delivered the opinion of the Court.

Petitioner Francisco Romero, a Spanish subject, signed on as a member of the crew of the S. S. Guadalupe for a voyage beginning about October 10, 1953. The Guadalupe was of Spanish registry, sailed under the Spanish *356flag and was owned by respondent Compañía Trasatlán-tica (also known as Spanish Line), a Spanish corporation. At the completion of the voyage for which he signed, Romero continued uninterruptedly to work on the Guadalupe. “Thereby, under the law of Spain, the terms and conditions of the original contract of hire remained in force. Subsequently the S. S. Guadalupe departed from the port of Bilbao in northern Spain, touched briefly at other Spanish ports, and sailed to the port of New York at Hoboken. From here the ship made a brief trip to the ports of Vera Cruz and Havana returning to Hoboken where, on May 12, 1954, Romero was seriously injured when struck by a cable on the deck of the Guadalupe. Thereupon petitioner filed suit on the law side of the District Court for the Southern District of New York.

The amended complaint claimed damages from four separate corporate defendants. Liability of Compañía Trasatlántica and Garcia & Diaz, Inc., a New York corporation which acted as the husbanding agent for Com-pania’s vessels while in the port of New York, was asserted under the Jones Act, 41 Stat. 1007, 46 U. S. C. § 688, and under the general maritime law of the United States for unseaworthiness of the ship, maintenance and cure1 and a maritime tort. Liability for a maritime tort was alleged against respondents International Terminal Operating Co. and Quin Lumber Co. These two companies were working on board the S. S. Guadalupe at the time of the injury pursuant to oral contracts with Garcia & Diaz, Inc. Quin, a New York corporation, was engaged in carpentry work preparatory to the receipt of a cargo *357of grain. International Terminal, incorporated in Delaware, was employed as stevedore to load the cargo. The jurisdiction of the District Court was invoked under the Jones Act and §§ 13312 and 13323 of the Judicial Code, 28 U. S. C.

Following a pre-trial hearing the District Court dismissed the complaint. 142 F. Supp. 570.4 The court *358held that the action under the Jones Act against Com-pania Trasatlántica must be dismissed for lack of jurisdiction since that Act provided no right of action for an alien seaman against a foreign shipowner under the circumstances detailed above. The claims under the general maritime law against Compania also were dismissed since the parties were not of diverse citizenship and 28 U. S. C. § 1331 did not confer jurisdiction on the federal law courts over claims rooted in federal maritime law. The District Court dismissed the Jones Act claim against Garcia & Diaz, Inc., pursuant to its finding that Garcia was not the employer of Romero nor, as a husbanding agent for Compania, did it have the operation and control of the vessel. The remaining claims, including those against the other respondents, were dismissed because of lack of the requisite complete diversity under the rule of Strawbridge v. Curtiss, 3 Cranch 267. Upon examination of the Spanish law the district judge also declined jurisdiction “even in admiralty as a matter of discretion.” 142 F. Supp., at 574. The Spanish law provides Romero with a lifetime pension of 35% to 55% of his seaman’s wages which may be increased by one-half if the negligence of the shipowner is established; it also allows the recovery of the Spanish counterpart of maintenance and cure. These rights under the Spanish law may be enforced through the Spanish consul in New York.

The Court of Appeals affirmed the dismissal of the complaint, 244 F. 2d 409. We granted certiorari, 355 U. S. 807, because of the conflict among Courts of Appeals as to the proper construction of the relevant provision of the Judiciary Act of 1875 (now 28 U. S. C. § 1331) and because of questions raised regarding the applicability of Lauritsen v. Larsen, 345 U. S. 571, to the situation before us. The case was argued during the last Term and restored to the calendar for reargument during the present Term. 356 U. S. 955.

*359I. Jurisdiction.

(a) Jurisdiction under the Jones Act. — The District Court dismissed petitioner’s Jones Act claims for lack of jurisdiction. “As frequently happens where jurisdiction depends on subject matter, the question whether jurisdiction exists has been confused with the question whether the complaint states a cause of action.” Montana-Dakota Utilities Co. v. Northwestern Public Service Co., 341 U. S. 246, 249. Petitioner asserts a substantial claim that the Jones Act affords him a right of recovery for the negligence of his employer. Such assertion alone is sufficient to empower the District Court to assume jurisdiction over the case and determine whether, in fact, the Act does provide the claimed rights. “A cause of action under our law was asserted here, and the court had power to determine whether it was or was not well founded in law and in fact.” Lauritzen v. Larsen, 345 U. S. 571, 575.

(b) Jurisdiction under 28 U. S. C. § 1881. — Petitioner, a Spanish subject, asserts claims under the general maritime law against Compania Trasatlántica, a Spanish corporation. The jurisdiction of the Federal District Court, sitting as a court of law, was invoked under the provisions ' of the Judiciary Act of 1875 which granted jurisdiction to the lower federal courts “of all suits of a civil nature at common law or in equity, . . . arising under the Constitution or laws of the United States, . . . .” (now 28 U. S. C. § 1331).5 Whether the Act of 1875 permits maritime claims rooted in federal law to be brought on *360the law side of the lower federal courts has recently been raised in litigation and has become the subject of conflicting decisions among Courts of Appeals. Jurisdiction has been sustained in the First Circuit, Doucette v. Vincent, 194 F. 2d 834, and denied in the Second and Third, Jordine v. Walling, 185 F. 2d 662; Paduano v. Yamashita Kisen Kabushiki Kaisha, 221 F. 2d 615. See also Jenkins v. Roderick, 156 F. Supp. 299. Such conflict in the construction of an old and important statute calls for a full exposition of the problem.

Abstractly stated, the problem is the ordinary task of a court to apply the words of a statute according to their proper construction. But “proper construction” is not satisfied by taking the words as if they were self-contained phrases. So considered, the words do not yield the meaning of the statute. The words we have to construe are not only words with a history. They express an enactment that is part of a serial, and a serial that must be related to Article III of the Constitution, the watershed of all judiciary legislation, and to the enactments which have derived from that Article. Moreover, Article III itself has its sources in history. These give content and meaning to its pithy phrases. Rationally construed, the Act of 1875 must be considered part of an organic growth — part of the evolutionary process of judiciary legislation that began September 24, 1789,-and projects into the future.

Article III, § 2, cl. 1 (3d provision) of the Constitution and section 9 of the Act of September 24, 1789, have from the beginning been the sources of jurisdiction in litigation based upon federal maritime law. Article III impliedly contained three grants. (1) It empowered Congress to confer admiralty and maritime jurisdiction on the “Tribunals inferior to the supreme Court” which were authorized by Art. I, § 8, cl. 9. (2) It empowered the federal courts in their exercise of the admiralty *361and maritime jurisdiction which had been conferred on them, to draw on the substantive law “inherent in the admiralty and maritime jurisdiction,” Crowell v. Benson, 285 U. S. 22, 55, and to continue the development of this law within constitutional limits. (3) It empowered Congress to revise and supplement the maritime law within the limits of the Constitution. See Crowell v. Benson, supra, at 55.

Section 9 of the First Judiciary Act6 granted the District Courts maritime jurisdiction. This jurisdiction has remained unchanged in substance to the present day.7 Indeed it was recognition of the need for federal tribunals to exercise admiralty jurisdiction that was one of the controlling considerations for the establishment of a system of lower federal courts.8 Such a system is not an inherent requirement of a federal government. There was strong opposition in the Constitutional Convention to any such inferior federal tribunals.9 No comprehensive system of lower federal courts has *362been established in Canada or Australia. Congress could leave the enforcement of federal rights to state courts,10 and indeed the state courts, in large measure, now exercise concurrent jurisdiction over a wide field of matters of federal concern, subject to review of federal issues by the Supreme Court.11

Section 9 not only established federal courts for the administration of maritime law; it recognized that some remedies in matters maritime had been traditionally administered by common-law courts of the original States.12 This role of the States in the administration of maritime law was preserved in the famous “saving clause” — “saving to suitors, in all cases, the right of a common-law remedy, where the common law is competent to give it.” 13 Since the original Judiciary Act also endowed the federal courts with diversity jurisdiction, common-law remedies for maritime causes could be enforced by the then Circuit Courts when the proper diversity of parties afforded access.

Up to the passage of the Judiciary Act of 187514 these jurisdictional bases provided the only claim for jurisdic*363tion in the federal courts in maritime matters.15 The District Courts, endowed with “exclusive original cognizance of all civil causes of admiralty and maritime jurisdiction,” sat to enforce the comprehensive federal interest in the law of the sea which had been a major reason for their creation. This jurisdiction was exercised according to the historic procedure in admiralty, by a judge without a jury. In addition, common-law remedies were, under the saving clause, enforcible in the courts of the States and on the common-law side of the lower federal courts when the diverse citizenship of the parties permitted. Except in diversity cases, maritime litigation brought in state courts could not be removed to the federal courts.16

The Judiciary Act of 1875 effected an extensive enlargement of the jurisdiction of the lower federal courts. For the first time their doors were opened to “all suits of a civil nature at common law or in equity, . . . arising under the Constitution or laws of the United States, or treaties made, or which shall be made, under their authority .. . .” 17 From 1875 to 1950 there is not to be found a hint or suggestion to cast doubt on the conviction that the language of that statute was taken straight from Art. Ill, § 2, cl. 1, extending the judicial power of the United States “to all Cases, in Law and Equity, arising under this Constitution, the Laws of the United States, and Treaties made, or which shall be made, under their Authority.” Indeed what little legislative history there is *364affirmatively indicates that this was the source.18 Thus the Act of 1875 drew on the scope of this provision of Clause 1, just as the Judiciary Act of 1789 reflected the constitutional authorization of Clause 1 of Section 2, which extended the judicial power “to all Cases of admiralty and maritime Jurisdiction.”

These provisions of Article III are two of the nine separately enumerated classes of cases to which “judicial power” was extended by the Constitution and which thereby authorized grants by Congress of “judicial Power” to the “inferior” federal courts. The vast stream of litigation which has flowed through these courts from the beginning has done so on the assumption that, in dealing with a subject as technical as the jurisdiction of the courts, the Framers, predominantly lawyers, used precise, differentiating and not redundant language. This assumption, reflected in The Federalist Papers,19 was authoritatively confirmed by Mr. Chief Justice Marshall in American Ins. Co. v. Canter, 1 Pet. 511, 544:

“We are therefore to inquire, whether cases in admiralty, and cases arising under the laws and Constitution of the United States, are identical.

“If we have recourse to that pure fountain from which all the jurisdiction of the Federal Courts is derived, we find language employed which cannot well be misunderstood. The Constitution declares, that The judicial power shall extend to all cases in law and equity, arising under this Constitution, the laws of the United States, and treaties made, or which shall be made, under their authority; to all cases affecting ambassadors, or other public ministers, and *365consuls; to all cases of admiralty and maritime jurisdiction.’

“The Constitution certainly contemplates these as three distinct classes of cases; and if they are distinct, the grant of jurisdiction over one of them does not confer jurisdiction over either of the other two. The discrimination made between them, in the Constitution, is, we think, conclusive against their identity.” See also The Sarah, 8 Wheat. 391.

This lucid principle of constitutional construction, embodied in one of Marshall’s frequently quoted opinions, was never brought into question until 1952.20 It *366had been treated as black-letter law in leading treatises.21 It was part of the realm of legal ideas in which the authors of the Act of 1875 moved. Certainly the accomplished lawyers who drafted the Act of 187522 drew on *367the language of the constitutional grant on the assumption that they were dealing with a distinct class of cases, that the language incorporated in their enactment precluded “identity” with any other class of cases contained in Article III. Thus the grant of jurisdiction over “suits of a civil nature at common law or in equity . . . arising under the Constitution or laws of the United States . . . in the Act of 1875, as derived from Article III, could not reasonably be thought of as comprehending an entirely separate and distinct class of cases — “Cases of admiralty and maritime Jurisdiction.”23

*368Of course all cases to which “judicial power” extends “arise,” in a comprehensive, non-jurisdictional sense of the term, “under this Constitution.” It is the Constitution that is the ultimate source of all “judicial Power” — defines grants and implies limits — and so “all Cases of admiralty and maritime Jurisdiction” arise under the Constitution in the sense that they have constitutional sanction. But they are not “Cases, in Law and Equity, arising under this Constitution, the Laws of the United States . . . .”

Not only does language and construction point to the rejection of any infusion of general maritime jurisdiction into the Act of 1875, but history and reason powerfully support that rejection. The far-reaching extension of national power resulting from the victory of the North, and the concomitant utilization of federal courts for the vindication of that power in the Reconstruction Era, naturally led to enlarged jurisdiction of the federal courts over federal rights. But neither the aim of the Act of 1875 to provide a forum for the vindication of new federally created rights, nor the pressures which led to its enactment, suggest, even remotely, the inclusion of maritime claims within the scope of that statute. The provision of the Act of 1875 with which we are concerned was designed to give a new content of jurisdiction to the federal courts, not to reaffirm one long-established, smoothly functioning since 1789.24 We have uncovered no basis for finding the additional design of changing the method by which federal courts had administered admiralty law from the *369beginning. The federal admiralty courts had been completely adequate to the task of protecting maritime rights rooted in federal law. There is not the slightest indication of any intention, or of any professional or lay demands for a change in the time-sanctioned mode of trying suits in admiralty without a jury, from which it can be inferred that by the new grant of jurisdiction of cases “arising under the Constitution or laws” a drastic innovation was impliedly introduced in admiralty procedure, whereby Congress changed the method by which federal courts had administered admiralty law for almost a century. To draw such an inference is to find that a revolutionary procedural change had undesignedly come to pass. If we are now to attribute such a result to Congress the sole remaining justification for the federal admiralty courts which have played such a vital role in our federal judicial system for 169 years will be to provide a federal forum for the small number of maritime claims which derive from state law, and to afford the ancient remedy of a libel in rem in those limited instances when an in personam judgment would not suffice to satisfy a claim.25

Indeed, until 1950, in a dictum in Jansson v. Swedish American Line, 185 F. 2d 212, 217-218 (C. A. 1st Cir.), followed by an opinion in Doucette v. Vincent, 194 F. 2d 834, judges, scholars and lawyers alike made the unquestioned assumption that the original maritime jurisdiction of the federal courts had, for all practical purposes, been left unchanged since the Act of 1789. Thus Mr. Justice Clifford, an experienced admiralty judge, in 1876, one year after the passage of the Act here in question, could reiterate the classic formulation without the » faintest indication of doubt as to its continued vitality. *370“Parties in maritime cases are not . . . compelled to proceed in the admiralty at all, as they may resort to their common-law remedy in the State courts, or in the Circuit Court, if the party seeking redress and the other party are citizens of different States.” 26 On the basis of an examination of sixty-six treatises on federal jurisdiction and on admiralty, and of a search of the reports it can be confidently asserted that for the seventy-four years following Mr. Justice Clifford’s opinion there is not a single professional utterance of legal opinion — by judges, lawyers, or commentators — disagreeing with his formulation.27 Negative testimony is often as compelling as bits of affirmative evidence. It is especially compelling when it comes from those whose scholarly or professional specialty was the jurisdiction of the federal courts and the practice of maritime law. Petitioner now asks us to hold that no student of the jurisdiction of the federal courts or of admiralty, no judge, and none of the learned and alert members of the admiralty bar were able, for seventy-five years, to discern the drastic change now asserted to have been contrived in admiralty jurisdiction by the Act of 1875. In light of such impressive testimony from the past the claim of a sudden discovery of a hidden latent meaning in an old technical phrase is surely suspect.

The history of archeology is replete with the unearthing of riches buried for centuries. Our legal history does not, however, offer a single archeological discovery of new, revolutionary meaning in reading an old judiciary enactment.27a The presumption is powerful that such a far-reaching, dislocating construction as petitioner would now have us find in the Act of 1875 was not uncovered by *371judges, lawyers or scholars for seventy-five years because it is not there.

It is also significant that in the entire history of federal maritime legislation, whether before the passage of the Act of 1875 (e. g., the Great Lakes Act — also a general jurisdictional statute and one often termed an anomaly in the maritime law because of its jury trial provision), or after (the Jones Act), Congress has not once left the availability of a trial on the law side to inference. It has made specific provision.28 It is difficult to accept that in 1875, and in 1875 alone, a most far-reaching change was made subterraneously.

Not only would the infusion of general maritime jurisdiction into the Act of 1875 disregard the obvious construction of that statute. Important difficulties of judicial policy would flow from such an interpretation, an interpretation which would have a disruptive effect on the traditional allocation of power over maritime affairs in our federal system.

Thus the historic option of a maritime suitor pursuing a common-law remedy to select his forum, state or federal, would be taken away by an expanded view of § 1331,29 since saving-clause actions would then be freely *372removable under § 1441 of Title 28.30 The interpretation of the Act of 1875 contended for would have consequences more deeply felt than the elimination of a suitor’s traditional choice of forum. By making maritime cases removable to the federal courts it would make considerable inroads into the traditionally exercised concurrent jurisdiction of the state courts in admiralty matters — a jurisdiction which it was the unquestioned aim of the saving clause of 1789 to preserve. This disruption of principle is emphasized by the few cases actually involved.31 This small number of cases is only important in that it negatives the pressure of any practical consideration for the subversion of a principle so long-established and so deeply rooted. The role of the States in the development of maritime law is a role whose significance is rooted in the Judiciary Act of 1789 and the decisions of this Court.32 Recognition of the part the States have played from the beginning has a dual significance. It indicates the extent to which an expanded view of the Act of 1875 would eviscerate the postulates of the saving clause, and it undermines the theoretical basis for giving the Act of 1875 a brand new meaning.

*373Although the corpus of admiralty law is federal in the sense that it derives from the implications of Article III evolved by the courts, to claim that all enforced rights pertaining to matters maritime are rooted in federal law is a destructive oversimplification of the highly intricate interplay of the States and the National Government in their regulation of maritime commerce. It is true that state law must yield to the needs of a uniform federal maritime law when this Court finds inroads on a harmonious system.33 But this limitation still leaves the States a wide scope. State-created liens are enforced in admiralty.34 State remedies for wrongful death and state statutes providing for the survival of actions, both historically absent from the relief offered by the admiralty,35 have been upheld when applied to maritime causes of action.36 Federal courts have enforced these statutes.37 State rules for the partition and sale of ships,38 state laws governing the specific performance of arbitration agreements,39 state laws regulating the effect of a breach of warranty under contracts of maritime insurance40 — all these laws and others have been accepted as rules of *374decision in admiralty cases, even, at times, when they conflicted with a rule of maritime law which did not require uniformity. “In the field of maritime contracts,” this Court has said, “as in that of maritime torts, the National Government has left much regulatory power in the States.”41 Thus, if one thing is clear it is that the source of law in saving-clause actions cannot be described in absolute terms. Maritime law is not a monistic system. The State and Federal Governments jointly exert regulatory powers today as they have played joint roles in the development of maritime law throughout our history.42 This sharing of competence in one aspect of our federalism has been traditionally embodied in the saving clause of the Act of 1789. Here, as is so often true in our *375federal system, allocations of jurisdiction have been carefully wrought to correspond to the realities of power and interest and national policy. To give a novel sweep to the Act would disrupt traditional maritime policies and quite gratuitously disturb a complementary, historic interacting federal-state relationship.

An infusion of general maritime jurisdiction into the “federal question” grant would not occasion merely an isolated change; it would generate many new complicated problems. If jurisdiction of maritime claims were allowed to be invoked under § 1331, it would become necessary for courts to decide whether the action “arises under federal law,” and this jurisdictional decision would largely depend on whether the governing law is state or federal. Determinations of this nature are among the most difficult and subtle that federal courts are called upon to make.43 Last Term’s decision in McAllister v. Magnolia Petroleum Co., 357 U. S. 221, illustrates the difficulties raised by the attempted application of a state statute of limitations to maritime personal injury actions. These problems result from the effort to fit state laws into the scheme of federal maritime law.

These difficulties, while nourishing academic speculation, have rarely confronted the courts. This Court has been able to wait until an actual conflict between state and federal standards has arisen, and only then proceed to resolve the problem of whether the State was free to *376regulate or federal law must govern. For example, if a State allowed the survival of a cause of action based on unseaworthiness as defined in the maritime law it was immaterial whether the standard was federal and governed by decisions of this Court, or was subject to state variations.44 Thus we have been able to deal with such conceptual problems in the context of a specific conflict and a specific application of policy, as is so well illustrated by the McAllister case. However, such practical considerations for adjudication would be unavailable under an expanded view of § 1331. Federal courts would be forced to determine the respective spheres of state and federal legislative competence, the source of the governing law, as a preliminary question of jurisdiction; for only if the applicable law is “federal” law would jurisdiction be proper under § 1331. The necessity for jurisdictional determinations couched in terms of “state” or “federal law” would destroy that salutary flexibility which enables the courts to deal with source-of-law problems in fight of the necessities illuminated by the particular question to be answered. Certainly sound judicial policy does not encourage a situation which necessitates constant adjudication of the boundaries of state and federal competence.

Typical also of the consequences that are implicit in this proposed modification of maritime jurisdiction, is the restriction of venue that would result from this novel interpretation of § 1331 of the Act of 1875. Liti*377gants of diverse citizenship are now able to invoke the federal law forum for the trial of saving-clause cases. Such litigants are aided in their search for a federal forum by the liberality of the venue provisions applicable to actions based on diversity of citizenship. These provisions allow the action to be brought either “where all plaintiffs or all defendants reside.” 45 If saving-clause actions were to be brought within the scope of § 1331, this choice could be no longer made. Plaintiffs would be subject to the rigid requirement that suit must be “brought only in the judicial district where all defendants reside . . . ,” 46 and this would be so even where there is, in fact, diversity of citizenship.47

In the face of the consistent and compelling inferences to be drawn from history and policy against a break with a long past in the application of the Act of 1875, what justification is offered for this novel view of the statute? Support is ultimately reduced, one is compelled to say, to empty logic, reflecting a formal syllogism. The argument may thus be fairly summarized. It was not until recently, in a line of decision culminating in Pope & Talbot, Inc., v. Hawn, 346 U. S. 406, that it became apparent that the source of admiralty rights was a controlling body of federal admiralty law. This development led to a deepened consideration of the jurisdictional consequences of the federal source of maritime law. And so one turns to the Act of 1875. The Act of 1875 gave orig*378inal jurisdiction to the federal courts over all cases arising under the Constitution and laws of the United States. Maritime law was federal law based on a constitutional grant of jurisdiction. Thus maritime cases arose under the Constitution or federal laws. By this mode of reasoning the words of the jurisdictional statute are found to “fit like a glove.” 48

Although it is true that the supremacy of federal maritime law over conflicting state law has recently been greatly extended, the federal nature of the maritime law administered in the federal courts has long been an accepted part of admiralty jurisprudence. The classic statement of Mr. Justice Holmes in The Western Maid, 257 U. S. 419, 432, summed up the accepted view that maritime law derived its force from the National Government and was part of the laws of the United States; and this was merely a restatement of a view which was clearly set forth in 1874 in The Lottawanna, 21 Wall. 558.49 Thus the theory which underlies the effort to infuse general maritime jurisdiction into the Act of 1875 rests on no novel development in maritime law, but on premises as available in 1875 as they are today.

The simple language of the Act of 1875 conceals complexities of construction and policy which have been already examined. When we apply to the statute, and to the clause of Article III from which it is derived, commonsensical and lawyer-like modes of construction, and the evidence of history and logic, it becomes clear that the words of that statute do not extend, and could not reasonably be interpreted to extend, to cases of admiralty and maritime jurisdiction. The statute is phrased in *379terms which, as a matter of inert language, lifeless words detached from the interpretive setting of history, legal lore, and due regard for the interests of our federal system, may be used as playthings with which to reconstruct the Act to include cases of admiralty and maritime jurisdiction. If the history of the interpretation of judiciary legislation teaches anything, it teaches the duty to reject treating such statutes as a wooden set of self-sufficient words — a failing to which the Court has not been subject since the Pacific Railroad Removal Cases.50 The Act of 1875 is broadly phrased, but it has been continuously construed and limited in the light of the history that produced it, the demands of reason and coherence, and the dictates of sound judicial policy which have emerged from the Act’s function as a provision in the mosaic of federal judiciary legislation. It is a statute, not a Constitution, we are expounding.51

The considerations of history and policy which investigation has illuminated are powerfully reinforced by the deeply felt and traditional reluctance of this Court to expand the jurisdiction of the federal courts through a broad reading of jurisdictional statutes. A reluctance which must be even more forcefully felt when the expansion is proposed, for the first time, eighty-three years after the jurisdiction has been conferred. Mr. Justice Stone, speaking of the Act of 1875, pointed out that “[t]he policy *380of the statute calls for its strict construction. . . . Due regard for the rightful independence of state governments, which should actuate federal courts, requires that they scrupulously confine their own jurisdiction to the precise limits which the statute has defined.” 52 Certainly this wise counsel is deeply persuasive when we are asked to accept a doctrine which would cut into a jurisdiction exercised by the States since Colonial days. Of course if compelling reasons can be found for redefining the statute, if an ancient error cries out for rectification, we should not be deterred from applying new illuminations to the interpretation of past enactments. However, in our examination of the manifold considerations of history, of construction, of the policy which underlies the allocation of competence over maritime matters in our federal system, and the considerations of judicial administration and procedure called into question — all of which direct us to the rejection of the proposed infusion of general maritime jurisdiction into the Act of 1875 — we are pointed to no considerations which lead us to overturn the existing maritime jurisdictional system — a system which is as old, and as justified by the experience of history, as the federal courts themselves.

(c) “Pendent” and Diversity Jurisdiction. — Rejection of the proposed new reading of § 1331 does not preclude consideration of petitioner’s claims under the general maritime law. These claims cannot, we have seen, be justified under § 1331. However, the District Court may have jurisdiction of them “pendent” to its jurisdiction under the Jones Act. Of course the considerations which call for the exercise of pendent jurisdiction of a state claim related to a pending federal cause of action within the appropriate scope of the doctrine of Hum v. Oursler, 289 U. S. 238, are not the same when, as here, what is involved *381are related claims based on the federal maritime law. We perceive no barrier to the exercise of “pendent jurisdiction” in the very limited circumstances before us. Here we merely decide that a district judge has jurisdiction to determine whether a cause of action has been stated if that jurisdiction has been invoked by a complaint at law rather than by a libel in admiralty, as long as the complaint also properly alleges a claim under the Jones Act. We are not called upon to decide whether the District Court may submit to the jury the “pendent” claims under the general maritime law in the event that a cause of action be found to exist.

Respondents Garcia & Diaz and Quin Lumber Company, New York corporations, and International Terminal Operating Company, a Delaware corporation, are of diverse citizenship from the petitioner, a Spanish subject. Since the Jones Act provides an independent basis of federal jurisdiction over the non-diverse respondent, Com-pania Trasatlántica, the rule of Strawbridge v. Curtiss, 3 Cranch 267, does not require dismissal of the claims against the diverse respondents. Accordingly, the dismissal of these claims for lack of jurisdiction was erroneous.

II. The Claims Against Compañía Trasatlántica— The Choice-of-Law Problem.

We now turn to the claims against Compania Tras-atlántica under the Jones Act and the general maritime law. In light of our recent decision in Lauritzen v. Larsen, 345 U. S. 571, these claims present the narrow issue, whether the maritime law of the United States may be applied in an action involving an injury sustained in an American port by a foreign seaman on board a foreign vessel in the course of a voyage beginning and ending in a foreign country.

*382While Lauritzen v. Larsen involved claims asserted under the Jones Act, the principles on which it was decided did not derive from the terms of that statute. We pointed out that the Jones Act had been written “not on a clean slate, but as a postscript to a long series of enactments governing shipping. All were enacted with regard to a seasoned body of maritime law developed by the experience of American courts long accustomed to dealing with admiralty problems in reconciling our own with foreign interests and in accommodating the reach of our own laws to those of other maritime nations.” 345 U. S., at 577. Thus the Jones Act was applied “to foreign events, foreign ships and foreign seamen only in accordance with the usual doctrine and practices of maritime law.” 345 U. S., at 581. The broad principles of choice of law and the applicable criteria of selection set forth in Lauritzen were intended to guide courts in the application of maritime law generally. Of course, due regard must be had for the differing interests advanced by varied aspects of maritime law. But the similarity in purpose and function of the Jones Act and the general maritime principles of compensation for personal injury, admit of no rational differentiation of treatment for choice of law purposes. Thus the reasoning of Lauritzen v. Larsen governs all claims here.53

We are not here dealing with the sovereign power of the United States to apply its law to situations involving one or more foreign contacts.54 But in the absence of a contrary congressional direction, we must apply those principles of choice of law that are consonant with the needs of a general federal maritime law and with due *383recognition of our self-regarding respect for the relevant interests of foreign nations in the regulation of maritime commerce as part of the legitimate concern of the international community. These principles do not depend upon a mechanical application of a doctrine like that of lex loci delicti commissi. The controlling considerations are the interacting interests of the United States and of foreign countries, and in assessing them we must move with the circumspection appropriate when this Court is adjudicating issues inevitably entangled in the conduct of our international relations. We need not repeat the exposition of the problem which we gave in Lauritzen v. Larsen. Due regard for the relevant factors we there enumerated, and the weight we indicated to be given to each, preclude application of American law to the claims here asserted.

In this case, as in Lauritzen v. Larsen, the ship is of foreign registry and sails under a foreign flag. Both the injured seaman and the owner of the ship have a Spanish status: Romero is a Spanish subject and Compania Trasatlántica a Spanish corporation. Unlike the contract in Lauritzen, Romero’s agreement of hire was entered into in Spain. By noting this fact, we do not mean to qualify our earlier view that the place of contracting is largely fortuitous and of little importance in determining the applicable law in an action of marine tort. Here, as in Lauritzen, the foreign law provides a remedy for the injury, and claims under that law may be conveniently asserted before the Spanish consul in New York.55

In Lauritzen v. Larsen the injury occurred in the port of Havana and the action was brought in New York. Romero was injured while temporarily in American territorial waters. This difference does not call for a difference in result. Discussing the significance of the place of the *384wrongful act, we pointed out in Lauritzen that “[t]he test of location of the wrongful act or omission, however sufficient for torts ashore, is of limited application to shipboard torts, because of the varieties of legal authority over waters she may navigate. . . . the territorial standard is so unfitted to an enterprise conducted under many territorial rules and under none that it usually is modified by the more constant law of the flag.” 345 U. S., at 583-584. Although the place of injury has often been deemed determinative of the choice of law in municipal conflict of laws, such a rule does not fit the accommodations that become relevant in fair and prudent regard for the interests of foreign nations in the regulation of their own ships and their own nationals, and the effect upon our interests of our treatment of the legitimate interests of foreign nations. To impose on ships the duty of shifting from one standard of compensation to another as the vessel passes the boundaries of territorial waters would be not only an onerous but also an unduly speculative burden, disruptive of international commerce and without basis in the expressed policies of this country. The amount and type of recovery which a foreign seaman may receive from his foreign employer while sailing on a foreign ship should not depend on the wholly fortuitous circumstance of the place of injury.

Thus we hold that the considerations found in Laurit-zen v. Larsen to preclude the assertion of a claim under the Jones Act apply equally here, and affirm the dismissal of petitioner’s claims against Compania Trasatlántica.

III. The Claims Against the Other Respondents.

(a) Petitioner made claims based both on the Jones Act and the general maritime law against Garcia & Diaz, Inc. At the pre-trial hearing the District Court concluded that Garcia & Diaz was not Romero’s employer and did not operate and control the vessel at the time of *385the injury. These issues were properly adjudicated, and thus the claims for unseaworthiness and maintenance and cure were properly dismissed. However, the District Court did not consider, and its disposition of the case did not require it to consider, whether petitioner was asserting a claim based upon the negligence of Garcia & Diaz; a claim independent of the employment relationship or operation and control. Thus it is necessary to remand the case for further proceedings as to this respondent.

(b) The claims against International Terminal Operating Co., and Quin Lumber Co., for a maritime tort, were dismissed for lack of jurisdiction. Our decision on the jurisdictional issues necessitates the return of the claims against these respondents for further adjudication.

The judgment of the Court of Appeals is vacated and the cause remanded to the District Court for further proceedings not inconsistent with this opinion.

Vacated and remanded.

APPENDIX TO OPINION OF THE COURT.

The following is the list of treatises on federal procedure and jurisdiction and admiralty law which were examined to determine if any commentator gave any intimation that the Act of 1875 had swept admiralty jurisdiction within its scope. No such intimation is found in a single treatise. On the contrary, all those which dealt with the subject specifically assumed that the federal courts on the law side had jurisdiction over a maritime cause after the Act of 1875 as before only when the parties were of diverse citizenship.

Boyce, Manual of the Practice in the Circuit Courts (1869).

Abbott, The United States Courts and Their Practice (1877).

Phillips, Statutory Jurisdiction and Practice of the Supreme Court of the United States (1878).

*386Desty, Manual of the Law Relating to Shipping and Admiralty (1879).

CuRTis, Jurisdiction, Practice and Peculiar Jurisprudence of the Courts in the United States (1880).

Bump, Federal Procedure (1881).

Miller and Field, Federal Practice (1881).

CoheN, Admiralty — Jurisdiction, Law and Practice (1883).

Field, Constitution and Jurisdiction of the Courts of the United States (1883).

Spear, Law of the Federal Judiciary (1883).

Thatcher (Thatcher’s Practice) — A Digest of Statutes, Equity Rules and Decisions upon the Jurisdiction, Pleadings and Practice of the Circuit Courts of the United States (1883).

Thatcher (Thatcher’s Practice) — A Digest of Statutes, Admiralty Rules and Decisions upon the Jurisdiction, Pleadings and Practice of the District Courts of the United States (1884).

HeNRY, Jurisdiction and Procedure of the Admiralty Courts (1885).

Holt, The Concurrent Jurisdiction of the Federal and State Courts (1888).

Curtis, Jurisdiction, Practice and Peculiar Jurisprudence of the Courts of the United States (rev. ed. 1896).

Benedict, The American Admiralty (3d ed. 1898).

Garland and Ralston, Constitution and Jurisdiction of the U- S. Courts (1898).

Simonton, Charles H. (U. S. Circuit Judge), The Federal Courts, Their Organization, Jurisdiction and Procedure (2d ed. 1898).

Carter, The Jurisdiction of Federal Courts as Limited by the Citizenship and Residence of the Parties (1899).

Desty, Manual of Practice in the Courts of the United States (9th ed. 1899).

May, Practice and Procedure of the U. S. Supreme Court (1899).

Dwyer, The Law and Procedure of United Stajtes Courts (1901).

Hughes, Handbook of Admiralty Law (1901).

Taylor, Jurisdiction and Procedure of the Supreme Court of the U.S. (1905).

Rose, Code of Federal Procedure (1907).

Bates, Federal Procedure at Law (1908).

Encyclopedia oe United States Supreme Court Reports (1908).

*387Benedict, The American Admiralty (4th ed. 1910).

Loveland, Appellate Jurisdiction of the Federal Courts (1911).

Hughes, Handbook of Jurisdiction and Procedure in United States Courts (2d ed. 1913).

Bunn, Jurisdiction and Practice of the Courts of the United States (1914) (also 3d ed. 1927 ; 4th ed. 1939; 5th ed. 1949).

Thayer, Jurisdiction of the Federal Courts (1914).

Chaplin, Principles of the Federal Law (1917).

Long, Outline of the Jurisdiction and Procedure of the Federal Courts (3d ed. 1917).

Foster, Federal Practice (6th ed. 1920).

Hughes, Handbook of Admiralty Law (2d ed. 1920).

Loveland, Annotated Forms of Federal Procedure (3d ed. 1922).

Rose, Jurisdiction and Procedure of the Federal Courts (2d ed. 1922).

Montgomery, Manual of Federal Jurisdiction and Procedure (3d ed. 1927).

Williams, Federal Practice (2d ed. 1927).

Dobie, Handbook of Federal Jurisdiction and Procedure (1928).

Longsdorf, Cyclopedia of Federal Procedure (1928).

Zoline, Federal Appellate Jurisdiction and Procedure (3d ed. 1928).

Hughes, Federal Practice, Jurisdiction and Procedure (1931).

Rose, Jurisdiction and Procedure of the Federal Courts (4th ed. 1931).

Browne, Federal Appellate Practice and Procedure (1932).

Brown, Guide to Federal and Bankruptcy Practice (1933).

Hopkins, Federal Judicial Code and the Judiciary (4th ed. 1934).

Marker, Federal Appellate Jurisdiction and Procedure (1935).

Rose, Jurisdiction and Procedure of the Federal Courts (5th ed. 1938).

Simkins, Federal Practice (3d ed. 1938) (also 1942 Supplement).

Robinson, Handbook of Admiralty Law in the United States (1939).

Benedict, Law of American Admiralty (Knauth ed. 1940).

Pound, Organization of Courts (1940).

Kirshbaum, Outline of Federal Practice and Procedure (1941).

O’Brien, Manual of Federal Appellate Procedure (3d ed. 1941).

Montgomery, Manual of Federal Jurisdiction and Procedure (4th ed. 1942).

*388Federal Redbook AND Practice Annual (Schweitzer ed. 2d ed. 1943).

Bender, Federal Practice Manual (1948).

Sunderland, Judicial Administration (1948).

Guandolo, Federal Procedure Forms (1949)

Moore, Á. Commentary on the Judicial Code (1949).

Wendell, Relations Between the Federal and State Courts (1949).

Barron and Holtzoef, Federal Practice and Procedure (1950).

Fins, Federal Practice Guide (1950).

Ohlinger, Federal Practice (rev. ed. 1950), Replacement Vol. One-A.

dissenting in part and concurring in part.

I.

I regret that I cannot agree with the Court’s holding that § 1331 of the Judicial Code does not give jurisdiction to a Federal District Court, sitting at law, over a seaman’s claims against his employer for maintenance and cure and for indemnity damages for injury caused by unseaworthiness, where the claims are asserted in the manner of a *390suit at common law and the requisite jurisdictional amount is in controversy. I believe that the jurisdictional statute and the logic of the principles of this Court’s decisions construing it compel a contrary result. I think the Court’s opinion attempts to turn aside the statutory language and the thrust of this Court’s decisions with reasoning that is altogether too insubstantial.

The point on which the Court and I are at issue is one which has been much mooted in the Courts of Appeals, and I agree that it is appropriate that a thorough expression of views on it be presented. I propose first to explain why jurisdiction should be sustained under § 1331, and then to offer some reply to specific arguments set forth by the Court which apparently proceed from supposed practical inconveniences that are thought to arise from sustaining the jurisdiction.

The petitioner brought this suit in a Federal District Court. The element in his action with which I am dealing is his claim for money damages from Compania Tras-atlántica, his employer, for breach of the shipowner’s duty to maintain a seaworthy ship and for maintenance and cure. Since there was no diversity of citizenship between petitioner and Compania Trasatlántica,1 jurisdiction was predicated on the grant in 28 U. S. C. § 1331 of jurisdiction in “civil actions wherein the matter in controversy . . . arises under the Constitution, laws or treaties of the United States.”2 Jurisdiction of such claims *391could have been established on the admiralty side of the District Court since 28 U. S. C. § 1333 specifically grants jurisdiction in the District Courts in “case[s] of admiralty or maritime jurisdiction.” The question is whether petitioner can bring this part of his action on the law side of a Federal District Court.

First. In a long series of decisions tracing from Southern Pacific Co. v. Jensen, 244 U. S. 205, this Court has made it clear that, in a seaman's action to recover damages for a maritime tort from his employer, the substantive law to be applied is federal maritime law made applicable as part of the laws of the United States by the Constitution itself, and that the right of recovery, if any, is a federally created right.3 Chelentis v. Luckenbach S. S. Co., 247 U. S. 372; Knickerbocker Ice Co. v. Stewart, 253 U. S. 149; Garrett v. Moore-McCormack Co., 317 U. S. 239; Pope & Talbot, Inc., v. Hawn, 346 U. S. 406. Cf. Red Cross Line v. Atlantic Fruit Co., 264 U. S. 109, 124-125.

It is true that early in our history maritime law was thought to be an international law merchant which was impartially administered by the several maritime nations of the world. This concept was expressed by Chief Justice Marshall’s language in American Ins. Co. v. Canter, *3921 Pet. 511, 545-546: “A case in admiralty does not, in fact, arise under the Constitution or laws of the United States. These cases are as old as navigation itself; and the law, admiralty and maritime, as it has existed for ages, is applied by our Courts to the cases as they arise.” But that this did not mean that there was some supranational law, by which American courts were bound, was made clear by Mr. Justice Bradley in The Lottawanna, 21 Wall. 558, 572, where he said for the Court: “[I]t is hardly necessary to argue that the maritime law is only so far operative as law in any country as it is adopted by the laws and usages of that country. . . .” This teaching was emphasized in The Western Maid, 257 U. S. 419, 432, where Mr. Justice Holmes, speaking for the Court, said: “[W]e must realize that however ancient may be the traditions of maritime law, however diverse the sources from which it has been drawn, it derives its whole and only power in this country from its having been accepted and adopted by the United States. There is no mystic over-law to which even the United States must bow.”

The sovereign power which determines the rules of substantive law governing maritime claims of the sort which petitioner asserts here is federal power, speaking through Congress as in the case of the Jones Act, or through this Court in the case of judicially defined causes of action. Southern Pacific Co. v. Jensen, supra. This is an area where the federal courts have defined substantive rules themselves, and have not applied state law. Indeed, it is federal substantive law so created which the States must enforce in such actions brought in state courts, Garrett v. Moore-McCormack Co., supra, and which the federal courts have applied in actions at law in which diversity of citizenship has been relied upon as a jurisdictional basis, Pope & Talbot, Inc., v. Hawn, supra. The causes of action asserted against his employer by petitioner here present “no claim created by or arising out *393of [state] law. His right of recovery ... is rooted in federal maritime law.” Id., at 409.

Second. Since petitioner’s causes of action for unseaworthiness and for maintenance and cure are created by federal law, his case arises under “the laws ... of the United States” within the meaning of § 1331, for it is clear that “a suit arises under the law that creates the .cause of action.” Holmes, J., in American Well Works Co. v. Layne & Bowler Co., 241 U. S. 257, 260.4 The contention cannot be accepted that since petitioner’s rights are judicially defined, The Osceola, 189 U. S. 158, they are not created by “the laws ... of the United States” within the meaning of § 1331; or, in other words, that only maritime rights created by Act of Congress are created by “the laws ... of the United States.” In another context, that of state law, this Court has recognized that the statutory word “laws” includes court decisions. Erie R. Co. v. Tompkins, 304 U. S. 64. The converse situation is presented here in that federal courts have an extensive responsibility of fashioning rules of substantive law in maritime cases. See Wilburn Boat Co. v. Fireman’s Fund Ins. Co., 348 U. S. 310, 314. These rules are as fully “laws” of the United States as if they had been enacted by Congress. Cf. Garrett v. Moore-McCormack Co., supra; Warren v. United States, 340 U. S. 523, 526-528; and see Mater v. Holley, 200 F. 2d 123.5

*394Third. Notwithstanding these conclusions, jurisdiction under § 1331 would, of course, not lie if it were beyond the constitutional power of Congress to vest jurisdiction over this action of a seaman against his employer, a matter falling «admittedly within the “admiralty or maritime jurisdiction,” in a federal court sitting at law. But it is too late to make such an argument. The jurisdictional treatment of the rights of seamen under the Jones Act, a cause of action bound up with the cause of action in question here, is preclusive on the issue. The Jones Act was held in Panama R. Co. v. Johnson, 264 U. S. 375, to be authorized by the legislative power residing in the Admiralty Clause of Article III. The right of action granted was, however, specifically stated by Congress to be exercisable “at law, with the right of trial by jury” and in the Federal District Courts. This treatment was upheld, against constitutional challenge, by the Court, which held that jurisdiction properly lay, at the option of the plaintiff, either in admiralty or on the law side of the District Court. “[T]he constitutional provision interposes no obstacle to permitting rights founded on the maritime law or an admissible modification of it to be enforced as such through appropriate actions on the common-law side of the courts . . . .” Id., at 388. And the unchallenged maintenance of the very cause of action in question here at law in the District Courts under 28 U. S. C. § 1332, where diversity of citizenship is present, is further proof that no constitutional inhibition to the maintenance of such an action at law under § 1331 exists. Cf. The Belfast, 7 Wall. 624, 644.

*395But despite the constitutional power of Congress, jurisdiction under § 1331 may still be defeated if that power has not there been exercised; in other words, if that jurisdictional grant is to be read as containing an implied exception as to cases falling within the “admiralty or maritime jurisdiction.” See Paduano v. Yamashita Kisen Kabushiki Kaisha, 221 F. 2d 615; Jenkins v. Roderick, 156 F. Supp. 299, 302. This I take to be the net effect of the Court’s reasoning. The gist of the argument, as it has been developed in the Courts of Appeals, is that § 1331 was enacted “to insure the availability of a forum designed to minimize the danger of hostility toward, and specially suited to the vindication of, federally created rights . ...” Paduano v. Yamashita Kisen Kabushiki Kaisha, supra, at 618. Continuously since 1789 Congress has provided specially for admiralty courts in which rights under the federal maritime law could be asserted. The argument runs that it follows that claims under the maritime law were not intended to fall within the scope of § 1331. And here, the Court’s conclusion rests primarily on an analysis of the terms and background of the 1875 Act which was the ancestor of § 1331, and on various inferences drawn from silence after that Act’s passage.

The members of the First Congress, in agreement that national courts of admiralty were an imperative necessity of the times, 1 Annals of Cong. 797-798 (1789), gave to the District Courts in § 9 of the First Judiciary Act original jurisdiction over “all civil causes of admiralty and maritime jurisdiction . . . .” 1 Stat. 76, 77. Under § 21 the Circuit Courts were given appellate jurisdiction “in causes of admiralty and maritime jurisdiction . . . .” 1 Stat. 83. These phrases followed almost literally the wording of Art. Ill, § 2, of the Constitution, extending the federal judicial power “to all Cases of admiralty and maritime Jurisdiction . . . .” Significantly, the First *396Judiciary Act granted to the District and Circuit Courts no general federal-question jurisdiction.

Section 9 of the First Judiciary Act, however, contained the clause . . saving to suitors, in all cases, the right of a common law remedy, where the common law is competent to give it . . . ” The Saving Clause survives in 28 U. S. C. § 1333, phrased “. . . saving to suitors in all cases all other remedies to which they are otherwise entitled. . . This provision, plainly, was a recognition that there were, prior to 1789, maritime claims within the concurrent jurisdiction of courts of admiralty and law, 1 Benedict, American Admiralty (6th ed. 1940), § 20; Schoonmaker v. Gilmore, 102 U. S. 118, 119, and it was clearly the intention of Congress to perpetuate this duality of remedy. It is true that certain classes of cases, such as the traditional in rem, prize, and seizure cases, lay within the exclusive jurisdiction of the admiralty, 1 Benedict, American Admiralty, § 23; The Moses Taylor, 4 Wall. 411; The Hine v. Trevor, 4 Wall. 555; The Glide, 167 U. S. 606, but all other suits under the maritime law of an in personam nature might be brought as well in the state courts or, under the diversity jurisdiction, in the Federal Circuit Courts. § 11, 1 Stat. 78.

It is thus clear that any argument that § 1333 is an exclusive grant of jurisdiction would be false to the history of enactments allocating the judicial power of the United States. The fact that, in a diversity case under § 1332, the claimant is free to proceed on the law side of the federal court to enforce rights created by the federal maritime law, Seas Shipping Co. v. Sieracki, 328 U. S. 85, 88-89, clearly runs counter to any theory that the federal courts, because of § 9 of the Judiciary Act of 1789, can adjudicate maritime claims only while sitting in admiralty. There is no compelling reason why § 1333, which does not exclude maritime actions from being brought at law in a federal court under § 1332, should exclude them *397from being so brought under § 1331.6 Indeed, I find it a gross anomaly to hold, as the Court holds today, that an action rooted in federal law can be brought on the law side of a federal court only if the diversity jurisdiction, usually a vehicle for the enforcement of state-created rights, can be invoked.

Plainly there is nothing in the language of § 1331 which would exclude jurisdiction of maritime claims of the nature asserted by petitioner. Rather, in more than a manner of speaking, the language of that section fits the cause of action in question here “like a glove,” Jenkins v. Roderick, 156 F. Supp. 299, 301. But the Court reasons that the section must be read restrictively because the corresponding jurisdictional grant in the Constitution speaks of “Cases, in Law and Equity, arising under this Constitution, the Laws of the United States . . . .” This specification of “law and equity,” reflected in the 1875 ancestor7 of present § 1331 as “suits of a civil nature at common law or in equity . . . arising under the Constitution or laws . . . ,” is said to indicate that a suit arising under the substantive maritime law is not comprehended under the section. But the argument mistakes the nature of a Saving Clause action. An action brought under the Saving Clause is maintained “at law” or “in *398equity,” and the very action that Romero would assert here he would assert “at law.” The mere fact that the substantive claim a court enforces in a particular Saving Clause action is rooted in the general maritime law does not transform the proceedings from a suit “at law’ to one “in admiralty”; the state courts can hardly be said to sit “in admiralty” when they try actions under the Saving Clause. Cf. The Hamilton, 207 U. S. 398, 404. The Saving Clause itself, in its 1789 form, stated that what it was “saving” was “a common law remedy” to be available in maritime fact situations. It can readily be admitted that a suit “in admiralty” is not the same thing as a suit “at law.” But this is not to say that a suit involving a maritime cause of action cannot be the subject of a suit “at law” in the federal courts. Obviously Saving Clause actions brought on the law side of the federal court, with diversity of citizenship present, are actions “at law.” In fact, the grant of diversity jurisdiction in the 1875 Act was in the very same terms as the grant of the “arising under” jurisdiction; the same introductory phrase, “suits of a civil nature at common law or in equity,” governed both grants. It seems to me very odd to say that this phrase, introducing two grants of jurisdiction, had the effect of excluding maritime causes of action entirely from the one but not at all from the other.

The legislative history of § 1331 does not indicate any intent on the part of Congress to exclude claims asserted under federal maritime law from its ambit. The present section is but the latest recodification of the provisions of the Judiciary Act of 1875, 18 Stat. 470, alluded to above, which for the first time with any permanence vested in the federal courts an original general federal-question jurisdiction over any claim which “arises under the Constitution, laws or treaties of the United States.” The congressional debates focused so largely on proposed *399changes in the diversity jurisdiction that no considered scrutiny was given to the provisions which have become § 1331. See 2 Cong. Rec. 4978-4988; Frankfurter and Landis, The Business of the Supreme Court (1927 ed.), 65-69. Nothing appears which would indicate a congressional intent to modify, by implication or otherwise, the sweep of the language of this Act, embodying as it does substantially the words of the constitutional grant.8 And nothing appears which would indicate any intention that the Act's coverage be “frozen” to exclude federal causes of action which were not fully developed in 1875.

The Court argues, however, that Congress, aware of Chief Justice Marshall’s statement that Article III created the admiralty jurisdiction as “distinct” from the “arising under” jurisdiction,9 American Ins. Co. v. Canter, *400supra, at 545, intended that the jurisdictional statutes be mutually exclusive. The manager of the 1875 legislation in the Senate declared of the bill generally that it conferred “precisely the power which the Constitution confers — nothing more, nothing less.” 2 Cong. Rec. 4987. It is difficult to infer that Congress meant to crystallize any particular interpretation of the Constitution in the statute. But even if it were proper, in the absence of concrete indication, speculatively to breathe into our construction of § 1331 views of the Constitution10 which might have served as a silent premise of congressional *401action, I do not think that the Court here is called on to do so. Marshall’s statement is not, when understood in its context, contrary to my position, and in fact its proper scope was recognized before 1875.

Before discussing the Canter case, I think it wise to restate the precise nature of the issue before the Court. This is so because I fear the Court, in an expansive reading of Canter not justified either by what was decided there or by what was said there considered in the light of what was decided, has blurred the issue for decision today. The issue before us is not whether all cases “of admiralty and maritime jurisdiction” are per se encompassed in the statutory “arising under” jurisdiction. A suit seeking the sort of remedy that the common law is not competent to give could not be fairly contended to lie under § 1331; it would clearly be the sort of suit in which the jurisdictional grant of § 1333 was intended to be exclusive. The issue before us concerns only actions maintainable in some forum “at law” under the Saving Clause. And again, the issue is not even the narrower one whether Saving Clause actions are per se cognizable under § 1331. The tests of jurisdiction under § 1331 must still be met, and there is no contention that they are met merely by a showing that an action is one maintainable under the Saving Clause and involving the requisite jurisdictional amount. The plaintiff’s right to recovery must still be one rooted in federal substantive law, and it has quite recently been made clear that there are Saving Clause actions that do not meet that test. Wilburn Boat Co. v. Fireman’s Fund Ins. Co., 348 U. S. 310. The issue before us is only whether the fact that an action is a Saving-Clause action excludes it from § 1331 where it would otherwise be maintainable thereunder.

At issue in American Ins. Co. v. Canter was the power of a territorial court to make a decree selling cargo to satisfy a maritime lien in rem existing in favor of its *402salvors. A state court, even under the Saving Clause, could not pass such a decree at all; it is the enforcement of the classic admiralty remedy, and a matter solely within the competence of the federal admiralty courts. The Hine v. Trevor, 4 Wall. 555. In the passage from Marshall’s opinion relied upon, the Chief Justice was saying only that the Act of Congress which conferred on certain territorial courts jurisdiction in “cases arising under the laws and constitution of the United States,” § 8, 3 Stat. 752, did not by that token alone grant them power to enforce a remedy peculiarly within the competence of admiralty courts.11 In its broadest permissible interpretation, the dictum only means that the fact that the Constitution creates admiralty jurisdiction does not make all admiralty cases cases arising under the Constitution.12 But Marshall’s opinion does not say that an action seeking remedial relief of a sort which the common law is competent to give, and in which the plaintiff’s right to recover is rooted in federal law, ceases to be a suit arising under the laws of the United States merely because it is of a maritime .nature.13 No one is contending here, *403of course, that § 1331 is a grant of power to enforce remedies peculiar to the admiralty; the contention is solely that that section, which empowers a federal court to administer common-law remedies in vindication of rights of plaintiffs which take their origin from federal law, is not subject to an exception for rights taking their origin in federal maritime law. Marshall’s opinion simply is not addressed to this question or dispositive of it.

Much is made by the Court of Marshall’s language that the categories of actions he mentions are “distinct” and not “identical.” Of course this is so, in a real sense and the only sense in which Marshall meant it. A matter affecting an ambassador or a consul is not per se an action “arising under,” just as it is not per se a maritime action. But could not a case involving a consul be also a case of admiralty jurisdiction, under certain fact situations? And could not a suit by or against a consul happen, perchance, to be also one “arising under” ? The fact that the jurisdictional categories are separate and distinct, as Marshall demonstrates, does not mean that a particular action could not come under the heading of more than one of them. Everyone recognizes that this is the case in a maritime matter in which the parties are of diverse citizenship. I see no reason why it should not be true here of Romero’s general maritime law claims against his employer.

*404It appears also to be clear that even before 1875 Marshall's opinion was not thought of as creating a situation in which it was impossible to say that there were maritime cases that could be also attributed to other categories of federal jurisdiction. Long before Congress contemplated the jurisdictional grant of 1875, this Court in Taylor v. Carryl, 20 How. 583, made it clear that there were fact situations which were of maritime cognizance, giving rise to rights for which the admiralty could supply a remedy, or for which alternatively proceedings at the course of common law lay. Maritime torts were specifically conceived of as within this category. The Court in that case followed the view of Mr. Justice Story expressed in his Commentaries on the Constitution, which were quoted with approval:

“ 'Mr. Chancellor Kent and Mr. Rawle seem to think that the admiralty jurisdiction given by the Constitution is, in all cases, necessarily exclusive. But it is believed that this opinion is founded on mistake. It is exclusive in all matters of prize, for the reason that, at the common law, this jurisdiction is vested in the courts of admiralty, to the exclusion of the courts of common law. But in cases where the jurisdiction of common law and admiralty are concurrent, (as in cases of possessory suits, mariners’ wages, and marine torts,) there is nothing in the Constitution necessarily leading to the conclusion that the jurisdiction was intended to be exclusive; and there is no better ground, upon general reasoning, to contend for it. The reasonable interpretation . . . would seem to be, that it conferred on the national judiciary the admiralty and maritime jurisdiction exactly according to the nature and extent and modifications in which it existed in the jurisprudence of the common law. When the jurisdiction was exclusive, it remained so; when it was con*405current, it remained so. Hence the States could have no right to create courts of admiralty as such, or to confer on their own courts the cognizance of such cases as were exclusively cognizable in admiralty courts. But the States might well retain and exercise the jurisdiction in cases of which the cognizance was previously concurrent in the courts of common law. This latter class of cases can be no more deemed cases of admiralty and maritime jurisdiction than cases of common-law jurisdiction.' (3 Story’s Com., sec. 1666, note.)” 20 How., at 598.

And it was understood before 1875 that this concurrent jurisdiction at law was not one merely existent in the state courts, but one available to suitors in the federal courts. See The Belfast, 7 Wall. 624, 644, infra, pp. 406-407.

Accordingly, I cannot see how it can be concluded that Congress in 1875 read Marshall’s opinion as creating some sort of gulf that would make it impossible for any maritime case to be also one “arising under the Constitution or laws of the United States.” 14

*406Of course, one cannot rely, to prove the Court’s thesis, on dicta in cases decided before 1875 to the effect that Saving Clause actions could be brought on the law side of a federal court only when there is diversity of citizenship, and the Court does not so rely. The Belfast, 7 Wall. 624, 643-644; Leon v. Galceran, 11 Wall. 185, 188; Steamboat Co. v. Chase, 16 Wall. 522, 533. The 1875 Act for the first time with any permanence granted general federal-question jurisdiction to the federal courts of first instance. It can hardly be denied that these statements were correct when made, but it is equally plain that they are no authority for limiting the law-side jurisdiction to diversity cases once the 1875 Act had been passed. Moreover, I cannot seriously attach any significance, as the Court does, to the repetition, obiter, of their formulation in a case decided shortly after the Act’s passage, where the effect of the new statute was not at all presented or discussed. Norton v. Switzer, 93 U. S. 355, 356. In fact, the approach this Court followed in the interpretation of the Saving Clause during this period supports, rather than detracts from, my conclusion here. It was observed in 1869 that the remedies saved by the Saving Clause were saved “to suitors, and not to the State courts, nor to the Circuit Courts 15 of the United States. . . . Congress intended by that provision to allow the party to seek redress in the admiralty if he saw fit to do so, but not to make it compulsory in any *407case where the common law is competent to give him a remedy. Properly construed, a party under that provision may proceed in rem in the admiralty, or he may bring a suit in personam in the same jurisdiction, or he may elect not to go into admiralty at all, and may resort to his common law remedy in the State courts or in the Circuit Court of the United States, if he can make proper parties to give that court jurisdiction of his case.” The Belfast, 7 Wall. 624, 644. It is clear from the Court’s language that the common-law remedies saved to suitors could properly be enforced in any tribunal otherwise having jurisdiction; the remedies saved were saved generally to suitors without discrimination as to any tribunal.

Nor can I consider it sound to place the reliance the Court has placed on the fact that the arguments we are considering today were not raised until 1950. Till then no court ever considered the problem that we discuss here at great length. None of the assortment of commentators listed in the Court’s Appendix ever discussed it. The Court’s argument, in fact, claims to draw force from the fact that it was not discussed at all. From the fact that the issue was never explored or tried at all till 1950, when Judge Magruder, in a dictum in Jansson v. Swedish American Line,16 185 F. 2d 212, 216-218, took a point of view similar to the one expressed here, we are asked to infer that the argument for jurisdiction should not succeed when finally raised. I cannot accept this as a convincing argument in the construction of a broadly written statute which was intended, at least in some aspects, to be as broad and dynamic as the Constitution itself, and which has served as the basic jurisdictional entitlement for the vindication of the numerous and increasing types of federally created rights in the lower federal courts ever since its *408enactment. It is a modern development in legal science in this country’s federal system that increasing concern is taken with the source of the substantive law administered by the courts. Southern Pacific Co. v. Jensen, supra, and notably Erie R. Co. v. Tompkins, supra, are indications of this trend. When lawyers and judges in our federal system came to concentrate more and more on the source of the substantive law administered in the courts, and when this Court’s opinions made it increasingly clear that there were kinds of maritime actions where the underlying right to recover was rooted in federally created law, inadmissible of significant modification by the States, it was an inevitable consequence that the relation of § 1331 to maritime matters would come for the first time to be examined, as Judge Magruder examined it in the Jansson and Doucette cases. If one views the history of the common-law system of adjudication as the history of a process, one must conclude that the “historical” material relied upon by the Court has nothing to do with this sort of history at all, except to illustrate its antithesis.

It is, finally, true that this Court has adhered to a policy of construing jurisdictional statutes narrowly. Healy v. Ratta, 292 U. S. 263, 270; Thomson v. Gaskill, 315 U. S. 442, 446. In regard to the grant of federal-question jurisdiction to the District Courts, this Court has insisted that a claim created under federal law be a necessary part of the plaintiff’s case, Louisville & Nashville R. Co. v. Mottley, 211 U. S. 149, and that this claim be truly federal in nature, Gully v. First National Bank, 299 U. S. 109. But the present problem is apart from this line of cases, for here it is clear that petitioner is presenting to a federal court a claim created by federal law, and the objection is that somehow Congress intended to exclude claims of this particular sort from the grant in § 1331. But the arguments presented for such a narrow construction appear to me too insubstantial to withstand the logic of petitioner’s *409position. However willing one might be to resolve doubtful language against jurisdiction, exceptions to statutory language cannot be manufactured in a manner unwarranted by the words themselves and derived from the pertinent history only by a process of futile speculation. I am compelled to the conclusion that it is the effect of the 1875 Act and its intent, judged by the lights by which the courts must discern legislative intent, that the federal courts possess original jurisdiction, at law, to determine claims arising under federal substantive maritime law, where the common law is competent to afford the remedy sought by the plaintiff.

Fourth. The Court envisions various unfortunate results, from a practical standpoint, that would ensue from a holding on the jurisdictional issue under § 1331 contrary to its own. I shall comment briefly on its arguments.

It is first argued that the recognition of jurisdiction under § 1331 would, combined with the removal provisions of § 1441 (b) of the Judicial Code, operate to destroy the competence of the States in maritime matters altogether. A source cited by the Court itself17 indicates that in the five-year period 1953 to 1957 inclusive only about 150 decisions in Saving Clause actions have been rendered in all of the state courts of the country. As I have developed, resolution of the jurisdictional issue contrary to the majority’s view would not mean that all these cases would be assertable originally in the federal court or removable there, even present $10,000 in controversy. It is apparent then that the removability point addresses itself to a situation nearly de minimis. Saving Clause suitors seem long ago to have deserted the state courts. I therefore cannot share the concern that *410state judiciaries will be deprived of their historic active roles in the development of maritime law. Of the few actions that are left in the state courts, many may stay, for aught that can be predicted now. What sort of role do the state judiciaries now have in the development of the maritime law, with thirty-odd Saving Clause actions a year among them? Will the doctrine really put an end to this role, whatever it is? And it must be noted that such legislative competence as they possess remains to the States regardless of what may happen to the number of maritime cases in their courts; the view I have urged does not subtract one iota from the legislative competence of the States. And it is only because of an enlargement of removal that it affects their judicial competence; it does not take away their original jurisdiction at all, if suitors are content with it.

In further elaboration of the inroads on state competence which rejection of the Court’s view is supposed to entail, it is stated that it is a destructive oversimplification to claim that all enforced rights pertaining to maritime matters are rooted in federal law. So it is; and no one is so claiming. The point is not that all Saving Clause actions meet the “arising under” test of § 1331.18 It is, however, perfectly evident from the past holdings of this Court that the seaman’s action for unseaworthiness and maintenance and cure is rooted in federal law, and it is only this claim that need present the issue of the case as *411to § 1331. I agree perfectly with the Court’s observation that in our federal system allocations of jurisdiction have been carefully wrought to correspond to the realities of power and interest and national policy. I think that § 1331 embodies this approach by vesting in the federal courts, in civil actions, jurisdiction, at the option of the suitors, over all suits seeking a legal or equitable remedy, arising under federal law and involving a specified amount, and that this is so whether they involve maritime matters or not. I cannot see how it fits with the “realities of power and interest and national policy” to say that there is federal jurisdiction at common law over federally defined maritime causes of action only if there is diversity of citizenship among the parties involved in them.

The Court next argues that a holding to the contrary of its own will produce venue problems, and will in fact be unduly restrictive toward plaintiffs in their choice of forums. Where the District Courts have jurisdiction under § 1331 (even though diversity may also be present) i 1391 (b) of the Judicial Code rather than § 1391 (a) governs, and the suit must be brought in the defendants’ residence district, and may not be brought in the plaintiffs’ residence district, unless of course it also happens to be the defendants’. But one reading the discussion of the consequences this will have for plaintiffs is apt to forget (for the Court does not inform him) that defendants in maritime actions are most likely to be corporations (particularly in personal injury litigation, the sort of case we have at bar) and that § 1391 (c) declares that the residence of a corporation for venue purposes is any district where it is incorporated or any district in which it is licensed to do, or actually doing business. With corporate venue so widely defined, it will be a rare plaintiff (and a rarer personal injury plaintiff, for seamen and longshoremen are apt to live near where their employers carry on business, or where the vessel owners their employers *412serve do business) who can take much advantage from the fact that he can sue in the district of his own residence in an action based solely on diversity and not otherwise. And of course, the existence of proper venue at his own residence does not mean the plaintiff can sue the defendants there; he must still serve them with process. Except that process can be run throughout the limits of the State, while venue speaks in terms of the district, this means that the broader diversity venue only is of assistance where there is a defendant who, while not “doing business” in an area, is nonetheless amenable to process there. Of course there are some such, but I think by now the dimensions of this “practical” reason for the Court’s holding are patent.

rH hH

The Court, though it rejects Romero’s assertion of jurisdiction over his general maritime law claims against his employer under § 1331, proceeds to adjudicate them on the merits. It reaches them through a “pendent” jurisdiction theory analogous to Hum v. Oursler, 289 U. S. 238. The Court’s action appears unprecedented, as it appears to recognize. The prior applications of the doctrine recognized here have been limited to cases where claims arising under state law, over which there was no independent jurisdiction in the federal court, have been intertwined with federal claims. The theory has not been here applied to cases where there have been two types of claims, both admittedly within the District Court’s jurisdiction, one of which was admittedly cognizable according to the forms of the common law and the other, except for the theory, not. Here a plaintiff comes, into court desiring that his claims be adjudicated strictly according to the common law and disclaiming federal jurisdiction in admiralty. In short, he desires that a common-law jury pass upon his claims. If the federal *413courts do not have such jurisdiction over all his claims, there are state courts which do, and he may well prefer them in that event. The Court today tells him that though it is doubtful whether there is enough common-law jurisdiction in the federal courts to proceed to a plenary adjudication of his claim, there is enough certainly to award summary judgment against him on the merits. I must say I cannot understand a sort of jurisdiction that allows the federal courts to make a preliminary exploration of the merits of the case, and a binding adjudication upon them, but which may not allow them to go further.

Obviously what we have here, once the Court's view of § 1331 is accepted, and as claims are presented which can survive summary judgment, is not a problem in pendent jurisdiction but a glaring problem in judicial administration and in the separation of functions between judge and jury. Crew members’ maritime tort suits almost invariably urge claims under the Jones Act and under the general maritime law for breach of the duty to maintain a seaworthy vessel. These claims are legally, and generally factually, completely bound up with each other. McAl-lister v. Magnolia Petroleum Co., 357 U. S. 221; Baltimore S. S. Co. v. Phillips, 274 U. S. 316. It would be productive of extraordinary problems if the two elements of the claim are presented to different triers of fact at the same time, as would be one consequence of holding that there was no jurisdiction at law of any sort over unseaworthiness claims where diversity of citizenship was absent. Cf. Jenkins v. Roderick, 156 F. Supp. 299, 304-306. Should an advisory jury (with the same membership, doubtless, as the “mandatory” one hearing the Jones Act claim) hear the unseaworthiness claim? To what extent would its verdict bind the judge? If the judge passes on the issues himself, how to avoid overlapping damages, or contradictory findings? And what would be *414the effect of a finding of facts common to both claims made by the judge before the rendition of the jury’s'verdict, or vice versa? Would the doctrine of collateral estoppel apply? These problems arise in the wake of the Court’s rejection of jurisdiction under § 1331 and its restricted holding on any other jurisdictional basis (apart from § 1333) of Romero’s claims under the general maritime law against his employer. I cannot consider that the Court’s solution of the controversy among the lower courts that has prevailed since the Jansson dictum has shed much light on them.

III.

Since under my view there would be jurisdiction at law (the only jurisdiction Romero invoked) to consider all his claims, I arrive at the merits of his claims against his employer, Compania Trasatlántica. • As to them, I concur in the result set forth in Part II of the Court’s opinion. I also agree with the Court’s disposition of the claims against the other respondents, as set forth in Part III of its opinion.

dissenting.

Although this case has aroused much discussion about the scope of jurisdiction under 28 U. S. C. § 1331,1 cannot feel that the issue is either complex or earth-shaking. The real core of the jurisdictional controversy is whether a few more seamen can have their suits for damages passed on by federal juries instead of judges. For the reasons stated by Mr. Justice Brennan here and by Judge Magruder in Doucette v. Vincent, 194 F. 2d 834, 839, I believe that federal jurisdiction under 28 U. S. C. § 1331 lies and a federal jury trial is proper. In particular I feel that technical or esoteric readings should not be given to congressional language which is perfectly understandable in ordinary English.

Much the same reason leads me also to dissent from Part II of the Court's opinion. By its terms the Jones Act applies to “any seaman who shall suffer personal injury in the course of his employment.” 41 Stat. 1007, 46 U. S. C. § 688. (Italics added.) This Court in Lauritzen v. Larsen, 345 U. S. 571, held that the words “any seaman” did not include foreign seamen sailing foreign ships and injured in foreign waters. I dissented from that holding. It was based, I thought, on the Court’s concepts of what would be good or bad for the country *389internationally rather than on an actual interpretation of the language of the Jones Act. Thus, it seemed to me that the Lauritzen holding rested on notions of what Congress should have said, not on what it did say. Such notions, weak enough in Lauritzen, seem much weaker still in this case where the tort involved occurred in our own waters. I cannot but feel that, at least as to torts occurring within the United States, Congress knew what it was doing when it said “any seaman” and I must dissent from today’s further and, I believe, unjustifiable reduction in the scope of the Jones Act. Moreover since the tort occurred in the navigable waters of the United States, I think the complaint against Compania Trasatlántica stated a good cause of action under general maritime law whether jurisdiction of the cause is based, as I believe, on 28 U. S. C. § 1331, or, as the Court assumes, on some theory of “pendent jurisdiction.”