Petitioner-union entered into a collective bargaining agreement in 1953 with respondent-employer, the agreement to run one year and from year to year thereafter, unless terminated on specified notices. The agreement provided that there would be no strikes or work stoppages and that grievances would be handled pursuant to a specified procedure. The last step in the grievance procedure—a step that could be taken by either party—was arbitration.

This controversy involves several grievances that concern work loads and work assignments. The grievances were processed through the various steps in the grievance procedure and were finally denied by the employer. The union requested arbitration, and the employer refused. Thereupon the union brought this suit in the District Court to compel arbitration.

The District Court concluded that it had jurisdiction and ordered the employer to comply with the grievance arbitration provisions of the collective bargaining agreement. The Court of Appeals reversed by a divided vote. 230 F.2d 81. It held that, although the District Court had jurisdiction to entertain the suit, the court had no authority founded either in federal or state law to grant the relief. The case is here on a petition for a writ of certiorari which we granted because of the importance of the problem and the contrariety of views in the courts. 352 U.S. 821, 77 S.Ct. 54, 1 L.Ed.2d 46.

The starting point of our inquiry is § 301 of the Labor Management Relations Act of 1947, 61 Stat. 156, 29 U.S.C. § 185, 29 U.S.C.A. § 185, which provides:

'(a) Suits for violation of contracts between an employer and a labor organization representing employees in an industry affecting commerce as defined in this chapter, or between any such labor organizations, may be brought in any district court of the United States having jurisdiction of the parties, without respect to the amount in controversy or without regard to the citizenship of the parties.

'(b) Any labor organization which represents employees in an industry affecting commerce as defined in this chapter and any employer whose activities affect commerce as defined in this chapter shall be bound by the acts of its agents. Any such labor organization may sue or be sued as an entity and in behalf of the employees whom it represents in the courts of the United States. Any money judgment against a labor organization in a district court of the United States shall be enforceable only against the organization as an entity and against its assets, and shall not be enforceable against any individual member or his assets.'

There has been considerable litigation involving § 301 and courts have construed it differently. There is one view that § 301(a) merely gives federal district courts jurisdiction in controversies that involve labor organizations in industries affecting commerce, without regard to diversity of citizenship or the amount in controversy.1 Under that view § 301(a) would not be the source of substantive law; it would neither supply federal law to resolve these controversies nor turn the federal judges to state law for answers to the questions. Other courts—the overwhelming number of them—hold that § 301(a) is more than jurisdictional2—that it authorizes federal courts to fashion a body of federal law for the enforcement of these collective bargaining agreements and includes within that federal law specific performance of promises to arbitrate grievances under collective bargaining agreements. Perhaps the leading decision representing that point of view is the one rendered by Judge Wyzanski in Textile Workers Union of America (C.I.O.) v. American Thread Co., D.C., 113 F.Supp. 137. That is our construction of § 301(a), which means that the agreement to arbitrate grievance disputes, contained in this collective bargaining agreement, should be specifically enforced.

From the face of the Act it is apparent that § 301(a) and § 301(b) supplement one another. Section 301(b) makes it possible for a labor organization, representing employees in an industry affecting commerce, to sue and be sued as an entity in the federal courts. Section 301(b) in other words provides the procedural remedy lacking at common law. Section 301(a) certainly does something more than that. Plainly, it supplies the basis upon which the federal district courts may take jurisdiction and apply the procedural rule of § 301(b). The question is whether § 301(a) is more than jurisdictional.

The legislative history of § 301 is somewhat cloudy and confusing. But there are a few shafts of light that illuminate our problem.

The bills, as they passed the House and the Senate, contained provisions which would have made the failure to abide by an agreement to arbitrate an unfair labor practice. S.Rep. No. 105, 80th Cong., 1st Sess., pp. 20—21, 23; H.R.Rep. No. 245, 80th Cong., 1st Sess., p. 21.3 This feature of the law was dropped in Conference. As the Conference Report stated, 'Once parties have made a collective bargaining contract, the enforcement of that contract should be left to the usual processes of the law and not to the National Labor Relations Board.' H.R.Conf.Rep. No. 510, 80th Cong., 1st Sess., p. 42.

Both the Senate and the House took pains to provide for 'the usual processes of the law' by provisions which were the substantial equivalent of § 301(a) in its present form. Both the Senate Report and the House Report indicate a primary concern that unions as well as employees should be bound to collective bargaining contracts. But there was also a broader concern—a concern with a procedure for making such agreements enforceable in the courts by either party. At one point the Senate Report, supra, p. 15, states, 'We feel that the aggrieved party should also have a right of action in the Federal courts. Such a policy is completely in accord with the purpose of the Wagner Act which the Supreme Court declared was 'to compel employers to bargain collectively with their employees to the end that an employment contract, binding on both parties, should be made * * *."

Congress was also interested in promoting collective bargaining that ended with agreements not to strike.4 The Senate Report, supra, p. 16 states:

'If unions can break agreements with relative impunity, then such agreements do not tend to stabilize industrial relations. The execution of an agreement does not by itself promote industrial peace. The chief advantage which an employer can reasonably expect from a collective labor agreement is assurance of uninterrupted operation during the term of the agreement. Without some effective method of assuring freedom from economic warfare for the term of the agreement, there is little reason why an employer would desire to sign such a contract.

'Consequently, to encourage the making of agreements and to promote industrial peace through faithful performance by the parties, collective agreements affecting interstate commerce should be enforceable in the Federal courts. Our amendment would provide for suits by unions as legal entities and against unions as legal entities in the Federal courts in disputes affecting commerce.'

Thus collective bargaining contracts were made 'equally binding and enforceable on both parties.' Id., p. 15. As stated in the House Report, supra, p. 6, the new provision 'makes labor organizations equally responsible with employers for contract violations and provides for suit by either against the other in the United States district courts.' To repeat, the Senate Report, supra, p. 17, summed up the philosophy of § 301 as follows: 'Statutory recognition of the collective agreement as a valid, binding, and enforceable contract is a logical and necessary step. It will promote a higher degree of responsibility upon the parties to such agreements, and will thereby promote industrial peace.' Plainly the agreement to arbitrate grievance disputes is the quid pro quo for an agreement not to strike. Viewed in this light, the legislation does more than confer jurisdiction in the federal courts over labor organizations. It expresses a federal policy that federal courts should enforce these agreements on behalf of or against labor oganizations and that industrial peace can be best obtained only in that way.

To be sure, there is a great medley of ideas reflected in the hearings, reports, and debates on this Act. Yet, to repeat, the entire tenor of the history indicates that the agreement to arbitrate grievance disputes was considered as quid pro quo of a no-strike agreement. And when in the House the debate narrowed to the question whether § 301 was more than jurisdictional, it became abundantly clear that the purpose of the section was to provide the necessary legal remedies. Section 302 of the House bill,5 the substantial equivalent of the present § 301, was being described by Mr. Hartley, the sponsor of the bill in the House:

'Mr. Barden. Mr. Chairman, I take this time for the purpose of asking the Chairman a question, and in asking the question I want it understood that it is intended to make a part of the record that may hereafter be referred to as history of the legislation.

'It is my understanding that section 302, the section dealing with equal responsibility under collective bargaining contracts in strike actions and proceedings in district courts contemplates not only the ordinary lawsuits for damages but also such other remedial proceedings, both legal and equitable, as might be appropriate in the circumstances; in other words, proceedings could, for example, be brought by the employers, the labor organizations, or interested individual employees under the Declaratory Judgments Act in order to secure declarations from the Court of legal rights under the contract.

'Mr. Hartley. The interpretation the gentleman has just given of that section is absolutely correct.' 93 Cong.Rec. 3656 3657.

It seems, therefore, clear to us that Congress adopted a policy which placed sanctions behind agreements to arbitrate grievance disputes,6 by implication rejecting the common-law rule, discussed in Red Cross Line v. Atlantic Fruit Co., 264 U.S. 109, 44 S.Ct. 274, 68 L.Ed. 582, against enforcement of executory agreements to arbitrate.7 We would undercut the Act and defeat its policy if we read § 301 narrowly as only conferring jurisdiction over labor organizations.

The question then is, what is the substantive law to be applied in suits under § 301(a)? We conclude that the substantive law to apply in suits under § 301(a) is federal law, which the courts must fashion from the policy of our national labor laws. See Mendelsohn, Enforceability of Arbitration Agreements Under Taft-Hartley Section 301, 66 Yale L.J. 167. The Labor Management Relations Act expressly furnishes some substantive law. It points out what the parties may or may not do in certain situations. Other problems will lie in the penumbra of express statutory mandates. dates. Some will lack express statutory sanction but will be solved by looking at the policy of the legislation and fashioning a remedy that will effectuate that policy. The range of judicial inventiveness will be determined by the nature of the problem. See Board of Commissioners of Jackson County v. United States, 308 U.S. 343, 351, 60 S.Ct. 285, 288, 84 L.Ed. 313. Federal interpretation of the federal law will govern, not state law. Cf. Jerome v. United States, 318 U.S. 101, 104, 63 S.Ct. 483, 485, 87 L.Ed. 640. But state law, if compatible with the purpose of § 301, may be resorted to in order to find the rule that will best effectuate the federal policy. See Board of Commissioners of Jackson County v. United States, supra, 308 U.S. at pages 351—352, 60 S.Ct. at pages 288—289. Any state law applied, however, will be absorbed as federal law and will not be an independent source of private rights.

It is not uncommon for federal courts to fashion federal law where federal rights are concerned. See Clearfield Trust Co. v. United States, 318 U.S. 363, 366—367, 63 S.Ct. 573, 574—575, 87 L.Ed. 838; National Metropolitan Bank v. United States, 323 U.S. 454, 65 S.Ct. 354, 89 L.Ed. 383. Congress has indicated by § 301(a) the purpose to follow that course here. There is no constitutional difficulty. Article III, § 2, extends the judicial power to cases 'arising under * * * the Laws of the United States * * *.' The power of Congress to regulate these labor-management controversies under the Commerce Clause is plain. Houston East & West Texas R. Co. v. United States, 234 U.S. 342, 34 S.Ct. 833, 58 L.Ed. 1341; National Labor Relations Board v. Jones & Laughlin Corp., 301 U.S. 1, 57 S.Ct. 615, 81 L.Ed. 893. A case or controversy arising under § 301(a) is, therefore, one within the purview of judicial power as defined in Article III.

The question remains whether jurisdiction to compel arbitration of grievance disputes is withdrawn by the Norris-LaGuardia Act, 47 Stat. 70, 29 U.S.C. § 101 at seq., 29 U.S.C.A. § 101 et seq. Section 7 of that Act prescribes stiff procedural requirements for issuing an injunction in a labor dispute. The kinds of acts which had given rise to abuse of the power to enjoin are listed in § 4. The failure to arbitrate was not a part and parcel of the abuses against which the Act was aimed. Section 8 of the Norris-LaGuardia Act does, indeed, indicate a congressional policy toward settlement of labor disputes by arbitration, for it denies injunctive relief to any person who has failed to make 'every reasonable effort' to settle the dispute by negotiation, mediation, or 'voluntary arbitration.' Though a literal reading might bring the dispute within the terms of the Act (see Cox, Grievance Arbitration in the Federal Courts, 67 Harv.L.Rev. 591, 602—604), we see no justification in policy for restricting § 301(a) to damage suits, leaving specific performance of a contract to arbitrate grievance disputes to the inapposite8 procedural requirements of that Act. Moreover, we held in Virginia R. Co. v. System Federation, 300 U.S. 515, 57 S.Ct. 592, 81 L.Ed. 789, and in Graham v. Brotherhood of Firemen, 338 U.S. 232, 237, 70 S.Ct. 14, 17, 94 L.Ed. 22 that the Norris-LaGuardia Act does not deprive federal courts of jurisdiction to compel compliance with the mandates of the Railway Labor Act, 45 U.S.C.A. § 151 et seq. The mandates there involved concerned racial discrimination. Yet those decisions were not based on any peculiarities of the Railway Labor Act. We followed the same course in Syres v. Oil Workers International Union, 350 U.S. 892, 76 S.Ct. 152, 100 L.Ed. 785, which was governed by the National Labor Relations Act, 29 U.S.C.A. § 151 et seq. There an injunction was sought against racial discrimination in application of a collective bargaining agreement; and we allowed the injunction to issue. The congressional policy in favor of the enforcement of agreements to arbitrate grievance disputes being clear,9 there is no reason to submit them to the requirements of § 7 of the Norris-LaGuardia Act.

A question of mootness was raised on oral argument. It appears that since the date of the decision in the Court of Appeals respondent has terminated its operations and has contracted to sell its mill properties. All work in the mill ceased in March, 1957. Some of the grievances, however, ask for back pay for increased work loads; and the collective bargaining agreement provides that 'the Board of Arbitration shall have the right to adjust compensation retroactive to the date of the change.' Insofar as the grievances sought restoration of workloads and job assignments, the case is, of course, moot. But to the extent that they sought a monetary award, the case is a continuing controversy.

The judgment of the Court of Appeals is reversed and the cause is remanded to that court for proceedings in conformity with this opinion.

Reversed.

Mr. Justice BLACK took no part in the consideration or decision of this case.

Mr. Justice BURTON, whom Mr. Justice HARLAN joins, concurring in the result.

This suit was brought in a United States District Court under § 301 of the Labor Management Relations Act of 1947, 61 Stat. 156, 29 U.S.C. § 185, 29 U.S.C.A. § 185, seeking specific enforcement of the arbitration provisions of a collective-bargaining contract. The District Court had jurisdiction over the action since it involved an obligation running to a union—a union controversy—and not uniquely personal rights of employees sought to be enforced by a union. Cf. Association of Westinghouse Salaried Employees v. Westinghouse Elec. Corp., 348 U.S. 437, 75 S.Ct. 489, 99 L.Ed. 510. Having jurisdiction over the suit, the court was not powerless to fashion an appropriate federal remedy. The power to decree specific performance of a collectively bargained agreement to arbitrate finds its source in § 301 itself,1 and in a Federal District Court's inherent equitable powers, nurtured by a congressional policy to encourage and enforce labor arbitration in industries affecting commerce.2

I do not subscribe to the conclusion of the Court that the substantive law to be applied in a suit under § 301 is federal law. At the same time, I agree with Judge Magruder in International Brotherhood v. W. L. Mead, Inc., 1 Cir., 230 F.2d 576, that some federal rights may necessarily be involved in a § 301 case, and hence that the constitutionality of § 301 can be upheld as a congressional grant to Federal District Courts of what has been called 'protective jurisdiction.'

Mr. Justice FRANKFURTER dissenting. For dissenting opinion see 353 U.S. 448, 77 S.Ct. 923.

The Court has avoided the difficult problems raised by § 301 of the Taft-Hartley Act, 61 Stat. 156, 29 U.S.C. § 185, 29 U.S.C.A. s 185,1 by attributing to the section an occult content. This plainly procedural section is transmuted into a mandate to the federal courts to fashion a whole body of substantive federal law appropriate for the complicated and touchy problems raised by collective bargaining. I have set forth in my opinion in Association of Westinghouse Salaried Employees v. Westinghouse Electric Corp. the detailed reasons why I believe that § 301 cannot be so construed, even if constitutional questions cannot be avoided. 348 U.S. 437, 441—449, 452—459, 75 S.Ct. 489, 490—494, 496—500, 99 L.Ed. 510. But the Court has a 'clear' and contrary conclusion emerge from the 'somewhat,' to say the least, 'cloudy and confusing legislative history.' This is more than can be fairly asked even from the alchemy of construction. Since the Court relies on a few isolated statements in the legislative history which do not support its conclusion, however favoringly read, I have deemed in necessary to set forth in an appendix the entire relevant legislative history of the Taft-Hartley Act and its predecessor, the Case Bill. This legislative history reinforces the natural meaning of the statute as an exclusively procedural provision, affording, that is, an accessible federal forum for suits on agreements between labor organizations and employers, but not enacting federal law for such suits. See also Wollett and Wllington, Federalism and Breach of the Labor Agreement, 7 Stan.L.Rev. 445.

I have also set forth in my opinion in the Westinghouse case an outline of the vast problems that the Court's present decision creates by bringing into conflict state law and federal law, state courts and federal courts. 348 U.S. at pages 454—455, 75 S.Ct. at pages 497—498; see also Judge Wyzanski's opinion in Textile Workers Union of America (C.I.O.) v. American Thread Co., D.C., 113 F.Supp. 137, 140. These problems are not rendered non-existent by disregard of them. It should also be noted that whatever may be a union's ad hoc benefit in a particular case, the meaning of collective bargaining for labor does not remotely derive from reliance on the sanction of litigation in the courts. Restrictions made by legislation like the Clayton Act of 1914, 38 Stat. 738, §§ 20, 22, 29 U.S.C.A. § 52, 18 U.S.C.A. § 402, and the Norris-LaGuardia Act of 1932, 47 Stat. 70, 29 U.S.C.A. § 101 et seq., upon the use of familiar remedies theretofore available in the federal courts, reflected deep fears of the labor movement of the use of such remedies against labor. But a union, like any other combatant engaged in a particular fight, is ready to make an ally of an old enemy, and so we also find unions resorting to the otherwise much excoriated labor injunction. Such intermittent yielding to expediency does not change the fact that judicial intervention is ill-suited to the special characteristics of the arbitration process in labor disputes; nor are the conditions for its effective functioning thereby altered.

'The arbitration is an integral part of the system of self-government. And the system is designed to aid management in its quest for efficiency, to assist union leadership in its participation in the enterprise, and to secure justice for the employees. it is a means of making collective bargaining work and thus preserving private enterprise in a free government. When it works fairly well, it does not need the sanction of the law of contracts or the law of arbitration. It is only when the system breaks down completely that the courts' aid in these respects is invoked. But the courts cannot, by occasional sporadic decision, restore the parties' continuing relationship; and their intervention in such cases may seriously affect the going systems of self-government. When their autonomous system breaks down, might not the parties better be left to the usual methods for adjustment of labor disputes rather than to court actions on the contract or on the arbitration award?' Shulman, Reason, Contract, and Law in Labor Relations, 68 Harv.L.Rev. 999, 1024.

These reflections summarized the vast and extraordinarily successful experience of Dean Harry Shulman as labor arbitrator, especially as umpire under the collective-bargaining contract between the Ford Motor Co. and the UAW—CIO. (See his Opinions of the Umpire, Ford Motor Co. and UAW—CIO, 1943—1946, and the review by E. Merrick Dodd in 60 Harv.L.Rev. 486.) Arbitration agreements are for specific terms, generally much shorter than the time required for adjudication of a contested lawsuit through the available stages of trial and appeal. Renegotiation of agreements cannot await the outcome of such litigation; nor can the parties' continuing relation await it. Cases under § 301 will probably present unusual rather than representative situations. A 'rule' derived from them is more likely to discombobulate than to compose. A 'uniform corpus' cannot be expected to evolve, certainly not within a time to serve its assumed function.

The prickly and extensive problems that the supposed grant would create further counsel against a finding that the grant was made. They present hazardous opportunities for friction in the regulation of contracts between employers and unions. They involve the division of power between State and Nation, between state courts and federal courts, including the effective functioning of this Court. Wisdom suggests self-restraint in undertaking to solve these problems unless the Court is clearly directed to do so. Section 301 is not such a direction. The legislative history contains no suggestion that these problems were considered; the terms of the section do not present them.

One word more remains to be said. The earliest declaration of unconstitutionality of an act of Congress—by the Justices on circuit—involved a refusal by the Justices to perform a function imposed upon them by Congress because of the non-judicial nature of that function. Hayburn's Case, 2 Dall. 409, 1 L.Ed. 436. Since then, the Court has many times declared legislation unconstitutional because it imposed on the Court powers or functions that were regarded as outside the scope of the 'judicial power' lodged in the Court by the Constitution. See, e.g., Marbury v. Madison, 1 Cranch 137, 2 L.Ed. 60; United States v. Ferreira, 13 How. 40, 14 L.Ed. 42; Muskrat v. United States, 219 U.S. 346, 31 S.Ct. 250, 55 L.Ed. 246; Keller v. Potomac Electric Power Co., 261 U.S. 428, 43 S.Ct. 445, 67 L.Ed. 731.

One may fairly generalize from these instances that the Court has deemed itself peculiarly qualified, with due regard to the contrary judgment of Congress, to determine what is meet and fit for the exercise of 'judicial power' as authorized by the Constitution. Solicitude and respect for the confines of 'judicial power,' and the difficult problem of marking those confines, apply equally in construing precisely what duties Congress has cast upon the federal courts, especially when, as in this case, the most that can be said in support of finding a congressional desire to impose these 'legislative' duties on the federal courts is that Congress did not mention the problem in the statute and that, insofar as purpose may be gathered from congressional reports and debates, they leave us in the dark.

The Court, however, sees no problem of 'judicial power' in casting upon the federal courts, with no guides except 'judicial inventiveness,' the task of applying a whole industrial code that is as yet in the bosom of the jidiciary. There are severe limits on 'judicial inventiveness' even for the most imaginative judges. The law is not a 'brooding omnipresence in the sky,' (Mr. Justice Holmes, dissenting, in Southern Pacific Co. v. Jensen, 244 U.S. 205, 222, 37 S.Ct. 524, 531, 61 L.Ed. 1086), and it cannot be drawn from there like nitrogen from the air. These problems created by the Court's interpretation of § 301 cannot 'be solved by resort to the established canons of construction that enable a court to look through awkward or clumsy expression, or language wanting in precision, to the intent of the legislature. For the vice of the statute here lies in the impossibility of ascertaining, by any reasonable test, that the legislature meant one thing rather than another * * *.' Connally v. General Construction Co., 269 U.S. 385, 394, 46 S.Ct. 126, 128, 70 L.Ed. 322. But the Court makes § 301 a mountain instead of a molehill and, by giving an example of 'judicial inventiveness,' it thereby solves all the constitutional problems that would otherwise have to be faced.

Even on the Court's attribution to § 301 of a direction to the federal courts to fashion, out of bits and pieces elsewhere to be gathered, a federal common law of labor contracts, it still does not follow that Congress has enacted that an agreement to arbitrate industrial differences be specifically enforceable in the federal courts. On the contrary, the body of relevant federal law precludes such enforcement of arbitration clauses in collective-bargaining agreements.

Prior to 1925, the doctrine that executory agreements to arbitrate any kind of dispute would not be specifically enforced still held sway in the federal courts. See, e.g., Judge Hough's opinion in United States Asphalt Refining Co. v. Trinidad Lake Petroleum Co., D.C., 222 F. 1006; Judge Mack's opinion in Atlantic Fruit Co. v. Red Cross Line, D.C., 276 F. 319; and Mr. Justice Brandeis' opinion in Red Cross Line v. Atlantic Fruit Co., 264 U.S. 109, 123, 125, 44 S.Ct. 274, 276, 277, 68 L.Ed. 582. Legislation was deemed necessary to assure such power to the federal courts. In 1925, Congress passed the United States Arbitration Act, 9 U.S.C. § 1 et seq., 9 U.S.C.A. § 1 et seq., making executory agreements to arbitrate specifically enforceable in the federal courts, but explicitly excluding 'contracts of employment' of workers engaged in interstate commerce from its scope. Naturally enough, I find rejection, though not explicit, of the availability of the Federal Arbitration Act to enforce arbitration clauses in collective-bargaining agreements in the silent treatment given that Act by the Court's opinion. If an Act that authorizes the federal courts to enforce arbitration provisions in contracts generally, but specifically denies authority to decree that remedy for 'contracts of employment,' were available, the Court would hardly spin such power out of the empty darkness of § 301. I would make this rejection explicit, recognizing that when Congress passed legislation to enable arbitration agreements to be enforced by the federal courts, it saw fit to exclude this remedy with respect to labor contracts. See Amalgamated Association, etc. v. Pennsylvania Greyhound Lines, 3 Cir., 192 F.2d 310; United Electrical, Radio & Machine Workers of America v. Miller Metal Products, Inc., 4 Cir., 215 F.2d 221; Lincoln Mills of Alabama v. Textile Workers Union, 5 Cir., 230 F.2d 81; United Steelworkers of America, C.I.O. v. Galland-Henning Mfg. Co., 7 Cir., 241 F.2d 323; and the legislative history set forth by the parties in the present cases. Congress heeded the resistance of organized labor, uncompromisingly led in its hostility to this measure by Andrew Furuseth, president of the International Seamen's Union and most powerful voice expressing labor's fear of the use of this remedy against it.2

Even though the Court glaringly ignores the Arbitration Act, it does at least recognize the common-law rule against enforcement of executory agreements to arbitrate. It nevertheless enforces the arbitration clause in the collective-bargaining agreements in these cases. It does so because it finds that Congress 'by implication' rejected the common-law rule. I would add that the Court, in thus deriving power from the unrevealing words of the Taft-Hartley Act, has also found that Congress 'by implication' repealed its own statutory exemption of collective-bargaining agreements in the Arbitration Act, an exemption made as we have seen for well-defined reasons of policy.

The Court of Appeals for the First Circuit, which reached the conclusion that arbitration clauses in collective-bargaining agreements were enforceable under the Arbitration Act, nevertheless found that such clauses would not have been enforceable by virtue of § 301:

'A number of courts have held that § 301 itself is a legislative authorization for decrees of specific performance of arbitration agreements. * * * We think that is reading too much into the very general language of § 301. The terms and legislative history of § 301 sufficiently demonstrate, in our view, that it was not intended either to create any new remedies or to deny applicable existing remedies. See H.R.Rep.No.245, 80th Cong., 1st Sess. 46 (1947); H.R.Rep.No.510 (Conference Report), 80th Cong., 1st Sess. 42 (1947); 93 Cong.Rec. 3734, 6540 (daily ed. 1947). Arbitration was scarcely mentioned at all in the legislative history. Furthermore, the same practical consideration that militates against judicial overruling of the common law doctrine applies against interpreting § 301 to give that effect. The most that could be read into it would be that it authorizes equitable remedies in general, including decrees for specific performance of an arbitration agreement. Lacking are the procedural specifications needed for administration of the power to compel arbitration. * * * Thus it seems to us that a firmer statutory basis than § 301 should be found to justify departure from the judicially formulated doctrines with reference to arbitration agreements.' Local 205, United Electrical, Radio and Mach. Workers of America (UE) v. General Electric Co., 233 F.2d 85, 96—97.

I would put the conclusion even more strongly because, contrary to the view of the Court of Appeals for the First Circuit, the rule that is departed from 'by implication' had not only been 'judicially formulated' but had purposefully been congressionally formulated in the Arbitration Act of 1925. And it is being departed from on the tenuous basis of the legislative history of § 301, for which the utmost that can be claimed is that insofar as there was any expectation at all, it was only that conventional remedies, including equitable remedies, would be available. But of course, as we have seen, 'equitable remedies' in the federal courts had traditionally excluded specific performance of arbitration clauses, except as explicitly provided by the 1925 Act. Thus, even assuming that § 301 contains directions for some federal substantive law of labor contracts, I see no justification for translating the vague expectation concerning the remedies to be applied into an overruling of previous federal common law and, more particularly, into the repeal of the previous congressional exemption of collective-bargaining agreements from the class of agreements in which arbitration clauses were to be enforced.

The second ground of my dissent from the Court's action is more fundamental.3 Since I do not agree with the Court's conclusion that federal substantive law is to govern in actions under § 301, I am forced to consider the serious constitutional question that was adumbrated in the Westinghouse case, 348 U.S. at 449—452, 75 S.Ct. 494—496, the constitutionality of a grant of jurisdiction to federal courts over contracts that came into being entirely by virtue of state substantive law, a jurisdiction not based on diversity of citizenship, yet one in which a federal court would, as in diversity cases, act in effect merely as another court of the State in which it sits. The scope of allowable federal judicial power that this grant must satisfy is constitutionally described as '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.' Art. III, § 2. While interpretive decisions are legion under general statutory grants of jurisdiction strikingly similar to this constitutional wording, it is generally recognized that the full constitutional power has not been exhausted by these statutes. See, e.g., Mishkin, The Federal 'Question' in the District Courts, 53 Col.L.Rev. 157, 160; Shulman and Jaegerman, Some Jurisdictional Limitations on Federal Procedure, 45 Yale L.J. 393, 405, n. 47; Wechsler, Federal Jurisdiction and the Revision of the Judicial Code, 13 Law & Contemp. Prob., 216, 224—225.

Almost without exception, decisions under the general statutory grants have tested jurisdiction in terms of the presence, as an integral part of plaintiff's cause of action, of an issue calling for interpretation or application of federal law. E.g., Gully v. First National Bank, 299 U.S. 109, 57 S.Ct. 96, 81 L.Ed. 70. Although it has sometimes been suggested that the 'cause of action' must derive from federal law, see American Well Works Co. v. Layne and Bowler Co., 241 U.S. 257, 260, 36 S.Ct. 585, 586, 60 L.Ed. 987, it has been found sufficient that some aspect of federal law is essential to plaintiff's success. Smith v. Kansas City Title & Trust Co., 255 U.S. 180, 41 S.Ct. 243, 65 L.Ed. 577. The litigation-provoking problem has been the degree to which federal law must be in the forefront of the case and not collateral, peripheral or remote.

In a few exceptional cases, arising under special jurisdictional grants, the criteria by which the prominence of the federal question is measured against constitutional requirements have been found satisfied under circumstances suggesting a variant theory of the nature of these requirements. The first, and the leading case in the field, is Osborn v. Bank of United States, 9 Wheat. 738, 6 L.Ed. 204. There, Chief Justice Marshall sustained federal jurisdiction in a situation—hypothetical in the case before him but presented by the companion case of Bank of United States v. Planters' Bank, 9 Wheat. 904, 6 L.Ed. 244—involving suit by a federally incorporated bank upon a contract. Despite the assumption that the cause of action and the interpretation of the contract would be governed by state law, the case was found to 'arise under the laws of the United States' because the propriety and scope of a federally granted authority to enter into contracts and to litigate might well be challenged. This reasoning was subsequently applied to sustain jurisdiction in actions against federally chartered railroad corporations. Pacific Railroad Removal Cases (Union Pac. Ry. Co. v. Myers), 115 U.S. 1, 5 S.Ct. 1113, 29 L.Ed. 319. The traditional interpretation of this series of cases is that federal jurisdiction under the 'arising' clause of the Constitution, though limited to cases involving potential federal questions, has such flexibility that Congress may confer it whenever there exists in the background some federal proposition that might be challenged, despite the remoteness of the likelihood of actual presentation of such a federal question.4

The views expressed in Osborn and the Pacific Railroad Removal Cases were severely restricted in construing general grants of jurisdiction. But the Court later sustained this jurisdictional section of the Bankruptcy Act of 1898:

'The United States district courts shall have jurisdiction of all controversies at law and in equity, as distinguished from proceedings in bankruptcy, between trustees as such and adverse claimants concerning the property acquired or claimed by the trustees, in the same manner and to the same extent only as though bankruptcy proceedings had not been instituted and such controversies had been between the bankrupts and such adverse claimants.' § 23, sub. a, as amended, 44 Stat. 664, 11 U.S.C.A. § 46, sub. a.

Under this provision the trustee could pursue in a federal court a private cause of action arising under and wholly governed by state law. Schumacher v. Beeler, 293 U.S. 367, 55 S.Ct. 230, 79 L.Ed. 433; Williams v. Austrian, 331 U.S. 642, 67 S.Ct. 1443, 91 L.Ed. 1718 (Chandler Act of 1938, 52 Stat. 840, 11 U.S.C.A. § 1 et seq.). To be sure, the cases did not discuss the basis of jurisdiction. It has been suggested that they merely represent an extension of the approach of the Osborn case; the trustee's right to sue might be challenged on obviously federal grounds—absence of bankruptcy or irregularity of the trustee's appointment or of the bankruptcy proceedings. National Mutual Ins. Co. of Dist. of Col. v. Tidewater Transfer Co., 337 U.S. 582, 611—613, 69 S.Ct. 1173, 1187—1188, 93 L.Ed. 1556 (Rutledge, J., concurring). So viewed, this type of litigation implicates a potential federal question.

Apparently relying on the extent to which the bankruptcy cases involve only remotely a federal question, Mr. Justice Jackson concluded in National Mutual Insurance Co. of Dist. of Col. v. Tidewater Transfer Co., 337 U.S. 582, 69 S.Ct. 1173, 93 L.Ed. 1556, that Congress may confer jurisdiction on the District Courts as incidental to its powers under Article I. No attempt was made to reconcile this view with the restrictions of Article III; a majority of the Court recognized that Article III defined the bounds of valid jurisdictional legislation and rejected the notion that jurisdictional grants can go outside these limits.

With this background, many theories have been proposed to sustain the constitutional validity of § 301. In Textile Workers Union of America (C.I.O.) v. American Thread Co., D.C., 113 F.Supp. 137, 140, Judge Wyzanski suggested, among other possibilities, that § 301 might be read as containing a direction that controversies affecting interstate commerce should be governed by federal law incorporating state law by reference, and that such controversies would then arise under a valid federal law as required by Article III. Whatever may be said of the assumption regarding the validity of federal jurisdiction under an affirmative declaration by Congress that state law should be applied as federal law by federal courts to contract disputes affecting commerce, we cannot argumentatively legislate for Congress when Congress has failed to legislate. To do so disrespects legislative responsibility and disregards judicial limitations.

Another theory, relying on Osborn and the bankruptcy cases, has been proposed which would achieve results similar to those attainable under Mr. Justice Jackson's view, but which purports to respect the 'arising' clause of Article III. See Hart and Wechsler, The Federal Courts and the Federal System, pp. 744—747; Wechsler, Federal Jurisdiction and the Revision of the Judicial Code, 13 Law & Contemp.Prob. 216, 224—225; International Brotherhood of Teamsters, etc. v. W. L. Mead, Inc., 1 Cir., 230 F.2d 576. Called 'protective jurisdiction,' the suggestion is that in any case for which Congress has the constitutional power to prescribe federal rules of decision and thus confer 'true' federal question jurisdiction, it may, without so doing, enact a jurisdictional statute, which will provide a federal forum for the application of state statute and decisional law. Analysis of the 'protective jurisdiction' theory might also be attempted in terms of the language of Article III—construing 'laws' to include jurisdictional statutes where Congress could have legislated substantively in a field. This is but another way of saying that because Congress could have legislated substantively and thereby could give rise to litigation under a statute of the United States, it can provide a federal forum for statecreated rights although it chose not to adopt state law as federal law or to originate federal rights.

Surely the truly technical restrictions of Article III are not met or respected by a beguiling phrase that the greater power here must necessarily include the lesser. In the compromise of federal and state interests leading to distribution of jealously guarded judicial power in a federal system, see 13 Cornell L.Q. 499, it is obvious that very different considerations apply to cases involving questions of federal law and those turning solely on state law. It may be that the ambiguity of the phrase 'arising under the laws of the United States' leaves room for more than traditional theory could accommodate. But, under the theory of 'protective jurisdiction,' the 'arising under' jurisdiction of the federal courts would be vastly extended. For example, every contract or tort arising out of a contract affecting commerce might be a potential cause of action in the federal courts, even though only state law was involved in the decision of the case. At least in Osborn and the bankruptcy cases, a substantive federal law was present somewhere in the background. See 353 U.S. 470—472, 77 S.Ct. 928, 929, supra, and 353 U.S. 480—484, 77 S.Ct. 933—936, infra. But this theory rests on the supposition that Congress could enact substantive federal law to govern the particular case. It was not held in those cases, nor is it clear, that federal law could be held to govern the transactions of all persons who subsequently become bankrupt, or of all suits of a Bank of the United States. See Mishkin, The Federal 'Question' in the District Courts, 53 Col.L.Rev. 157, 189.

'Protective jurisdiction,' once the label is discarded, cannot be justified under any view of the allowable scope to be given to Article III. 'Protective jurisdiction' is a misused label for the statute we are here considering. That rubric is properly descriptive of safeguarding some of the indisputable, staple business of the federal courts. It is a radiation of an existing jurisdiction. See Adams v. United States ex rel. McCann, 371 U.S. 269, 63 S.Ct. 236, 87 L.Ed. 268; 28 U.S.C. § 2283, 28 U.S.C.A. § 2283. 'Protective jurisdiction' cannot generate an independent source for adjudication outside of the Article III sanctions and what Congress has defined. The theory must have as its sole justification a belief in the inadequacy of state tribunals in determining state law. The Constitution reflects such a belief in the specific situation within which the Diversity Clause was confined. The intention to remedy such supposed defects was exhausted in this provision of Article III.5 That this 'protective' theory was not adopted by Chief Justice Marshall at a time when conditions might have presented more substantial justification strongly suggests its lack of constitutional merit. Moreover, Congress in its consideration of § 301 nowhere suggested dissatisfaction with the ability of state courts to administer state law properly. Its concern was to provide access to the federal courts for easier enforcement of state-created rights.

Another theory also relies on Osborn and the bankruptcy cases as an implicit recognition of the propriety of the exercise of some sort of 'protective jurisdiction' by the federal courts. Mishkin, op. cit. supra, 53 Col.L.Rev. 157, 184 et seq. Professor Mishkin tends to view the assertion of such a jurisdiction, in the absence of any exercise of substantive powers, as irreconcilable with the 'arising' clause since the case would then arise only under the jurisdictional statute itself, and he is reluctant to find a constitutional basis for the grant of power outside Article III. Professor Mishkin also notes that the only purpose of such a statute would be to insure impartiality to some litigant, an objection inconsistent with Article III's recognition of 'protective jurisdiction' only in the specified situation of diverse citizenship. But where Congress has 'an articulated and active federal policy regulating a field, the 'arising under' clause of Article III apparently permits the conferring of jurisdiction on the national courts of all cases in the area including those substantively governed by state law.' Id., at 192. In such cases, the protection being offered is not to the suitor, as in diversity cases, but to the 'congressional legislative program.' Thus he supports § 301: 'even though the rules governing collective bargaining agreements continue to be state-fashioned, nonetheless the mode of their application and enforcement may play a very substantial part in the labor-management relations of interstate industry and commerce—an area in which the national government has labored long and hard.' Id., at 196.

Insofar as state law governs the case, Professor Mishkin's theory is quite similar to that advanced by Professors Hart and Wechsler and followed by the Court of Appeals for the First Circuit: The substantive power of Congress, although not exercised to govern the particular 'case,' gives 'arising under' jurisdiction to the federal courts despite governing state law. The second 'protective jurisdiction' theory has the dubious advantage of limiting incursions on state judicial power to situations in which the State's feelings may have been tempered by early substantive federal invasions.

Professor Mishkin's theory of 'protective jurisdiction' may find more constitutional justification if there is not merely an 'articulated and active' congressional policy regulating the labor field but also federal rights existing in the interstices of actions under § 301. See Wollett and Wellington, Federalism and Breach of the Labor Agreement, 7 Stan.L.Rev. 445, 475—479. Therefore, before resting on an interpretation of § 301 that would compel a declaration of unconstitutionality, we must, as was stated in Westinghouse, defer to the strong presumption even as to such technical matters as federal jurisdiction—that Congress legislated in accordance with the Constitution. The difficult nature of the problem of construction to be faced if some federal rights are sought was set forth in Westinghouse, where the constitutional questions were involved only in their bearing on the construction of the statute. Now that the constitutional questions themselves must be faced, the nature of the problem bears repeating.

Legislation must, if possible, be given a meaning that will enable it to survive. This rule of constitutional adjudication is normally invoked to narrow what would otherwise be the natural but constitutionally dubious scope of the language. E.g., United States v. Delaware & Hudson Co., 213 U.S. 366, 29 S.Ct. 527, 53 L.Ed. 836; United States v. Jin Fuey Moy, 241 U.S. 394, 401, 36 S.Ct. 658, 659, 60 L.Ed. 1061; United States v. Rumely, 345 U.S. 41, 73 S.Ct. 543, 97 L.Ed. 770. Here the endeavor of some lower courts and of this Court has resulted in adding to the section substantive congressional regulation even though Congress saw fit not to exercise such power or to give the courts any concrete guidance for defining such regulation.

To be sure, the full scope of a substantive regulation is frequently in dispute and must await authoritative determination by courts. Congress declares its purpose imperfectly or partially, and compatible judicial construction completes it. But in this case we start with a provision that is wholly jurisdictional and as such bristles with constitutional problems under Article III. To avoid them, interpolation of substantive regulation has been proposed. From what materials are we to draw a determination that § 301 is something other than what it declares itself? Is the Court justified in creating all the difficult problems of choice within a sphere of delicate policy without any direction from Congress and merely for the sake of giving effect to a provision that seems to deal with a different subject? The somewhat Delphic wisdom of Mr. Justice Cardozo, speaking for the whole Court, pulls us here in the opposite direction: 'We think the light is so strong as to flood whatever places in the statute might otherwise be dark. Courts have striven mightily at times to canalize construction along the path of safety. * * * When a statute is reasonably susceptible of two interpretations, they have preferred the meaning that preserves to the meaning that destroys. * * * 'But avoidance of a defficulty will not be pressed to the point of disingenuous evasion.' * * * 'Here the intention of the Congress is revealed too distinctly to permit us to ignore it because of mere misgivings as to power." Hopkins Federal Savings & Loan Ass'n v. Cleary, 296 U.S. 315, 334—335, 56 S.Ct. 235, 239—240, 90 L.Ed. 251.

Assuming, however, that we would be justified in pouring substantive content into a merely procedural vehicle, what elements of federal law could reasonably be put into the provisions of § 301? The suggestion that the section permits the federal courts to work out, without more, a federal code governing collective-bargaining contracts must, for reasons that have already been stated, be rejected. Likewise the suggestion that § 301 may be viewed as a congressional authorization to the federal courts to work out a concept of the nature of the collective-bargaining contract, leaving detailed questions of interpretation to state law. See 348 U.S. at pages 455—459, 75 S.Ct. at pages 498—500.

Nor will Congress' objective be furthered by an attempt to limit the grant of a federal forum to certain types of actions between unions and employers. It would be difficult to find any basis for, or principles of, selection, either in the terms of § 301 or in considerations relevant to promotion of stability in labor relations. It is true that a fair reading of § 301 in the context of its enactment shows that the suit that Congress primarily contemplated was the suit against a union for strike in violation of contract. From this it might be possible to imply a federal right to bring an action for damages based on such an event. In the interest of mutuality, so close to the heart of Congress, we might in turn find a federal right in the union to sue for a lockout in violation of contract. But neither federal right would be involved in the present cases. Moreover, it bears repetition that Congress chose not to make this the basis of federal law, i.e., it chose not to make such conduct an unfair labor practice.

There is a point, however, at which the search may be ended with less misgiving regarding the propriety of judicial infusion of substantive provisions into § 301. The contribution of federal law might consist in postulating the right of a union, despite its amorphous status as an unincorporated association, to enter into binding collective-bargaining contracts with an employer. The federal courts might also give sanction to this right by refusing to comply with any state law that does not admit that collective bargaining may result in an enforceable contract. It is hard to see what serious federal-state conflicts could arise under this view. At most, a state court might dismiss the action, while a federal court would entertain it. Moreover, such a function of federal law is closely related to the removal of the procedural barriers to suit. Section 301 would be futile if the union's status as a contracting party were not recognized. The statement in § 301(b) that the acts of the agents of the union are to be regarded as binding upon the union may be used in support of this conclusion. This provision, not confined in its application to suits in the District Court under § 301(a), was primarily directed to responsibility of the union for its agents' actions in authorizing strikes or committing torts. It can be construed, however, as applicable to the formation of a contract. So applied, it would imply that a union must be regarded as contractually bound by the acts of its agents, which in turn presupposes that the union is capable of contract relations.

Of course, the possibility of a State's law being counter to such a limited federal proposition is hypothetical, and to base an assertion of federal law on such a possibility, one never considered by Congress, is an artifice. And were a State ever to adopt a contrary attitude, its reasons for so doing might be such that Congress would not be willing to disregard them. But these difficulties are inherent in any attempt to expand § 301 substantively to meet constitutional requirements.

Even if this limited federal 'right' were read into § 301, a serious constitutional question would still be present. It does elevate the situation to one closely analogous to that presented in Osborn v. Bank of United States, 9 Wheat. 738, 6 L.Ed. 204.6 Section 301 would, under this view, imply that a union is to be viewed as a juristic entity for purposes of acquiring contract rights under a collective-bargaining agreement, and that it has the right to enter into such a contract and to sue upon it. This was all that was immediately and expressly involved in the Osborn case, although the historical setting was vastly different and the juristic entity in that case was completely the creature of federal law, one engaged in carrying out essential governmental functions. Most of these special considerations had disappeared, however, at the time and in the circumstances of the decision of the Pacific Railroad Removal Cases (Union Pac. R. Co. v. Myers), 115 U.S. 1, see page 11, 5 S.Ct. 1113, page 1117, 29 L.Ed. 319, supra. There is force in the view that regards the latter as a 'sport' and finds that the Court has so viewed it. See Mishkin, 53 Col.L.Rev., at 160, n. 24, citing Gully v. First National Bank, 299 U.S. 109, 113—114, 57 S.Ct. 96, 98, 81 L.Ed. 70 ('Only recently we said after full consideration that the doctrine of the charter cases was to be treated as exceptional, though within their special field there was no thought to disturb them.'), and People of Puerto Rico v. Russell & Co., 288 U.S. 476, 485, 53 S.Ct. 447, 450, 77 L.Ed. 903; see also Mr. Justice Holmes, in Smith v. Kansas City Title & Trust Co., 255 U.S. 180, 214—215, 41 S.Ct. 243, 250, 65 L.Ed. 577 (dissenting opinion). The question is whether we should now so consider it and refuse to apply its holding to the present situation.

I believe that we should not extend the precedents of Osborn and the Pacific Railroad Removal Cases to this case even though there be some elements of analytical similarity. Osborn, the foundation for the Removal Cases, appears to have been based on premises that today, viewed in the light of the jurisdictional philosophy of Gully v. First National Bank, supra, are subject to criticism. The basic premise was that every case in which a federal question might arise must be capable of being commenced in the federal courts, and when so commenced it might, because jurisdiction must be judged as the outset, be concluded there despite the fact that the federal question was never raised. Marshall's holding was undoubtedly influenced by his fear that the bank might suffer hostile treatment in the state courts that could not be remedied by an appeal on an isolated federal question. There is nothing in Article III that affirmatively supports the view that original jurisdiction over cases involving federal questions must extend to every case in which there is the potentiality of appellate jurisdiction. We also have become familiar with removal procedures that could be adapted to alleviate any remaining fears by providing for removal to a federal court whenever a federal question was raised. In view of these developments, we would not be justified in perpetuating a principle that permits assertion of original federal jurisdiction on the remote possibility of presentation of a federal question. Indeed, Congress, by largely withdrawing the jurisdiction that the Pacific Railroad Removal Cases recognized, and this Court, by refusing to perpetuate it under general grants of jurisdiction, see Gully v. First National Bank, supra, have already done much to recognize the changed atmosphere.

Analysis of the bankruptcy power also reveals a superficial analogy to § 301. The trustee enforces a cause of action acquired under state law by the bankrupt. Federal law merely provides for the appointment of the trustee, vests the cause of action in him, and confers jurisdiction on the federal courts. Section 301 similarly takes the rights and liabilities which under state law are vested distributively in the individual members of a union and vests them in the union for purposes of actions in federal courts, wherein the unions are authorized to sue and be sued as an entity. While the authority of the trustee depends on the existence of a bankrupt and on the propriety of the proceedings leading to the trustee's appointment, both of which depend on federal law, there are similar federal propositions that may be essential to an action under § 301. Thus, the validity of the contract may in any case be challenged on the ground that the labor organization negotiating it was not the representative of the employees concerned, a question that has been held to be federal, La Crosse Telephone Corp. v. Wisconsin Employment Relations Board, 336 U.S. 18, 69 S.Ct. 379, 93 L.Ed. 463, or on the ground that subsequent change in the representative status of the union has affected the continued validity of the agreement. Perhaps also the qualifications imposed on a union's right to utilize the facilities of the National Labor Relations Board, dependent on the filing of non-Communist affidavits required by § 9(h), 29 U.S.C.A. § 159(h), and the information and reports required by § 9(f) and (g), might be read as restrictions on the right of the union to sue under § 301, again providing a federal basis for challenge to the union's authority. Consequently, were the bankruptcy cases to be viewed as dependent solely on the background existence of federal questions, there would be little analytical basis for distinguishing actions under § 301. But the bankruptcy decisions may be justified by the scope of the bankruptcy power, which may be deemed to sweep within its scope interests analytically outside the 'federal question' category, but sufficiently related to the main purpose of bankruptcy to call for comprehensive treatment. See National Mutual Ins. Co. of Dist. of Col. v. Tidewater Transfer Co., 337 U.S. 582, 652, Note 3, 69 S.Ct. 1173, 1198, 93 L.Ed. 1556 (concurring in part, dissenting in part). Also, although a particular suit may be brought by a trustee in a district other than the one in which the principal proceedings are pending, if all the suits by the trustee, even though in many federal courts, are regarded as one litigation for the collection and apportionment of the bankrupt's property, a particular suit by the trustee, under state law, to recover a specific piece of property might be analogized to the ancillary or pendent jurisdiction cases in which, in the disposition of a cause of action, federal courts may pass on state grounds for recovery that are joined to federal grounds. See Hurn v. Oursler, 289 U.S. 238, 53 S.Ct. 586, 77 L.Ed. 1148; Siler v. Louisville & Nashville, R. Co., 213 U.S. 175, 29 S.Ct. 451, 53 L.Ed. 753; but see Mishkin, 53 Col.L.Rev., at 194, n. 161.

If there is in the phrase 'arising under the laws of the United States' leeway for expansion of our concepts of jurisdiction, the history or Article III suggests that the area is not great and that it will require the presence of some substantial federal interest, one of greater weight and dignity than questionable doubt concerning the effectiveness of state procedure. The bankruptcy cases might possibly be viewed as such an expansion. But even so, not merely convenient judicial administration but the whole purpose of the congressional legislative program—conservation and equitable distribution of the bankrupt's estate in carrying out the constitutional power over bankruptcy—required the availability of federal jurisdiction to avoid expense and delay. Nothing pertaining to § 301 suggests vesting the federal courts with sweeping power under the Commerce Clause comparable to that vested in the federal courts under the bankruptcy power.

In the wise distribution of governmental powers, this Court cannot do what a President sometimes does in returning a bill to Congress. We cannot return this provision to Congress and respectfully request that body to face the responsibility placed upon it by the Constitution to define the jurisdiction of the lower courts with some particularity and not to leave these courts at large. Confronted as I am, I regretfully have no choice. For all the reasons elaborated in this dissent, even reading into § 301 the limited federal rights consistent with the purposes of that section, I am impelled to the view that it is unconstitutional in cases such as the present ones where it provides the sole basis for exercise of jurisdiction by the federal courts.7

I. The Case Bill (H.R. 4908, 79th Cong., 2d Sess.)

(The Federal Mediation Act of 1946.)

1. Hearings before the Committee on Labor on H.R. 4908, 79th Cong., 1st Sess.:

a. H.R. 4908, as considered by committee, provided for fact-finding boards. It had no provision concerning suits on collective-bargaining contracts.

b. During these hearings, there was, however, some concern with breach of such contracts. Despite the filing of two memoranda detailing the problems of enforcement of agreements against a union (pp. 89, 96) there was no elucidation of the problem. The prevalence of violation was noted and the desire to do something to promote enforceability expressed. (Pp. 15, 27, 28, 38, 41, 68, 73, 84, 88, 97, 101, 113.)

2. The House Report contained no comment on the problem. (H.R.Rep. No. 1493, 79th Cong., 2d Sess.)

3. The bill, as introduced on the House floor (92 Cong.Rec. 765):

'Sec. 10. Binding effect of collective-bargaining contracts: All collective-bargaining contracts shall be mutually and equally binding and enforceable either at law or in equity against each of the parties thereto, any other law to the contrary notwithstanding. In the event of a breach of any such contract or of any agreement contained in such contract by either party thereto, then, in addition to any other remedy or remedies existing either in law or equity, a suit for damages for such breach or for injunctive relief in equity may be maintained by the other party or parties in any United States district court having jurisdiction of the parties. If the defendant against whom action is sought to be commenced and maintained is a labor organization, such action may be filed in the United States district court of any district wherein any officer of such labor organization resides or may be found.'

a. General comment on the desirability of mutual enforceability of contracts: 92 Cong.Rec. 662, 668, 677, 679, 684, 686, 753, 767.

b. Representative Francis Case's only comments were not pertinent. Id., at 680, 765.

c. Representative Vorys, in offering corrective amendments to Section 10, stated:

'We do create, if there is any doubt about its present existence, an action for damages for breach of contract against a labor organization or an employer, which means that either party, the labor organization or the employer, may have the benefit of a trial by jury in any such action.

'* * * Since we are attempting to create no new right in the equity side, there is no reason to refer to the equity side * * *.

'* * * It will take away any particular benefits or advantages of one party or the other that now exist under other laws which keep the obligations from being equal and mutual; will not give any new rights by way of injunction to either party but will specifically provide for an action at law for damages to enforce any act of violation of the contracts.' (Id., at 853.) d. Representative Thom, opposing Section 10 as an incursion on States' rights, appears to have been the only member to have felt that collective bargaining contracts already were enforceable in state and federal courts. Id., at 847.

5. As it finally passed the House, Section 10 read:

'Sec. 10. Binding effect of collective-bargaining contracts: All collective-bargaining contracts shall be mutually and equally binding and enforceable against each of the parties thereto, any other law to the contrary notwithstanding. In the event of a breach of any such contract or of any agreement contained in such contract by either party thereto, then, in addition to any other remedy or remedies existing, a suit for damages for such breach may be maintained by the other party or parties in any State or United States district court having jurisdiction of the parties.'

1. Hearings before Senate Committee on Education and Labor on S. 1661:

a. Hearings were held on a companion bill to the fact-finding bill on which House hearings were held. The Case bill had not yet passed the House.

b. As in the House, however, concern was expressed over a general impression that unions were not subject to suits for damages for breach of contract to the same extent as employers. (Pp. 138, 168, 354, 383, 400, 554, 623, 662, 740.) For the first time, however, oral testimony directed the legislators to the primary source of the problem. This testimony, with a supporting memorandum, indicated that the problem lay in the status of the union as an unincorporated association. This memorandum, however, also pointed out that in some jurisdictions the union was viewed as acting as agent of the individual employees in negotiating the collective agreement, and thus was not viewed as having, even in theory, any rights or obligations on the contract. (P. 411.)

2. Hearings before a Subcommittee of the Senate Committee on Education and Labor on H.R. 4908 (as it had passed the House):

a. 'Mr. Case. Section 10, opening the miscellaneous provisions of the bill, is very brief, and I would like to read it because I think it speaks for itself.

'It almost would seem unnecessary to say that contracts entered into between two parties call for mutual obligations and mutual observance. That is implicit in all contracts, whether expressed or not, by statutory provisions saying they are equally binding and enforceable; but because of some interpretations or some theories that they are not binding where labor organizations are involved, I thought in harmony with what the President said here—recognizing a practical situation for which we have to find methods not only of peaceful negotiation but also of insuring that the contracts once made must be lived up to—we should have a section in the bill on that subject.

'This section was modified somewhat in the consideration in the House. Originally I think we had in a provision authorizing restraining orders, but that was eliminated * * * with the consent of myself and others who had been studying the bill, with the thought that this possibly met the situation by authorizing a suit for damages for breach of contract.

'Senator Taft. Mr. Case, there are one or two minor questions on that. It says:

"All collective-bargaining contracts shall be * * * enforceable by a suit in a State or a United States district, court having jurisdiction of the parties.'

'Would you intend to give jurisdiction to the United States district court in purely local collective-bargaining contracts not dealing with interstate commerce? Ought not that be limited in some way?

'Mr. Case. That may be. There are other places in the bill where we worked in, first of all, a declaration of public interest, and my thought there was that resting on the general welfare clause, it was clearly within the authority of Congress where substantial public interest was involved to take cognizance—

'Senator Taft * * * I would think that probably the jurisdiction of the United States court should be confined to the type of dispute which is interstate in character and under the jurisdiction of the National Labor Relations Board.

'Mr. Case. I certainly would have no objection to a clarification on that point. * * *

'Of course, the Senator is aware of the fact that the interpretation of the Supreme Court recently, on the subject of interstate commerce, has been so broad that anything which affects interstate commerce, which is technically interstate commerce has been ruled to be interstate commerce.

'Senator Taft. On the other hand, there is still a field of intrastate commerce. It doesn't destroy it, although it goes a long way.

'Mr. Case. As far as I am concerned, I would be glad to protect that vanishing field of intrastate commerce, and if the Senators wanted to clarify that I would have no objection.

'Senator Taft. The other question relating to that section which occurs to me is the effect of this 'binding and enforceable,' as related to the incorporation of unions. As I understand it, a collective-bargaining agreement is already mutually and equally binding and enforceable. I can't think of any circumstance where it would not be. But the problem seems to be a practical problem, in many States, of successfully suing a union which is not incorporated. I don't take it that this section would change the requirements in a State, we will say, to make every member of the union a party. It doesn't really meet that particular difficulty, which seems to be the chief difficulty in enforcing collective-bargaining agreements.

'Mr. Case. I think that what this would do would be to make the officers of a collective-bargaining agent suable. I do not think it would extend to individual members of the union, because the language is that the contract is made 'mutually and equally binding and enforceable against each of the parties thereto' and under the Wagner Act the party to the contract is the recognized bargaining agent.

'Senator Taft. No; the recognized bargaining agent is the union, not the officers of the union; it is the union, and the union is usually an unincorporated association. The point I was trying to make is that if you want to sue an unincorporated association in many States it is almost impossible to get them into court, because they have requirements that you have to serve every member, and you can't reach them, you can't find them in many cases. So the whole thing is delayed, and it is a long and tedious problem, if it can be done at all. In some States that is not so. You can sue an unincorporated association by serving the officers. But we have before us a bill from Senator Byrd requiring that unions be incorporated for the purpose of carrying out, as I see it, the same purpose you have here, at least partially, and I just wondered whether this really was effective to meet the actual difficulty today in enforcing collective-bargaining agreements.

'Mr. Case. Well, the intent of it is to make it possible to bring the action against the union in the common name of the association.

'Senator Taft. I don't say it doesn't have that effect, but I don't think it does. I don't think that 'other law to the contrary notwithstanding' would affect the method by which you had to bring a suit against an unincorporated association.

'Senator Ball. Would that make it possible to sue the union as an entity in the Federal court by simply serving the officers?

'Senator Taft. I don't think so, unless you said so. You might conceivably say so.

'Senator Ball. Then the other question is: Could we, in effect, waive State laws and make the same provision apply in State courts?

'Senator Taft. No: I don't think you can. But this would authorize suit to be brought in the Federal court and I thought that should be confined to interstate cases.

'Mr. Case. Following the procedure probably, however, of the law of the State in which the action was brought.

'Senator Taft. I don't know; that is a complicated question, as you know, in a Federal court as to when you have to comply with the State law and when you do not.

'Mr. Case. The Senator is probably more familiar than I with what is called the Second Coronado case. I am not familiar with the details of it, but as I have read references to it I think there the Supreme Court held that an injunction could be obtained against a union as such where the United States was the party. Now of course, this section doesn't carry any injunctive procedure; but there, at least, the Supreme Court seemed to recognize that a union might be made the object of an action as an association without reaching the individual members.

'Senator Taft. This would carry to a civil injunction process if it were one generally usable under laws of equity, I think, when you say it is binding and enforceable; whatever the equitable remedies might be in that State, or might be considered proper by the court, could be used.

'Senator Smith. I would like to ask Mr. Case one question on section 10. I have difficulty in seeing how that changes what the situation would be if it just wasn't in there at all. I don't quite see why we need to put that in this bill.

'Let me say first that I have not studied this carefully, so I am raising the question as it comes to me.

'Mr. Case. The intent primarily was to meet the technical legal difficulty that at the present time the union isn't suable or actionable as an association unless it is incorporated, and to avoid the necessity of joining all members of the union as parties.

'Senator Taft. I would suggest if that is your purpose it ought to say so in so many words.

'Mr. Case. I appreciate the Senator's suggestion.

'Senator Smith. That was my difficulty. It didn't seem to me it did make that clear, if that is the intent.' (Pp. 8—11.)

b. The technical, procedural nature of the problem was also stressed in testimony of some other witnesses. See pp. 175—179; 198—200; 248—249.

c. Other less discerning discussion: pp. 34, 47, 48, 110, 125, 129, 135, 144, 148, 157, 225, 237, 240, 266, 371, 372, 378, 385, 409.

3. Senate Report (No. 1177, 79th Cong., 2d Sess.):

a. Majority:

'Your committee has also considered and rejected section 10 of H.R. 4908, as passed by the House, which would explicitly declare that all collective-bargaining contracts shall be mutually binding and enforceable by the parties thereto. In the first place, this proposal appears to be based upon a misapprehension as to the legal responsibility of the parties under such contracts. Collective-bargaining agreements are at present legally enforceable in the courts, and in the Federal courts, if jurisdiction is otherwise established according to applicable law, unions may be sued in their own names under the doctrine of the first Coronado case (United Mine Workers v. Coronado Coal Company, 259 U.S. 344, (42 S.Ct. 570, 66 L.Ed. 975)). Legally, therefore, the proposed provision is unnecessary. Practically, it presents serious dangers. By offering easy access to the courts in cases where a breach of a collective-bargaining agreement is alleged, it would act as an inducement to litigate every alleged grievance, and would result in a flood of litigation making the courts again the battlefield for industrial disputes, increasing rather than eliminating the acrimony and conflict between the parties. In addition, your committee wishes to point out the fallacy of a widely held notion that breaches of contract are most often committed by labor organizations and employees. On the contrary, most breaches of contract are by employers. As has been elsewhere stated at greater length in this report your committee feels that labor disputes should be settled by conference, negotiation, and compromise, and not by the use of mandatory judicial processes. * * *' (Pp. 8—9.)

b. Minority (Senators Ball, Taft, and H. Alexander Smith):

'Amendment No. 3 (the minority proposal) would make unions suable as legal entities in the Federal courts for violation of contract, with liability limited to union assets and not enforceable against individual members or their property. A subsection would provide that where individual employees participated in a 'wildcat' strike in violation of contract, not sanctioned or approved by the union, the union itself would not be liable but such employees would lose their legal status as employees under the Wagner Act, leaving the employer free to discharge them or not. While collective-bargaining agreements theoretically are legally enforceable contracts, as a practical matter, because of the many obstacles to suits against unions imposed by most States, they are actually binding on only one party, the employer. The minority believes this provision, imposing equal responsibility on both parties to such contracts, is absolutely essential to the stability of labor relations. The only argument so far advanced against it is that some employers might embarrass unions by suits for enforcement of contract. This hardly is a valid ground for continuing to regard such contracts as one-way arrangements, wherein one party receives benefits and assumes no binding obligations whatsoever.' (Pp. 3—4.)

'Text of Amendment No. 3

'Amend H.R. 4908 by inserting at the proper place the following new section:

"Sec.—. (a) Suits for violation of a contract concluded as the result of collective bargaining between an employer and a labor organization if such contract affects commerce as defined in this Act may be brought in any district court of the United States having jurisdiction of the parties.

"(b) Any labor organization whose activities affect commerce as defined in this Act shall be bound by the acts of its duly authorized agents acting within the scope of their authority from the said labor organization and may sue or be sued as an entity and in behalf of the employees whom it represents in the courts of the United States: Provided, That any money judgment against such labor organization shall be enforceable only against the organization as an entity and against its assets, and shall not be enforceable against any individual member or his assets.

"(c) For the purposes of this section district courts shall be deemed to have jurisdiction of a labor organization (1) in the district in which such organization maintains its principal office, or (2) in any district in which its duly authorized officers or agents are engaged in promoting or protecting the interests of employee members. The service of summons, subpena or other legal process upon such officer or agent shall constitute service upon the labor organization.

"(d) Any employee who participates in a strike or other stoppage of work in violation of an existing collective-bargaining agreement, if such strike or stoppage is not ratified or approved by the labor organization party to such agreement and having exclusive bargaining rights for such employee, shall lose his status as an employee of the employer party to such agreement for the purposes of sections 8, 9, and 10 of the National Labor Relations Act: Provided, That such loss of status for such employee shall cease if and when he is reemployed by such employer.'

'The purpose of this amendment is simple: to make collective-bargaining contracts equally binding and enforceable on both parties to them. The courts have held that the purpose of the Wagner Act was—

"to compel employers to bargain collectively with their employees to the end that an employment contract, binding on both parties, should be made.' (H. J. Heinz Co., (v. N.L.R.B.), 311 U.S. 514 (,61 S.Ct. 320, 85 L.Ed. 309)—1941.)

'But neither the Wagner Act nor any other Federal statute makes labor unions legally responsible for carrying out their agreements.

'The laws of many States make it difficult to sue effectively and to recover a judgment against an unincorporated labor union. It is difficult to reach the funds of a union to satisfy a judgment against it. In some States it is necessary to serve all the members before an action can be maintained against the union. This is an almost impossible process (see testimony of Raymond S. Smethurst before committee, February 25, 1946). Despite these practical difficulties in the collection of a judgment against a union, the National Labor Relations Board has held it an unfair labor practice for an employer to insist that a union incorporate or post a bond to establish some sort of legal responsibility under a collective agreement.

'President Truman, in opening the management-labor conference in November 1945, took cognizance of this condition. He said very plainly that collective agreements should be mutually binding on both parties to the contract:

"We shall have to find methods not only of peaceful negotiation of labor contracts, but also of insuring industrial peace for the lifetime of such contracts. Contracts once made must be lived up to and should be changed only in the manner agreed upon by the parties. If we expect confidence in agreements made, there must be responsibility and integrity on both sides in carrying them out.'

'If unions can break agreements with relative impunity, then such agreements do not tend to stabilize industrial relations. The execution of an agreement does not by itself promote industrial peace. The main reason for an employer to sign a collective labor agreement is to assure uninterrupted operation during the term of the agreement. Without some effective method of assuring freedom from economic warfare for the term of the agreement, there is little reason why an employer would desire to enter into an agreement.

'Consequently, to encourage the making of agreements and to promote industrial peace through faithful performance by the parties, collective agreements dealing with interstate commerce should be enforceable in the Federal courts. Our amendment would provide for suits by unions as legal entities and against unions as legal entitles in the Federal courts in disputes affecting commerce (as defined by sec. 2(a)(1) of this act).

'The amendment specifically provides that only the assets of the union can be attached to satisfy a money judgment against it. The property of the individual members of the organization would become absolutely free from any liability under such a judgment. Thus the members of the union would secure all the advantages of limited liability without incorporation of the union.

'The proposed amendment relative to union liability specifically provides that a labor organization would be bound by the acts of its authorized agents only. Thus a labor organization would not be liable for damages arising as a result of an unauthorized strike carried on in violation of a contract. If a union or an officer thereof did not authorize the strike or participate in it, or support it, or subsequently approve it, no liability would be imposed on the union as a consequence of the work stoppage. To protect the employer against work stoppages in violation of an agreement but not approved by the union, employees who take part in such strikes would lose their status as employees under sections 8, 9, and 10 of the National Labor Relations Act. The employer could refuse to rehire them after the strike. Besides providing a remedy for the employer for irresponsible interruptions of production, such a provision would tend to strengthen sound union discipline.

'The initial obstacle in enforcing the terms of a collective agreement against a union which has breached its provisions is the difficulty of subjecting the union to process. The great majority of labor unions are unincorporated associations and at common law voluntary associations are not suable as such (Wilson v. Airline Coal Company, 215 Iowa 855 (246 N.W. 753); Iron Molders' Union (No. 125 of Milwaukee, Wis.) v. Allis-Chalmers Company, 7 (Cir.), 166 F. 45). As a consequence the rule in all jurisdictions, in the absence of statute, is that unincorporated labor unions cannot be sued in their common name (Grant v. Carpenters' District Council, 322 Pa. 62 (185 A. 273)). Accordingly, the difficulty or impossibility of enforcing the terms of a collective agreement in a suit at law against a union arises from the fact that each individual member of the union must be named and made a party to the suit.

'Some States have enacted statutes which subject unincorporated associations to the jurisdiction of law courts. These statutes are by no means uniform; some pertain to fraternal societies, welfare organizations, associations doing business, etc., and in some States the courts have excluded labor unions from their application.

'On the other hand, some States, including California and Montana, have construed statutes permitting common name suits against associations doing business to apply to labor unions (Armstrong v. Superior Court, 173 Cal. 341 (159 P. 1176); Vance v. McGinley, 39 Mont. 46 (101 P. 247)). Similarly, but more restrictive, in a considerable number of States the action is permitted against the union or representatives in proceedings in which the plaintiff could have maintained such an action against all the associates. Such States include Alabama, California, Connecticut, Delaware, Maryland, Montana, Nevada, New Jersey, New York, Rhode Island, South Carolina, and Vermont.

'In at least one jurisdiction, the District of Columbia, the liberal view is held that unincorporated labor unions may be sued as legal entities, even in the absence of statute (Busby v. Elec. Util. Emp. Union, U.S. Court of Appeals for the District of Columbia, No. 8548, January 22, 1945 (79 U.S.App.D.C. 336, 147 F.2d 865)).

'In the Federal Courts, whether an unincorporated union can be sued depends upon the procedural rules of the State in which the action is brought (Busby v. Elec. Uth. Empl. Union, U.S. Supreme Court, 89 Law.Adv.Op. 108, December 4, 1944 (323 U.S. 72, 65 S.Ct. 142)).

'The Norris-LaGuardia Act has insulated labor unions, in the field of injunctions, against liability for breach of contract. It has been held by a Federal court that strikes, picketing, or boycotting, when carried on in breach of a collective agreement, involve a 'labor dispute' under the act so as to make the activity not enjoyable (sic) without a showing of the requirements which condition the issuance of an injunction under the act (Wilson & Co. v. Birl, 105 F.2d 948, C.C.A. 3).

'A great number of States have enacted anti-injunction statutes modeled after the Norris-LaGuardia Act, and the courts of many of these jurisdictions have held that a strike in violation of a collective agreement is a 'labor dispute' and cannot be enjoined (The Nevins (Inc.), v. Kasmach, 279 N.Y. 323 (18 N.E.2d 294); Bulkin v. Sacks, 31 Pa. D & C 501).

'There are no Federal laws giving either an employer or even the Government itself any right of action against a union for any breach of contract. Thus there is no 'substantive right' to enforce, in order to make the union suable as such in Federal courts.

'Even where unions are suable, the union funds may not be reached for payment of damages, and any judgments or decrees rendered against the association as an entity may be unenforceable. (See Aalco Laundry (& Cleaning) Co. v. Laundry Linen Union, 115 S.W.2d 89 Mo.App.) However, only where statutes provide for recognition of the legal status of associations do as