Sinclair Refining Company v. Atkinson/Dissent Brennan
Mr. Justice BRENNAN, with whom Mr. Justice DOUGLAS and Mr. Justice HARLAN join, dissenting.
I believe that the Court has reached the wrong result because it has answered only the first of the questions which must be answered to decide this case. Of course § 301 of the Taft-Hartley Act did not, for purposes of actions brought under it, 'repeal' § 4 of the Norris-LaGuardia Act. But the two provisions do coexist, and it is clear beyond dispute that they apply to the case before us in apparently conflicting senses. Our duty, therefore, is to seek out that accommodation of the two which will give the fullest possible effect to the central purposes of both. Since such accommodation is possible, the Court's failure to follow that path leads it to a result-not justified by either the language or history of § 301 which is wholly at odds with our earlier handling of directly analogous situations and which cannot be woven intelligibly into the broader fabric of related decisions.
Section 301 of the Taft-Hartley Act, enacted in 1947, authorizes Federal District Courts to entertain '(s)uits for violation of contracts between an employer and a labor organization * * *.' It does not in terms address itself to the question of remedies. As we have construed § 301, it casts upon the District Courts a special responsibility to carry out contractual schemes for arbitration, by holding parties to that favored process for settlement when it has been contracted for, and by then regarding its result as conclusive.  At the same time, § 4 of the Norris-LaGuardia Act, enacted in 1932, proscribes the issuance by federal courts of injunctions against various concerted activities 'in any case involving or growing out of any labor dispute.' But the enjoining of a strike over an arbitrable grievance may be indispensable to the effective enforcement of an arbitration scheme in a collective agreement; thus the power to grant that injunctive remedy may be essential to the uncrippled performance of the Court's function under § 301.  Therefore, to hold that § 301 did not repeal § 4 is only a beginning. Having so held, the Court should but does not-go on to consider how it is to deal with the surface conflict between the two statutory commands.
The Court has long acted upon the premise that the Norris-LaGuardia Act does not stand in isolation. It is one of several statutes which, taken together, shape the national labor policy. Accordingly, the Court has recognized that Norris-LaGuardia does not invariably bar injunctive relief when necessary to achieve an important objective of some other statute in the pattern of labor laws. See Brotherhood of Railroad Trainmen v. Chicago River R. Co., 353 U.S. 30, 77 S.Ct. 635, 1 L.Ed.2d 622; Graham v. Brotherhood of Locomotive Firemen, 338 U.S. 232, 70 S.Ct. 14, 94 L.Ed. 22; Virginian R. Co. v. System Federation, 300 U.S. 515, 562-563, 57 S.Ct. 592, 81 L.Ed. 789. In Chicago River we insisted that there 'must be an accommodation of (the Norris-LaGuardia Act) and the Railway Labor Act so that the obvious purpose in the enactment of each is preserved.' 
These decisions refusing inflexible application of Norris-LaGuardia point to the necessity of a careful inquiry whether the surface conflict between § 301 and § 4 is irreconcilable in the setting before us: a strike over a grievance which both parties have agreed to settle by binding arbitration. I think that there is nothing in either the language of § 301 or its history to prevent § 4's here being accommodated with it, just as § 4 was accommodated with the Railway Labor Act.
It cannot be denied that the availability of the injunctive remedy in this setting is far more necessary to the accomplishment of the purposes of § 301 than it would be detrimental to those of Norris-LaGuardia. Chicago River makes this plain. We there held that the federal courts, notwithstanding Norris-LaGuardia, may enjoin strikes over disputes as to the interpretation of an existing collective agreement, since such strikes flout the duty imposed on the union by the Railway Labor Act to settle such 'minor disputes' by submission to the National Railroad Adjustment Board rather than by concerted economic pressures. We so held, even though the Railway Labor Act contains no express prohibition of strikes over 'minor disputes,' because we found it essential to the meaningful enforcement of that Act-and because the existence of mandatory arbitration eliminated one of the problems to which Norris-LaGuardia was chiefly addressed, namely, that 'the injunction strips labor of its primary weapon without substituting any reasonable alternative.' 
That reasoning is applicable with equal force to an injunction under § 301 to enforce a union's contractual duty, also binding on the employer, to submit certain disputes to terminal arbitration and to refrain from striking over them. The federal law embodied in § 301 stresses the effective enforcement of such arbitration agreements. When one of them is about to be sabotaged by a strike, § 301 has as strong a claim upon an accommodating interpretation of § 4 as does the compulsory arbitration law of the Railway Labor Act. It is equally true in both cases that '(an injunction) alone can effectively guard the plaintiff's right,' International Ass'n of Machinists v. Street, 367 U.S. 740, 773, 81 S.Ct. 1784, 1802, 6 L.Ed.2d 1141. It is equally true in both cases that the employer's specifically enforceable obligation to arbitrate provides a 'reasonable alternative' to the strike weapon. It is equally true in both cases that a major contributing cause for the enactment of Norris-LaGuardia-the at-largeness of federal judges in enjoining activities thought to seek 'unlawful ends' or to constitute 'unlawful means' -is not involved. Indeed, there is in this case a factor weighing in favor of the issuance of an injunction which was not present in Chicago River:  the express contractual commitment of the union to refrain from striking, viewed in light of the overriding purpose of § 301 to assist the enforcement of collective agreements.
In any event, I should have thought that the question was settled by Textile Workers v. Lincoln Mills, 353 U.S. 448, 77 S.Ct. 912, 1 L.Ed.2d 972. In that case, the Court held that the procedural requirements of Norris-LaGuardia's § 7, although in terms fully applicable, would not apply so as to frustrate a federal court's effective enforcement under § 301 of an employer's obligation to arbitrate. It is strange, I think, that § 7 of the Norris-LaGuardia Act need not be read, in the face of § 301, to impose inapt procedural restrictions upon the specific enforcement of an employer's contractual duty to arbitrate; but that § 4 must be read, despite § 301, to preclude absolutely the issuance of an injunction against a strike which ignores a union's identical duty.
The legislative history of § 301 affords the Court no refuge from the compelling effect of our prior decisions. That history shows that Congress considered and rejected 'the advisability of repealing the Norris-LaGuardia Act insofar as suits based upon breach of collective bargaining agreements are concerned * * *.'  But congressional rejection of outright repeal certainly does not imply hostility to an attempt by the courts to accommodate all statutes pertinent to the decision of cases before them. Again, the Court's conclusion stems from putting the wrong question. When it is appreciated that there is no question here of 'repeal,' but rather one of how the Court is to apply the whole statutory complex to the case before it, it becomes clear that the legislative history does not support the Court's conclusion. First, however, it seems appropriate to discuss, as the Court has done, the language of § 301 considered in light of other provisions of the statute.
There is nothing in the words of § 301 which so much as intimates any limitation to damage remedies when the asserted breach of contract consists of concerted activity. The section simply authorizes the District Courts to entertain and decide suits for violation of collective contracts. Taking the language alone, the irresistible implication would be that the District Courts were to employ their regular arsenal of remedies appropriately to the situation. That would mean, of course, that injunctive relief could be afforded when damages would not be an adequate remedy. This much, surely, is settled by Lincoln Mills. But the Court reasons that the failure of § 301 explicitly to repeal § 4 of Norris-LaGuardia completely negates the availability of injunctive relief in any case where that provision-in the absence of § 301-would apply. That reasoning stems from attaching undue significance to the fact that express repeal of Norris-LaGuardia provisions may be found in certain other sections of the Taft-Hartley Act-from which the Court concludes 'not only that Congress was completely familiar with those provisions but also that it regarded an express declaration of inapplicability as the normal and proper manner of repealing them in situations where such repeal seemed desirable.'  Even on this analysis the most that can be deduced from such a comparative reading is that while repeal of Norris-LaGuardia seemed desirable to Congress in certain other contexts, repeal did not seem desirable in connection with § 301.
Sound reasons explain why repeal of Norris-LaGuardia provisions, acceptable in other settings, might have been found ill-suited for the purpose of § 301. And those reasons fall far short of a design to preclude absolutely the issuance under § 301 of any injunction against an activity included in § 4 of Norris-LaGuardia. Section 10(h) of the Act  simply lifts the § 4 barrier in connection with proceedings brought by the National Labor Relations Board-in the Courts of Appeals for enforcement of Board cease-and-desist orders against unfair labor practices, and in the District Courts for interlocutory relief against activities being prosecuted before the Board as unfair labor practices. This repeal in aid of government litigation to enforce carefully drafted prohibitions already in the Act as unfair labor practices was, obviously, entirely appropriate, definitely limited in scope, predictable in effect, and devoid of any risk of abuse or misunderstanding. Much the same is true of § 208(b) of Taft-Hartley,  which simply repeals Norris-LaGuardia in a case where the Attorney General seeks an injunction at the direction of the President, who must be of the opinion-after having been advised by a board of inquiry-that continuation of the strike in question would imperil the national health and safety.
Only in § 302(e) of Taft-Hartley  is there found a repeal of Norris-LaGuardia's anti-injunction provisions in favor of a suit by a private litigant.  The District Courts are there authorized to restrain the payment by employers and the acceptance by employee representatives of unauthorized payments in the nature of bribes. Not only is the problem thus dealt with 'unusually sensitive and important,' as the Court notes,  but the repeal of Norris-LaGuardia is clearly, predictably, and narrowly confined to one kind of suit over one kind of injury; and obviously it presents no possible threat to the important purposes of that Act.
How different was the problem posed by § 301, which broadly authorized District Courts to decide suits for breach of contract. The Congress understandably may not have felt able to predict what provisions would crop up in collective bargaining agreements, to foresee the settings in which these would become subjects of litigation, or to forecast the rules of law which the courts would apply. The consequences of repealing the anti-injunction provisions in this context would have been completely unknowable, and outright repeal, therefore, might well have seemed unthinkable. Congress, clearly, had no intention of abandoning wholesale the Norris-LaGuardia policies in contract suits; but it does not follow that § 301 is not the equal of § 4 in cases which implicate both provisions.
Indeed, it might with as much force be said that Congress knew well how to limit remedies against employee activities to damages when that was what it intended, as that Congress knew how to repeal Norris-LaGuardia when that was what it intended. Section 303 of Taft-Hartley  authorizes private actions for damages resulting from certain concerted employee activities. When that section was introduced on the Senate floor, it provided for injunctive relief as well. Extended debate revealed strong sentiment against the injunction feature, which incorporated a repeal of Norris-LaGuardia. The section's supporters, therefore, proposed a different version which provided for damages only. In this form, the section was adopted by the Senate-and later by the Conference and the House.  Certainly, after this experience Congress would have used language confining § 301 to damage remedies when it was invoked against concerted activity, if such had been the intention.
The statutory language thus fails to support the Court's position. The inference is at least as strong that Congress was content to rely upon the courts to resolve any seeming conflicts between § 301 and § 4 as they arose in the relatively manageable setting of particular cases, as that Congress intended to limit to damages the remedies courts could afford against concerted activities under § 301. The Court then should so exercise its judgment as best to effect the most important purposes of each statute. It should not be bound by inscrutable congressional silence to a wooden preference for one statute over the other.
Nor does the legislative history of § 301 suggest any different conclusion. As the Court notes, the House version would have repealed Norris-LaGuardia in suits brought under the new section.  The Senate version of § 301, like the section as enacted, did not deal with Norris-LaGuardia, but neither did it limit the remedies available against concerted activity.  Thus any attempt to ascertain the Senate's intention would face the same choices as those I have suggested in dealing with the language of § 301 as finally enacted. It follows that to construe the Conference Committee's elimination of the House repeal as leaving open the possibility of judicial accommodation is at least as reasonable as to conclude that Congress, by its silence, was directing the courts to disregard § 301 whenever opposition from § 4 was encountered. 
I emphasize that the question in this case is not whether the basic policy embodied in Norris-LaGuardia against the injunction of activities of labor unions has been abandoned in actions under § 301; the question is simply whether injunctions are barred against strikes over grievances which have been routed to arbitration by a contract specifically enforceable against both the union and the employer. Enforced adherence to such arbitration commitments has emerged as a dominant motif in the developing federal law of collective bargaining agreements. But there is no general federal anti-strike policy; and although a suit may be brought under § 301 against strikes which, while they are breaches of private contracts, do not threaten any additional public policy, in such cases the anti-injunction policy of Norris-LaGuardia should prevail. Insistence upon strict application of Norris-LaGuardia to a strike over a dispute which both parties are bound by contract to arbitrate threatens a leading policy of our labor relations law. But there may be no such threat if the union has made no binding agreement to arbitrate; and if the employer cannot be compelled to arbitrate, restraining the strike would cut deep into the core of Norris-LaGuardia. Therefore, unless both parties are so bound, limiting an employer's remedy to damages might well be appropriate. The susceptibility of particular concrete situations to this sort of analysis shows that rejection of an outright repeal of § 4 was wholly consistent with acceptance of a technique of accommodation which would lead, in some cases, to the granting of injunctions against concerted activity. Accommodation requires only that the anti-injunction policy of Norris-LaGuardia not intrude into areas, not vital to its ends, where injunctive relief is vital to a purpose of § 301; it does not require unconditional surrender.
Today's decision cannot be fitted harmoniously into the pattern of prior decisions on analogous and related matters. Considered in their light, the decision leads inescapably to results consistent neither with any imaginable legislative purpose nor with sound judicial administration.
We have held that uniform doctrines of federal labor law are to be fashioned judicially in suits brought under § 301, Textile Workers v. Lincoln Mills, 353 U.S. 448, 77 S.Ct. 912, 1 L.Ed.2d 972; that actions based on collective agreements remain cognizable in state as well as federal courts, Charles Dowd Box Co. v. Courtney, 368 U.S. 502, 82 S.Ct. 519, 7 L.Ed.2d 483; and that state courts must apply federal law in such actions, Local 174 Teamsters v. Lucas Flour Co., 369 U.S. 95, 82 S.Ct. 571, 7 L.Ed.2d 593.
The question arises whether today's prohibition of injunctive relief is to be carried over to state courts as a part of the federal law governing collective agreements. If so, § 301, a provision plainly designed to enhance the responsibility of unions to their contracts, will have had the opposite effect of depriving employers of a state remedy they enjoyed prior to its enactment.
On the other hand if, as today's literal reading suggests  and as a leading state decision holds,  States remain free to apply their injunctive remedies against concerted activities in breach of contract, the development of a uniform body of federal contract law is in for hard times. So long as state courts remain free to grant the injunctions unavailable in federal courts, suits seeking relief against concerted activities in breach of contract will be channeled to the States whenever possible. Ironically, state rather than federal courts will be the preferred instruments to protect the integrity of the arbitration process, which Lincoln Mills and the Steelworkers decisions forged into a kingpin of federal labor policy. Enunciation of uniform doctrines applicable in such cases will be severely impeded. Moreover, the type of relief available in a particular instance will turn on fortuities of locale and susceptibility to process-depending upon which States have anti-injunction statutes and how they construe them.
I have not overlooked the possibility that removal of the state suit to the federal court might provide the answer to these difficulties. But if § 4 is to be read literally, removal will not be allowed.  And if it is allowed, the result once again is that § 301 will have had the strange consequence of taking away a contract remedy available before its enactment.
The decision deals a crippling blow to the cause of grievance arbitration itself. Arbitration is so highly regarded as a proved technique for industrial peace that even the Norris-LaGuardia Act fosters its use.  But since unions cannot be enjoined by a federal court from striking in open defiance of their undertakings to arbitrate, employers will pause long before committing themselves to obligations enforceable against them but not against their unions. The Court does not deny the desirability, indeed, necessity, for injunctive relief against a strike over an arbitrable grievance.  The Court says only that federal courts may not grant such relief, that Congress must amend § 4 if those courts are to give substance to the congressional plan of encouraging peaceable settlements of grievances through arbitration.
A District Court entertaining an action under § 301 may not grant injunctive relief against concerted activity unless and until it decides that the case is one in which an injunction would be appropriate despite the Norris-LaGuardia Act. When a strike is sought to be enjoined because it is over a grievance which both parties are contractually bound to arbitrate, the District Court may issue no injunctive order until it first holds that the contract does have that effect; and the employer should be ordered to arbitrate, as a condition of his obtaining an injunction against the strike. Beyond this, the District Court must, of course, consider whether issuance of an injunction would be warranted under ordinary principles of equity-whether breaches are occurring and will continue, or have been threatened and will be committed; whether they have caused or will cause irreparable injury to the employer; and whether the employer will suffer more from the denial of an injunction than will the union from its issuance.
In the case before us, the union enjoys the contractual right to make the employer submit to final and binding arbitration of any employee grievance. At the same time, the union agrees that '(T)here shall be no strikes * * * for any cause which is or may be the subject of a grievance.'  The complaint alleged that the union had, over the past several months, repeatedly engaged in 'quickie' strikes over arbitrable grievances. Under the contract and the complaint, then, the District Court might conclude that there have occurred and will continue to occur breaches of contract of a type to which the principle of accommodation applies. It follows that rather than dismissing the complaint's request for an injunction, the Court should remand the case to the District Court with directions to consider whether to grant the relief sought-an injunction against future repetitions. This would entail a weighing of the employer's need for such an injunction against the harm that might be inflicted upon legitimate employee activity. It would call into question the feasibility of setting up in futuro contempt sanctions against the union (for striking) and against the employer (for refusing to arbitrate) in regard to prospective disputes which might fall more or less clearly into the adjudicated category of arbitrable grievances. In short, the District Court will have to consider with great care whether it is possible to draft a decree which would deal equitably with all the interests at stake.
I would reverse the Court of Appeals and remand to the District Court for further proceedings consistent with this dissenting opinion.
^1 Textile Workers v. Lincoln Mills, 353 U.S. 448, 77 S.Ct. 912, 1 L.Ed.2d 972; United Steelworkers v. American Mfg. Co., 363 U.S. 564, 80 S.Ct. 1343, 4 L.Ed.2d 1403; United Steelworkers v. Warrior & Gulf Co., 363 U.S. 574, 80 S.Ct. 1347, 4 L.Ed.2d 1409; United Steelworkers v. Enterprise Corp., 363 U.S. 593, 80 S.Ct. 1358, 4 L.Ed.2d 1424.
^2 In Local 174, Teamsters v. Lucas Flour Co., 369 U.S. 95, 82 S.Ct. 571, 7 L.ed.2d 593, we held that a strike over a dispute which a contract provides shall be settled exclusively by binding arbitration is a breach of contract despite the absence of a no-strike clause, saying, at p. 105, 82 S.Ct. at p. 579: 'To hold otherwise would obviously do violence to accepted principles of traditional contract law. Even more in point, a contrary view would be completely at odds with the basic policy of national labor legislation to promote the arbitral process as a substitute for economic warfare.' And in Brotherhood of Railroad Trainmen v. Chicago River R. Co., 353 U.S. 30, 39, 77 S.Ct. 635, 639, 1 L.Ed.2d 622, we recognized that allowing a strike over an arbitrable dispute would effectively 'defeat the jurisdiction' of the arbitrator.
^3 353 U.S. at 40, 77 S.Ct. at 640.
^4 Id., at 41, 77 S.Ct. at 640.
^5 See, e.g., S.Rep. No. 163, 72d Cong., 1st Sess., p. 18; Frankfurter and Greene, The Labor Injunction, pp. 24-46, 200, 202.
^6 It is worth repeating that the Railway Labor Act incorporates no express prohibition of strikes over 'minor disputes.'
^7 Ante, 370 U.S., p. 205, 82 S.Ct., p. 1334.
^8 Ante, 370 U.S., p. 205, 82 S.Ct., p. 1334. (Emphasis added.)
^12 Section 301(e), 61 Stat. 157, 29 U.S.C. § 185(e), 29 U.S.C.A. § 185(e), also mentioned by the Court, has no bearing on injunction problems. It repeals, for its purposes, § 6 of the Norris-LaGuardia Act, which deals with agency responsibility for concerted activities. Its only relevance here is in showing what is clear anyway: That § 301 effected no repeal of the anti- injunction provisions of Norris-LaGuardia.
^13 Ante, 370 U.S., p. 205, 82 S.Ct., p. 1334, n. 19.
^15 See II Leg.Hist. 1323-1400; I Leg.Hist. 571.
^16 I Leg.Hist. 221-222.
^17 I Leg.Hist. 279-280.
^18 There is nothing in any Committee Report, or in any floor debate, which even intimates a confinement of § 301 remedies to damages in cases involving concerted activities. The only bit of legislative history which could is the statement of Senator Taft, quoted by the Court at note 27 of its opinion, which he inserted into the Congressional Record. What little significance that isolated insertion might have had has, of course, been laid to rest by Lincoln Mills.
^19 Section 4 commences: 'No courts of the United States shall have jurisdiction to issue any restraining order * * *.'
^21 Compare note 19, supra, with the language of the removal statute, 28 U.S.C. § 1441, 28 U.S.C.A. § 1441, allowing removal in cases 'of which the district courts of the United States have original jurisdiction.'
^23 The Court acknowledges, of course, that an employer may obtain an order directing a union to comply with its contract to arbitrate. Consistently with what we said in Lucas, supra, note 2, a strike in the face of such an order would risk a charge of contempt.