delivered the opinion of the Court.

This case requires us to decide whether state courts have concurrent jurisdiction over civil actions brought under the *457Racketeer Influenced and Corrupt Organizations Act (RICO), Pub. L. 91-452, Title IX, 84 Stat. 941, as amended, 18 U. S. C. §§ 1961-1968.

I

The underlying litigation arises from the failure of Old Court Savings & Loan, Inc. (Old Court), a Maryland savings and loan association, and the attendant collapse of the Maryland Savings-Share Insurance Corp. (MSSIC), a state-chartered nonprofit corporation created to insure accounts in Maryland savings and loan associations that were not federally insured. See Brandenburg v. Seidel, 859 F. 2d 1179, 1181-1183 (CA4 1988) (reviewing history of Maryland’s savings and loan crisis). Petitioners are nonresidents of Maryland who hold unpaid certificates of deposit issued by Old Court. Respondents are the former officers and directors of Old Court, the former officers and directors of MSSIC, the law firm of Old Court and MSSIC, the accounting firm of Old Court, and the State of Maryland Deposit Insurance Fund Corp., the state-created successor to MSSIC. Petitioners allege various state law causes of action as well as claims under the Securities Exchange Act of 1934 (Exchange Act), 48 Stat. 881, 15 U. S. C. §78a et seq., and RICO.

The District Court granted respondents’ motions to dismiss, concluding that petitioners had failed to state a claim under the Exchange Act and that, because state courts have concurrent jurisdiction over civil RICO claims, federal abstention was appropriate for the other causes of action because they had been raised in pending litigation in state court. The Court of Appeals for the Fourth Circuit affirmed. 865 F. 2d 595 (1989). The Court of Appeals agreed with the District Court that the Old Court certificates of deposit were not “securities” within the meaning of the Exchange Act, see 15 U. S. C. §78c(a)(10), and that petitioners’ Exchange Act claims were therefore properly dismissed. 865 F. 2d, at 598-599. The Court of Appeals further held, in reliance on its prior decision in Brandenburg v. Seidel, *458supra, that “a RICO action could be instituted in a state court and that Maryland’s ‘comprehensive scheme for the rehabilitation and liquidation of insolvent state-chartered savings and loan associations,’ 859 F. 2d at 1191, provided a proper basis for the district court to abstain under the authority of Burford v. Sun Oil Co., 319 U. S. 315 (1943).” 865 F. 2d, at 600 (citations omitted).

To resolve a conflict among the federal appellate courts and state supreme courts,1 we granted certiorari limited to the question whether state courts have concurrent jurisdiction over civil RICO claims. 490 U. S. 1089 (1989). We hold that they do and accordingly affirm the judgment of the Court of Appeals.

II

We begin with the axiom that, under our federal system, the States possess sovereignty concurrent with that of the Federal Government, subject only to limitations imposed by the Supremacy Clause. Under this system of dual sovereignty, we have consistently held that state courts have inherent authority, and are thus presumptively competent, to adjudicate claims arising under the laws of the United States. See, e. g., Houston v. Moore, 5 Wheat. 1, 25-26 *459(1820); Claflin v. Houseman, 93 U. S. 130, 136-137 (1876); Plaquemines Tropical Fruit Co. v. Henderson, 170 U. S. 511, 517 (1898); Charles Dowd Box Co. v. Courtney, 368 U. S. 502, 507-508 (1962); Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 477-478 (1981). As we noted in Claflin, “if exclusive jurisdiction be neither express nor implied, the State courts have concurrent jurisdiction whenever, by their own constitution, they are competent to take it.” 93 U. S., at 136; see also Dowd Box, supra, at 507-508 (“We start with the premise that nothing in the concept of our federal system prevents state courts from enforcing rights created by federal law. Concurrent jurisdiction has been a common phenomenon in our judicial history, and exclusive federal court jurisdiction over cases arising under federal law has been the exception rather than the rule”). See generally 1 J. Kent, Commentaries on American Law *400; The Federalist No. 82 (A. Hamilton); F. Frankfurter & J. Landis, The Business of the Supreme Court 5-12 (1927); H. Friendly, Federal Jurisdiction: A General View 8-11 (1973).

This deeply rooted presumption in favor of concurrent state court jurisdiction is, of course, rebutted if Congress affirmatively ousts the state courts of jurisdiction over a particular federal claim. See, e. g., Claflin, supra, at 137 (“Congress may, if it see[s] fit, give to the Federal courts exclusive jurisdiction”) (citations omitted); see also Houston, supra, at 25-26. As we stated in Gulf Offshore:

“In considering the propriety of state-court jurisdiction over any particular federal claim, the Court begins with the presumption that state courts enjoy concurrent jurisdiction. Congress, however, may confine jurisdiction to the federal courts either explicitly or implicitly. Thus, the presumption of concurrent jurisdiction can be rebutted by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear in*460compatibility between state-court jurisdiction and federal interests.” 453 U. S., at 478 (citations omitted).

See also Clafiin, supra, at 136 (state courts have concurrent jurisdiction “where it is not excluded by express provision, or by incompatibility in its exercise arising from the nature of the particular case”). The parties agree that these principles, which have “remained unmodified through the years,” Dowd Box, supra, at 508, provide the analytical framework for resolving this case.

Ill

The precise question presented, therefore, is whether state courts have been divested of jurisdiction to hear civil RICO claims “by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.” Gulf Offshore, supra, at 478. Because we find none of these factors present with respect to civil claims arising under RICO, we hold that state courts retain their presumptive authority to adjudicate such claims.

At the outset, petitioners concede that there is nothing in the language of RICO — much less an “explicit statutory directive” — to suggest that Congress has, by affirmative enactment, divested the state courts of jurisdiction to hear civil RICO claims. The statutory provision authorizing civil RICO claims provides in full:

“Any person injured in his business or property by reason of a violation of section 1962 of this chapter may sue therefor in any appropriate United States district court and shall recover threefold the damages he sustains and the cost of the suit, including a reasonable attorney’s fee.” 18 U. S. C. § 1964(c) (emphasis added).

This grant of federal jurisdiction is plainly permissive, not mandatory, for “[t]he statute does not state nor even suggest that such jurisdiction shall be exclusive. It provides that suits of the kind described ‘may’ be brought in the federal dis*461trict courts, not that they must be.” Dowd Box, supra, at 506. Indeed, “[i]t is black letter law. . . that the mere grant of jurisdiction to a federal court does not operate to oust a state court from concurrent jurisdiction over the cause of action.” Gulf Offshore, supra, at 479 (citing United States v. Bank of New York & Trust Co., 296 U. S. 463, 479 (1936)).

Petitioners thus rely solely on the second and third factors suggested in Gulf Offshore, arguing that exclusive federal jurisdiction over civil RICO actions is established “by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests,” 453 U. S., at 478.

Our review of the legislative history, however, reveals no evidence that Congress even considered the question of concurrent state court jurisdiction over RICO claims, much less any suggestion that Congress affirmatively intended to confer exclusive jurisdiction over such' claims on the federal courts. As the Courts of Appeals that have considered the question have concluded, “[t]he legislative history contains no indication that Congress ever expressly considered the question of concurrent jurisdiction; indeed, as the principal draftsman of RICO has remarked, ‘no one even thought of the issue.’” Brandenburg, 859 F. 2d, at 1193 (quoting Flaherty, Two States Lay Claim to RICO, Nat. L. J., May 7, 1984, p. 10, col. 2); see also Lou v. Belzberg, 834 F. 2d 730, 736 (CA9 1987) (“The legislative history provides ‘no evidence that Congress ever expressly considered the question of jurisdiction; indeed, the evidence establishes that its attention was focused solely on whether to provide a private right of action’”) (citation omitted), cert. denied, 485 U. S. 993 (1988); Chivas Products Ltd. v. Owen, 864 F. 2d 1280, 1283 (CA6 1988) (“There is no ‘smoking gun’ legislative history in which RICO sponsors indicated an express intention to commit civil RICO to the federal courts”). Petitioners nonetheless insist that if Congress had considered the issue, it would have granted federal courts exclusive jurisdiction over civil *462RICO claims. This argument, however, is misplaced, for even if we could reliably discern what Congress’ intent might have been had it considered the question, we are not at liberty to so speculate; the fact that Congress did not even consider the issue readily disposes of any argument that Congress unmistakably intended to divest state courts of concurrent jurisdiction.

Sensing this void in the legislative history, petitioners rely, in the alternative, on our decisions in Sedima, S. P. R. L. v. Imrex Co., 473 U. S. 479 (1985), and Agency Holding Corp. v. Malley-Duff & Assocs., 483 U. S. 143 (1987), in which we noted that Congress modeled § 1964(c) after § 4 of the Clayton Act, 15 U. S. C. § 15(a). See Sedima, supra, at 489; Agency Holding, supra, at 151-152. Petitioners assert that, because we have interpreted § 4 of the Clayton Act to confer exclusive jurisdiction on the federal courts, see, e. g., General Investment Co. v. Lake Shore & M. S. R. Co., 260 U. S. 261, 286-288 (1922), and because Congress may be presumed to have been aware of and incorporated those interpretations when it used similar language in RICO, cf. Cannon v. University of Chicago, 441 U. S. 677, 694-699 (1979), Congress intended, by implication, to grant exclusive federal jurisdiction over claims arising under § 1964(c).

This argument is also flawed. To rebut the presumption of concurrent jurisdiction, the question is not whether any intent at all may be divined from legislative silence on the issue, but whether Congress in its deliberations may be said to have affirmatively or unmistakably intended jurisdiction to be exclusively federal. In the instant case, the lack of any indication in RICO’s legislative history that Congress either considered or assumed that the importing of remedial language from the Clayton Act into RICO had any jurisdictional implications is dispositive. The “mere borrowing of statutory language does not imply that Congress also intended to incorporate all of the baggage that may be attached to the borrowed language.” Lou, supra, at 737. Indeed, to the *463extent we impute to Congress knowledge of our Clayton Act precedents, it makes no less sense to impute to Congress knowledge of Claflin and Dowd Box, under which Congress, had it sought to confer exclusive jurisdiction over civil RICO claims, would have had every incentive to do so expressly.

Sedima and Agency Holding are not to the contrary. Although we observed in Sedima that “[t]he clearest current in [the legislative] history [of § 1964(c)] is the reliance on the Clayton Act model,” 473 U. S., at 489, that statement was made in the context of noting the distinction between “private and governmental actions” under the Clayton Act. Ibid. We intimated nothing as to whether Congress’ reliance on the Clayton' Act implied any intention to establish exclusive federal jurisdiction for civil RICO claims, and in Sedima itself we rejected any requirement of proving “racketeering injury,” noting that to borrow the “antitrust injury” requirement from antitrust law would “creat[e] exactly the problems Congress sought to avoid.” Id., at 498-499. Likewise, in Agency Holding we were concerned with “borrowing,” in light of legislative silence on the issue, an appropriate statute of limitations period from an “analogous” statute. 483 U. S., at 146. Under such circumstances, we found it appropriate to borrow the statute of limitations from the Clayton Act. Id., at 152. In this case, by contrast, where the issue is whether jurisdiction is exclusive or concurrent, we are not free to add content to a statute via analogies to other statutes unless the legislature has specifically endorsed such action. Under Gulf Offshore, legislative silence counsels, if not compels, us to enforce the presumption of concurrent jurisdiction. In short, in both Sedima and Agency Holding we looked to the Clayton Act in interpreting RICO without the benefit of a background juridical presumption of the type present in this case. Thus, to whatever extent the Clayton Act analogy may be relevant to our interpretation of RICO generally, it has no place in our inquiry into the jurisdiction of state courts.

*464Petitioners finally urge that state court jurisdiction over civil RICO claims would be clearly incompatible with federal interests. We noted in Gulf Offshore that factors indicating clear incompatibility “include the desirability of uniform interpretation, the expertise of federal judges in federal law, and the assumed greater hospitality of federal courts to peculiarly federal claims.” 453 U. S., at 483-484 (citation and footnote omitted). Petitioners’ primary contention is that concurrent jurisdiction is clearly incompatible with the federal interest in uniform interpretation of federal criminal laws, see 18 U. S. C. § 3231,2 because state courts would be required to construe the federal crimes that constitute predicate acts defined as “racketeering activity,” see 18 U. S. C. §§ 1961(1)(B), (C), and (D). Petitioners predict that if state courts are permitted to interpret federal criminal statutes, they will create a body of precedent relating to those statutes and that the federal courts will consequently lose control over the orderly and uniform development of federal criminal law.

We perceive no “clear incompatibility” between state court jurisdiction over civil RICO actions and federal interests. As a preliminary matter, concurrent jurisdiction over § 1964(c) suits is clearly not incompatible with § 3231 itself, for civil RICO claims are not “offenses against the laws of the United States,” §3231, and do not result in the imposition of criminal sanctions — uniform or otherwise. See Shearson/American Express Inc. v. McMahon, 482 U. S. 220, 240-241 (1987) (civil RICO intended to be primarily remedial rather than punitive).

More to the point, however, our decision today creates no significant danger of inconsistent application of federal crimi*465nal law. Although petitioners’ concern with the need for uniformity and consistency of federal criminal law is well taken, see Ableman v. Booth, 21 How. 506, 517-518 (1859); cf. Musser v. Utah, 333 U. S. 95, 97 (1948) (vague criminal statutes may violate the Due Process Clause), federal courts, pursuant to § 3231, would retain full authority and responsibility for the interpretation and application of federal criminal law, for they would not be bound by state court interpretations of the federal offenses constituting RICO’s predicate acts. State courts adjudicating civil RICO claims will, in addition, be guided by federal court interpretations of the relevant federal criminal statutes, just as federal courts sitting in diversity are guided by state court interpretations of state law, see, e. g., Commissioner v. Estate of Bosch, 387 U. S. 456, 465 (1967). State court judgments misinterpreting federal criminal law would, of course, also be subject to direct review by this Court. Thus, we think that state court adjudication of civil RICO actions will, in practice, have at most a negligible effect on the uniform interpretation and application of federal criminal law, cf. Pan-American Petroleum Corp. v. Superior Court of Delaware, Newcastle County, 366 U. S. 656, 665-666 (1961) (rejecting claim that uniform interpretation of the Natural Gas Act will be jeopardized by concurrent jurisdiction), and will not, in any event, result in any more inconsistency than that which a multimembered, multi-tiered federal judicial system already creates, cf. H. J. Inc. v. Northwestern Bell Telephone Co., 492 U. S. 229, 235, n. 2 (1989) (surveying conflict among federal appellate courts over RICO’s “pattern of racketeering activity” requirement).

Moreover, contrary to petitioners’ fears, we have full faith in the ability of state courts to handle the complexities of civil RICO actions, particularly since many RICO cases involve asserted violations of state law, such as state fraud claims, over which state courts presumably have greater expertise. See 18 U. S. C. § 1961(1)(A) (listing state law offenses constituting predicate acts); Gulf Offshore, supra, at 484 *466(“State judges have greater expertise in applying” laws “whose governing rules are borrowed from state law”); see also Sedima, 473 U. S., at 499 (RICO “has become a tool for everyday fraud cases”); BNA, Civil RICO Report, Vol. 2, No. 44, p. 7 (Apr. 14, 1987) (54.9% of all RICO cases after Sedima involved “common law fraud” and another 18% involved either “nonsecurities fraud” or “theft or conversion”). To hold otherwise would not only denigrate the respect accorded coequal sovereigns, but would also ignore our “consistent history of hospitable acceptance of concurrent jurisdiction,” Dowd Box, 368 U. S., at 508. Indeed, it would seem anomalous to rule that state courts are incompetent to adjudicate civil RICO suits when we have recently found no inconsistency in subjecting civil RICO claims to adjudication by arbitration. See Shear son/American Express, supra, at 239 (rejecting argument that “RICO claims are too complex to be subject to arbitration” and that “there is an irreconcilable conflict between arbitration and RICO’s underlying purposes”).

Petitioners further note, as evidence of incompatibility, that RICO’s procedural mechanisms include extended venue and service-of-process provisions that are applicable only in federal court, see 18 U. S. C. § 1965. We think it sufficient, however, to observe that we have previously found concurrent state court jurisdiction even where federal law provided for special procedural mechanisms similar to those found in RICO. See, e. g., Dowd Box, supra (finding concurrent jurisdiction over Labor Management Relations Act § 301(a) suits, despite federal enforcement and venue provisions); Maine v. Thiboutot, 448 U. S. 1, 3, n. 1 (1980) (finding concurrent jurisdiction over 42 U. S. C. § 1983 suits, despite federal procedural provisions in § 1988); cf. Hathorn v. Lovorn, 457 U. S. 255, 269 (1982) (finding concurrent jurisdiction over disputes regarding the applicability of §5 of the Voting Rights Act of 1965, 42 U. S. C. § 1973c, despite provision for a three-judge panel). Although- congressional specification *467of procedural mechanisms applicable only in federal court may tend to suggest that Congress intended exclusive federal jurisdiction, it does not by itself suffice to create a “clear incompatibility” with federal interests.

Finally, we note that, far from disabling or frustrating federal interests, “[permitting state courts to entertain federal causes of action facilitates the enforcement of federal rights.” Gulf Offshore, 453 U. S., at 478, n. 4; see also Dowd Box, supra, at 514 (conflicts deriving from concurrent jurisdiction are “not necessarily unhealthy”). Thus, to the extent that Congress intended RÍCO to serve broad remedial purposes, see, e. g., Pub. L. 91-452, § 904(a), 84 Stat. 947 (RICO must “be liberally construed to effectuate its remedial purposes”); Sedima, supra, at 492, n. 10 (“[I]f Congress’ liberal-construction mandate is to be applied anywhere, it is in § 1964, where RICO’s remedial purposes are most evident”), concurrent state court jurisdiction over civil RICO claims will advance rather than jeopardize federal policies underlying the statute.

For all of the above reasons, we hold that state courts have concurrent jurisdiction to consider civil claims arising under RICO. Nothing in the language, structure, legislative history, or underlying policies of RICO suggests that Congress intended otherwise. The judgment of the Court of Appeals is accordingly

Affirmed.

concurring.

I agree that state courts have concurrent jurisdiction over civil RICO actions and join the opinion and judgment of the Court. I add a few words only because this Court has rarely considered contentions that civil actions based on federal criminal statutes must be heard by the federal courts. As the Court observes, ante, at 465, the uniform construction of federal criminal statutes is no insignificant matter, particularly because Congress has recognized potential dangers in nonuniform construction and has confined jurisdiction over *468federal criminal cases to the federal courts. There is, therefore, reason for caution before concluding that state courts have jurisdiction over civil claims related to federal criminal statutes and for assessing in each case the danger to federal interests presented by potential inconsistent constructions of federal criminal statutes.

RICO is an unusual federal criminal statute. It borrows heavily from state law; racketeering activity is defined in terms of numerous offenses chargeable under state law, 18 U. S. C. § 1961(1)(A), as well as various federal offenses. To the extent that there is any danger under RICO of nonuniform construction of criminal statutes, it is quite likely that the damage will result from federal misunderstanding of the content of state law — a problem, to be sure, but not one to be solved by exclusive federal jurisdiction. Many of the federal offenses named as racketeering activity under RICO have close, though perhaps not exact, state-law analogues, cf. Durland v. United States, 161 U. S. 306, 312 (1896), which construed the federal mail fraud statute, and it is unlikely that the state courts will be incompetent to construe those federal statutes. Nor does incorrect state-court construction of those statutes present as significant a threat to federal interests as that posed by improper interpretation of the federal antitrust laws, which could have a disastrous effect on interstate commerce, a particular concern of the Federal Government. Racketeering activity as defined by RICO includes other federal offenses without state-law analogues, but given the history as written until now of civil RICO litigation, I doubt that state-court construction of these offenses will be greatly disruptive of important federal interests.

There is also the possibility that the state courts will disrupt the uniform construction of criminal RICO by launching new interpretations of the “pattern” and “enterprise” elements of that offense when hearing civil RICO suits. This possibility, though not insubstantial, cf. H. J. Inc. v. North*469western Bell Telephone Co., 492 U. S. 229 (1989), is not enough to require exclusive federal jurisdiction of civil RICO claims. Even though varying interpretations of the “pattern” and “enterprise” elements of RICO may drastically change the consequences that flow from particular acts, these variations cannot make an act criminal in one court system but blameless in another and therefore do not implicate the core due process concerns identified by the Court, ante, at 464, as underlying the need for uniform construction of criminal statutes. Moreover, we have the authority to reduce the risk of, and to set aside, incorrect interpretations of these elements of RICO liability.

with whom Justice Kennedy joins, concurring.

I join the opinion of the Court, addressing the issues before us on the basis argued by the parties, which has included acceptance of the dictum in Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473, 478 (1981), that “‘the presumption of concurrent jurisdiction can be rebutted by an explicit statutory directive, by unmistakable implication from legislative history, or by a clear incompatibility between state-court jurisdiction and federal interests.’” Ante, at 459-460. Such dicta, when repeatedly used as the point of departure for analysis, have a regrettable tendency to acquire the practical status of legal rules. I write separately, before this one has become too entrenched, to note my view that in one respect it is not a correct statement of the law, and in another respect it may not be.

State courts have jurisdiction over federal causes of action not because it is “conferred” upon them by the Congress; nor even because their inherent powers permit them to entertain transitory causes of action arising under the laws of foreign sovereigns, see, e. g., McKenna v. Fisk, 1 How. 241, 247-249 (1843); but because “[t]he laws of the United States are laws in the several States, and just as much binding on the citizens and courts thereof as the State laws are. . . . The two *470together form one system of jurisprudence, which constitutes the law of the land for the State; and the courts of the two jurisdictions are not foreign to each other . . . Claflin v. Houseman, 93 U. S. 130, 136-137 (1876); see also Minneapolis & St. Louis R. Co. v. Bombolis, 241 U. S. 211, 221-223 (1916).

It therefore takes an affirmative act of power under the Supremacy Clause to oust the States of jurisdiction — an exercise of what one of our earliest cases referred to as “the power of congress to withdraw” federal claims from state-court jurisdiction. Houston v. Moore, 5 Wheat. 1, 26 (1820) (emphasis added). See also Bombolis, supra, at 221 (concurrent jurisdiction exists “unless excepted by express constitutional limitation or by valid legislation”); Missouri ex rel. St. Louis, B. & M. R. Co. v. Taylor, 266 U. S. 200, 208 (1924) (“As [Congress] made no provision concerning the remedy, the federal and the state courts have concurrent jurisdiction”).

As an original proposition, it would be eminently arguable that depriving state courts of their sovereign authority to adjudicate the law of the land must be done, if not with the utmost clarity, cf. Atascadero State Hospital v. Scanlon, 473 U. S. 234, 243 (1986) (state sovereign immunity can be eliminated only by “clear statement”), at least expressly. That was the view of Alexander Hamilton:

“When ... we consider the State governments and the national governments, as they truly are, in the light of kindred systems, and as parts of ONE WHOLE, the inference seems to be conclusive that the State courts would have a concurrent jurisdiction in all cases arising under the laws of the Union, where it was not expressly prohibited.” The Federalist No. 82, p. 132 (E. Bourne ed. 1947).

See also Galveston, H. & S. A. R. Co. v. Wallace, 223 U. S. 481, 490 (1912) (“[jurisdiction is not defeated by implication”). Although as early as Claflin, see 93 U. S., at 137, *471and as late as Gulf Offshore, we have said that the exclusion of concurrent state jurisdiction could be achieved by implication, the only cases in which to my knowledge we have acted upon such a principle are those relating to the Sherman Act and the Clayton Act — where the full extent of our analysis was the less than compelling statement that provisions giving the right to sue in United States District Court “show that [the right] is to be exercised only in a ‘court of the United States.’” General Investment Co. v. Lake Shore & Michigan Southern R. Co., 260 U. S. 261, 287 (1922) (emphasis added). See also Blumenstock Bros. Advertising Agency v. Curtis Publishing Co., 252 U. S. 436, 440 (1920) (dictum); Freeman v. Bee Machine Co., 319 U. S. 448, 451, n. 6 (1943) (dictum); Hathom v. Lovom, 457 U. S. 255, 267, n. 18 (1982) (dictum). In the standard fields of exclusive federal jurisdiction, the governing statutes specifically recite that suit may be brought “only” in federal court, Investment Company Act of 1940, as amended, 84 Stat. 1429, 15 U. S. C. § 80a-35(b) (5); that the jurisdiction of the federal courts shall be “exclusive,” Securities Exchange Act of 1934, as amended, 48 Stat. 902,15 U. S. C. § 78aa; Natural Gas Act of 1938, 52 Stat. 833, 15 U. S. C. §717u; Employee Retirement Income Security Act of 1974, 88 Stat. 892, 29 U. S. C. § 1132(e)(1); or indeed even that the jurisdiction of the federal courts shall be “exclusive of the courts of the States,” 18 U. S. C. §3231 (criminal cases); 28 U. S. C. §§ 1333 (admiralty, maritime, and prize cases), 1334 (bankruptcy cases), 1338 (patent, plant variety protection, and copyright cases), 1351 (actions against consuls or vice consuls of foreign states), 1355 (actions for recovery or enforcement of fine, penalty, or forfeiture incurred under Act of Congress), 1356 (seizures on land or water not within admiralty and maritime jurisdiction).

Assuming, however, that exclusion by implication is possible, surely what is required is implication in the text of the statute, and not merely, as the second part of the Gulf Offshore dictum would permit, through “unmistakable implica*472tion from legislative history.” 453 U. S., at 478. Although Charles Dowd Box Co. v. Courtney, 368 U. S. 502 (1962), after concluding that the statute “does not state nor even suggest that [federal] jurisdiction shall be exclusive,” id., at 506, proceeded quite unnecessarily to examine the legislative history, it did so to reinforce rather than contradict the conclusion it had already reached. We have never found state jurisdiction excluded by “unmistakable implication” from legislative history. It is perhaps harmless enough to say that it can be, since one can hardly imagine an “implication from legislative history” that is “unmistakable” — i. e., that demonstrates agreement to a proposition by a majority of both Houses and the President — unless the proposition is embodied in statutory text to which those parties have given assent. But harmless or not, it is simply wrong in principle to assert that Congress can effect this affirmative legislative act by simply talking about it with unmistakable clarity. What is needed to oust the States of jurisdiction is congressional action (i. e., a provision of law), not merely congressional discussion.

It is perhaps also true that implied preclusion can be established by the fact that a statute expressly mentions only federal courts, plus the fact that state-court jurisdiction would plainly disrupt the statutory scheme. That is conceivably what was meant by the third part of the Gulf Offshore dictum, “clear incompatibility between state-court jurisdiction and federal interests.” 453 U. S., at 478. If the phrase is interpreted more broadly than that, however — if it is taken to assert some power on the part of this Court to exclude state-court jurisdiction when systemic federal interests make it undesirable — it has absolutely no foundation in our precedent.

Gulf Offshore cited three cases to support its “incompatibility” formulation. The first was Dowd Box, swpra, at 507-508, which contains nothing to support any “incompatibility” principle, except a quotation from the second case Gulf Off*473shore cited, Claflin. Indeed, in response to the argument that “[o]nly the federal judiciary . . . possesses both the familiarity with federal labor legislation and the monolithic judicial system necessary” to elaborate a coherent system of national labor laws, the Dowd Box opinion said: “Whatever the merits of this argument as a matter of policy, we find nothing to indicate that Congress adopted such a policy in enacting § 301.” 368 U. S., at 507. The second case cited was Claflin, which said that concurrent jurisdiction exists “where it is not excluded by express provision or by incompatibility in its exercise arising from the nature of the particular case.” 93 U. S., at 136. The subsequent discussion makes it entirely clear, however, that what the Court meant by “incompatibility in its exercise arising from the nature of the particular case” was that the particular statute at issue impliedly excluded state-court jurisdiction. “Congress,” the Court said, “may, if it sees fit, give to the Federal courts exclusive jurisdiction,” which it does “sometimes ... by express enactment and sometimes by implication.” Id., at 137. The third case cited, Garner v. Teamsters, 346 U. S. 485 (1953), had nothing to do with state-court jurisdiction over a federal cause of action. It held that the National Labor Relations Act, whose express provision that the jurisdiction of the National Labor Relations Board shall be exclusive had already been held to prevent federal courts from assuming primary jurisdiction over labor disputes, see Myers v. Bethlehem Shipbuilding Corp., 303 U. S. 41, 48 (1938), prevented state courts as well.

In sum: As the Court holds, the RICO cause of action meets none of the three tests for exclusion of state-court jurisdiction recited in Gulf Offshore. Since that is so, the proposition that meeting any one of the tests would have sufficed is dictum here, as it was there. In my view meeting the second test is assuredly not enough, and meeting the third may not be.