JUSTICE SCALIA delivered the opinion of the Court.
The question presented in this case is whether, in a suit brought by a limited partnership, the citizenship of the limited partners must be taken into account to determine diversity of citizenship among the parties.
Respondent Arkoma Associates (Arkoma), a limited partnership organized under the laws of Arizona, brought suit on a contract dispute in the United States District Court for the Eastern District of Louisiana, relying upon diversity of citizenship for federal jurisdiction. The defendants, C. Tom Carden and Leonard L. Limes, citizens of Louisiana, moved to dismiss, contending that one of Arkoma's limited partners was also a citizen of Louisiana. The District Court denied the motion but certified the question for interlocutory appeal, which the Fifth Circuit declined. Thereafter Magee Drilling Company intervened in the suit and, together with the original defendants, counterclaimed against Arkoma under Texas law. Following a bench trial, the District Court awarded Arkoma a money judgment plus interest and attorney's fees; it dismissed Carden and Limes' counterclaim as well as Magee's intervention and counterclaim. Carden, Limes, and Magee (petitioners here) appealed, and the Fifth Circuit affirmed. F.2d 226 (1988). With respect to petitioners' jurisdictional challenge, the Court of Appeals found complete diversity, reasoning that Arkoma's citizenship should be determined by reference to the citizenship of the general, but not the limited, partners. We granted certiorari. 490 U.S. 1045 (1989).
Article III of the Constitution provides, in pertinent part, that "the judicial Power shall extend to . . . Controversies . . . between Citizens of different States." Congress first authorized the federal courts to exercise diversity jurisdiction in the Judiciary Act of 1789, ch. 20, § 11, 1 Stat. 78. In its current form, the diversity statute provides that "the district courts shall have original jurisdiction of all civil actions where the matter in controversy exceeds . . . $50,000 . . ., and is between . . . citizens of different States . . . ." 28 U.S.C. § 1332(a). Since its enactment, we have interpreted the diversity statute to require "complete diversity" of citizenship. See Strawbridge v. Curtiss, 3 Cranch 267 (1806). The District Court erred in finding complete diversity in this case unless (1) a limited partnership may be considered in its own right a "citizen" of the State that created it, or (2) a federal court must look to the citizenship of only its general, but not its limited, partners to determine whether there is complete diversity of citizenship. We consider these questions in turn.
We have often had to consider the status of artificial entities created by state law insofar as that bears upon the existence of federal diversity jurisdiction. The precise question posed under the terms of the diversity statute is whether such an entity may be considered a "citizen" of the State under whose laws it was created.1 A corporation is the paradigmatic artificial "person," and the Court has considered its proper characterization under the diversity statute on more than one occasion -- not always reaching the same conclusion. Initially, we held that a corporation "is certainly not a citizen," so that to determine the existence of diversity jurisdiction the Court must "look to the character of the individuals who compose [it]." Bank of United States v. Deveaux, 5 Cranch 61, 86, 91-92 (1809). We overruled Deveaux 35 years later in Louisville, C. & C. R. Co. v. Letson, 2 How. 497, 558 (1844), which held that a corporation is "capable of being treated as a citizen of [the State which created it], as much as a natural person." Ten years later, we reaffirmed the result of Letson, though on the somewhat different theory that "those who use the corporate name, and exercise the faculties conferred by it," should be presumed conclusively to be citizens of the corporation's State of incorporation. Marshall v. Baltimore & Ohio R. Co., 16 How. 314, 329 (1854). While the rule regarding the treatment of corporations as "citizens" has become firmly established, we have (with an exception to be discussed presently) just as firmly resisted extending that treatment to other entities. For example, in Chapman v. Barney, 129 U.S. 677 (1889), a case involving an unincorporated "joint stock company," we raised the question of jurisdiction on our own motion, and found it to be lacking:
"On looking into the record we find no satisfactory showing as to the citizenship of the plaintiff. The allegation of the amended petition is, that the United States Express Company is a joint stock company organized under a law of the State of New York, and is a citizen of that State. But the express company cannot be a citizen of New York, within the meaning of the statutes regulating jurisdiction, unless it be a corporation. The allegation that the company was organized under the laws of New York is not an allegation that it is a corporation. In fact the allegation is, that the company is not a corporation, but a joint stock company -- that is, a mere partnership." Id., at 682.
Similarly, in Great Southern Fire Proof Hotel Co. v. Jones, 177 U.S. 449 (1900), we held that a "limited partnership association" -- although possessing "some of the characteristics of a corporation" and deemed a "citizen" by the law creating it -- may not be deemed a "citizen" under the jurisdictional rule established for corporations. Id., at 456. "That rule must not be extended." Id., at 457. As recently as 1965, our unanimous opinion in Steelworkers v. R. H. Bouligny, Inc., 382 U.S. 145, reiterated that "the doctrinal wall of Chapman v. Barney," id., at 151, would not be breached.
The one exception to the admirable consistency of our jurisprudence on this matter is Puerto Rico v. Russell & Co., 288 U.S. 476 (1933), which held that the entity known as a sociedad en comandita, created under the civil law of Puerto Rico, could be treated as a citizen of Puerto Rico for purposes of determining federal-court jurisdiction. The sociedad 's juridical personality, we said, "is so complete in contemplation of the law of Puerto Rico that we see no adequate reason for holding that the sociedad has a different status for purposes of federal jurisdiction than a corporation organized under that law." Id., at 482. Arkoma fairly argues that this language, and the outcome of the case, "reflect the Supreme Court's willingness to look beyond the incorporated/unincorporated dichotomy and to study the internal organization, state law requirements, management structure, and capacity or lack thereof to act and/or sue, to determine diversity of citizenship." Brief for Respondent 14. The problem with this argument lies not in its logic, but in the fact that the approach it espouses was proposed and specifically rejected in Bouligny. There, in reaffirming "the doctrinal wall of Chapman v. Barney," we explained Russell as a case resolving the distinctive problem "of fitting an exotic creation of the civil law . . . into a federal scheme which knew it not." 382 U.S., at 151. There could be no doubt, after Bouligny, that at least common-law entities (and likely all entities beyond the Puerto Rican sociedad en comandita) would be treated for purposes of the diversity statute pursuant to what Russell called "the tradition of the common law," which is "to treat as legal persons only incorporated groups and to assimilate all others to partnerships." 288 U.S., at 480.2 Arkoma claims to have found another exception to our Chapman tradition in Navarro Savings Assn. v. Lee, 446 U.S. 458 (1980). That case, however, did not involve the question whether a party that is an artificial entity other than a corporation can be considered a "citizen" of a State, but the quite separate question whether parties that were undoubted "citizens" (viz., natural persons) were the real parties to the controversy. The plaintiffs in Navarro were eight individual trustees of a Massachusetts business trust, suing in their own names. The defendant, Navarro Savings Association, disputed the existence of complete diversity, claiming that the trust beneficiaries rather than the trustees were the real parties to the controversy, and that the citizenship of the former and not the latter should therefore control. In the course of rejecting this claim, we did indeed discuss the characteristics of a Massachusetts business trust -- not at all, however, for the purpose of determining whether the trust had attributes making it a "citizen," but only for the purpose of establishing that the respondents were "active trustees whose control over the assets held in their names is real and substantial," thereby bringing them under the rule, "more than 150 years" old, which permits such trustees "to sue in their own right, without regard to the citizenship of the trust beneficiaries." Id., at 465-466. Navarro, in short, has nothing to do with the Chapman question, except that it makes available to respondent the argument by analogy that, just as business reality is taken into account for purposes of determining whether a trustee is the real party to the controversy, so also it should be taken into account for purposes of determining whether an artificial entity is a citizen. That argument is, to put it mildly, less than compelling.
As an alternative ground for finding complete diversity, Arkoma asserts that the Fifth Circuit correctly determined its citizenship solely by reference to the citizenship of its general partners, without regard to the citizenship of its limited partners. Only the general partners, it points out, "manage the assets, control the litigation, and bear the risk of liability for the limited partnership's debts," and, more broadly, "have exclusive and complete management and control of the operations of the partnership." Brief for Respondent 30, 36. This approach of looking to the citizenship of only some of the members of the artificial entity finds even less support in our precedent than looking to the State of organization (for which one could at least point to Russell). We have never held that an artificial entity, suing or being sued in its own name, can invoke the diversity jurisdiction of the federal courts based on the citizenship of some but not all of its members. No doubt some members of the joint stock company in Chapman, the labor union in Bouligny, and the limited partnership association in Great Southern exercised greater control over their respective entities than other members. But such considerations have played no part in our decisions.
To support its approach, Arkoma seeks to press Navarro into service once again, arguing that just as that case looked to the trustees to determine the citizenship of the business trust, so also here we should look to the general partners, who have the management powers, in determining the citizenship of this partnership. As we have already explained, however, Navarro had nothing to do with the citizenship of the "trust," since it was a suit by the trustees in their own names.
The dissent supports Arkoma's argument on this point, though, as we have described, under the rubric of determining which parties supposedly before the Court are the real parties, rather than under the rubric of determining the citizenship of the limited partnership. See n. 1, supra. The dissent asserts that "the real party to the controversy approach," post, at 201 -- by which it means an approach that looks to "control over the conduct of the business and the ability to initiate or control the course of litigation," post, at 204 -- "has been implemented by the Court both in its oldest and in its most recent cases examining diversity jurisdiction with respect to business associations." Post, at 201. Not a single case the dissent discusses, either old or new, supports that assertion. Deveaux, which was in any event overruled by Letson, seems to be applying not a "real party to the controversy" test, but rather the principle that for jurisdictional purposes the corporation has no substance, and merely "represents" its shareholders, see 5 Cranch, at 90-91; but even if it can be regarded as applying a "real party to the controversy" test, it deems that test to be met by all the shareholders of the corporation, without regard to their "control over the operation of the business." Marshall, which as we have discussed rerationalized Letson 's holding that a corporation was a "citizen" in its own right, contains language quite clearly adopting a "real party to the controversy" approach, and arguably even adopting a "control" test for that status. ("The court . . . will look behind the corporate or collective name . . . to find the persons who act as the representatives, curators, or trustees. . . ." 16 How., at 328-329 (emphasis added). "The presumption arising from the habitat of a corporation in the place of its creation [is] conclusive as to the residence or citizenship of those who use the corporate name and exercise the faculties conferred by it . . . ." Id., at 329 (emphasis added).) But as we have also discussed, and as the last quotation shows, that analysis was a complete fiction; the real citizenship of the shareholders (or the controlling shareholders) was not consulted at all.3 From the fictional Marshall, the dissent must leap almost a century and a third to Navarro to find a "real party to the controversy" analysis that discusses "control." That case, as we have said, is irrelevant, since it involved not a juridical person but the distinctive common-law institution of trustees.
The dissent finds its position supported, rather than contradicted, by the trilogy of Chapman, Great Southern, and Bouligny -- cases that did involve juridical persons but that did not apply "real party to the controversy" analysis, much less a "control" test as the criterion for that status. In those cases, the dissent explains, "the members of each association held equivalent power and control over the association's assets, business, and litigation." Post, at 202. It seeks to establish this factual matter, however, not from the text of the opinions (where not the slightest discussion of the point appears) but, for Chapman, by citation of scholarly commentary dealing with the general characteristics of joint stock company agreements, with no reference to (because the record does not contain) the particular agreement at issue in the case, post, at 202-203; for Great Southern, by citation of scholarly commentary dealing with the general characteristics of Pennsylvania limited partnership associations, and citation of Pennsylvania statutes, post, at 203; and, for Bouligny, by nothing more than the observation that "there was no indication that any of the union members had any greater power over the litigation or the union's business and assets than any other member, and, therefore, as in Chapman and Great Southern, the Court was not called upon to decide" the issue, post, at 204. This will not do. Since diversity of citizenship is a jurisdictional requirement, the Court is always "called upon to decide" it. As the Court said in Great Southern itself:
"The failure of parties to urge objections [to diversity of citizenship] cannot relieve this court from the duty of ascertaining from the record whether the Circuit Court could properly take jurisdiction of this suit. . . . 'The rule . . . is inflexible and without exception, which requires this court, of its own motion, to deny its own jurisdiction, and, in the exercise of its appellate power, that of all other courts of the United States, in all cases where such jurisdiction does not affirmatively appear in the record on which, in the exercise of that power, it is called to act.'" 177 U.S., at 453 (quoting Mansfield, C. & L. M. R. Co. v. Swan, 111 U.S. 379, 382 (1884)).
If, as the dissent contends, these three cases were applying a "real party to the controversy" test governed by "control" over the associations, so that the citizenship of all members would be consulted only if all members had equivalent control, it is inconceivable that the existence of equivalency, or at least the absence of any reason to suspect nonequivalency, would not have been mentioned in the opinions. Given what 180 years of cases have said and done, as opposed to what they might have said, it is difficult to understand how the dissent can characterize as "newly formulated" the "rule that the Court will, without analysis of the particular entity before it, count every member of an unincorporated association for purposes of diversity jurisdiction." Post, at 199.
In sum, we reject the contention that to determine, for diversity purposes, the citizenship of an artificial entity, the court may consult the citizenship of less than all of the entity's members. We adhere to our oft-repeated rule that diversity jurisdiction in a suit by or against the entity depends on the citizenship of "all the members," Chapman, 129 U.S., at 682, "the several persons composing such association," Great Southern, 177 U.S., at 456, "each of its members," Bouligny, 382 U.S., at 146.
The resolutions we have reached above can validly be characterized as technical, precedent-bound, and unresponsive to policy considerations raised by the changing realities of business organization. But, as must be evident from our earlier discussion, that has been the character of our jurisprudence in this field after Letson. See Currie, The Federal Courts and the American Law Institute, 36 U. Chi. L. Rev. 1, 35 (1968). Arkoma is undoubtedly correct that limited partnerships are functionally similar to "other types of organizations that have access to federal courts," and is perhaps correct that "considerations of basic fairness and substance over form require that limited partnerships receive similar treatment." Brief for Respondent 33. Similar arguments were made in Bouligny. The District Court there had upheld removal because it could divine "'no common sense reason for treating an unincorporated national labor union differently from a corporation,'" 382 U.S., at 146, and we recognized that that contention had "considerable merit," id., at 150. We concluded, however, that "whether unincorporated labor unions ought to be assimilated to the status of corporations for diversity purposes," id., at 153, is "properly a matter for legislative consideration which cannot adequately or appropriately be dealt with by this Court," id., at 147. In other words, having entered the field of diversity policy with regard to artificial entities once (and forcefully) in Letson, we have left further adjustments to be made by Congress.
Congress has not been idle. In 1958 it revised the rule established in Letson, providing that a corporation shall be deemed a citizen not only of its State of incorporation but also "of the State where it has its principal place of business." 28 U.S.C. § 1332(c). No provision was made for the treatment of artificial entities other than corporations, although the existence of many new, post- Letson forms of commercial enterprises, including at least the sort of joint stock company at issue in Chapman, the sort of limited partnership association at issue in Great Southern, and the the sort of Massachusetts business trust at issue in Navarro, must have been obvious.
Thus, the course we take today does not so much disregard the policy of accommodating our diversity jurisdiction to the changing realities of commercial organization, as it honors the more important policy of leaving that to the people's elected representatives. Such accommodation is not only performed more legitimately by Congress than by courts, but it is performed more intelligently by legislation than by interpretation of the statutory word "citizen." The 50 States have created, and will continue to create, a wide assortment of artificial entities possessing different powers and characteristics, and composed of various classes of members with varying degrees of interest and control. Which of them is entitled to be considered a "citizen" for diversity purposes, and which of their members' citizenship is to be consulted, are questions more readily resolved by legislative prescription than by legal reasoning, and questions whose complexity is particularly unwelcome at the threshold stage of determining whether a court has jurisdiction. We have long since decided that, having established special treatment for corporations, we will leave the rest to Congress; we adhere to that decision.
Arkoma argues that even if this Court finds complete diversity lacking with respect to Carden and Limes, we should nonetheless affirm the judgment with respect to Magee because complete diversity indisputably exists between Magee and Arkoma. This question was not considered by the Court of Appeals. We decline to decide it in the first instance, and leave it to be resolved by the Court of Appeals on remand. The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion.
It is so ordered.
874 F.2d 226, reversed and remanded.
JUSTICE O'CONNOR, with whom JUSTICE BRENNAN, JUSTICE MARSHALL, and JUSTICE BLACKMUN join, dissenting.
The only potentially nondiverse party in this case is a limited partner. All other parties, including the general partners and the limited partnership itself, assuming it is a citizen, are diverse. Thus, the Court has before it a single question -- whether the citizenship of a limited partner must be counted for purposes of diversity jurisdiction. The Court first addresses whether the limited partnership is a "citizen." I do not consider that issue, because even if we were to hold that a limited partnership is a citizen, we are still required to consider which, if any, of the other citizens before the Court as members of Arkoma Associates are real parties to the controversy, i. e., which parties have control over the subject of and litigation over the controversy. See Marshall v. Baltimore & Ohio R. Co., 16 How. 314, 328 (1854). Application of that test leads me to conclude that limited partners are not real parties to the controversy and, therefore, should not be counted for purposes of diversity jurisdiction.
The Court asserts that "we have long since decided" to leave to Congress the issue of the proper treatment of unincorporated associations for diversity purposes, because the issue of which business association "is entitled to be considered a 'citizen' for diversity purposes, and which of their members' citizenship is to be consulted, are questions more readily resolved by legislative prescription than by legal reasoning." Ante, at 197. That assertion is insupportable in light of Navarro Savings Assn. v. Lee, 446 U.S. 458 (1980) (determination of which members of unincorporated business trust must be considered for purposes of diversity jurisdiction), and even Steelworkers v. R. H. Bouligny, Inc., 382 U.S. 145 (1965) (determination of proper treatment of union for diversity jurisdiction purposes according to settled law; Congress has power to change result), on which the Court relies. Ante, at 196. Indeed, the Court in this case does not leave the issue to Congress, but rather decides the issue and then invokes deference to Congress to justify its newly formulated rule that the Court will, without analysis of the particular entity before it, count every member of an unincorporated association for purposes of diversity jurisdiction. In my view, the Court properly tackles the issue, because "application of statutes to situations not anticipated by the legislature is a pre-eminently judicial function." Currie, Federal Courts and the American Law Institute, 36 U. Chi. L. Rev. 1, 35 (1968); see also Bank of United States v. Deveaux, 5 Cranch 61, 87 (1809) ("The duties of this Court, to exercise jurisdiction where it is conferred, and not to usurp it where it is not conferred, are of equal obligation. The constitution, therefore, and the law, are to be expounded, without a leaning the one way or the other, according to those general principles which usually govern in the construction of fundamental or other laws").
The starting point for any analysis of who must be counted for purposes of diversity jurisdiction is Strawbridge v. Curtiss, 3 Cranch 267 (1806), in which the Court held that "complete diversity" is required among "citizens" of different States. Complete diversity, however, is not constitutionally mandated. See State Farm Fire & Casualty Co. v. Tashire, 386 U.S. 523, 530-531 (1967) (statutory interpleader need not satisfy complete diversity requirement as long as there is diversity between two or more claimants); see also American Law Institute, Study of the Division of Jurisdiction Between State and Federal Courts § 1301(b)(2), Supporting Memorandum A, pp. 426-436 (1969). For example, in a class action authorized pursuant to Federal Rule of Civil Procedure 23, only the citizenship of the named representatives of the class is considered, without regard to whether the citizenship of other members of the class would destroy complete diversity or to the class members' particular stake in the controversy. See Snyder v. Harris, 394 U.S. 332, 340 (1969); C. Wright, Law of Federal Courts 484 (4th ed. 1983); see also Owen Equipment & Erection Co. v. Kroger, 437 U.S. 365, 375, and n. 18 (1978) (citizenship of parties joined under ancillary jurisdiction not taken into account for purposes of determining diversity jurisdiction); Wright, supra, at 28 (same).
Since the early 19th century, one of the benchmarks for determining whether a particular party among those involved in the litigation must be counted for purposes of diversity jurisdiction has been whether the party has a "real interest" in the suit or, in other words, is a "real party" to the controversy. See 6 C. Wright & A. Miller, Federal Practice and Procedure § 1556, p. 711 (1971) (well settled "citizenship rule testing diversity in terms of the real party in interest is grounded in notions of federalism"). See generally Note, Diversity Jurisdiction over Unincorporated Business Entities: The Real Party in Interest as a Jurisdictional Rule, 56 Texas L. Rev. 243, 247-250 (1978). In Wormley v. Wormley, 8 Wheat. 421 (1823), for example, the Court stated:
"This Court will not suffer its jurisdiction to be ousted by the mere joinder or non-joinder of formal parties; but will rather proceed without them, and decide upon the merits of the case between the parties, who have the real interests before it, whenever it can be done without prejudice to the rights of others." Id., at 451 (footnote omitted).
See also Wood v. Davis, 18 How. 467, 469 (1856) ("It has been repeatedly decided by this Court, that formal parties, or nominal parties, or parties without interest, united with the real parties to the litigation, cannot oust the federal courts of jurisdiction . . ."). The real party to the controversy approach has been implemented by the Court both in its oldest and in its most recent cases examining diversity jurisdiction with respect to business associations. In the Court's first examination of the corporate form to determine who must be counted for purposes of diversity jurisdiction, the Court invoked the real party to the controversy test and concluded that the citizenship of each shareholder must be counted for purposes of diversity jurisdiction. Bank of United States v. Deveaux, 5 Cranch, at 91-92. In Deveaux, the Court recognized that corporations had been considered as possessing "corporeal qualities," id., at 89, but concluded that the actual parties to the controversy were "the members of the corporation . . . who come into court, in this case, under their corporate name." Id., at 91. By 1854, the Court no longer characterized the corporation as merely possessing "corporeal qualities," but rather as a "juridical person," which made an even stronger case for recognizing a corporation as a proper party in its own right before the Court. See Marshall v. Baltimore & Ohio R. Co., 16 How., at 328; see also Louisville, C. & C. R. Co. v. Letson, 2 How. 497, 558-559 (1844) (corporation is a person; shareholders' citizenship will not be counted).
In Mashall, as in Deveaux, however, the determination whether the corporation was a citizen did not signal the end of the diversity jurisdiction inquiry. 16 How., at 328. Rather, the Court engaged in a two-part inquiry: (1) is the corporation a "juridical person" which can serve as a real party to the controversy, see id., at 327-329; and (2) are the shareholders real parties to the controversy. See id., at 328. To determine whether the corporation or the shareholders were real parties to the controversy, the Court considered which citizens held control over the business decisions and assets of the corporation and over the initiation and course of litigation involving the corporation. The corporation, as the representative body of the shareholders, itself had such power. The shareholders did not.
"For all the purposes of acting, contracting, and judicial remedy, [shareholders] can speak, act, and plead, only through their representatives or curators. For the purposes of a suit or controversy, the persons represented by a corporate name can appear only by attorney, appointed by its constitutional organs. . . . They are not really parties to the suit or controversy." Ibid.
Having concluded that the shareholders were not the real parties to the controversy, the Court held that only the State of incorporation of the corporate entity need be counted for purposes of diversity jurisdiction and that the citizenship of the shareholders would be presumed to be that of the State of incorporation. Id., at 328-329. As the Court makes plain in Marshall, consideration of whether the shareholders were real parties to the controversy was a necessary prerequisite to the creation of the legal fiction that their citizenship would be deemed that of the corporation.
In a series of three cases considering the citizenship of business associations following Marshall, the Court was not called upon to determine which of the citizens before it were the real parties to the controversy because the business associations were not citizens themselves and the members of each association held equivalent power and control over the association's assets, business, and litigation. In Chapman v. Barney, 129 U.S. 677 (1889), the Court addressed the issue whether a joint stock company was a citizen for purposes of diversity jurisdiction. A joint stock company, now a historical anomaly, see A. Bromberg, Crane and Bromberg on Partnership 178, and n. 16 (1968), had several features of the corporate form, e. g., centralized management and transferability of shares, but was more like a general partnership in that each partner was personally liable and there was only one class of partners. See Comment, Limited Partnerships and Federal Diversity Jurisdiction, 45 U. Chi. L. Rev. 384, 389, and n. 32 (1978). Each "partner" had equal power over the conduct of the business by virtue of his power to elect and control the company's managers. Bromberg, supra, at 179, n. 19. The Court held that a joint stock company was a "mere partnership" and therefore not sufficiently similar to a corporation to justify designating it as a citizen. 129 U.S., at 682. Hence, the citizenship of each owner had to be counted for purposes of diversity jurisdiction. Because the joint stock company owners were similarly situated in terms of power and control over the company and possessed all of the power that could be exercised over the company's business and litigation, and the company itself was not a citizen, the Court was not called upon to determine which of the citizens before it were the real parties to the controversy.
The Court applied a similar approach in Great Southern Fire Proof Hotel Co. v. Jones, 177 U.S. 449 (1900), when it examined a limited partnership association. Quite unlike the modern limited partnership, the limited partnership association at issue in Great Southern, recognized by very few States, Comment, 45 U. Chi. L. Rev., supra, at 389, n. 36, was a species of business association involving a single class of partners with limited liability who exercised control over the operation of the business by annually electing the managers of the association. See, e. g., 1874 Pa. Laws, Act. No. 153, §§ 2, 5; Comment, 45 U. Chi. L. Rev., supra, at 389, n. 36. Not surprisingly, the Court viewed such an organization as more like a partnership than a corporation. See F. Burdick, Law of Partnership 361-362 (1899) (limited partnership association like corporation in some respects, but generally treated by the courts as a general partnership). As in the case of the joint stock company, because all partners were similarly situated in terms of power and control over the company, there was no reason for the Court to inquire who, among the partners, were the real parties to the controversy. In Steelworkers v. R. H. Bouligny, Inc., 382 U.S. 145 (1965), the Court addressed whether a labor union could be treated as an entity for purposes of diversity jurisdiction. The Court held that a labor union is not a juridical person and, therefore, not a citizen for purposes of diversity jurisdiction. See Mesa Operating Limited Partnership v. Louisiana Intrastate Gas Corp., 797 F.2d 238, 240-241 (CA5 1986) (union in Bouligny failed to meet party to controversy test). There was no indication that any of the union members had any greater power over the litigation or the union's business and assets than any other member, and, therefore, as in Chapman and Great Southern, the Court was not called upon to decide which of the citizens before it were real parties to the controversy.
In the next case, in which application of the real party to the controversy test was appropriate, the Court unanimously applied it. See Navarro Savings Assn. v. Lee, 446 U.S., at 460, 464-465; id., at 469, 475 (BLACKMUN, J., dissenting). In that case, the Court addressed the question whether the beneficiaries' citizenship must be counted when the trustees brought suit involving the assets of the trust. See id., at 458. Because the trust beneficiaries lacked both control over the conduct of the business and the ability to initiate or control the course of litigation, the Court held that the citizenship of the trust beneficiaries should not be counted. Id., at 464-465.
As Navarro makes clear, the nature of the named party does not settle the question of who are the real parties to the controversy. In fact, if the Court's characterization of the issue before us were correct, ante, at 187-188, n. 1, then we seriously erred in Navarro Savings Assn. v. Lee, supra, at 464-466, when we considered whether the trust beneficiaries were the real parties to the controversy, in light of the fact that they were not named parties to the litigation.
The Court attempts to distinguish Navarro on the ground that it involved not a juridical person, but rather the "distinctive common-law institution of trustees." Ante , at 194. Such a view is consonant with the Court's new diversity jurisdiction analysis announced in this case, but fails to take into account the actual language and analysis in Navarro. If the nature of the institution of trustees was sufficient to answer the question of which parties to count for diversity jurisdiction purposes in that case, the Court's discussion of whether the trust beneficiaries were real parties to the controversy would have been wholly superfluous. Given that the Court granted certiorari in that case on the very issue whether the citizenship of trust beneficiaries must be counted, and then unanimously applied the real parties to the controversy test, the discussion clearly was not superfluous.
Application of the parties to the controversy test to the limited partnership yields the conclusion that limited partners should not be considered for purposes of diversity jurisdiction. Like the trust beneficiary in Navarro, the limited partner "can neither control the disposition of this action nor intervene in the affairs of the trust except in the most extraordinary situations." Navarro, supra, at 464-465. See Uniform Limited Partnership Act § 26, 6 U. L. A. 614 (1969) (limited partner "is not a proper party to proceedings by or against a partnership, except where the object is to enforce a limited partner's right against or liability to the partnership"); Uniform Limited Partnership Act § 1001, 6 U. L. A. 371 (Supp. 1989) (derivative actions); Ariz. Rev. Stat. Ann. § 29-324 (1989) (general partners of limited partnership have duties and obligations of partners to general partnership); § 29-209 (general partner is agent of partnership); § 29-356 (limited partners limited to derivative actions); Arkoma Associates Partnership Agreement, Art. VI, § 6.1 (general partners have "exclusive and complete control of the operations"); id., § 7.1 (limited partners "shall not take any part in the control or management of . . . Partnership"). And like the shareholder in Marshall, "for all the purposes of acting, contracting, and judicial remedy, [limited partners] can speak, act, and plead, only through [others]." Marshall, 16 How., at 328. In fact, the limited partner has even less power in the limited partnership than the shareholder does in a corporation. "The shareholder . . . retain[s] some measure of control over management through his voting power, while the more restricted role of the limited partner permits restraint [of management] only by his refusal to concur in certain acts for which his consent is required by law." See Note, Standing of Limited Partners to Sue Derivatively, 65 Colum. L. Rev. 1463, 1478 (1965). Without the power to "control . . . the assets" or to initiate or "control the litigation," Navarro, supra, at 465, the limited partner is not a real party to the controversy and, therefore, should not be counted for purposes of diversity jurisdiction. Because the majority of States have adopted the Uniform Limited Partnership Act, this rule would result in uniform treatment of limited partners for purposes of diversity jurisdiction. See Uniform Limited Partnership Act, 6 U. L. A. 172, 220 (Supp. 1989).
The commentators are in agreement that the party to the controversy test is the appropriate test to be applied to determine diversity jurisdiction with respect to limited partnerships and that the citizenship of limited partners should not be counted. See, e. g., Comment, 45 U. Chi. L. Rev., at 418 (citizenship of limited partners should not be counted for purposes of diversity jurisdiction); Note, Who Are the Real Parties In Interest for Purposes of Determining Diversity Jurisdiction for Limited Partnerships?, 61 Wash. U. L. Q. 1051, 1066-1067 (1984) (same); Note, 56 Texas L. Rev., at 250-251 (real party in interest test should be applied to unincorporated business associations to determine whom to count for diversity); see also Colonial Realty Corp. v. Bache & Co., 358 F.2d 178, 183 (1966) (Friendly, J.) (citizenship of limited partner should not be counted where state law declares partner is not "proper party to proceedings by or against a partnership"). The concern perhaps implicit in the Court's holding today is that failure to consider the citizenship of all the members of an unincorporated business association will expand diversity jurisdiction at a time when our federal courts are already seriously overburdened. This concern is more illusory than real in the context of unincorporated business associations. For, despite the Court's holding today, unincorporated associations may gain access to the federal courts by bringing or defending suit as a Rule 23 class action, in which case the citizenship of the members of the class would not be considered. See Federal Diversity Jurisdiction -- Citizenship for Unincorporated Associations, 19 Vand. L. Rev. 984, 991-992 (1966). Thus, I see little reason to depart in this case from our long settled practice of applying the real parties to the controversy test.
Because there is complete diversity between petitioners and the limited partnership (assuming that it should be considered a citizen) and each of the general partners, the issue presented by this case is fully resolved by application of the parties to the controversy test.
Even though the case does not directly relate to the issue before us, the Court takes pains to address and distinguish Puerto Rico v. Russell & Co., 288 U.S. 476 (1933). See ante, at 189-190. The issue in Russell was not diversity, but whether the suit against the sociedad en comandita could be removed from the Insular Court to the United States District Court for Puerto Rico on the ground that no party on one side was a citizen of or domiciled in Puerto Rico. See 288 U.S., at 478. None of the partners were citizens of Puerto Rico, but the Court determined that the sociedad was and, therefore, removal was precluded. Thus, Russell tells us nothing about whether the citizenship of the sociedad 's members, unlimited or limited, should be considered for purposes of diversity jurisdiction. In any event, the Court's attempts to distinguish Russell are seriously flawed. In Russell, the Court examined the Puerto Rican sociedad en comandita, which is the civil law version of the modern limited partnership. The Court delineated a series of factors and concluded that, under civil law, the sociedad was a "juridical person." Id., at 481. Ironically, the Court in this case endorses the holding of Russell, despite the fact that virtually all of the factors listed are equally applicable to the modern limited partnership. The Court fails to acknowledge that our modern limited partnership, like the sociedad, finds its origins in the civil law. The limited partnership originated in Europe in the middle ages, first appearing in France "under the name of la Societe en comandite, . . . mention being made of it in the most ancient commercial records, and in the early mercantile regulations of Marseilles and Montpelier." Ames v. Downing, 1 Bradf. Surr. 321, 329 (N. Y. 1850). The limited partnership did not find acceptance in the United Kingdom and was not a creature of the common law. F. Burdick, Law of Partnership 384-385 (2d ed. 1906). It was first introduced into this country in Louisiana and then New York. See Note, 65 Colum. L. Rev., at 1464. Although a "'creation of the civil law,'" the Puerto Rican sociedad was hardly "'exotic.'" Ante, at 190 (quoting Bouligny, 382 U.S., at 151). Rather, it is yet one of many forms of the limited partnership descended from the ancient French Societe as is the modern limited partnership adopted in this country. See Ames, supra, at 329-330 (American limited "partnership is, in fact, no novelty, but an institution of considerable antiquity, well known, understood and regulated"). It is hardly an answer to the history of the limited partnership in this country and abroad to assert that it appears 25 years after Steelworkers v. R. H. Bouligny, Inc., 382 U.S. 145 (1965). See ante, at 190-191, n. 2. The "admirable consistency of our jurisprudence," ante, at 189, is not blemished by distinguishing between unions and limited partnerships. It is, however, severely marred by holding that an association within the continental United States is not afforded the same treatment as its virtually identical Puerto Rican counterpart. See also ante, at 191, n. 2 ("operative word in both the diversity statute and the removal statute at issue in Russell is 'citizens'"). The Court's decision today, endorsing treatment of a Puerto Rican business association as an entity while refusing to treat as an entity its virtually identical stateside counterpart, is justified neither by our precedents nor by historical and commercial realities.
For the foregoing reasons, I respectfully dissent.
1. The dissent reaches a conclusion different from ours primarily because it poses, and then answers, an entirely different question. It "do[es] not consider" "whether the limited partnership is a 'citizen,'" but simply "assum[es] it is a citizen," because even if we hold that it is, "we are still required to consider which, if any, of the other citizens before the Court as members of Arkoma Associates are real parties to the controversy." Post, at 198 (emphasis added). Furthermore, "the only potentially non-diverse party in this case is a limited partner" because "all other parties, including the general partners and the limited partnership itself, assuming it is a citizen, are diverse." Ibid. (emphasis added). That is the central fallacy from which, for the most part, the rest of the dissent's reasoning logically follows. The question presented today is not which of various parties before the Court should be considered for purposes of determining whether there is complete diversity of citizenship, a question that will generally be answered by application of the "real party to the controversy" test. There are not, as the dissent assumes, multiple respondents before the Court, but only one: the artificial entity called Arkoma Associates, a limited partnership. And what we must decide is the quite different question of how the citizenship of that single artificial entity is to be determined -- which in turn raises the question whether it can (like a corporation) assert its own citizenship, or rather is deemed to possess the citizenship of its members, and, if so, which members. The dissent fails to cite a single case in which the citizenship of an artificial entity, the issue before us today, has been decided by application of the "real party to the controversy" test that it describes. See infra, at 192-195.
2. The dissent correctly observes that " Russell tells us nothing about whether the citizenship of the sociedad 's members, unlimited or limited, should be considered for purposes of diversity jurisdiction." Post, at 207. Rather, as is evident from our discussing the case here instead of in Part B below, Russell (according to respondent) tells us something about whether an artificial entity other than a corporation can be considered a "citizen" in its own right. That "the issue in Russell was not diversity, but whether the suit against the sociedad en comandita could be removed from the Insular Court to the United States District Court for Puerto Rico," post, at 207, does not affect Russell 's arguable relevance to that question because the operative word in both the diversity statute and the removal statute at issue in Russell is "citizens." The dissent goes on to criticize as "seriously flawed," post, at 208, our attempt to distinguish Russell in connection with the issue we do address, whether a partnership can be considered a "citizen." We point out, not by way of complaint but to prevent confusion, that the criticism is gratuitous, inasmuch as the dissent itself takes no position on this issue, announcing at the very outset that it "do[es] not consider" the question "whether the limited partnership is a 'citizen.'" Post, at 198. In any event, the dissent's evidence bearing on the historical pedigree of partnerships comes to our attention at least 25 years too late. For the reasons stated in the text, Bouligny considered and rejected applying Russell beyond its facts.
3. Marshall 's fictional approach appears to have been abandoned. Later cases revert to the formulation of Louisville, C. & C. R. Co. v. Letson, 2 How. 497 (1844), that the corporation has its own citizenship. See Great Southern Fire Proof Hotel Co. v. Jones, 177 U.S. 449, 456 (1900) ("For purposes of jurisdiction . . . a corporation was to be deemed a citizen of the State creating it") (citing Letson); Chapman v. Barney, 129 U.S. 677, 682 (1889) ("express company cannot be a citizen of New York, within the meaning of the statutes regulating jurisdiction, unless it be a corporation").