Natural professional interest in trying to disentangle the legal snarl presented by this case would not justify me in enlarging my dissent from the Court’s views. But the state of the law regarding litigation brought formally against an official but intrinsically against the Government is so compounded of confusion and artificialities that an analysis differing from the Court’s may not be futile.
At the outset it is desirable to dispel a misconception regarding the legislation on abatement of suits in the federal courts. In 1899, Congress for the first time made provision for the continuance of a suit involving official conduct which abated by a succession in office during pendency of the suit. 30 Stat. 822. By § 11 of the Judiciary Act of 1925, Congress again dealt with this problem. 43 Stat. 936, 941. The Court finds that the provision of the 1925 Act “effected a substantial change.” It does this on the basis of the analysis of the first enactment made in Defense Supplies Corp. v. Lawrence Warehouse Co., 336 U. S. 631, 637-638. According to what was there said, the Act of 1899 had a categorical command that “no action shall abate,” which was eliminated in 1925. So to interpret the relation between the 1899 and the 1925 provisions is to misread legislation by quoting out of context and disregarding authoritative legislative history.
*23So far as concerns the legal effect upon the pendency of an action due to change in the occupancy of an office, a reading of the provisions of the 1899 and 1925 Acts can leave not a shadow of doubt as to their identity of purpose and procedure for its accomplishment. The difference between the two acts is a matter of English and not of law. In both, Congress assumed that a proceeding by or against an officer of the United States in relation to his official conduct would abate unless within a time certain the court authorized continuance of the proceeding by or against the successor in office. Only the phrasing of this rule differs. In the 1899 Act, Congress said that such an action shall abate unless leave is given for its continuance; in the 1925 Act, Congress said that unless leave is given for the continuance of such a suit it is at an end. To say, as we said in Defense Supplies Corp. v. Lawrence Warehouse Co., that the 1899 Act “categorically” provided that “no action shall abate” is a mutilating reading. The dominant thought of an enactment controls the primary import of isolated words. To find that the 1925 Act “eliminated” this provision has significance only if what is meant is that certain words of the 1899 Act were “eliminated” while the thought was retained. The full texts of the two provisions, set forth in the margin, speak for themselves.1 What emerges is that the two enactments have essentially the same function regarding *24the abatement and mechanism for securing survival of an action by or against an officer of the United States. The only difference is that the thought is expressed more felicitously in the later enactment, as would be expected from Mr. Justice Van Devan ter, who, as is well known, was the chief draftsman of the Judiciary Act of 1925.
The range of the 1899 Act was changed in 1925, which may have stimulated its redrafting. The change concerned not in the slightest the legal consequences to pending suits where the occupancy of an office of the United States was involved. The only modification made by the 1925 Act, apart from cutting down the time for substitu*25tion to six months from twelve, was to extend the Act of 1899 so as to permit the substitution of successors of state and local officers as well as those of federal officials. The legislative histories of the 1899 and 1925 enactments, confirming the face of the legislation, demonstrate that the two enactments were conceived for the same purpose, were intended to have the same consequences, and are to be given the same significance, excepting only that the 1925 Act extended the range of applicability.
The Act of 1899 was a response to this Court’s suggestion. See United States ex rel. Bernardin v. Butterworth, 169 U. S. 600, 605.2 This was likewise true of the Act of 1925. See Irwin v. Wright, 258 U. S. 219, 223-224.3 The opinion in that case was rendered on March 20,1922, but while it was in the bosom of the Court, having been submitted on January 24, Chief Justice Taft sent to Senator Cummins a résumé of what was known as the “Judges’ Bill,” which became the Act of 1925. As to the matter *26here under discussion, the Chief Justice said that the proposed bill “extends the right now given by statute, to substitute the successors of certain officers of the United States where the latter have died, resigned, or otherwise vacated their offices pending suit, so as to embrace the successors of officers of the District of Columbia, the Canal Zone, and the Territories and insular possessions of the United States, as well as of a State or political subdivision or agency thereof.” Confidential Committee Print entitled “Jurisdiction of Circuit Courts of Appeals and of the Supreme Court,” Senate Committee on the Judiciary, 67th Cong., 2d Sess. 4. The formulation of what was thus summarized by Chief Justice Taft is the present § 11, and that formulation was in the Judges’ Bill from the time it was introduced in the two Houses by Senator Cummins and Congressman Walsh, respectively, on February 17, 19224
*27The correctness of the result in Defense Supplies Corp. v. Lawrence Warehouse Co., supra, does not depend on the misconceived relation indicated in its opinion. But it ought not to form a part of the chain of reasoning in disposing of this case. Therefore, insofar as § 11 of the Act of 19255 is relevant to our present problem, we must reject the notion that, while under the 1899 Act such an action as this, brought against Paymaster General Buck, “did not abate,” the 1925 Act eliminated this “command.”
This brings us to the circumstances of the case. The petitioner claims to be the lawful widow of a naval officer. She brought this action to recover a death gratuity allowance, amounting to $1,365, payable under the Act of June 4, 1920. 41 Stat. 824, as amended, 34 U. S. C. § 943. Jurisdiction was alleged under the Tucker Act, 24 Stat. 505, as amended, and other statutes. Nominally, the action was for mandamus to compel Buck, the Paymaster General of the Navy, to make payment. The District Court refused to grant relief by mandamus, but, in accordance with modern practice, granted what it thought to be the proper remedy. The judgment, after enjoining Buck from persisting in his refusal to make payment, concluded: “. . . and the defendant is directed to pay the plaintiff Thirteen Hundred and Sixty-five Dollars ($1,365.00) which is the amount equal to six months’ pay at the rate received by the deceased at the time of his death.”
*28.The District Court judgment was entered on January 30, 1948. Admiral Buck was retired as Paymaster General on March 1. Notice of appeal was, nevertheless, filed in his name by Government attorneys on March 18. The issue of abatement was not raised until the Government attorney called the fact of Buck’s retirement to the attention of the Court of Appeals upon oral argument, which occurred after the six-month period for substitution had passed. The Court of Appeals vacated the judgment of the District Court and remanded with directions to dismiss the complaint as abated.
1. I agree with the Court that this was not a personal action against Admiral Buck, and that the judgment was not against him as an individual. That suits against a collector of revenue for illegal exactions under the Revenue Acts are deemed personal actions enforceable as such against the collector is an anomalous situation in our law which calls for abrogation instead of extension. For the history of these actions, see Cary v. Curtis, 3 How. 236, and United States v. Nunnally Investment Co., 316 U. S. 258.6
2. The starting point, then, is recognition of the fact that this was a suit to secure a money claim due from the United States, enforced against the officer who was the effective conduit for its payment. In short, this was a representative suit, and the crucial question, I submit, is the reach of the representative character of the suit.
The intrinsic and not merely formalistic answer to this question is of course entangled with the doctrine of sovereign immunity from suits. In scores of cases this Court *29has had to consider when a suit, though nominally against one holding public office, is in fact a suit against the Government and as such barred by want of the sovereign’s consent to be sued. See Larson v. Domestic & Foreign Commerce Corp., 337 U. S. 682, App. 729. The subject, it has been recognized, is not free from casuistry because of the natural, even if unconscious, pressure to escape from the doctrine of sovereign immunity which — whatever its historic basis — is hardly a doctrine based upon moral considerations. The trend of deep sentiment, reflected by legislation and adjudication, has looked askance at the doctrine. See Keifer & Keifer v. R. F. C., 306 U. S. 381, 390-392. If astuteness has been exercised to deny the representative character of an official in order to avoid his identification as the sovereign ad hoc, it runs counter to the rational administration of justice not to find an official the sovereign ad hoc and the suit against him, in effect, a suit against the sovereign when sovereign immunity is not circumscribed thereby.
Under the Court of Claims Act, 12 Stat. 765, as amended, the plaintiff here could have gone to the Court of Claims.7 By the Act of March 3, 1887, 24 Stat. 505, as amended, she could have brought suit in the District Court. When the sovereign has in fact given consent formally to be sued as such on the very claim and to allow, in the same court and by the same procedure (trial without a jury), precisely the same relief as was sought and obtained against the official in his representative capacity, it would needlessly enthrone formality to deny the intrinsic nature of the suit to be a suit against the sovereign. And that is this situation. Certainly those charged with the duty of defending the interests of the United States so *30conceived it. By denominating Admiral Buck as “Paymaster General of the Navy” in his notice of appeal, the United States Attorney recognized that Paymaster General Buck was, as it were, merely an alias for the United States, the real client of the United States Attorney. The Government, indeed, has consistently recognized that justice does not call for abatement of the suit. Both here and below it has disavowed a desire for abatement. Of course, if it were a fixed rule of law that a suit such as this should die when the nominal defendant dies, the Court would have to bow to it, however harsh and futile the rule. It required legislation represented by Lord Campbell's Act to make tort liability survive the death of the victim. But it is not the controlling policy of the law that such actions die upon change of office-holders. The policy of the law is to the contrary, even as to suits which could not be brought against the Government directly. So also, it has long been the policy of our law to look behind an office-holder nominally a party litigant in order to find that, for all practical purposes, it is a suit against the Government and therefore not maintainable. Justice should be equally open-eyed in order to find behind the nominal official defendant the United States as the real defendant.
This seems to me to be the spirit of the decision in Thompson v. United States, 103 U. S. 480. To be sure, Mr. Justice Bradley there differentiated his identification of an officer of a municipality with the municipality from the situation of an officer of the United States because normally the Government could not be sued. But when the Government does allow itself to be sued for the same cause of action for which suit was brought against him who for the purposes of the litigation is the United States, the reason for the differentiation disappears.
The differentiation remains in actions brought against officials for remedies which could not be got in a direct *31suit against the United States. These are the situations in which substitution cannot come into play automatically and involve recourse to the remedial legislation of 1899 and 1925 in their present form. This gives ample scope to the legislation and at the same time avoids treating procedural requirements as tyrannical commands satisfying no other end except sterile formality.
Accordingly, I would recognize that the judgment of the District Court is in effect a money judgment against the United States and would allow the Government’s notice of appeal the force it was intended to have as an effective instrument whereby the United States might obtain a review of that judgment. It would be nothing novel in the observance of decorous form by courts to note as a matter of record that the name of the Paymaster General of the Navy is now Fox and to proceed with the appeal on that basis.8
A final question has to be faced — a question which should, in logic, have been treated first, for it concerns the *32power of this Court to decide the case. Section 2105 of 28 U. S. C. provides: “There shall be no reversal in the Supreme Court or a court of appeals for error in ruling upon matters in abatement which do not involve jurisdiction.” I agree with the Court that this statute is not applicable, but not on the ground that lack of substitution is a question of “jurisdiction.” Section 2105 relates only to the modern equivalent of a common law plea in abatement, which was made in the trial court before issue was joined on the merits of the case.9 It can have no effect upon a decision by an appellate court that a suit has abated.
Chapter 121 of the Laws of 1899, 30 Stat. 822, provided: “. . . That no suit, action, or other proceeding lawfully commenced by or against the head of any Department or Bureau or other officer of the United States in his official capacity, or in relation to the discharge of his official duties, shall abate by reason of his death, or the expiration of his term of office, or his retirement, or resignation, or removal from office, but, in such event, the Court, on motion or supplemental petition filed, at any time within twelve months thereafter, showing a necessity for the survival thereof to obtain a settlement of the questions involved, may allow the same to be maintained *24by or against his successor in office, and the Court may make such order as shall be equitable for the payment of costs.”
Section 11 of the Judiciary Act of 1925, 43 Stat. 936, 941, provided:
“(a) That where, during the pendency of an action, suit, or other proceeding brought by or against an officer of the United States, or of the District of Columbia, or the Canal Zone, or of a Territory or an insular possession of the United States, or of a county, city, or other governmental agency of such Territory or insular possession, and relating to the present or future discharge of his official duties, such officer dies, resigns, or otherwise ceases to hold such office, it shall be competent for the court wherein the action, suit, or proceeding is pending, whether the court be one of first instance or an appellate tribunal, to permit the cause to be continued and maintained by or against the successor in office of such officer, if within six months after his death or separation from the office it be satisfactorily shown- to the court that there is a substantial need for so continuing and maintaining the cause and obtaining an adjudication of the questions involved.
“(b) Similar proceedings may be had and taken where an action, suit, or proceeding brought by or against an officer of a State, or of a county, city, or other governmental agency of a State, is pending in a court of the United States at the time of the officer’s death or separation from the office.
“(c) Before a substitution under this section is made, the party or officer to be affected, unless expressly consenting thereto, must be given reasonable notice of the application therefor and accorded an opportunity to present any objection which he may have.”
“In view of the inconvenience, of which the present case is a striking instance, occasioned by this state of the law, it would seem desirable that Congress should provide for the difficulty by enacting that, in the case of suits against the heads of departments abating by death or resignation, it should be lawful for the successor in office to be brought into the case by petition, or some other appropriate method.”
“It may not be improper to say that it would promote justice if Congress were to enlarge the scope of the Act of February 8, 1899, so as to permit the substitution of successors for state officers suing or sued in the federal courts, who cease to be officers by retirement or death, upon a sufficient showing in proper cases. Under the present state of the law, an important litigation may be begun and carried through to this court after much effort and expense, only to end in dismissal because, in the necessary time consumed in reaching here, state officials, parties to the action, have retired from office. It is a defect which only legislation can cure.” Chief Justice Taft used Irwin v. Wright as an illustration in his testimony before the House Judiciary Committee on March 30, 1922. Hearings before House Committee on the Judiciary on H. R. 10479, 67th Cong., 2d Sess. 7.
See 62 Cong. Rec. 2686, 2737. The language concerning abatement was the same in the bills introduced in 1922 (S. 3164 and H. R. 10479, 67th Cong., 2d Sess.), in the bills introduced by Senator Cummins and Congressman Graham, respectively, in 1924 (S. 2060 and H. R. 8206, 68th Cong., 1st Sess.) and in the statute as enacted, 43 Stat. 936, 941.
I am partly responsible for the misconception of finding a substantive change between the significance of the 1899 Act and the 1925 Act because I joined in Defense Supplies Corp. v. Lawrence Warehouse Co., 336 U. S. 631. It is not by way of extenuating my responsibility that I deem it pertinent to suggest that the nature and volume of the Court’s business preclude examination of all the judicial and legislative materials of ail opinions in which one concurs. In order that the energies of the Court may be concentrated on those eases for which adjudication by this Court is indispensable, I have been insistent in my view that the Court should be rigorous in limiting the cases which it will allow to come here. That it may so control its business, the Congress, by the Act of 1925, gave the Court — for all practical purposes — a free hand. See Ex parte Peru, 318 U. S. 578, 602-604.
Section 11 of the 1925 Judiciary Act, 43 Stat. 936, 941, was repealed as of September 1, 1948. 62 Stat. 992, 1000. The repealing Act, however, preserved any rights or liabilities existing under the laws repealed. 62 Stat. 869, 992. Since the six-month statutory period within which substitution can be made expired on September 1, 1948, the repeal of § 11 does not affect the case at bar. Abatement of actions brought against officials is now governed, in the District Courts, by Rule 25 (d) of the Federal Rules of Civil Procedure, which also provides a six-month period for substitution.
The problems raised by the personal liability of collectors have necessitated special legislation. See I. R. C., § 3770 (b), 26 U. S. C. § 3770 (b), R. S. § 3220, as amended (authority to reimburse collectors), and I. R. C., § 3772 (d), 26 U. S. C. § 3772 (d), 56 Stat. 956. (Suits against collectors are treated as suits against the United States for purposes of res judicata.)
Campbell v. United States, 80 Ct. Cl. 836; Hill v. United States, 68 Ct. Cl. 740; Maxwell v. United States, 68 Ct. Cl. 727; Thomson v. United States, 58 Ct. Cl. 207; Phillips v. United States, 49 Ct. Cl. 703.
Davis v. Preston, 280 U. S. 406, was a suit under the Federal Employers’ Liability Act of 1908, 35 Stat. 65, brought to recover for the death of a railroad employee occurring while the railroad was being operated by the Director General. The Court held a petition for cer-tiorari ineffective when taken in the name of an agent of the Government who had retired from office. A statute there applicable, however, required that the United States conduct the litigation in the name of a special agent. 41 Stat. 461. There is no such requirement in the case at bar, for suit could have been brought against the United States itself. Since this is true, the Court can scarcely refuse to give effect to the notice of appeal filed by the Government attorneys in the name of Buck as Paymaster General.
Mr. Justice Van Devanter discussed only the effectiveness of the appeal, for the Court was faced with no problem of abatement. Congress had made clear its policy of protecting suitors against the pitfalls of abatement by passing the Winslow Act, 42 Stat. 1443, to make certain that the 1899 statute would not prevent recovery for persons injured or killed during the Government operation of the railroads. This statute allowed substitution of a successor agent at *32“any time before satisfaction of such final judgment, decree, or award.” The broad legislative policy reflected in the Winslow Act points to a reliance upon substance, rather than form, in the present case.
The predecessor section, 28 U. S. C. (1946 ed.) § 879, R. S. § 1011, as amended, provided:
“There shall be no reversal in the Supreme Court or in a circuit court of appeals upon a writ of error, for error in ruling any plea in abatement, other than a plea to the jurisdiction of the court, or for any error in fact.”
Section 22 of the Judiciary Act of 1789, 1 Stat. 84, provided:
“. . . But there shall be no reversal in either court [i. e., the Circuit Court or Supreme Court] on such writ of error for error in ruling any plea in abatement, other than a plea to the jurisdiction of the court, or such plea to a petition or bill in equity, as is in the nature of a demurrer, or for any error in fact. . . .”
The Reviser’s Note to § 2105 indicates that “matters in abatement” was substituted for “plea in abatement” because of the change in terminology under the Federal Rules.