Opinion
No. 84-5460.
Order May 30, 1986.
Richard K. Willard, Asst. Atty. Gen., Dept. of Justice, Joseph E. diGenova, U.S. Atty., Washington, D.C., Jeffrey Axelrad, Barbara L. Herwig and Marc Johnston, Attys., Dept. of Justice, Washington, D.C., were on suggestion for rehearing en banc.
Appeal from the United States District Court for the District of Columbia (Civil Action No. 83-750).
ON APPELLEE'S SUGGESTION FOR REHEARING EN BANCBefore ROBINSON, Chief Judge; WRIGHT, WALD, MIKVA, EDWARDS, GINSBURG, BORK, SCALIA, STARR, SILBERMAN and BUCKLEY, Circuit Judges.
ORDER
Appellee's suggestion for rehearing en banc has been transmitted to the full court. A vote was requested. A majority of the judges of the court in regular active service did not vote in favor of the suggestion. Upon consideration of the foregoing, it is
ORDERED, by the Court en banc, that appellee's suggestion is denied.
Circuit Judges BORK, SCALIA, STARR, SILBERMAN and BUCKLEY would grant the suggestion for rehearing en banc. A statement is attached.
A statement of Circuit Judges J. SKELLY WRIGHT and GINSBURG is also attached.
The dissenters indicate their readiness to scrutinize pleas of "military necessity" with due rigor and care, even in "times of emergency in our future," even when "utter capitulation" is "supported by popular passion." See diss. at 305, 309. We praise that stance, concur in it, and write only to inter the dissenters' most grave misunderstanding that our opinion holds anything to the contrary.
The panel majority opinion deals particularly and precisely with the special facts of an extraordinary episode of injustice. It most assuredly "creates [ no] rule of absolute and permanent judicial deference to any claim of `military necessity.'" Id. at 305 (emphasis altered). Rather, the opinion simply describes and turns on what we find to be the situation-specific holding of Hirabayashi v. United States, 320 U.S. 81, 63 S.Ct. 1375, 87 L.Ed. 1774 (1943), and Korematsu v. United States, 323 U.S. 214, 65 S.Ct. 193, 89 L.Ed. 194 (1944): courts must defer to the judgment of Congress and the Executive that sufficient military necessity existed to justify the World War II internment policy. That Supreme Court holding seems to us clear, pin-pointed, and definite. We therefore concluded that the former internees faced an insuperable obstacle to the present suit until the "war-making branches," Korematsu, 323 U.S. at 218-19, 65 S.Ct. at 195, released the federal courts from the grasp of Korematsu and Hirabayashi by indicating that deference was no longer due to the wartime judgment of military necessity for the mass evacuation.
The dissenters also maintain that, although Korematsu and Hirabayashi may have established the military necessity of confining the Japanese-Americans to internment camps, those cases did not establish the military necessity of the takings at issue in this case. See diss. at 308. In making this argument, the dissenters presume that the taking of property and the confining of the Japanese-Americans are properly analyzed as separate and distinct actions. In fact, however, the taking of property was part and parcel of the internment policy; that policy included not simply confining United States citizens but, necessarily adjunct to that action, forcing them to leave their property behind under military supervision. The losses that occurred under military supervision were therefore sustained under the very internment policy that the Supreme Court held justified by military necessity.
The dissenters also argue that the wartime cases either held only that the specific — non-takings clause — claims then before the court were not ones upon which relief could be granted, or that courts owe absolute and permanent deference to military judgments about military necessity at all times. We have adopted neither view. Instead, we read the wartime cases to have established the military necessity of the internment policy for both the particular claims at issue in those cases and the takings clause claims now before the court. See Hohri v. United States, 782 F.2d 227, 250-53 (D.C. Cir. 1986). We believe our reading to be historically faithful to the argument and ethos of the wartime cases, while avoiding the gross over-generalizations or artificial narrowness suggested by the dissenters.
Our opinion, read as we conceived it and not as the dissenters would have it, does not stretch beyond the setting in which it is embedded. We did not announce a general rule of automatic judicial capitulation to the military's claims of military necessity. We ruled narrowly, specifically, and only that when the Supreme Court has definitively held that deference to a military judgment is due in a particular case, litigants may not reasonably be required to re-litigate that issue in advance of a green light from the "war-making branches."
The dissenters argue that President Ford's Proclamation No. 4417, 3 C.F.R. 8 (1977), provided such a green light. See diss. at 5 n. 1. As the panel majority opinion points out, that proclamation announces merely that the internment policy was morally, not legally, wrong. See Hohri, 782 F.2d at 253 n. 67. The dissent, however, maintains that if the policy were morally wrong, then it must not have been "necessary." Diss. at 306 n. 1. This argument confuses the nature of the "necessity" involved. It may be true, as the dissent suggests, that necessity in the sense of absolute coercion is a complete moral excuse. But the government has never claimed that it was under absolute coercion to implement the policy. Rather, the "war-making branches" made the judgment that sufficient military need existed to justify the policy. The policy was thus necessary, not in an absolute sense, but in a relative one: it was necessary to military ends then deemed important enough to outweigh the harm to the internees. Therefore, the policy may well have been both morally wrong (in the sense that the moral wrongs outweighed the military need) and legally "necessary" (in the sense that the military need outweighed the harm to legal rights).
The dissenters also suggest that Korematsu and Hirabayashi may best be understood as erroneous decisions made under the pressures of wartime. If these cases had come up in peacetime or if their validity were reconsidered in peacetime, the dissenters appear to advise, the Supreme Court might well have ruled the internment policy unconstitutional. Therefore, the former internees should presumably have brought this case at some undefined point after the war when the "emergency [was] long past" and the mood of the Court had sufficiently changed. See diss. at 305, 306. In making that argument, the dissenters overlook this reality: litigants do not have the academic luxury of indulging the belief that they can lay a solid foundation for their in-court pleas by insisting that the Supreme Court does not really mean what it says, or that a peacetime Court should not hesitate to repudiate a wartime Court for ignoring the Constitution's requirements. The dissenters' double standard would thus preclude the former internees from ever obtaining judicial redress: the validity of the internment may have been tested originally by the deferential standard imposed by wartime pressures, but we should nonetheless measure the tolling of the statute of limitations with the dissenters' more searching standard in mind. We do not believe that the policies served by a statute of limitations inexorably require courts to subject litigants to such a vicious whipsaw.
As to remaining portions of the dissent from denial of rehearing en banc, we pass by restatements of the panel dissenter's opinion, along with much of the current dissenters' rhetorical excess, and make only these points. First, the legislative provision on the proper forum for appellate review, 28 U.S.C. § 1295(a)(2), all will agree, is densely composed. As Judge Markey observed in his dissent from the panel decision, see Hohri v. United States, 782 F.2d 227, 260 (D.C. Cir. 1986), the section has been a source of confusion. See Professional Mangers' Association v. United States, 761 F.2d 740, 745 (D.C. Cir. 1985) (Section 1295(a)(2) "has recently been the source of much confusion in our court."). The differences of view presented in this case and others, compare Squillacote v. United States, 747 F.2d 432 (7th Cir. 1984), cert. denied, ___ U.S. ___, 105 S.Ct. 2021, 85 L.Ed.2d 302 (1985) (courts may depart from strict statutory language to promote statutory goals of judicial efficiency and fairness) with Professional Managers' Association, 761 F.2d at 745 (declining to adopt the Squillacote approach), suggest that Congress should attend to the technical amendment of section 1295(a)(2) that would obviate court conflicts.
The dissenters lean with a heavy hand on the district court's "lack of jurisdiction" to hear the tort claims appellants sought to present. We note that the flaw blocking presentation of the tort claims is a textbook illustration of "jurisdiction writ small." See United States v. Kember, 648 F.2d 1354, 1357-59 (D.C. Cir. 1980). We did not label "frivolous" "[a]ppellants' failure to grasp in full the distinction between [administrative complaint] filing requirements that are non-waivable and those that are subject to waiver on equitable grounds." Hohri, 782 F.2d at 240 n. 27. Whether we were correct or incorrect in that judgment, however, we certainly did not lay down "law of this circuit" that plaintiffs who plainly "lack standing," or present claims that are clearly "moot," or flout "the most basic procedural requirements" can nonetheless escape characterization of their case as "frivolous." But see diss. at 312.
On another day, in another case, the author of the present dissent from denial of rehearing en banc took his colleagues to task for extravagant attribution of significance to his panel opinion. See Dronenburg v. Zech, 746 F.2d 1579, 1582-84 (1984) (en banc) (separate statement of Bork, J.). We agreed with him on that occasion that the temptation to exaggerate a decision with which one disagrees, thereby to make it an easy target for slings and arrows, ought to be resisted.
We note finally the dissenters' apparent misperception of Restatement black letter. The Restatement (Second) of Judgments §§ 11, 12 (1982), discourages second hearings even on questions of "subject matter jurisdiction," and warns against expansive reading of the subsections set out by the dissenters. See diss. at 312. Suffice it to say that the question of D.C. Circuit or Federal Circuit review with which we deal entails no "[m]anifest defect" such as the granting of a divorce by a justice of the peace, a federal court entertaining what is plainly a common law tort action between citizens of the same state, intrusion upon the jurisdiction of a tribunal of legally superior authority, improper judicial interference with a non-judicial agency, or disturbance of other interests that genuinely warrant classification as "fundamental." See id. at § 11 comment e; § 12 Reporter's Note. Here again we believe that the dissenters have succumbed to the temptation to overstate and overwrite. We do not think our view of this singular case warrants the extravagant attack mounted against it.
This case should be reheard en banc. The panel majority has created an unprecedented rule of absolute deference to the political branches whenever "military necessity" is claimed, even where the claim is irrelevant and however spurious the claim is shown to be. The court did this, moreover, in a case in which it clearly had no jurisdiction. Both errors warrant reconsideration by the full court. I am in complete agreement with the criticisms of the majority opinion expressed in Chief Judge Markey's excellent dissent; I write separately to advance some additional grounds why the majority decision should not be allowed to stand.
Plaintiffs in this case are nineteen individuals, all of whom were either Japanese-Americans subjected to internment during World War II or the representatives of such internees. They sought money damages and a declaratory judgment on twenty-two claims, based upon a variety of alleged constitutional violations, torts, and breaches of contract and fiduciary duties. The district court dismissed each of these claims. The court of appeals affirmed, except as to one claim founded on the fifth amendment to the Constitution. With respect to that claim, virtually every step of the panel majority's reasoning either adopts broad and troublesome propositions or is plainly wrong.
I.
Plaintiffs alleged that the government internment program effected an uncompensated taking of their property. The statute of limitations requires that such claims be brought no later than six years after the right of action accrues. 28 U.S.C. § 2401(a) (1982). The alleged taking occurred approximately forty years before this lawsuit was filed. The district court properly held that the statute of limitations barred the claim. That conclusion would seem inescapable, but this court reversed and remanded.
A.
In an effort to escape the statute of limitations that plainly bars this action, the panel majority engaged in contrived reasoning that creates a rule of absolute and permanent judicial deference to any claim of "military necessity." Judges owe deference to such claims, of course, particularly in wartime, but never before has a court enunciated a deference so great that it requires utter capitulation. So sweeping is the panel majority's new rule, the executive branch may remove American citizens from their homes and impound them in camps, solely on the grounds of race, and courts will not interfere, no matter what facts are shown. So powerful is this rule that courts will not reexamine what was done even when facts establishing the absence of military necessity, or of any plausible belief in its existence, become public and the period of military emergency is long past. So potent is the rule that it applies to associated actions or neglects as to which no claim of military necessity was made or could be made.
I am certain that the majority intended none of this but that is what their argument inevitably leads to. It is easily demonstrated that I do not overstate the rule the panel majority has inadvertently created.
To summarize the majority's reasoning: In Hirabayashi v. United States, 320 U.S. 81, 63 S.Ct. 1375, 87 L.Ed. 1774 (1943), and Korematsu v. United States, 323 U.S. 214, 65 S.Ct. 193, 89 L.Ed. 194 (1944), the Supreme Court upheld the racially-based curfew and the internment regulations. The Court did so because it deferred completely to the judgment of the military authorities that the programs were justified by military necessity. The Court's acceptance of the claim of military necessity for these purposes also had the effect of vitiating any future claims for compensation that might arise under the fifth amendment, since "[w]hen the government impinges on property rights in the midst of a military emergency, there is no compensable taking." Hohri v. United States, 782 F.2d 227, 251 (D.C. Cir. 1986). However, according to plaintiffs' allegations, the government had concealed from the Supreme Court both internal memoranda disputing the necessity for the program and "the fact that there were no intelligence reports contradicting" those memoranda. Id. at 252. Had both the memoranda and the fact been disclosed, the outcomes in Hirabayashi and Korematsu would have been different and the finding of military necessity would not have been made. Therefore, the fraudulent concealment of the memoranda and the fact tolled the statute of limitations on the takings claim. Finally, since the Supreme court had based its reasoning on an irrebuttable presumption of deference to the political branches and the military, "nothing less than an authoritative statement by one of the political branches, purporting to review the evidence when taken as a whole, could rebut" this presumption. Id. at 251. The required statement was made in 1980 when Congress created a Commission to investigate the internment. Thus, it was not until 1980 that the basis for this takings claim existed, and, consequently, not until 1980 that the statute of limitations began to run.
In the course of this reasoning, the panel majority remade important law in more than one way. Hirabayashi and Korematsu are read to reflect an absolute deference to military judgment, though the Supreme Court did not express any such extreme position. Moreover, military necessity is said to justify uncompensated takings of private property even in the United States and well outside the battle zone, regardless of the fact that no one ever claimed that the takings, as opposed to the internment, were necessary. Third, the fact that Hirabayashi and Korematsu were decided during the height of World War II, a circumstance that must certainly figure in calculating their weight as precedent during peacetime, is overlooked in order that their supposed rationale of absolute deference may be made permanent, unless and until one of the political branches admits the absence of military necessity. Surely we must recognize that courts are likely to accord a claim of military necessity greater deference during a major war than would be proper years later when the emergency is long past and a conventional takings claim is advanced. Finally, courts may not reconsider prior holdings in light of new evidence until "released . . . from the grasp" of those holdings by Congress. Statement by Circuit Judges Wright and Ginsburg at 1. This means that in this context, at least, Congress may dictate the results of lawsuits to the courts.
The truth is that, had plaintiffs filed their claim earlier, they would have been able to use the relevant documents, most of which were already in the public domain, in building their case, as well as anything else accessible through discovery. As Chief Judge Markey pointed out, the essential facts for a legal challenge were well known by 1950. Hohri, 782 F.2d at 261-62 (Markey, C.J., dissenting). The government would have borne the burden of persuasion in establishing its affirmative defense and it would not have been able to meet that burden simply by citing Hirabayashi and Korematsu. See infra p. 8. It is only by announcing that no claim existed until Congress opened a new inquiry that the majority is able to justify tolling the statute of limitations until 1980.
In so doing, the panel has conducted something more than a mere historical analysis of the reasoning embodied in a pair of Court decisions from the 1940's. It has indicated that the doctrine of absolute deference applies today as well. As recently as 1979, we are told, plaintiffs had no case to bring, because the law laid down in Korematsu would have required automatic acquiescence to the expressed judgment of the political branches regardless of whatever factual evidence plaintiffs might have brought forth. Indeed, no set of facts in the public domain could possibly be sufficient to form the basis of a lawsuit in the absence of some sort of political retraction. The statement of Congress has therefore become a "crucial element" of the claim; without it, plaintiffs would have been unable to "survive a threshold motion to dismiss for failure to tender a claim that would advance beyond the pleading stage." Hohri, 782 F.2d at 249-50 n. 57.
And, if a statement by a political branch were required, as it clearly should not be, President Ford provided that in a Presidential Proclamation. See Proclamation No. 4417, 3 C.F.R. 8 (1977). The only difficulty is that the statement is inconveniently early, for the panel majority's purposes, since it was made more than six years before this suit was filed. Although the Presidential Proclamation describes the evacuation as "wrong" and a "national mistake," the panel majority found it insufficient to cure the concealment on the ground that it did not expressly declare that a legal wrong had been committed. This is the worst sort of hairsplitting. It is difficult to see how a declaration that the evacuation was "wrong" could fail to undercut the finding that it was "necessary." Moreover, if an "authoritative statement" is required, the clear language of this Proclamation fills the need far more naturally than the Act of Congress on which the panel majority relied.
This means that a claim of military necessity, once made and upheld, may never be challenged in court, no matter what the facts are proved to be, until a political branch states that the claim was known to be baseless when made. To make matters worse, the majority describes its rule of absolute and permanent judicial deference to claims of military necessity as resting on "constitutional underpinnings." Id. at 251.
This is one reason the majority's theory of fraudulent concealment seems so contrived. This doctrine, as the majority explains, concerns the concealment of "material facts," Hohri, 782 F.2d at 246 ( quoting Fitzgerald v. Seamans, 553 F.2d 220, 228 (D.C. Cir. 1977)), and permits the statute to be tolled only until "a `duly diligent' plaintiff would have discovered that which was concealed." 782 F.2d at 249. That doctrine was warped badly out of shape here, for what was "concealed," apparently, was an opinion by Congress, rather than any material facts or information. It is the source of the disclosure (Congress), and not the facts themselves, on which the majority's theory rests. Congress' failure to express earlier a formal judgment questioning the justification for the internment was not concealment, and its decision to express its doubts in 1980 has no effect on the statute of limitations.
This new doctrine confuses governmental decisions which warrant a degree of deference with those that are unreviewable. Until now, doctrines of deference to the Executive or Congress had never required the unthinking acceptance of unsupported assertions suggested by the majority opinion. Having now held that a statement by one of the political branches is a necessary element to any legal challenge to an assertion of military need, the majority has established a doctrine far more threatening to legitimate civil liberties and to judicial review of government action than any that would have been accomplished through an affirmance of the district court's decision.
B.
A word must be said about the panel majority's Statement in response to this dissent. The Statement's attempt to resuscitate the panel majority's original decision only makes that decision's error clearer. The heart of the panel majority's Statement is that the original opinion
most assuredly "creates [no] rule of absolute and permanent judicial deference to any claim of `military necessity.'" . . . Rather, the opinion simply describes and turns on what we find to be the situation-specific holding of Hirabayashi v. United States, 320 U.S. 81 [ 63 S.Ct. 1375, 87 L.Ed.2d 1774] (1943), and Korematsu v. United States, 323 U.S. 214 [ 65 S.Ct. 193, 89 L.Ed. 194] (1944): courts must defer to the judgment of Congress and the Executive that sufficient military necessity existed to justify the World War II internment policy. That Supreme Court holding seems to us clear, pin-pointed, and definite. We therefore concluded that the former internees faced an insuperable obstacle to the present suit until the "war-making branches," . . . released the federal courts from the grasp of Korematsu and Hirabayashi by indicating that deference was no longer due to the wartime judgment of military necessity for the mass evacuation.
Statement at 1.
This explanation of the original opinion will not do. If the majority's holding really turned on a "situation-specific holding" of Hirabayashi and Korematsu, which is "clear, pin-pointed, and definite," then those decisions would have posed no obstacle whatever to the bringing of this action ten, twenty, thirty, or even forty years ago. Hirabayashi upheld a curfew imposed upon persons of Japanese descent and Korematsu upheld their relocation and internment. Neither case holds, or even remotely suggests, that military necessity also required that the internees' property be taken. It is, in fact, perfectly apparent that the taking of property was not the object of, nor was it in any way necessary to, the relocation program. Therefore, on the rationale of the panel majority's Statement, these plaintiffs could have made a takings claim at any time without being in the least hampered, much less absolutely barred, by the holding of Hirabayashi and Korematsu.
The majority equates the showing required to prove necessity for the internment with the showing required to render a wartime taking noncompensable. These are two distinct inquiries. The cases cited by the majority on the latter issue each dealt with property deliberately destroyed by American troops in battle in order to keep it from falling into the hands of approaching enemy troops. United States v. Caltex, Inc., 344 U.S. 149, 73 S.Ct. 200, 97 L.Ed. 157 (1952); United States v. Pacific Railroad, 120 U.S. 227, 75 S.Ct. 490, 30 L.Ed. 634 (1887). There are other cases in which regulatory programs to ration or divert national resources in time of war have been held not to require compensation. See, e.g., United States v. Central Eureka Mining Co., 357 U.S. 155, 78 S.Ct. 1097, 2 L.Ed.2d 1228 (1958). Neither of those circumstances describes the taking alleged in Hohri. Each involves a situation in which the taking was itself deemed necessary to the war effort. That is not the case here. No one has claimed that the government, having decided to conduct the relocation, could not have protected the property rights of those it evacuated.
Any indication in the legislative history of the American-Japanese Evacuation Claims Act, 50 U.S.C. App., § 1981 et seq. (1982), that Congress did not believe compensation was required is of course irrelevant. The question of whether a taking has occurred is a purely legal one — unless the majority meant to suggest that absolute deference to the political branches is required in this context as well. See Hohri, 782 F.2d at 237-39.
It seems unlikely to me, therefore, that the government could have won a takings suit on the claim of military necessity. It clearly would have been impossible for the government to win such a suit simply by citing Korematsu and moving to dismiss. Indeed, in Central Eureka Mining Co., the Court emphasized that the question of whether a taking has occurred turns "upon the particular circumstances of each case." 357 U.S. at 168, 78 S.Ct. at 1104. The majority notes that certainty of success is not a necessary prerequisite to the running of the statute of limitations. Hohri, 782 F.2d at 253 n. 68. For the panel majority to persist in stating that any such claim would have foundered at the pleading stage, it must go so far as to hold that there was certainty of defeat. That cannot be true without the sort of broad holding it now denies having made.
That being so, one of two conclusions follows. Either the panel majority rests on a narrow view of Korematsu, as it now claims, which means that the statute of limitations has long since run and this case was wrongly decided, or the panel majority has indeed created a rule of absolute and permanent deference to a claim of military necessity. The deference must be so sweeping that any harm attendant upon the relocation, however unnecessarily inflicted and however unrelated itself to any military necessity, is also utterly immune from any lawsuit. A principle that broad, unfortunately, is essential to the result the panel majority reached. I thus do not exaggerate the holding of this case.
The panel majority claims that a confession of error by either Congress or the Executive was necessary before the present suit could be maintained. If a takings claim had been brought years ago, and if the court did not hold that anything done to persons subject to relocation was immunized by Hirabayashi and Korematsu, then the government would have been put to the proof of its defenses. As I have said, I cannot imagine that the government would have claimed that military necessity also required the loss of homes and businesses. Indeed, if the government reacted as it has now, it would have had to admit that there was no evidence supporting the claim of military necessity for the relocation, much less for the taking. Thus, even if the legal justification for the relocation were identical to that needed to render a taking noncompensable, which it is not, and a statement from the political branches necessary, which it is not, and the statement from President Ford irrelevant, which it is not, the statement by one of the "war-making branches" the panel majority requires could have been extracted through litigation. This means that this suit could have been brought successfully at any time within the past forty years and that the six-year limitations period has long since passed.
* * * * * *
I am entirely confident that the panel majority would not follow its rule of absolute deference should a similar circumstance arise in the future. That prediction is certain because the rule was created to rectify, so far as that can now be done, an injustice in the past. But, if that is true, the evasion of the statute of limitations stands revealed as unprincipled. Worse, there may be other times of emergency in our future, times when racial or ethnic animosities surface, and today's precedent will be available to any court reluctant to examine a claim of military necessity supported by popular passion. The panel majority has purchased freedom from the statute of limitations at an unacceptable price. A panel majority that so obviously disapproves of the wartime internment ought to have been more reluctant to create a legal basis for rendering any similar future incident forever unreviewable in any of its consequences.
II.
There are other grounds for rehearing this case as well. This court was without jurisdiction. The majority has completely reordered Congress' division of jurisdiction between the United States Court of Appeals for the Federal Circuit and the regional courts of appeals. The panel majority's conclusion that this court had jurisdiction over the appeal rested on its construction of the Federal Courts Improvement Act, 28 U.S.C. § 1295(a)(2) (1982), which provides that the United States Court of Appeals for the Federal Circuit shall have exclusive jurisdiction over an appeal from a final decision by a district court when
the jurisdiction of that court was based, in whole or in part, on section 1346 of this title, except that jurisdiction of an appeal in a case brought in a district court under section 1346(a)(1), 1346(b), 1346(e), or 1346(f) of this title or under section 1346(a)(2) when the claim is founded upon an Act of Congress or a regulation of an executive department providing for internal revenue shall be governed by sections 1291, 1292, and 1294 of this title.
28 U.S.C. §§ 1291, 1292, 1294 (1982) all provide for appeals to the regional courts of appeals, each provision noting as well, however, that the Federal Circuit retains jurisdiction in all cases so described in the Federal Courts Improvement Act. The jurisdictional controversy in Hohri arose from the fact that plaintiffs' original complaint included both a Takings Clause claim, under 28 U.S.C. § 1346(a)(2) (1982) (the Tucker Act), and a Federal Tort Claims Act claim, under 28 U.S.C. § 1346(b) (1982).
A.
The initial question is whether a suit based upon both the Tucker Act and the Federal Tort Claims Act must be appealed to the Federal Circuit or to a regional court of appeals. It is clear, both upon textual analysis and analysis of congressional policy, that this appeal belonged in the Federal Circuit. The majority concluded that while the general rule provides that the Court of Appeals for the Federal Circuit has exclusive jurisdiction over any appeal from a claim based "in whole or in part" on the Tucker Act — as this one was — the general rule is nevertheless inapplicable when the claim is also based in part on any of the provisions listed after the word "except" in the portion of the Federal Courts Improvement Act quoted above. Thus, the statute is read, most implausibly, to say that a suit based in whole or in part on the Tucker Act must be appealed to the Federal Circuit, unless it is also based in part on the Federal Tort Claims Act, in which case it must be appealed to one of the twelve regional courts of appeals. Though no coherent policy underlies such a jurisdictional scheme, this conclusion was based entirely on the majority's understanding of the "plain meaning" of the Federal Courts Improvement Act.
I believe the better, indeed the only plausible, interpretation of the "except" clause is that it states an exception, not, as the majority supposes, a new, independent, and superseding rule. The controlling sentence of the Federal Courts Improvement Act, read in its entirety, simply provides that a claim based in part on section 1346 — other than those parts of section 1346 listed in the "except" clause — may only be appealed to the Federal Circuit. Save when taxation is involved, the Tucker Act is not found in the "except" clause. Therefore, the appeal in this case, which is based in part on the Tucker Act and does not involve taxation, belonged in the Federal Circuit. The majority, however, interpreted the statute so that a Federal Tort Claims Act count overrides the presence of a Tucker Act count and affirmatively requires the case to be heard by the regional courts of appeals and not by the Federal Circuit. That stands the statute on its head.
The majority's reading also fails to explain the remarkable coincidence that all the provisions listed after the word "except" are subsections of the provision listed before the word "except" — section 1346. (This is true as well of the jurisdictional grant in the preceding paragraph of the Federal Courts Improvement Act, 28 U.S.C. § 1295(a)(1) (1982), which is similarly structured and provides for exclusive jurisdiction in the Federal Circuit for appeals arising under 28 U.S.C. § 1338 (1982), except for certain claims arising under section 1338(a).) If the purpose of the "except" clause, as the majority believed, were to list those provisions whose presence in a complaint Congress felt ought to shift review of a suit based in part on the Tucker Act to the regional courts of appeals, there is absolutely no reason to suppose that all those provisions would happen to be found within section 1346. Under the reading I suggest, of course, it would have made no sense to include any provision outside of section 1346 in the "except" clause, since that clause simply carves out of a general rule that applies by its terms to all of section 1346 those subsections of section 1346 to which that rule is not to apply.
An examination of the provisions listed in the "except" clause demonstrates how unlikely it is that Congress intended the meaning adopted in the majority opinion. Those provisions deal with tort claims against the government, suits to quiet title, and certain types of tax cases. One of the principal purposes of the Federal Courts Improvement Act was to centralize jurisdiction in one forum "over appeals in areas of law where Congress determines there is a special need for nationwide uniformity." S. Rep. No. 275, 97th Cong., 1st Sess. 2 (1981), reprinted in 1982 U.S. Code Cong. Ad. News 11, 12. Under the majority's interpretation, we must presume that the Congress thought nationwide jurisdiction and uniform decisionmaking with respect to Tucker Act claims to be so important that it provided that any case including such a claim would go, in its entirety, straight to the Federal Circuit, even though the case contains a due process claim, or an equal protection claim, or any of numerous other important constitutional and statutory claims — unless it also contains a claim to quiet title. Members of Congress must therefore have thought that the vesting of jurisdiction over appeals involving actions to quiet title in the twelve regional courts of appeals was so important that it overrode their clearly articulated desire to place Tucker Act appeals within the exclusive jurisdiction of the Federal Circuit. Such a jurisdictional scheme, as Chief Judge Markey pointed out in his dissent, would be "senseless." 782 F.2d at 257 (Markey, C.J., dissenting).
The majority opinion nonetheless seeks to bring some theoretical sense to its interpretation by suggesting that Congress placed Federal Tort Claims Act claims within the "except" clause because it "did not want to centralize adjudication of cases involving tort claims." 782 F.2d at 241 n. 30. That is undoubtedly correct, and it explains why Congress provided in the Federal Courts Improvement Act that a federal tort claim, standing alone, would be heard by the regional courts of appeals. Similarly, Congress apparently saw no need to require the centralization of actions to quiet title, or of those tax cases it exempted from the general rule by listing them in the "except" clause. The majority takes a substantial and unwarranted step, however, when it reasons that Congress not only "did not want to centralize" such claims, but affirmatively sought to decentralize them as well, to the point of sacrificing the principal goal of the Federal Courts Improvement Act — to centralize patent and Tucker Act claims by vesting exclusive jurisdiction over such appeals in the Federal Circuit. In so doing, the majority ignored language to the contrary in both the Senate and the House Reports accompanying the Federal Courts Improvement Act. The Senate Report expressly states that the Federal Circuit will have exclusive jurisdiction over " all patent appeals and all appeals in federal contract cases brought against the United States that are presently heard in the regional courts of appeals." S. Rep. No. 275, supra, at 7 1982 U.S. Code Cong. Ad.News, at 17 (emphasis added). The House Report formulates the rule much the way I do: the Federal Courts Improvement Act "gives the Court of Appeals for the Federal Circuit jurisdiction of any appeal from a trial court where the jurisdiction of the district court was based, in whole or in part, on section 1346 of title 28, United States Code, except 1346(a)(1) and (e) (tax appeals), 1346(b) (Federal Tort Claims), 1346(f) (quiet title actions), or 1346(a)(2) when the claim is founded upon an Act of Congress or a regulation of an executive department providing for internal revenue." H.R. Rep. No. 312, 97th Cong., 1st Sess. 42 (1981). This makes it clear that the claims in the "except" clause are those whose presence do not vest jurisdiction in the Federal Circuit and not, as the majority thinks, those whose presence removes jurisdiction that would otherwise vest in that court. These congressional descriptions of the workings of the Federal Courts Improvement Act are much more consistent with the interpretation I offer than with that suggested by the majority. I think those descriptions are the correct ones. For these reasons, and for those cogently stated by Chief Judge Markey, I believe that the panel decided this issue wrongly.
B.
The majority's disposition of the jurisdictional question was wrong for an additional reason: the federal tort claim, which under the majority's analysis was the sole reason for finding jurisdiction in this court to hear the appeal, was itself dismissed by the district court for lack of jurisdiction. Hohri v. United States, 586 F.Supp. 769, 793 (D.D.C. 1984). That dismissal was affirmed by the panel that decided this appeal. Hohri, 782 F.2d at 245. Our jurisdiction to hear the entire case therefore rested entirely on a claim over which we had no jurisdiction.
The majority's rationale is impossible to understand. It held that although the court was without jurisdiction to hear the federal tort claim, that claim was not frivolous, except for the lack of jurisdiction, and therefore could serve as the predicate for jurisdiction in this circuit over the remainder of the case. The majority never justified or supported its reliance on the presumed absence of frivolity in the tort claim, regardless of its jurisdictional deficiency, in asserting jurisdiction over the entire case by virtue of the presence of that claim in the original complaint.
The only case cited in support of this argument was Doe v. United States Department of Justice, 753 F.2d 1092 (D.C. Cir. 1985). As the majority presumably recognized, since it cited Doe with a cf., the case is probably inapposite. Doe involved the interpretation of a different statutory clause than did Hohri. Moreover, to the extent that it is relevant, it undermines the majority's frivolity-based rule, since Doe adopted a rule based on the presence or absence of jurisdiction. See Von Drasek v. Lehman, 762 F.2d 1065, 1069 (D.C. Cir. 1985) (noting that in Doe, "[b]ecause [the district court] did not have jurisdiction to hear the back pay claim, the jurisdiction of the district court could not have been based, even in part, on the Tucker Act").
In any event, the federal tort claim in this case was clearly frivolous. As the panel noted, the plaintiffs failed to comply with the explicit and mandatory statutory directive requiring that federal tort claims be filed first with the appropriate government agency. It may be true, as the majority stated, that there was no bad faith involved, although the basis for that conclusion escapes me. The claim remains frivolous, however, because no non-frivolous legal arguments may be made in its defense. That is why, I suppose, the frivolity-based rule as formulated by the majority permits jurisdiction in this court when the federal tort claim is not " substantively farfetched." Hohri, 782 F.2d at 240 n. 27 (emphasis added). It is thus now the law of this circuit that such claims can be predicates for jurisdiction when, as to that claim, plaintiffs lack standing, or the claim is moot, or the most basic procedural requirements are ignored, provided the merits of the underlying claim are not "farfetched." I cannot imagine a rationale for this inventive rule, save that it allowed the majority to decide this appeal.
C.
I think this decision was an unfortunate one, and would have preferred to see its plain errors corrected by this court sitting en banc. Since only five judges — one short of the necessary majority — voted to rehear, the task falls instead to the Supreme Court, or perhaps to the United States Court of Appeals for the Federal Circuit. As the majority indicates, any appeal from the proceedings on remand will be to the Federal Circuit, since the Tucker Act claim is all that remains. The majority suggested in dicta that its decision constitutes "law of the case," binding the Federal Circuit unless the panel of that court that hears the second appeal finds both "clear error" and "manifest injustice" in the prior opinion. Hohri, 782 F.2d at 241 n. 31. The majority failed to consider what effect the deficiency of its jurisdictional holding has on the respect due its holding on the merits by another circuit court. As the Supreme Court has explained, "[l]aw of the case directs a court's discretion, it does not limit the tribunal's power." Arizona v. California, 460 U.S. 605, 618, 103 S.Ct. 1382, 1391, 75 L.Ed.2d 318 (1983). This court has held that a decision by a prior panel that subject-matter jurisdiction exists over a case may be departed from by a later panel in the same case. There need have been no intervening changes in the facts or the law; it is sufficient justification that the second panel determines that jurisdiction is lacking. See Potomac Passengers Association v. Chesapeake O.R.R., 520 F.2d 91 (D.C. Cir. 1975). For much the same reason that a court will be unwilling to issue a judgment when it lacks jurisdiction itself, so too, it would seem, may it exercise its discretion to refuse to resolve a case on the basis of a prior opinion issued by another court without jurisdiction. Indeed, even within the doctrine of res judicata, which accords the court far less discretion and which generally gives preclusive effect even to judgments issued by courts without subject-matter jurisdiction, such effect will not be given when "[t]he subject matter of the action was so plainly beyond the court's jurisdiction that its entertaining the action was a manifest abuse of authority" or when "[a]llowing the judgment to stand would substantially infringe the authority of another tribunal." Restatement (Second) of Judgments § 12 (1982). I do not pretend to advise the Federal Circuit whether it should exercise its discretion to depart from the law of the case in circumstances such as those here. Should the proceedings on remand be appealed, the issue will undoubtedly be briefed and argued. I merely note that the matter is not at all as simple as the majority suggests, and that on this as on so many of the other issues resolved by the majority, I would have reached the opposite conclusion.
This case illustrates the costs to the legal system when compassion displaces law. The panel majority says it is not too late for justice to be done. But we administer justice according to law. Justice in a larger sense, justice according to morality, is for Congress and the President to administer, if they see fit, through the creation of new law. The wartime internment around which this case revolves is undeniably a very troublesome part of our history. It is within the authority of the political branches to make whatever reparations they deem appropriate, and it is my understanding that such legislation is presently under consideration. The issue of whether an additional remedy is available from a court, and, if so, which court, should only be resolved on the basis of a sober and fair assessment of the legal claims presented. When a court relies instead on a plainly deficient analysis, it fails to do justice to the parties before it, and inevitably establishes those deficiencies as precedent. The temptation to do so, in service of an attractive outcome, is often strong. The panel opinion in this case, which completely disrupts a carefully crafted jurisdictional scheme while establishing several unfounded and undesirable precedents as law, demonstrates why such temptations ought to be resisted.
Statement of Circuit Judge WRIGHT and Circuit Judge GINSBURG.