank, Mengler, Compounding or Creating Confusion About Supplemental Jurisdiction? A Reply to Professor Freer, 40 Emory L. J. 943, 960, n. 90 (1991). Significantly, the reference to Zahn in the House Report does not at all appear to be tacked on or out of place; indeed, it is wholly consistent with the Report's broader explanation of Congress' goal of overruling Finley and preserving pre- Finley law. To suggest that these professors participated in a "deliberate effort to amend a statute through a committee report," ante, at 570, reveals an unrealistic view of the legislative process, not to mention disrespect for three law professors who acted in the role of public servants. To be sure, legislative history can be manipulated. But, in the situation before us, there is little reason to fear that an unholy conspiracy of "unrepresentative committee members," ante, at 568, law professors, and "unelected staffers and lobbyists," ibid., endeavored to torpedo Congress' attempt to overrule (without discussion) two longstanding features of this Court's diversity jurisprudence.
The professors' account of the challenges they faced in drafting § 1367 gives some sense, I think, of why that statute has proved difficult to interpret: "More broadly, codifying a complex area like supplemental jurisdiction — as Professor Freer's discussion illustrates — is itself complex business. A danger is that the result of the effort to deal with all the foreseeables will be a statute too prolix and baroque for everyday use and application by practitioners and judges. Section 1367 reflects an effort to provide sufficient detail without overdoing it. The statute is concededly not perfect. What it accomplishes, however, is to change the direction taken by the Supreme Court in Finley, to provide basic guidance (in particular the legislative history's general approval of pre- Finley case law, which has treated some specific issues Professor Freer raises), and then to trust the federal courts under the changed direction to interpret the statute sensibly. . . ." 40 Emory L. J., at 961.
After nearly 20 pages of complicated analysis, which explores subtle doctrinal nuances and coins various neologisms, the Court announces that § 1367 could not reasonably be read another way. See ante, at 567. That conclusion is difficult to accept. Given JUSTICE GINSBURG'S persuasive account of the statutory text and its jurisprudential backdrop, and given the uncommonly clear legislative history, I am confident that the majority's interpretation of § 1367 is mistaken. I respectfully dissent.
These cases present the question whether Congress, by enacting 28 U. S. C. § 1367, overruled this Court's decisions in Clark v. Paul Gray, Inc., 306 U. S. 583, 589 (1939) (reaffirming the holding of Troy Bank v. G A. Whitehead Co., 222 U. S. 39, 40 (1911)), and Zahn v. International Paper Co., 414 U. S. 291 (1973). Clark held that, when federal-court jurisdiction is predicated on a specified amount in controversy, each plaintiff joined in the litigation must independently meet the jurisdictional amount requirement. Zahn confirmed that in class actions governed by Federal Rule of Civil Procedure 23(b)(3), "[e]ach [class member]. . . must satisfy the jurisdictional amount, and any [class member] who does not must be dismissed from the case." 414 U. S., at 301.
Section 1367, all agree, was designed to overturn this Court's decision in Finley v. United States, 490 U. S. 545 (1989). Finley concerned not diversity-of-citizenship jurisdiction ( 28 U. S. C. § 1332), but original federal-court jurisdiction in cases arising under federal law ( 28 U. S. C. § 1331). The plaintiff in Finley sued the United States under the Federal Tort Claims Act (FTCA), 28 U. S. C. § 1346(b), to recover for the death of her husband and children in an airplane crash. She alleged that the Federal Aviation Administration's negligence contributed to the fatal accident. She later amended her complaint to add state-law tort claims against two other defendants, a municipality and a utility company. 490 U. S., at 546-547. No independent basis for federal subject-matter jurisdiction existed over the state-law claims. The plaintiff could not have brought her entire action in state court, because federal jurisdiction in FTCA actions is exclusive. § 1346(b). Hence, absent federal jurisdiction embracing the state-law claims, she would be obliged to pursue two discrete actions, one in federal court, the other in state court. This Court held, nevertheless, that the District Court lacked jurisdiction over the "pendent-party" state-law claims. Id., at 555-556. In so holding, the Court stressed that Congress held the control rein. Id., at 547-549. Congress could reverse the result in Finley, and permit pendent jurisdiction over state-law claims against additional defendants, if it so chose. Id., at 556. Congress did so in § 1367.
What more § 1367 wrought is an issue on which courts of appeals have sharply divided. Compare Stromberg Metal Works, Inc. v. Press Mechanical, Inc., 77 F. 3d 928, 930 (CA7 1996) (§ 1367 "supersedes Clark and allows pendent-party jurisdiction when the additional parties have claims worth less than [the jurisdictional minimum]"), and In re Abbott Labs., 51 F. 3d 524, 529 (CA5 1995) ("[U]nder § 1367 a district court can exercise supplemental jurisdiction over members of a class, although they did not meet the amount-in-controversy requirement, as did the class representatives."), with Merit-care Inc. v. St. Paul Mercury Ins. Co., 166 F. 3d 214, 222 (CA3 1999) (§ 1367 "preserves the prohibition against aggregation outlined in [ Zahn and Clark]"), and Leonhardt v. Western Sugar Co., 160 F. 3d 631, 641 (CA10 1998) (§ 1367 does not alter "the historical rules prohibiting aggregation of claims, including Zahn's prohibition of such aggregation in diversity class actions"). The Court today holds that § 1367, although prompted by Finley, a case in which original access to federal court was predicated on a federal question, notably enlarges federal diversity jurisdiction. The Court reads § 1367 to overrule Clark and Zahn, thereby allowing access to federal court by coplaintiffs or class members who do not meet the now in excess of $75,000 amount-in-controversy requirement, so long as at least one coplaintiff, or the named class representative, has a jurisdictionally sufficient claim. Ante, at 549.
The Court adopts a plausibly broad reading of § 1367, a measure that is hardly a model of the careful drafter's art. There is another plausible reading, however, one less disruptive of our jurisprudence regarding supplemental jurisdiction. If one reads § 1367(a) to instruct, as the statute's text suggests, that the district court must first have "original jurisdiction" over a "civil action" before supplemental jurisdiction can attach, then Clark and Zahn are preserved, and supplemental jurisdiction does not open the way for joinder of plaintiffs, or inclusion of class members, who do not independently meet the amount-in-controversy requirement. For the reasons that follow, I conclude that this narrower construction is the better reading of § 1367.
Section 1367, captioned "Supplemental jurisdiction," codifies court-recognized doctrines formerly labeled "pendent" and "ancillary" jurisdiction. Pendent jurisdiction involved the enlargement of federal-question litigation to include related state-law claims. Ancillary jurisdiction evolved primarily to protect defending parties, or others whose rights might be adversely affected if they could not air their claims in an ongoing federal-court action. Given jurisdiction over the principal action, federal courts entertained certain matters deemed ancillary regardless of the citizenship of the parties or the amount in controversy.
Mine Workers v. Gibbs, 383 U. S. 715 (1966), the leading pendent jurisdiction case, involved a claim against a union for wrongfully inducing the plaintiff's discharge. The plaintiff stated a federal claim under the Taft-Hartley Act, and an allied state-law claim of unlawful conspiracy to interfere with his employment contract. This Court upheld the joinder of federal and state claims. "[T]here is power in federal courts to hear the whole," the Court said, when the state and federal claims "derive from a common nucleus of operative fact" and are so linked that the plaintiff "would ordinarily be expected to try them all in one judicial proceeding." Id., at 725.
Gibbs involved the linkage of federal and state claims against the same defendant. In Finley v. United States, 490 U. S. 545, the Court contained Gibbs. Without congressional authorization, the Court admonished, the pendent jurisdiction umbrella could not be stretched to cover the joinder of additional parties. Gibbs had departed from earlier decisions recognizing that "jurisdiction [must] be explicitly conferred," the Court said. 490 U. S., at 556. Aldinger v. Howard, 427 U. S. 1 (1976), the Court observed, although resting "on a much narrower basis," R. Fallon, D. Meltzer, D. Shapiro, Hart and Wechsler's The Federal Courts and the Federal System 925 (5th ed. 2003) (hereinafter Hart Wechsler), had already signaled that "the Gibbs approach would not be extended to the pendent-party field," Finley, 490 U. S., at 556. While the Finley Court did not "limit or impair" Gibbs itself, 490 U. S., at 556, for further development of pendent jurisdiction, the Court made it plain, the initiative would lie in Congress' domain, id., at 555-556.
"[B]oth the Finley result and its implications" sparked "considerable criticism." Hart Wechsler 926; see also 13B C. Wright, A. Miller, E. Cooper, R. Freer, Federal Practice and Procedure § 3567.2, p. 91 (2d ed., Supp. 2005) (hereinafter Wright Miller) (characterizing the Finley decision as "surprising").
Ancillary jurisdiction, which evolved as a more sprawling doctrine than pendent jurisdiction, was originally rooted in "the notion that [when] federal jurisdiction in [a] principal suit effectively controls the property or fund under dispute, other claimants thereto should be allowed to intervene in order to protect their interests, without regard to jurisdiction." Aldinger, 427 U. S., at 11; see, e. g., Freeman v. Howe, 24 How. 450 (1861). In Owen Equipment Erection Co. v. Kroger, 437 U. S. 365 (1978), the Court addressed the permissible scope of the doctrine in relation to the liberal provisions of the Federal Rules of Civil Procedure for joinder of parties and claims.
Kroger commenced as a suit between a citizen of Iowa and a Nebraska corporation. When the Nebraska defendant impleaded an Iowa corporation as a third-party defendant under Rule 14(a), the plaintiff asserted state-law claims against the impleaded party. No independent basis of federal jurisdiction existed over the newly asserted claims, for both plaintiff and impleaded defendant were citizens of Iowa. 437 U. S., at 370. The Court held that the plaintiff could not draw in a co-citizen defendant in this manner. Id., at 377. Federal courts, by the time of Kroger, were routinely exercising ancillary jurisdiction over compulsory counterclaims, impleader claims, cross-claims among defendants, and claims of parties who intervened "of right." See id., at 375, n. 18 (collecting cases). In Kroger, however,
"the nonfederal claim . . . was asserted by the plaintiff, who voluntarily chose to bring suit upon a state-law claim in a federal court. By contrast, ancillary jurisdiction typically involve[d] claims by a defending party haled into court against his will, or by another person whose rights might be irretrievably lost unless he could assert them in an ongoing action in a federal court." Id., at 376.
Having "chosen the federal rather than the state forum," the Court said, the plaintiff had to "accept its limitations." Ibid.
In sum, in federal-question cases before § 1367's enactment, the Court recognized pendent-claim jurisdiction, Gibbs, 383 U. S., at 725, but not pendent-party jurisdiction, Finley, 490 U. S., at 555-556. As to ancillary jurisdiction, the Court adhered to the limitation that in diversity cases, throughout the litigation, all plaintiffs must remain diverse from all defendants. See Kroger, 437 U. S., at 374.
Although pendent jurisdiction and ancillary jurisdiction evolved discretely, the Court has recognized that they are "two species of the same generic problem: Under what circumstances may a federal court hear and decide a state-law claim arising between citizens of the same State?" Id., at 370. Finley regarded that question as one properly addressed to Congress. See 490 U. S., at 549, 556; 13 Wright Miller § 3523, p. 127 (2d ed., Supp. 2005); Hart Wechsler 924-926.
See generally 13B Wright Miller §§ 3567, 3567.1, 3567.2 (2d ed. 1984) (discussing pendent jurisdiction); 13 id., § 3523 (discussing ancillary jurisdiction); Hart Wechsler 922-926 (discussing pendent jurisdiction); id., at 1488-1490 (discussing ancillary jurisdiction).
Shortly before the Court decided Finley, Congress had established the Federal Courts Study Committee to take up issues relating to "the federal courts' congestion, delay, expense, and expansion." Judicial Conference of the United States, Report of the Federal Courts Study Committee 3 (Apr. 2, 1990) (hereinafter Committee Report). The Committee's charge was to conduct a study addressing the "crisis" in federal courts caused by the "rapidly growing" case-load. Id., at 6 (internal quotation marks omitted).
Among recommendations, the Committee urged Congress to "authorize federal courts to assert pendent jurisdiction over parties without an independent federal jurisdictional base." Id., at 47. If adopted, this recommendation would overrule Finley. Earlier, a Subcommittee had recommended that Congress overrule both Finley and Zahn. Report of the Subcommittee on the Role of the Federal Courts and Their Relationship to the States 547, 561, n. 33 (Mar. 12, 1990), reprinted in 1 Judicial Conference of the United States, Federal Courts Study Committee, Working Papers and Subcommittee Reports (July 1, 1990) (hereinafter Subcommittee Report). In the Subcommittee's view, "[f]rom a policy standpoint," Zahn "ma[de] little sense." Subcommittee Report, at 561, n. 33. The full Committee, however, urged only the overruling of Finley and did not adopt the recommendation to overrule Zahn. Committee Report, at 47-48.
Anomalously, in holding that each class member "must satisfy the jurisdictional amount," Zahn v. International Paper Co., 414 U. S. 291, 301 (1973), the Zahn Court did not refer to Supreme Tribe of Ben-Hur v. Cauble, 255 U. S. 356, 366 (1921), which established that in a class action, the citizenship of the named plaintiff is controlling. But see Zahn, 414 U. S., at 309-310 (Brennan, J., dissenting) (urging Zahn's inconsistency with Ben-Hur).
As a separate matter, a substantial majority of the Committee "strongly recommend[ed]" the elimination of diversity jurisdiction, save for "complex multi-state litigation, interpleader, and suits involving aliens." Id., at 38-39; accord Subcommittee Report, at 454-458. "[N]o other step," the Committee's Report maintained, "will do anywhere nearly as much to reduce federal caseload pressures and contain the growth of the federal judiciary." Committee Report, at 39.
Congress responded by adopting, as part of the Judicial Improvements Act of 1990, 104 Stat. 5089, recommendations of the Federal Courts Study Committee ranked by the House Committee on the Judiciary as "modest" and "noncontroversial." H. R. Rep. No. 101-734, pp. 15-16 (1990) (hereinafter H. R. Rep.); see also 136 Cong. Rec. 36288 (1990). Congress did not take up the Study Committee's immodest proposal to curtail diversity jurisdiction. It did, however, enact a supplemental jurisdiction statute, codified as 28 U. S. C. § 1367.
The omnibus Act encompassed the Civil Justice Reform Act of 1990 (Title I), the creation of new judgeships (Title II), the Federal Courts Study Committee Implementation Act of 1990 (Title III), and the establishment of the National Commission on Judicial Discipline and Removal (Title IV).
Section 1367, by its terms, operates only in civil actions "of which the district courts have original jurisdiction." The "original jurisdiction" relevant here is diversity-of-citizenship jurisdiction, conferred by § 1332. The character of that jurisdiction is the essential backdrop for comprehension of § 1367.
The Constitution broadly provides for federal-court jurisdiction in controversies "between Citizens of different States." Art. III, § 2, c1. 1. This Court has read that provision to demand no more than "minimal diversity," i. e., so long as one party on the plaintiffs' side and one party on the defendants' side are of diverse citizenship, Congress may authorize federal courts to exercise diversity jurisdiction. See State Farm Fire Casualty Co. v. Taskire, 386 U. S. 523, 530-531 (1967). Further, the Constitution includes no amount-in-controversy limitation on the exercise of federal jurisdiction. But from the start, Congress, as its measures have been construed by this Court, has limited federal-court exercise of diversity jurisdiction in two principal ways. First, unless Congress specifies otherwise, diversity must be "complete," i e., all parties on plaintiffs' side must be diverse from all parties on defendants' side. Strawbridge v. Curtiss, 3 Cranch 267 (1806); see 13B Wright Miller § 3605 (2d ed. 1984). Second, each plaintiff's stake must independently meet the amount-in-controversy specification: "When two or more plaintiffs, having separate and distinct demands, unite for convenience and economy in a single suit, it is essential that the demand of each be of the requisite jurisdictional amount." Troy Bank, 222 U. S., at 40.
The statute today governing federal-court exercise of diversity jurisdiction in the generality of cases, § 1332, like all its predecessors, incorporates both a diverse-citizenship requirement and an amount-in-controversy specification. As to the latter, the statute reads: "The district courts shall have original jurisdiction [in diversity-of-citizenship cases] where the matter in controversy exceeds the sum . . . of $75,000." § 1332(a). This Court has long held that, in determining whether the amount-in-controversy requirement has been satisfied, a single plaintiff may aggregate two or more claims against a single defendant, even if the claims are unrelated. See, e. g., Edwards v. Bates County, 163 U. S. 269, 273 (1896). But in multiparty cases, including class actions, we have unyieldingly adhered to the nonaggregation rule stated in Troy Bank. See Clark, 306 U. S., at 589 (reaffirming the "familiar rule that when several plaintiffs assert separate and distinct demands in a single suit, the amount involved in each separate controversy must be of the requisite amount to be within the jurisdiction of the district court, and that those amounts cannot be added together to satisfy jurisdictional requirements"); Snyder v. Harris, 394 U. S. 332, 339-340 (1969) (abandonment of the nonaggregation rule in class actions would undercut the congressional "purpose . . . to check, to some degree, the rising caseload of the federal courts").
Endeavoring to preserve the "complete diversity" rule first stated in Strawbridge v. Curtiss, 3 Cranch 267 (1806), the Court's opinion drives a wedge between the two components of 28 U. S. C. § 1332, treating the diversity-of-citizenship requirement as essential, the amount-in-controversy requirement as more readily disposable. See ante, at 553, 562. Section 1332 itself, however, does not rank order the two requirements. What "[o]rdinary principl[e] of statutory construction" or "sound canon of interpretation," ante, at 558, allows the Court to slice up § 1332 this way? In partial explanation, the Court asserts that amount in controversy can be analyzed claim-by-claim, but the diversity requirement cannot. See ante, at 554. It is not altogether clear why that should be so. The cure for improper joinder of a nondiverse party is the same as the cure for improper joinder of a plaintiff who does not satisfy the jurisdictional amount. In both cases, original jurisdiction can be preserved by dismissing the nonqualifying party. See Caterpillar Inc. v. Lewis, 519 U. S. 61, 64 (1996) (diversity); Newman-Green, Inc. v. Alfonzo-Larrain, 490 U. S. 826, 836-838 (1989) (same); Zahn, 414 U. S., at 295, 300 (amount in controversy); Clark v. Paul Gray, Inc., 306 U. S. 583, 590 (1939) (same).
This Court most recently addressed "[t]he meaning of [§ 1332's] `matter in controversy' language" in Zahn, 414 U. S., at 298. Zahn, like Snyder decided four years earlier, was a class action. In Snyder, no class member had a claim large enough to satisfy the jurisdictional amount. But in Zahn, the named plaintiffs had such claims. 414 U. S., at 292. Nevertheless, the Court declined to depart from its "longstanding construction of the `matter in controversy' requirement of § 1332." Id., at 301. The Zahn Court stated:
"Snyder invoked the well-established rule that each of several plaintiffs asserting separate and distinct claims must satisfy the jurisdictional-amount requirement if his claim is to survive a motion to dismiss. This rule plainly mandates not only that there may be no aggregation and that the entire case must be dismissed where none of the plaintiffs claims [meets the amount-in-controversy requirement] but also requires that any plaintiff without the jurisdictional amount must be dismissed from the case, even though others allege jurisdictionally sufficient claims." Id., at 300.
The rule that each plaintiff must independently satisfy the amount-in-controversy requirement, unless Congress expressly orders otherwise, was thus the solidly established reading of § 1332 when Congress enacted the Judicial Improvements Act of 1990, which added § 1367 to Title 28.
These cases present the question whether Congress abrogated the nonaggregation rule long tied to § 1332 when it enacted § 1367. In answering that question, "context [should provide] a crucial guide." Rosario Ortega v. StarKist Foods, Inc., 370 F. 3d 124, 135 (CA1 2004). The Court should assume, as it ordinarily does, that Congress legislated against a background of law already in place and the historical development of that law. See National Archives and Records Admin, v. Favish, 541 U. S. 157, 169 (2004). Here, that background is the statutory grant of diversity jurisdiction, the amount-in-controversy condition that Congress, from the start, has tied to the grant, and the nonaggregation rule this Court has long applied to the determination of the "matter in controversy."
Section 1367(a) provides:
"Except as provided in subsections (b) and (c) or as expressly provided otherwise by Federal statute, in any civil action of which the district courts have original jurisdiction, the district courts shall have supplemental jurisdiction over all other claims that are so related to claims in the action within such original jurisdiction that they form part of the same case or controversy under Article III of the United States Constitution. Such supplemental jurisdiction shall include claims that involve the joinder or intervention of additional parties."
The Court is unanimous in reading § 1367(a) to permit pendent-party jurisdiction in federal-question cases, and thus, to overrule Finley. The basic jurisdictional grant, § 1331, provides that "[t]he district courts shall have original jurisdiction of all civil actions arising under the Constitution, laws, or treaties of the United States." Since 1980, § 1331 has contained no amount-in-controversy requirement. See 94 Stat. 2369 (eliminating § 1331's amount-in-controversy requirement). Once there is a civil action presenting a qualifying claim arising under federal law, § 1331's sole requirement is met. District courts, we have held, may then adjudicate, additionally, state-law claims "deriving] from a common nucleus of operative fact." Gibbs, 383 U. S., at 725. Section 1367(a) enlarges that category to include not only state-law claims against the defendant named in the federal claim, but also "[state-law] claims that involve the joinder or intervention of additional parties."
The Court noted in Zahn, 414 U. S., at 302, n. 11, that when the exercise of § 1331 federal-question jurisdiction and § 1332 diversity jurisdiction were conditioned on the same jurisdictional-amount limitation, the same nonaggregation rule applied under both heads of federal jurisdiction. But cf. ante, at 562. The Court added, however, that "Congress ha[d] exempted major areas of federal-question jurisdiction from any jurisdictional-amount requirements," thus diminishing the impact of § 1331's "matter in controversy" specification in cases arising under federal law. Zahn, 414 U. S., at 302, n. 11.
The Court divides, however, on the impact of § 1367(a) on diversity cases controlled by § 1332. Under the majority's reading, § 1367(a) permits the joinder of related claims cut loose from the nonaggregation rule that has long attended actions under § 1332. Only the claims specified in § 1367(b) would be excluded from § 1367(a)'s expansion of § 1332's grant of diversity jurisdiction. And because § 1367(b) contains no exception for joinder of plaintiffs under Rule 20 or class actions under Rule 23, the Court concludes, Clark and Zahn have been overruled.
Title 28 U. S. C. § 1367(b) provides:
"In any civil action of which the district courts have original jurisdiction founded solely on section 1332 of this title, the district courts shall not have supplemental jurisdiction under subsection (a) over claims by plaintiffs against persons made parties under Rule 14, 19, 20, or 24 of the Federal Rules of Civil Procedure, or over claims by persons proposed to be joined as plaintiffs under Rule 19 of such rules, or seeking to intervene as plaintiffs under Rule 24 of such rules, when exercising supplemental jurisdiction over such claims would be inconsistent with the jurisdictional requirements of section 1382."
Under the Court's construction of § 1367, see ante, at 560, 566-567, Beatriz Ortega's family members can remain in the action because their joinder is merely permissive, see Fed. Rule Civ. Proc. 20. If, however, their presence was "needed for just adjudication," Rule 19, their dismissal would be required. The inclusion of those who may join, and exclusion of those who should or must join, defies rational explanation, but cf. ante, at 565, and others adopting the interpretation the Court embraces have so acknowledged, see Stromberg Metal Works, Inc. v. Press Mechanical, Inc., 77 F. 3d 928, 932 (CA7 1996) (recognizing the anomaly and inquiring: "What sense can this make?"); cf 14B Wright Miller § 3704, p. 168 (3d ed. 1998) (distinction between Rule 19 and Rule 20 "seems incongruous, and serves no apparent public policy purpose").
The Court's reading is surely plausible, especially if one detaches § 1367(a) from its context and attempts no reconciliation with prior interpretations of § 1332's amount-in-controversy requirement. But § 1367(a)'s text, as the First Circuit held, can be read another way, one that would involve no rejection of Clark and Zahn.
As explained by the First Circuit in Ortega, and applied to class actions by the Tenth Circuit in Leonhardt, see supra, at 578-579, § 1367(a) addresses "civil action[s] of which the district courts have original jurisdiction," a formulation that, in diversity cases, is sensibly read to incorporate the rules on joinder and aggregation tightly tied to § 1332 at the time of § 1367's enactment. On this reading, a complaint must first meet that "original jurisdiction" measurement. If it does not, no supplemental jurisdiction is authorized. If it does, § 1367(a) authorizes "supplemental jurisdiction" over related claims. In other words, § 1367(a) would preserve undiminished, as part and parcel of § 1332 "original jurisdiction" determinations, both the "complete diversity" rule and the decisions restricting aggregation to arrive at the amount in controversy. Section 1367(b)'s office, then, would be "to prevent the erosion of the complete diversity [and amount-in-controversy] requirement[s] that might otherwise result from an expansive application of what was once termed the doctrine of ancillary jurisdiction." See Pfander, Supplemental Jurisdiction and Section 1367: The Case for a Sympathetic Textualism, 148 U. Pa. L. Rev. 109, 114 (1999); infra, at 593-594. In contrast to the Court's construction of § 1367, which draws a sharp line between the diversity and amount-in-controversy components of § 1332, see ante, at 554; supra, at 585, n. 5, the interpretation presented here does not sever the two jurisdictional requirements.
On this reading of § 1367(a), it is immaterial that § 1367(b) "does not withdraw supplemental jurisdiction over the claims of the additional parties at issue here." Ante, at 560. Because those claims would not come within § 1367(a) in the first place, Congress would have had no reason to list them in § 1367(b). See infra, at 592-593.
The more restrained reading of § 1367 just outlined would yield affirmance of the First Circuit's judgment in Ortega, and reversal of the Eleventh Circuit's judgment in Exxon. It would not discard entirely, as the Court does, the judicially developed doctrines of pendent and ancillary jurisdiction as they existed when Finley was decided. Instead, it would recognize § 1367 essentially as a codification of those doctrines, placing them under a single heading, but largely retaining their substance, with overriding Finley the only basic change: Supplemental jurisdiction, once the district court has original jurisdiction, would now include "claims that involve the joinder or intervention of additional parties." § 1367(a).
The Court's opinion blends the two doctrines, according no significance to their discrete development. See ante, at 552-557.
Pendent jurisdiction, as earlier explained, see supra, at 579-580, applied only in federal-question cases and allowed plaintiffs to attach nonfederal claims to their jurisdiction-qualifying claims. Ancillary jurisdiction applied primarily, although not exclusively, in diversity cases and "typically involved] claims by a defending party haled into court against his will." Kroger, 437 U. S., at 376 (emphasis added); see also id., at 375, n. 18; supra, at 581-582. As the First Circuit observed, neither doctrine permitted a plaintiff to circumvent the dual requirements of § 1332 (diversity of citizenship and amount in controversy) "simply by joining her [jurisdictionally inadequate] claim in an action brought by [a] jurisdictionally competent diversity plaintiff." Ortega, 370 F. 3d, at 138.
Not only would the reading I find persuasive "alig[n] statutory supplemental jurisdiction with the judicially developed doctrines of pendent and ancillary jurisdiction," ibid., it would also synchronize § 1367 with the removal statute, 28 U. S. C. § 1441. As the First Circuit carefully explained:
"Section 1441, like § 1367, applies only if the `civil action' in question is one `of which the district courts . . . have original jurisdiction.' § 1441(a). Relying on that language, the Supreme Court has interpreted § 1441 to prohibit removal unless the entire action, as it stands at the time of removal, could have been filed in federal court in the first instance. See, e. g., Syngenta Crop Protection, Inc. v. Henson, 537 U. S. 28, 33 (2002); Okla. Tax Comm'n v. Graham, 489 U. S. 838, 840 (1989) (per curiam). Section 1441 has thus been held to incorporate the well-pleaded complaint rule, see City of Chicago [v. International College of Surgeons, 522 U. S. 156, 163 (1997)]; the complete diversity rule, see Caterpillar, Inc. v. Lewis, 519 U. S. 61, 73 (1996); and rules for calculating the amount in controversy, see St. Paul Mercury Indent. Co. v. Red Cab Co., 303 U. S. 283, 291-292 (1938)." 370 F. 3d, at 138 (citations omitted and footnote added).
The point of the Court's extended discussion of Chicago v. International College of Surgeons, 522 U. S. 156 (1997), in the instant cases, see ante, at 562-564, slips from my grasp. There was no disagreement in that case, and there is none now, that 28 U. S. C. § 1367(a) is properly read to authorize the exercise of supplemental jurisdiction in removed cases. International College of Surgeons was unusual in that the federal court there was asked to review a decision of a local administrative agency. Such review, it was unsuccessfully argued, was "appellate" in character, and therefore outside the ken of a court empowered to exercise "original" jurisdiction. Compare 522 U. S., at 166-168, with id., at 176-177 (GINSBURG, J., dissenting).
The less disruptive view I take of § 1367 also accounts for the omission of Rule 20 plaintiffs and Rule 23 class actions in § 1367(b)'s text. If one reads § 1367(a) as a plenary grant of supplemental jurisdiction to federal courts sitting in diversity, one would indeed look for exceptions in § 1367(b). Finding none for permissive joinder of parties or class actions, one would conclude that Congress effectively, even if unintentionally, overruled Clark and Zahn. But if one recognizes that the nonaggregation rule delineated in Clark and Zahn forms part of the determination whether "original jurisdiction" exists in a diversity case, see supra, at 590, then plaintiffs who do not meet the amount-in-controversy requirement would fail at the § 1367(a) threshold. Congress would have no reason to resort to a § 1367(b) exception to turn such plaintiffs away from federal court, given that their claims, from the start, would fall outside the court's § 1332 jurisdiction. See Pfander, supra, at 148.
Nor does the more moderate reading assign different meanings to "original jurisdiction" in diversity and federal-question cases. See ante, at 561. As the First Circuit stated:
"`[O]riginal jurisdiction' in § 1367(a) has the same meaning in every case: [An] underlying statutory grant of original jurisdiction must be satisfied. What differs between federal question and diversity cases is not the meaning of `original jurisdiction' but rather the [discrete] requirements of sections 1331 and 1332. Under § 1331, the sole issue is whether a federal question appears on the face of the plaintiff's well-pleaded complaint; the [citizenship] of the parties and the amounts they stand to recover [do not bear on that determination]. Section 1332, by contrast, predicates original jurisdiction on the identity of the parties ( i.e., [their] complete diversity) and their [satisfaction of the amount-in-controversy specification]. [In short,] the `original jurisdiction' language in § 1367 operates differently in federal-question and diversity cases not because the meaning of that term varies, but because the [jurisdiction-granting] statutes are different." 370 F. 3d, at 139-140.
What is the utility of § 1367(b) under my reading of § 1367(a)? Section 1367(a) allows parties other than the plaintiff to assert reactive claims once entertained under the heading ancillary jurisdiction. See supra, at 581 (listing claims, including compulsory counterclaims and impleader claims, over which federal courts routinely exercised ancillary jurisdiction). As earlier observed, see supra, at 590-591, § 1367(b) stops plaintiffs from circumventing § 1332's jurisdictional requirements by using another's claim as a hook to add a claim that the plaintiff could not have brought in the first instance. Kroger is the paradigm case. See supra, at 581-582. There, the Court held that ancillary jurisdiction did not extend to a plaintiff's claim against a nondiverse party who had been impleaded by the defendant under Rule 14. Section 1367(b), then, is corroborative of § 1367(a)'s coverage of claims formerly called ancillary, but provides exceptions to ensure that accommodation of added claims would not fundamentally alter "the jurisdictional requirements of section 1332." See Pfander, 148 U. Pa. L. rev., at 135-137.
While § 1367's enigmatic text defies flawless interpretation, see supra, at 589, n. 8, the precedent-preservative reading, I am persuaded, better accords with the historical and legal context of Congress' enactment of the supplemental jurisdiction statute, see supra, at 582-584, 587, and the established limits on pendent and ancillary jurisdiction, see supra, at 580-582. It does not attribute to Congress a jurisdictional enlargement broader than the one to which the legislators adverted, cf. Finley, 490 U. S., at 549, and it follows the sound counsel that "close questions of [statutory] construction should be resolved in favor of continuity and against change," Shapiro, Continuity and Change in Statutory Interpretation, 67 N. Y. U. L. Rev. 921, 925 (1992).
The Court notes the passage this year of the Class Action Fairness Act (CAFA), Pub.L. 109-2, 119 Stat. 4, ante, at 571-572, only to dismiss that legislation as irrelevant. Subject to several exceptions and qualifications, CAFA provides for federal-court adjudication of state-law-based class actions in which diversity is "minimal" (one plaintiff's diversity from one defendant suffices), and the "matter in controversy" is an aggregate amount in excess of $5,000,000. Significant here, CAFA's enlargement of federal-court diversity jurisdiction was accomplished, "clearly and conspicuously," by amending § 1332. Cf. Rosario Ortega v. Star-Kist Foods, Inc., 370 F. 3d 124, 142 (CA1 2004).
If § 1367(a) itself renders unnecessary the listing of Rule 20 plaintiffs and Rule 23 class actions in § 1367(b), see supra, at 592, then it is similarly unnecessary to refer, as § 1367(b) does, to "persons proposed to be joined as plaintiffs under Rule 19." On one account, Congress bracketed such persons with persons "seeking to intervene as plaintiffs under Rule 24" to modify pre-§ 1367 practice. Before enactment of § 1367, courts entertained, under the heading ancillary jurisdiction, claims of Rule 24(a) interveners "of right," see Owen Equipment Erection Co. v. Kroger, 437 U. S. 365, 375, n. 18 (1978), but denied ancillary jurisdiction over claims of "necessary" Rule 19 plaintiffs, see 13 Wright Miller § 3523, p. 127 (2d ed., Supp. 2005). Congress may have sought simply to underscore that those seeking to join as plaintiffs, whether under Rule 19 or Rule 24, should be treated alike, i. e., denied joinder when "inconsistent with the jurisdictional requirements of section 1332." See 370 F. 3d, at 140, and n. 15 (internal quotation marks omitted); H. R. Rep., at 29 ("Subsection (b) makes one small change in pre- Finley practice," i. e., it eliminates the Rule 19/Rule 24 anomaly.).
While the interpretation of § 1367 described in this opinion does not rely on the measure's legislative history, that history, as JUSTICE STEVENS has shown, see ante, at 573 (dissenting opinion), is corroborative of the statutory reading set out above.
* * *
For the reasons stated, I would hold that § 1367 does not overrule Clark and Zahn. I would therefore affirm the judgment of the Court of Appeals for the First Circuit and reverse the judgment of the Court of Appeals for the Eleventh Circuit.