In the spring of 1942 the government of the United States forcibly removed some 120,000 of its Japanese-American citizens from their homes and placed them in internment camps. There they remained for as long as four years. When the constitutionality of this action was challenged in the Supreme Court the government justified its actions on the grounds of “military necessity.” The Supreme Court deferred. Nearly forty years later, a congressional commission concluded that the government’s asserted justification was without foundation. It is now alleged that this fact was concealed from the Supreme Court when it rendered its historic decision in Korematsu v. United States. Yet today, now that the truth can be known, the government says that the time for justice has passed. We cannot agree.
This suit was brought by nineteen individuals, former internees or their representatives, against the United States.1 They seek money damages and a declaratory judgment on twenty-two claims, based upon a variety of constitutional violations, torts, and breach of contract and fiduciary duties. The United States moved to dismiss for lack of subject matter jurisdiction under Federal Rule of Civil Procedure 12(b)(1). In support of its motion the United States cited the applicable statutes of limitations, sovereign immunity, and the alleged exclusivity of the American-Japanese Evacuation Claims Act. The District Court granted appellee’s motion to dismiss. Hohri v. United States, 586 F.Supp. 769 (D.D.C.1984) (Hohri). We now affirm in part and reverse in part, remanding the Takings Clause claims of those appellants who never received awards under the Claims Act for further proceedings.
I. Background
A. Exclusion and Internment
In the wake of Pearl Harbor the United States immediately took steps to improve security on the West Coast. Initially, attention focused on the activities of Japanese nationals. See Proclamation No. 2525, 6 Fed.Reg. 6321 (1941). Internment of these “enemy aliens” began at once. These precautions, however, did not satisfy the Commanding General of the Western Defense Command, Lt. General John L. DeWitt. In his Final Recommendation of the Commanding General, Western Defense Command and Fourth Army, to the Secretary of War (Feb. 14, 1942) (Final Recommendation), he urged the evacuation of all Japanese-American citizens from the Pacific coast. Joint Appendix (JA) 109-110. DeWitt reasoned:
The Japanese race is an enemy race and while many second and third generation Japanese born on United States soil, possessed of United States citizenship, have become “Americanized”, the racial strains are undiluted * * *. There are indications that these [Japanese-Americans] are organized and ready for concerted action at a favorable opportunity. The very fact that no sabotage has taken place to date is a disturbing and confirming indication that such action will be taken.
Final Recommendation, JA 109.
On February 18, 1942 DeWitt received legal authority to carry out his policy of racial exclusion. On that date the President signed Executive Order 9066, authorizing the Secretary of War or his designees to prescribe “military areas” from which any person could be excluded. 7 Fed.Reg. 1407, JA 112.2 . DeWitt designat*232ed California, western Oregon and Washington, and southern Arizona as “military areas.” In so doing, he declared that all persons of Japanese ancestry were to be excluded from these areas. At first, relocation proceeded on a voluntary basis.3 When this proved inefficient, compulsion replaced exhortation.
The evacuees were given as little as forty-eight hours notice of their impending removal. They were allowed to bring only what they could carry.4 In the assembly centers — racetracks and fairgrounds — the evacuees were placed in mass barracks housing 600 to 800 people. Beginning in May 1942 they were transferred to permanent relocation centers: camps surrounded by barbed wire and guarded' by military police. They were housed one or two families to a tar-paper room. They ate and bathed in mass facilities.
The majority of the evacuees remained in these camps for the duration of the war.5 According to the Commission on Wartime Relocation and Internment of Civilians (CWRIC),6 detention continued after military authorities concluded that there was no further military justification for the internment.7 Motivated by a desire to capture Western votes in the 1944 election, President Roosevelt refused to take any “drastic” action. Report of the Commission on Wartime Relocation and Internment of Civilians, Personal Justice Denied 229 (1982) (Personal Justice Denied). Finally, on November 10, 1944 the cabinet decided to end the exclusion; the War Department publicly rescinded the exclusion order on December 17, 1944. Administrative delay, however, prolonged detention for many. It was not until March 1946 that the last camp closed.
B. Deference and Concealment
In Hirabayashi v. United States, 320 U.S. 81, 63 S.Ct. 1375, 87 L.Ed. 1774 (1943), the Supreme Court considered the constitutionality of the curfew regulations imposed pursuant to Executive Order 9066. In Korematsu v. United States, 323 U.S. 214, 65 5. Ct. 193, 89 L.Ed. 194 (1944), the Court considered the constitutionality of the decision to exclude Japanese-Amerieans from the West Coast. In both cases the Court based its decision on the government’s allegations of military necessity. In these two cases the Court erected a virtually insurmountable presumption of deference to the judgment of the military authorities. Appellants allege, however, that the application of this deferential standard was marred by the fraudulent concealment of evidence indicating that there was no ra*233tional basis for the mass evacuation program.
1. Hirabayashi: concealment of evidence and deference to the judgment of the “war-making branches. ” The Department of Justice’s basic argument in Hirabayashi rested on two propositions. First, various cultural characteristics suggested that there was a serious potential for disloyalty by some members of the Japanese-American community. Hirabayashi, Brief for the United States at 18-31.8 Second, under the exigencies imposed by the military emergency, it was impossible to segregate the loyal from the disloyal. Id. at 61-63. This double-barrelled argument proved decisive. After reviewing the factors suggesting that members of the Japanese-American community might be disloyal, Chief Justice Stone concluded:
Whatever views we may entertain regarding the loyalty to this country of the citizens of Japanese ancestry, we cannot reject as unfounded the judgment of the military authorities and of Congress that there were disloyal members of that population, whose number and strength could not be precisely and quickly ascertained. We cannot say that the war-making branches of the Government did not have ground for believing that in a critical hour such persons could not readily be isolated and separately dealt with * * *.
320 U.S. at 99, 63 S.Ct. at 1385. The Court, however, did not purport to make an independent assessment of the evidence. As the Chief Justice indicated, the Court’s decision rested first and foremost on a pivotal constitutional assumption: that where matters of national security are at issue, the Court must defer to the judgment of the military and of Congress9 as the “war-making branches.”
As the Justice Department prepared its brief, however, Edward Ennis, the Director of the Alien Enemy Control Unit, came into possession of the intelligence work of one Lt. Commander Kenneth D. Ringle, an expert on Japanese intelligence in the Office of Naval Intelligence.10 Ringle had *234reached conclusions directly contradicting the two key premises in the government’s argument. Ringle argued that the cultural characteristics of the Japanese-Americans had not resulted in a high risk of disloyalty by members of that group.11 Moreover, Ringle expressly concluded that individualized determinations could be made expeditiously:
[T]he entire “Japanese Problem” has been magnified out of its true proportion, largely because of the physical characteristics of the [Japanese] people * * *. [I]t should be handled on the basis of the individual, regardless of citizenship, and not on a racial basis.
K. Ringle, Report on the Japanese Question 3 (Jan. 26, 1942) (Ringle Report), JA 93 (emphasis in original).12
Ennis knew that Ringle’s views could not be dismissed as those of a solitary dissident; for Ennis had been informed that Ringle’s views were shared by his superiors at Naval Intelligence. E. Ennis, Memorandum for the Solicitor General (April 30, 1943) (Ennis I) at 2, JA 116. Ennis also knew that the Army and Navy had previously agreed that Naval Intelligence would assume responsibility for the Japanese issue.13 Nor did Ennis question the reliability of Ringle’s report; on the contrary, he found Ringle’s report the “most reasonable and objective discussion of the security problem presented by the presence of the Japanese minority” of all of the “great numbers of reports, memoranda, and articles” that he had perused over the previous year. Id. at 3, JA 117. And Ennis fully understood that Ringle’s conclusions directly undermined the government’s case.14 He therefore concluded:
I think we should consider very carefully whether we do not have a duty to advise the Court of the existence of the Ringle memorandum and of the fact that this represents the view of the Office of Naval Intelligence. It occurs to me that any other course of conduct might approximate the suppression of evidence.
Ennis I at 4, JA 118.
Notwithstanding Ennis’ plea, the Justice Department’s brief made no mention of Ringle’s analysis.15 Equally important, it *235is now apparent that there were no countervailing professional intelligence analyses justifying the need for a mass evacuation based on race.16 Thus the CWRIC concluded in 1982 that political pressure, not official intelligence analysis, produced the evacuation, that' “[ijntelligence opinions were disregarded or drowned out,” PERSONAL JUSTICE DENIED at 60, and that “[t]he promulgation of Executive Order 9066 was not justified by military necessity * * Id. at 18.
Mere disclosure of Ringle’s analysis to the Court, without more, would not likely have changed the result in Hirabayaski.17 But disclosure combined with a concession that the government had no data rebutting Ringle’s analysis would likely have influenced the outcome. And taken together, the suppression of the Ringle report and the absence of countervailing data suggest that the Justice Department misled the Supreme Court when it argued that “military necessity” justified a mass evacuation of Japanese-American citizens.18
2. Korematsu: the presumption of deference becomes nearly irrebuttable. In preparing its Korematsu brief the Justice Department simply followed the path cut by Hirabayaski. See Korematsu, Brief for the United States at 11-12, 26. Similarly, in upholding the evacuation the Korematsu Court simply reiterated the Hirabayashi rationale: time was short, the situation grave, and it was impossible readily to distinguish the loyal from the disloyal. 323 U.S. at 218-19, 65 S.Ct. at 194-95.19
In Korematsu, however, unlike Hirabayashi, the litigants provided the Court with a wealth of factual material attacking the factual predicates of the government’s argument. See, e.g., Korematsu, Brief of Japanese-American Citizens League. Yet for the majority the presumption of deference to the “war-making branches,” articulated in Hirabayaski, settled the matter. 323 U.S. at 218-19, 65 S.Ct. at 194-95.
By 1944 the Court could rest its presumption of deference to the military judgment on seemingly firmer ground than had been available in Hirabayaski. In the interim the War Department had issued an *236official analysis of the exclusion and internment program, General DeWitt’s Final Report, Japanese Evacuation from the West Coast, 1942 (1943) (Final Report), supplying “facts” supporting the conclusions of the Final Recommendation. Although much of the Final Report addressed the issue whether members of the Japanese-American community had actually engaged in espionage or sabotage, the Report did purport to provide factual support for the key premises of the Hirabayashi decision: there was widespread disloyalty in the Japanese-American community and it was impossible to separate the loyal from the disloyal in an efficient manner. See Hohri, 586 F.Supp. at 777.
Recently uncovered documents, however, suggest that the Justice Department was less than fully candid in revealing to the Court the untrustworthy character of the Final Report. For example, the Final Report alleged that Japanese-Americans had been engaged in shore-to-ship radio and light signaling to Japanese warships, facilitating attacks on American ships or shore installations. Id. at 4. By the spring of 1944, however, the Attorney General had learned the allegations of shore-to-ship signaling were baseless. See Letter from FCC Chairman Fly to Attorney General Biddle (April 4, 1944), JA 101-104 (noting that the evacuation appeared to have no effect on radio signaling); Burling, Memorandum for the Attorney General (April 12, 1944), JA 119 (discussing letter of FBI Director Hoover on shore-to-ship signaling). Once again, Ennis had demanded full disclosure and had drafted a footnote for the government’s brief to that effect, reading:
The Final Report of General DeWitt (which is dated June 5, 1943, but which was not made public until January 1944) is relied on in this brief for statistics and other details concerning the actual evacuation and the events that took place subsequent thereto. The recital of circumstances justifying the evacuation as a matter of military necessity, however, is in several respects, particularly with reference to the use of illegal radio transmitters and to shore-to-ship signalling by persons of Japanese ancestry, in conflict with information in the possession of the Department of Justice. In view of the contrariety of the reports on this matter we do not ask the Court to take judicial notice of the recital of those facts contained in the report.
Quoted in Hohri, 586 F.Supp. at 780. After heated negotiations with attorneys for the War Department, see E. Ennis, Memorandum for Mr. Herbert Wechsler (Sept. 30, 1944) (Ennis II), JA 120, however, Justice merely inserted the following, ambiguous, footnote in its brief:
The Final Report of General DeWitt (which is dated June 5, 1943 but which was not made public until January 1944), hereinafter cited as Final Report, is relied on in this brief for statistics and other details concerning the actual evacuation and the events that took place subsequent thereto. We have specifically recited in this brief the facts relating to the justification for the evacuation, of which we ask the Court to take judicial notice, and we rely on the Final Report only to the extent that it relates to such facts.
Korematsu, Brief for the United States at 11 n. 2. Thus the final footnote did not adequately alert the Justices to the lack of empirical data supporting the government’s claims.20 For the Korematsu majority, DeWitt’s statement was the official view of one of the “war-making branches,” 323 U.S. at 218, quoting Hirabayashi, 320 U.S. at 99, 63 S.Ct. at 1385. As noted in Hirabayashi, in times of military emergency the Court believed that such “war-making branches” need only have a “rational basis” for race-related classifications. 320 U.S. at 102, 63 S.Ct. at 1386.21 And the *237mere fact that the military’s conclusions were hotly disputed, see, e.g., Korematsu, Brief of Japanese-American Citizens League, did not make them irrational. Finally, the fact that Congress had “re-posted] its confidence in this time of war in our military leaders — as inevitably it must,” Korematsu, 323 U.S. at 223, 65 S.Ct. at 197, left little room for judicial reevaluation.22
Thus in Korematsu the Court crystallized the presumption of deference first articulated in Hirabayashi. Once again, the application of this presumption was marred by a failure on the part of the Justice Department to disclose the questionable credibility of the War Department pronouncements. The Court effectively announced that given this presumption of deference no mere incremental evidentiary showing could change its view of the case. Indeed, given the constitutional underpinnings of the Court’s holding, it would appear that only a statement by one of the political branches, purporting to assess the evidence as a whole, could have altered the result.
C. Extension of the Rule in Korematsu to Claims for Compensation
In 1948 Congress enacted the American-Japanese Evacuation Claims Act, 50 U.S.C. App. § 1981 et seq. (1982) (hereinafter the Claims Act). Under the Act the Attorney General was given jurisdiction to determine claims for “damage to or loss of real or personal property” filed by former evacuees that were a “reasonable and natural consequence of the evacuation * * 50 U.S.C.App. § 1981. The Act provided for specific limitations on the types of compensable losses for which claims could be filed.23 All awards were deemed “final and conclusive for all purposes.” 50 U.S.C. App. § 1984(d).
The Claims Act, however, was not passed in recognition of a legal wrong inflicted on *238the evacuees. On the contrary, the history of the Act reveals that Congress believed it was acting out of moral impulse, not legal obligation. Congress thereby signified its belief that although the Korematsu holding may only have applied to the validity of a criminal conviction, the Korematsu rationale effectively barred all claims for compensation as well.
The basic justification for the Act was provided in a 1947 letter written by the Secretary of the Interior to the Speaker of the House. This letter was incorporated in the House report, H.R.Rep. No. 732, 80th Cong., 1st Sess. (1947). It provided the sole explanation for the House bill, H.R. 3999, and provided the following insight into the contemporaneous view of the prevailing legal rights of the internees:
The only clear recourse which the evacuees now have, through the passage of private relief bills, is totally impractical. To provide for adjudication of the claims by the Court of Claims would be an imposition on that court, because of the small individual amounts involved and the potential volume of claims * * *.
H.R.Rep. No. 732, supra, at 3.24
In suggesting that the only “clear recourse” then available was through the passage of private bills, the House report indicated that the Committee did not believe the evacuees could state an actionable claim. Similarly, by rejecting the suggestion that the Congress vest jurisdiction in the Court of Claims the report suggests that the Court of Claims did not already have jurisdiction to hear such claims under the Tucker Act.25
This view was reaffirmed in the subsequent history of the Claims Act. In 1951 Congress amended the Act to allow the Attorney General to settle claims up to $2,500. Both the House and Senate reports affirmed that a perception of “military necessity” supported the evacuation. See S.Rep. No. 601, 82d Cong., 1st Sess. 2 (1951); H.R.Rep. No. 496, 82d Cong., 1st Sess. 2 (1951). The House Report once again reprinted the letter of the Secretary of the Interior, restating the view that the evacuees had no cognizable claims absent the Claims Act. See id. at 2-3.
In 1956 Congress amended the Claims Act for the last time, allowing the Attorney General to settle claims up to $100,000 and giving the Court of Claims jurisdiction over contested claims. Here the legislative history did not directly address the question of the civil liability of the United States absent the Claims Act. The only reference to this issue can be found in the House report, which merely referred to the legislative history of the 1948 Act itself. See H.R.Rep. No. 1809, 84th Cong., 2d Sess. 3 (1956). Thus, to the extent that they addressed the issue at all, the 1956 amendments indirectly evince a continuing belief in the legality of the evacuation policy.
Finally, in administering the Act the Attorney General took the position that the Claims Act was not predicated on the view that the evacuees had suffered a legal wrong. Thus in the leading case of Claim of Mary Sogawa, 1 Adjudications of the Attorney General 126 (December 20, 1950), the Attorney General explicitly rejected a *239claim for compensation for the expenses entailed by the claimant in preparing for evacuation and in obtaining return transportation. In reaching this decision the Attorney General expressly considered and rejected the view that the Claims Act was premised on the notion that the evacuees had suffered an actionable wrong. The opinion concluded:
The foregoing discussion of the legislative history of the Evacuation Claims Act makes it clear, we believe, that it was intended to be an act of bounty * * *. [I]t may not be adjudicated as if the claimant’s evacuation constituted a legal wrong, in the teeth of the decision of the Supreme Court in the Korematsu case, supra, to the contrary.
Id. at 134.
Thus the “war-making branches” once again reaffirmed their belief that military necessity had provided a legal justification for the exclusion program. And in no uncertain terms the Attorney General and Congress had concluded that Korematsu not only applied to a criminal conviction but that it also effectively barred claims for compensation arising out of the evacuation program.26
The foregoing narrative establishes three points relevant to our analysis. First, the government’s suppression of critical evidence in the Hirabayashi case contributed to the Supreme Court’s conclusion that it must defer to the judgment of Congress and the military authorities that the exclusion program was justified by military necessity. Second, Korematsu suggests that the mere presentation of facts contradicting the government’s claims could not rebut this presumption of deference; only an official statement by one of the political branches, purporting to assess the evidence when viewed as a whole, could carry that burden. Third, congressional action signalled a general assumption that Korematsu not only barred challenges to criminal convictions but applied to civil claims as well. It is against this backdrop that we evaluate the legal contentions of the parties to this suit.
II. Apellate Jurisdiction
Before turning to the merits, we are required to consider whether this court can take appellate jurisdiction over this case. 28 U.S.C. § 1295(a)(2) (1982) provides that the United States Court of Appeals for the Federal Circuit shall have exclusive jurisdiction
of an appeal from a final decision of a district court of the United States * * * if 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 * * * shall be governed by sections 1291, 1292, and 1294 of this title * * *.
(Emphasis added.)
Thus Section 1295(a) establishes a general rule that where original jurisdiction is based “in whole or in part” on Tucker Act claims {i.e., on Section 1346(a)(2)), the Federal Circuit has exclusive jurisdiction. But Section 1295(a)(2) also provides for an exception to this general rule. Specifically, Section 1295(a)(2) provides that where original jurisdiction is based, inter alia, on Federal Tort Claims Act (FTCA) claims {i.e., on 1346(b)) the general rule — that the Federal Circuit has *240exclusive appellate jurisdiction over all cases where original jurisdiction was based “in whole or in part” on Section 1346 — does not apply.
In this case jurisdiction in the District Court was based, in part, on Section 1346(a)(2) Tucker Act claims. As an initial matter it might therefore seem proper to apply the general rule stated in Section 1295(a) and assume that the Federal Circuit has exclusive appellate jurisdiction. But original jurisdiction in this case was also based on Section 1346(b) FTCA claims.27 This case therefore falls squarely within the “except” clause of Section 1295(a)(2), allowing for appellate jurisdiction in the regional Courts of Appeals.28
Appellee argues that the “except” clause should be read to provide appellate jurisdiction in the regional Courts of Appeals only in cases where jurisdiction is based solely on Section 1346(b). Brief of appellee at 63. Appellee argues that such a reading would render Section 1295(a)(2) “in accord” with Section 1295(a)(1). But as a comparison of subsections (1) and (2) of Section 1295(a) demonstrates, appellee’s argument proves too much.
In subsection (1) Congress indicated that the Federal Circuit would have appellate jurisdiction where original jurisdiction in the District Court was based “in whole or in part” on Section 1338(a) (providing jurisdiction for cases involving patent, copyright, trademark). As appellee notes, subsection (1) also includes an exception. This exception concerns those 1338(a) claims relating to copyrights or trademarks. But the “except” clause in subsection (1) does not contain the same words as the “except” clause in subsection (2). In subsection (1) Congress explicitly stated that the regional Courts of Appeals would only have appellate jurisdiction where the claims related to *241“copyrights or trademarks and no other claims under Section 1338(a)” (emphasis added).29 By contrast, subsection (2) does not limit the “except” clause to cases where jurisdiction is based on FTCA claims and “no other claims” under Section 1346. Given the proximity of subsection (1) to subsection (2), the absence of the phrase “and no other claims” is conspicuous indeed.
It seems that where Congress desired to craft a narrow exception, preventing the regional Courts of Appeals from hearing cases with mixed jurisdictional bases, it knew how to unambiguously effectuate its will: it included the phrase “and no other claims.” On the other hand, where Congress intended to craft a broad exception, allowing the regional Courts of Appeals to hear appeals of cases', with mixed jurisdictional bases, it also knew what to do: it simply dropped the words “and no other claims” from the terms of the “except” clause.30 The “except” clause governing our case is of the broader variety. We take appellate jurisdiction accordingly.31
III. Standard of Review
In deciding a motion to dismiss on the pleadings for want of subject matter jurisdiction “the allegations of the complaint should be construed favorably to the pleader.” Scheuer v. Rhodes, 416 U.S. 232, 236, 94 S.Ct. 1683, 1686, 40 L.Ed.2d 90 (1974). See also Walker v. Jones, 733 F.2d 923, 925-26 (D.C.Cir.1984). The District Court, however, is not limited to the allegations of the complaint in deciding a Rule 12(b)(1) motion. Here the District Court properly relied on extra-pleading material in deciding the motion. 5 C. Wright & A. Miller, Federal Practice and Procedure § 1350 at 549-550 & n. 77 (1969 & 1985 Supp.) (collecting citations).
The District Court, however, did not purport to make any factual findings on disputed issues. See Hohri, 586 F.Supp. at 773. To the degree it relied on extra-pleading material it did so only where such documents supplied undisputed facts. See, e.g., id. at 788 (relying on the “undisputed” facts in the Ennis and Burling memoranda to establish fraudulent concealment). In such circumstances we engage in an independent review of the legal sufficiency of the District Court’s views and of its application of the law to undisputed facts in the historical record. See Williamson v. Tucker, 645 F.2d 404, 413 (5th Cir.1981). In so *242doing we construe the allegations of the complaint most favorably to the appellants unless such allegations are contradicted by the undisputed historical documents on which the District Court based its judgment.
IV. Sovereign Immunity
It is well settled that the United States is amenable to suit only in those instances where it has specifically waived its immunity. Two such waivers are alleged in this case: the Tucker Act, 28 U.S.C. § 1346(a)(2) (1982),32 and the Federal Tort Claims Act, 28 U.S.C. § 2671 et seq. (1982). Although we find that the Tucker Act does provide a waiver for appellants’ claims founded upon the Takings Clause and upon contract, it appears that sovereign immunity bars the residue of appellants’ monetary claims.33
A. Waiver Under the Tucker Act
The Tucker Act waives sovereign immunity only for those claims founded on statutes, regulations, contracts, or provisions of the Constitution that create substantive rights to money damages. United States v. Mitchell, 463 U.S. 206, 216-17,103 S.Ct. 2961, 2967-68, 77 L.Ed.2d 580 (1983).34 Whether the Tucker Act waives sovereign immunity therefore turns on whether plaintiff’s claims are based on a statute, regulation, contract, or constitutional provision that “can fairly be interpreted as mandating compensation by the Federal Government for the damage sustained.” United States v. Testan, 424 U.S. 392, 400, 96 S.Ct. 948, 954, 47 L.Ed.2d 114 (1976) (quoting Eastport S.S. Corp. v. United States, 372 F.2d 1002, 1009 (0.0.1967)). We must therefore review each of appellants’ non-tort claims to determine which, if any, are based on statutes, constitutional provisions, contracts, or regulations that demand monetary compensation.
1. The Takings Clause claims. As the District Court noted, appellants’ Takings Clause claim “is in essence an inverse condemnation proceeding, in which a citizen is deprived of property by the government and then must initiate judicial action to obtain just compensation.” 586 F.Supp. at 783. It is well established that “an individual claiming that the United States has taken his property can seek just compensation under the Tucker Act * * * Ruckelshaus v. Monsanto Co., — U.S. -, 104 S.Ct. 2862, 2880, 81 L.Ed.2d 815 (1984). Given the alleged damage to appellants’ real and personal property directly caused by the evacuation program, there is no question that appellants have stated a claim cognizable under the Takings Clause.35
*243Appellee, however, argues that actions taken pursuant to a “perceived need to protect the national security” cannot constitute a taking. Brief of appellee at 57. There is no legal support for this proposition.36 Only a showing of actual (and not merely imagined) military emergency vitiates a Takings Clause claim. United States v. Caltex, 344 U.S. 149, 73 S.Ct. 200, 97 L.Ed. 157 (1952). Here the gravamen of appellants’ claim is that there was no such military emergency. The District Court concluded that, given the procedural posture of this case, the allegations of appellants (as plaintiffs below) were dispositive. We agree. See Scheuer v. Rhodes, supra.37
2. Contract claims. Appellants allege breach of express contracts, both oral and written, contracts implied in fact and contracts implied in law. Complaint at 67-68 If 133, JA 73-74. These contracts allegedly concerned the nature of detention, the services (including bailment) to be provided them during detention, and specific protections to be accorded the internees. The .contracts allegedly arose from promises 'made by the relevant authorities.and from official conduct.
The Tucker Act, however, waives '.sovereign immunity only for express contracts and contracts implied in fact. There | is no waiver for contracts implied in law or 'contracts based on equitable principles. See United States v. Mitchell, supra, 463 U.S. at 218, 103 S.Ct. at 2968-69. Consequently, only appellants’ claims for breach | of express contracts and contracts implied-in-fact appear to survive this threshold bar.
3. Fiduciary duty claims. By contrast, appellants’ fiduciary duty claims are barred by sovereign immunity. Appellants allege that the “statutes, regulations and orders” promulgated by the United States “established a system of comprehensive and pervasive federal control, management, and supervision” over the daily lives of the internees. Complaint at 68 If 134, JA 74. Appellants argue that such a fiduciary duty included an obligation to deal truthfully with the evacuees and that appellee breached its duty by failing to disclose the lack of military necessity for the evacuation. See Complaint at 69 If 135, JA 75.
Appellants’ argument is reducible to the proposition that whenever the United States imposes such a pervasive regulatory scheme it necessarily enters into a fiduciary relationship with the individuals whose lives it supervises. Brief of appellants at 42-43. Appellants cite Mitchell to support this proposition. We do not read Mitchell to go so far.
Mitchell construed the clause of the Tucker Act that waives sovereign immunity for claims founded on statute or regulation. 463 U.S. at 218, 103 S.Ct. at 2968-69. The Court held that this provision operated to waive sovereign immunity for claims of breach of fiduciary duty where specific statutes or regulations gave rise to the fiduciary duty in question. Mitchell, however, found that the relevant statutes and regulations, by their own terms, explicitly created a fiduciary relationship by requiring the Secretary of the Interior to manage the Quinalt Indians’ assets for the “best interests of the Indian owner * * *.” Id. at 224, 103 S.Ct. at 2972 (quoting 25 U.S.C. § 406(a) (1982)).38
*244In our case there are no analogous statutes or regulations. It is true that the government did sometimes speak of acting for the benefit of the evacuees. See, e.g., Plaintiffs Exhibit Q, War Relocation Authority Tentative Policy Statement, JA 141-146. Within this context the government may have undertaken to treat the internees in a responsible manner. But even assuming, without deciding, that the applicable regulations could be construed to create specific duties to the evacuees, such duties must be distinguished from a comprehensive obligation to provide for the “best interests” of the evacuees. We are reluctant to find that such a distinct, overarching duty is implicit in a narrower set of regulatory obligations.39
Appellants also rely on Juda v. United States, 6 Cl.Ct. 441 (1984). In Juda the court found a tacit contractual commitment by the United States to act as fiduciary for Bikini Islanders whom the United States removed from their atoll in 1946 while the government tested nuclear bombs on that site. Id. at 452. Our case is plainly distinguishable. Unlike Juda, appellants here have not alleged that the United States contracted, even tacitly, to act as a fiduciary. Their fiduciary duty argument is based solely on regulatory obligations. Complaint at 68-69 11134, JA 74-75. Moreover, even if, arguendo, the United States did enter into a contractual relationship with the evacuees, it was a contract to provide specific services. Just as we are loath to impute a regulatory commitment to act as a fiduciary on the basis of alleged narrow regulatory obligations, we are also reluctant to infer a broad contractual commitment to act as fiduciary on the basis of an alleged contract to provide specific services.
4. Other constitutional claims. Plaintiffs also allege sundry violations of their constitutional rights under the Due Process,40 Equal Protection, and Privileges and Immunities Clauses of the Fifth Amendment; the Search and Seizure Clause of the Fourth Amendment; the Cruel and Unusual Punishment Clause of the Eighth Amendment; the rights to fair trial and counsel under the Sixth Amendment; the Press, Speech, Religion, Petition, and Assembly Clauses of the First Amendment; the prohibition of Bills of Attainder and Ex Post Facto Laws and the right to the writ of habeas corpus under Art. I, Section 9; and the protection from involuntary servitude under the Thirteenth Amendment. Complaint at 60-66 ¶¶ 112-113, 116-127, JA 66-72. We find that sovereign immunity bars all such claims.
Appellants allege that the Tucker Act’s declaration that the United States is amenable to suit in actions “founded upon the Constitution” waives sovereign immunity for all of their constitutional claims. Brief of appellants at 47; reply brief of appel*245lants at 17. The law of this circuit41 and of other circuits42 is to the contrary.
Appellants argue, however, that because some of these constitutional provisions have been found to mandate compensation in Bivens actions against individual defendants, this court ought to find that they also mandate compensation in an action against the United States. Brief of appellants at 48, 50-51. This circuit has rejected that view. See Clark v. Library of Congress, 750 F.2d 89, 103 (D.C.Cir.1984); Monarch Ins. Co. of Ohio v. District of Columbia, 353 F.Supp. 1249, 1254 (D.D.C. 1973), aff'd, 497 F.2d 684 (D.C.Cir.), cert. denied, 419 U.S. 1021, 95 S.Ct. 497, 42 L.Ed.2d 295 (1974). See also Garcia v. United States, 666 F.2d 960, 966 (5th Cir.), cert. denied, 459 U.S. 832, 103 S.Ct. 73, 74 L.Ed.2d 72 (1982).43
B. Waiver Under the Federal Tort Claims Act
Appellants allege a series of common law44 torts, see Complaint at 66-67 ¶¶ 129-131, JA 72-73,45 for which they claim the Federal Tort Claims Act, 28 U.S.C. § 2671 et seq. (1982), waives sovereign immunity. Appellants’ failure, however, to comply with the unyielding administrative filing requirements of the FTCA bars their claims.
Under 28 U.S.C. § 2675(a) a plaintiff must file his claim with the appropriate government agency before bringing suit in federal court. This explicit statutory directive applies without exception and therefore has been termed “jurisdictional.” See Odin v. United States, 656 F.2d 798, 802 (D.C.Cir.1981). The FTCA’s mandatory administrative filing requirement is not to be confused with the prudential, judge-made exhaustion doctrine, or other requirements that indicate a general, but not an inexorable, rule. Unlike the exhaustion requirement, the jurisdictional FTCA filing requirement is not subject to equitable waiver.46 Moreover, whatever the equities *246affecting appellants’ claims before 1980, there was no reason why appellants should have failed to file their claims after 1980 and the congressional declaration releasing the courts from their presumption of deference to the findings of the political branches in this case.47 Appellants FTCA claims therefore must be dismissed for failure to meet the statute’s stringent “file first with the agency” instruction.48
V. Statute of Limitations
28 U.S.C. § 2401(a) (1982) is the statute of limitations governing appellants’ Taking Clause and contract claims. It provides that a claim must be filed within six years of the time that the “right of action first accrues.” Appellee argues that appellants’ cause of action first “accrued” when appellants’ were first subjected to the evacuation program. Brief of appellee at 16. For their part appellants argue that because the government fraudulently concealed essential elements of their cause of action the statute of limitations was tolled until they actually discovered the facts that had been concealed. Brief of appellants at 28. The law of this circuit supports neither view. Instead, our cases hold that when a defendant fraudulently conceals the basis of a plaintiff’s cause of action, the statute of limitations is tolled until the time that a reasonably diligent plaintiff could have discovered the elements of his claim. Applying this standard to the case at bar, we hold that although appellants’ contract claims are barred by the statute of limitations, appellants’ Takings Clause claims were timely filed. A. The Due Diligence Doctrine
1. The applicable rule. In Fitzgerald v. Seamans, 553 F.2d 220, 228 (D.C.Cir. 1977), this court stated:
Read into every federal statute of limitations * * * is the equitable doctrine that in the case of defendant’s fraud or deliberate concealment of material facts relating to his wrongdoing, time does not begin to run until plaintiff discovers, or by reasonable diligence could have discovered, the basis of the lawsuit.
The due diligence doctrine was reiterated in Richards v. Mileski, 662 F.2d 65, 71 (D.C. Cir.1981), where Judge Mikva, writing for the court, noted that the fraudulent concealment of a plaintiff’s “cause of action” would toll the statute of limitations until a plaintiff has, or through due diligence should have had, notice of his claim. See also Smith v. Nixon, 606 F.2d 1183, 1191 (D.C.Cir.1979). More recently, in Hobson v. Wilson, 737 F.2d 1, 35 (D.C.Cir.1984), cert. denied, — U.S.-, 105 S.Ct. 1843, *24785 L.Ed.2d 142 (1985), Judge Edwards refined this standard by stating that fraudulent concealment would toll the statute of limitations until a plaintiff could have discovered “facts giving notice of the particular cause of action at issue, not of just any cause of action.”
Appellee argues, however, that the due diligence doctrine is not applicable to this case because fraudulent concealment cannot toll a statute of limitations governing claims against the United States. Brief of appellee at 20. And although Fitzgerald declared that the doctrine of tolling for fraudulent concealment must be read into “every” statute of limitations, 553 F.2d at 228, this court has not previously addressed the question of whether “every” statute of limitations necessarily includes statutes of limitations governing claims against the United States.
Appellee largely rests its argument on Block v. North Dakota ex rel. Board of University and School Lands, 461 U.S. 273, 287-88, 103 S.Ct. 1811, 1819-20, 75 L.Ed.2d 840 (1983), and Soriano v. United States, 352 U.S. 270, 276, 77 S.Ct. 269, 273,1 L.Ed.2d 306 (1957). These cases firmly establish the proposition that statutes of limitations governing claims against the United States, as conditions on waivers of sovereign immunity, are to be strictly construed. Fully aware of this principle, we nonetheless believe that fraudulent concealment tolls 28 U.S.C. § 2401(a) (1982), the statute of limitations at issue in this case.
An analysis of the historical background of 28 U.S.C. § 2401(a) supports the view that fraudulent concealment does toll the statute. Long before the predecessor to Section 2401(a) was first enacted in 1863, 12 Stat. 765 (37th Cong., 3d Sess. March 3, 1863), a majority of United States jurisdictions has held that a defendant’s subsequent concealment of a fraud would toll the statute of limitations. See Bailey v. Glover, 88 U.S. (21 Wall.) 342, 348-49, 22 L.Ed. 636 (1875) (collecting cases).49 Not surprisingly, the Supreme Court has held that “[t]his equitable doctrine is read into every federal statute of limitation.” Holmberg v. Armbrecht, 327 U.S. 392, 397, 66 S.Ct. 582, 585, 90 L.Ed. 743 (1946). Several federal Courts of Appeals have therefore held that fraudulent concealment by the United States will toll the statute of limitations. See, e.g., Barrett v. United States, 689 F.2d 324, 329-30 (2d Cir.1982), cert. denied, 462 U.S. 1131, 103 S.Ct. 3111, 77 L.Ed.2d 1366 (1983); Welcker v. United States, 752 F.2d 1577, 1580 (Fed.Cir.1985). See also Japanese War Notes Claimants Ass’n of the Philippines, Inc. v. United States, 178 Ct.Cl. 630, 373 F.2d 356, 358-359, cert. denied, 389 U.S. 971, 88 S.Ct. 466, 19 L.Ed.2d 461 (1967).50
*248The foregoing suggests that it would have been inconceivable to the drafters of the statute to read it as exempting the United States from the doctrine of tolling for fraudulent concealment.51 This conclusion does not contradict the proposition that Section 2401(a) must be strictly construed.52 We do not interpolate a provision of tolling for fraudulent concealment on the basis of our notions of equity.53 Rather, we believe that the 1863 Congress simply assumed that this doctrine was incorporated in “every” statute of limitations and that it would do violence to the intent of Congress for us to hold to the contrary.
For rather different, reasons appellants also argue that the due diligence doctrine is not applicable to this case. Our previous cases applying the due diligence doctrine concerned wrongs that were “self-concealing.” See Hobson v. Wilson, supra, 737 F.2d at 34. Noting that this case concerns a wrong that is usually knowable but which has only been obscured by an alleged subsequent positive act of concealment, appellants argue that we should reject the due diligence rule in favor of a standard providing for tolling of the statute until a plaintiff had “actually discovered” what was concealed. Brief of appellants at 28. There appears to be a split in the circuits on this point. Compare Tornera v. Galt, 511 F.2d 504, 510 (7th Cir.1975) (applying an actual discovery rule), with Campbell v. Upjohn Co., 676 F.2d 1122, 1128 (6th Cir. 1982) (applying the due diligence rule).54
Given the particular facts of this case, we cannot accept the “actual discovery” standard suggested by appellants. It is the legal effect of fraudulent concealment that tolls the statute, not its immorality. It is one thing to toll the statute of limitations until a reasonable plaintiff could undo the effects of concealment. It is quite another matter to discharge a plaintiff completely from his usual obligations to conduct reasonable inquiries into the grounds supporting his cause. The former course merely nullifies the effect of concealment. It allows the statute of limitations to operate in the manner that Congress provided and under the assumption that Congress did not intend for the United States to abuse such statutes by engaging in conscious frauds. The latter approach, by contrast, serves as a punitive measure and perhaps as a deterrent of future fraud. Although *249such deterrence might make sound policy, we refuse to imply it in an action against the United States absent a congressional suggestion in that direction.
2. Interpreting the due diligence rule. Unfortunately, our cases do not provide operational definitions of the key terms of the governing standard. Thus tolling is triggered by concealment of the “facts giving notice of the particular cause of action at issue.” Hobson v. Wilson, supra, 737 F.2d at 35. What falls within the ambit of that phrase, however, is not self-evident. Similarly, the statute begins to run when a “duly diligent” plaintiff would have discovered that which was concealed; but “due diligence” is hardly self-explanatory. Because the facts of this case are sui generis, we will refrain from definitively restating the due diligence doctrine. This is not the occasion to establish a new rule to govern future cases. We seek only to clarify our prevailing formula so that there is no mystery as to the basis of our decision here.
(a) What tolls the statute: concealment of the “factual basis of a complaint.” Appellee argues that the “facts giving notice of the particular cause of action at issue” include only the fact of injury and the identity of the inflictor. Brief of appellee at 22, 24. We do not agree. As already noted, in assessing the import of fraudulent concealment we are first and foremost concerned with its legal effect.55 Once a defendant has effectively closed the courthouse door to all plaintiffs it is of little significance that that defendant has not also concealed his identity or the fact of injury.56
Thus, where a defendant concealed information that prevented a plaintiff from alleging a crucial element of his claim, the statute would be tolled.57 Nor would it *250change our analysis if a defendant had achieved the same effect by concealing facts that would prevent a plaintiff from overcoming a seemingly ironclad defense. For, as the District Court suggested, where the result is the same — to prevent a law-abiding plaintiff from filing a complaint — it matters little whether the issue is labeled a “claim” or a “defense.” 586 F.Supp. at 787.
(b) When the statute begins to run. “Due diligence” also lacks a precise definition. But unlike the concept of the “factual basis of the complaint,” the concept of “due diligence” is best left unfocused. As we read- our cases, “due diligence” refers to a fact-specific judgment in each case as to what a reasonable plaintiff could be expected to do. See Richards v. Mileski, supra, 662 F.2d at 71.
Nonetheless, two specific guidelines do emerge from our cases. First, in evaluating the extent of a plaintiffs constructive knowledge a court ought to pay careful attention to whether a plaintiff was ever put on notice that further inquiries might be appropriate. See Hobson v. Wilson, supra, 737 F.2d at 35 n. 107. Of course, a court must still make a situation-specific judgment as to when (or if) subsequent inquiries might have produced the “factual basis” of a good faith complaint. But an initial determination on when a plaintiff was put on “inquiry notice” will help to narrow the issue. On the other hand, the fact that a plaintiff is on “inquiry notice” does not, without more, begin the running of the statute. See id. at 35. Inquiry notice is merely a necessary, but not a sufficient, condition for the running of the statute. Whether such inquiries would lead a diligent plaintiff to discover that which was concealed will naturally vary with the facts of each case.
B. The Doctrine Applied
The foregoing suggests that not every act of concealment will toll the statute of limitations. Concealment must go to a critical element or defense attending each particular cause of action. See id. at 35. We must therefore analyze the disparate effect of appellee’s course of conduct on the only two claims that are not barred by sovereign immunity: the Takings Clause and contract claims.
1. The Takings Clause claims and the military emergency doctrine. In their complaint appellants alleged that the United States concealed the fact that there was no military necessity justifying the exclusion, evacuation, and internment program. Complaint at 52-53 11Q6, JA 58-59. The District Court, however, did. not restrict its. judgment to the pleadings. As previously noted, the District Court also looked to certain undisputed facts in the historical record.58 After reviewing this material the District Court concluded that it did appear that the United States had concealed critical evidence during the wartime legal challenges to the exclusion program, 586 F.Supp. at 787-788. The District Court assumed, however, that the government’s act of concealment was limited to its alleged suppression of the Hoover, Fly, and Ringle memoranda. See id. It noted that these documents were in the public domain as early as 1949. Id. at 788.59 It therefore concluded that although the statute of limitations may have been tolled for a time the statute had run long before appellants filed their claims in 1983.
We do not dispute the District Court’s reading of the historical record.60 But be*251cause we believe the District Court’s analysis to have rested on a legally defective premise, we reverse this aspect of its judgment.
(a) What was allegedly concealed. Paragraph 95 of appellants’ complaint, Complaint at 52, JA 58, alleges that the government “excluded from the record of pending court actions * * * evidence contradicting the so-called ‘military necessity’ for mass imprisonment.” The District Court credited this allegation, finding it consistent with the undisputed historical material before it. 586 F.2d at 787-88. But the District Court never considered the legal relevance of this allegation to the particular cause of action pleaded by appellants in this case.
When the government impinges on property rights in the midst of a military emergency, there is no compensable taking under the Fifth Amendment. United States v. Caltex, supra, 344 U.S. at 154-56, 73 S.Ct. at 202-04; United States v. Pacific Railroad, 120 U.S. 227, 234, 7 S.Ct. 490, 493, 30 L.Ed. 634 (1887). In Korematsu and Hirabayashi the Supreme Court addressed the question of military necessity as a justification for the evacuation program, albeit not in the context of a Takings Clause claim. In those cases the Court determined that it must defer to the military judgment that it was impossible, as a practical matter, to segregate the loyal from the disloyal. It is true that Ringle’s analysis contained evidence undermining that conclusion. But the Court did not lack for evidence arguing against the military judgment on this vital point. In Korematsu the Japanese-American Citizens League (JACL) had submitted a brief that raised substantial questions about the empirical basis of the claim of military necessity.61 In the face of such contrary evidence, however, the Court determined that it must defer to the military judgment. Korematsu, 323 U.S. at 218-19, 223-24, 65 S.Ct. at 194-95, 197-98.
For the government to have concealed the factual basis of appellants’ claims it would not merely have had to conceal evidence suggesting the absence of a military emergency. In addition, the concealed evidence would have had to be sufficient to rebut the presumption of deference to the military judgment articulated by the Supreme Court. Given the constitutional underpinnings of the presumption of deference articulated by the Court, however, nothing less than an authoritative statement by one of the political branches, purporting to review the evidence when taken as a whole, could rebut the presumption articulated in Korematsu.62
*252Thus to have concealed evidence going to the very basis of the evacuee’s Takings Clause claim, the government would have had to conceal both Ringle’s report and the fact that there were no intelligence reports contradicting Ringle. Although appellants alleged this further act of concealment in their complaint, see Complaint at 52. ¶ 96, JA 58, the District Court did not discuss whether the undisputed historical material on which it based its judgment contradicted this allegation. As noted previously, however, nothing prevents us from looking at those same historical documents to determine whether appellants’ allegations retain any credibility.
Our reading of the CWRIC report suggests that appellants’ allegation does have support in the historical record.63 At the very least, the CWRIC report suggests that contemporary official intelligence analysis firmly opposed a mass evacuation. See Personal Justice Denied at 51-60. Moreover, both the CWRIC Report, see id., and the Burling and Ennis memoranda, see Ennis I, JA 115-118; J. Burling, Memorandum for the Attorney General (April 12, 1944), JA 119, indicate that this information was available to the War Department and the Justice Department at the time it prepared its Hirabayashi and Korematsu briefs. Given the procedural posture of this case, we must credit the allegations of appellants’ complaint unless they are specifically contradicted by the historical documents before the District Court. We therefore conclude that concealment has been alleged sufficient to toll the statute of limitations.64
(b) When the statute began to run. The District Court found that the statute began to run when reference to the Ringle, Fly, and Hoover memoranda appeared in several books and articles. But just as we do not believe that the suppression of these materials, by itself, could have tolled the statute, we do not find that their disclosure could have started the running of the statute anew. None of these documents could have reversed the presumption of deference erected by the Supreme Court in Korematsu. Any court reviewing such documents would have concluded that it must defer to the judgment of the military authorities who often must be presumed to act on the basis of conflicting reports.
Not only would the Ringle Report have been discounted as a partial statement of the facts; but it could not pass as an authoritative statement of one of the political branches. The Korematsu and Hirabayashi Court grounded its deference to the “war-making branches’ ” special role in securing the national defense. Hirabayashi, 320 U.S. at 99, 63 S.Ct. at 1385. Consequently, only a statement by one of the political branches could have rebutted the presumption of deference.
Of course, there can be no question but that the publication of the Ringle Report should have put the evacuees on notice of the need to conduct further inquiries into possible claims they might have against the *253United States.65 It is wholly possible that further inquiries would have uncovered the Ennis and Burling memoranda.66 Nonetheless, even these memoranda would not likely have affected appellants’ legal rights. To be sure, these memoranda indicate that responsible Justice Department officials, who purported to have a wide view of the evidence, had serious doubts about the military necessity rationale. But' that is all they represent. They present one side of a heated debate within the Justice Department, and between Justice and the War Department, on the appropriateness of the evacuation policy. They cannot be understood to be an authoritative statement by one of the political branches that there ,was reason to doubt the basis of the military necessity rationale.
That statement came only in 1980 when Congress passed the Act creating the Commission on Wartime Relocation and Internment of Civilians (CWRIC). Pub.L. 96-317, 94 Stat. 964 (July 31, 1980), codified at 50 U.S.C.App. § 1981 note (1982). Section 2(a) provides a brief statement of Findings and Purpose. It states that the Act was passed because “no sufficient inquiry has been made into [the internment].” Pub.L. 96-317 2(a)(3), 94 Stat. 964. This reference is elucidated by the Act’s legislative history. According to the House report, “[T]he committee found that no significant study has been done by the Government to determine the extent of any civil rights violations * * *.” H.R.Rep. No. 1146, 96th Cong., 2d Sess. 5 (1980). The Senate report spoke in stronger terms, finding that the “[internees were deprived of their liberty and property apparently based on their ethnic origins alone.” S.Rep. No. 751, 96th Cong., 2d Sess. 2 (1980).
At a minimum, the Act can be understood to be a formal statement that Congress no longer believed that the explanation provided by the military authorities for the internment program was adequate and that the issue should be reopened. Moreover, Congress took this step fully cognizant of previous congressional and Supreme Court approval of the legality of internment program. See H.R.Rep. No. 1146, 96th Cong., 2d Sess. 11 (1980) (reprinting the letter of the Assistant Attorney General detailing previous Supreme Court and congressional review of Executive Order 9066). In so doing Congress finally removed the presumption of deference to the judgment of the political branches.67 With this step the statute of limitations began to run on appellants’ Takings Clause claims.68
2. The contract claims. Although not barred by sovereign immunity, the statute of limitations was never tolled for appellants’ contract claims. Unlike the “military *254emergency” doctrine of United States v. Caltex, supra, there is no analogous doctrine governing contract claims that suggests that “military necessity” is a defense to a contract claim.
It is true that when the United States has made promises to perform an “act of sovereign” no contract is formed. See United States v. Juda, supra, 6 Ct.Cl. at 454. But the mere fact that the government acts to further the national defense does not bring its conduct within the “act of sovereign” doctrine. See id. at 454-455 (finding that the evacuation of the Bikini Islanders from their homes to facilitate atomic tests did not constitute an “act of sovereign”). Indeed, the “act of sovereign” doctrine is only invoked where the government can allege that it never intended to form a contract but only sought to distribute public benefits without binding obligations. See id.
Thus there is no reason why appellants could not have brought their contract claims in the 1940’s. A judicial determination that the government had acted pursuant to a military emergency would have had no effect on their claims. Nor would it have affected their ability to attack the “act of sovereign” defense. That defense would stand or fall, in 1945 or 1985, on whether the United' States intended to undertake binding commitments to specific persons or whether it merely intended to distribute public benefits. The existence vel non of a military emergency could have only the most attenuated influence on this issue.
The concealment of the lack of military necessity therefore did not have any legal effect on appellants’ contract claims. Having failed to assert such claims within the statutory period, they may not do so at this later date.
VI. The AmericanJapanese Evacuation Claims Act
A. The Exclusivity of the Act
Appellee argues that appellants’ Takings Clause claims must fail because the AmericanJapanese Evacuation Claims Act constituted an exclusive remedy for all claims arising out of the evacuation and internment program. Brief of appellee at 43. Under Brown v. GSA, 425 U.S. 820, 96 S.Ct. 1961, 48 L.Ed.2d 402 (1976), a statute is deemed to provide an exclusive remedy where three conditions are met: (1) the statute provides a detailed and complete scheme for adjudicating claims arising out of a particular subject matter; (2) Congress, rightly or wrongly, did not believe that the affected individuals had alternative remedies at the time it enacted the statute; and (3) the statute addresses a specific injury or issue while alternative remedies address a broader grievance. The Claims Act is obviously specifically tailored to the conditions of the evacuation program. The Act thereby fulfills the third of the Brown conditions. But the Claims Act fails to fulfill the first two of the conditions articulated in Brown.
First, the Act fails to provide a complete remedy for the losses sustained. As the District Court found, the Act tended to exclude claims for compensation that would have been compensable under the Takings Clause at-the time the Act became law. See 586 F.Supp. at 785-786. For example, claimants were not paid for the interest that accrued between the time of the evacuation and the time their claims were paid. Compare Claim of George M. Kawaguchi, 1 Adjudications of the Attorney General 14, 19-20 (1956), with Seaboard Air Line R. Co. v. United States, 261 U.S. 299, 306, 43 S.Ct. 354, 356, 67 L.Ed. 664 (1923).
Second, it is true that at the time the Claims Act was passed Congress did not think that the evacuees had any alternative remedies. See H.R.Rep. No. 732, 80th Cong., 2d Sess. 3 (1948). But this fact does not have the same legal meaning in our case that it had in Brown. Brown turned on the premise that Congress had assumed that the remedies provided by Section 717 of the Civil Rights Act and the government’s waiver of sovereign immunity were coterminous. 425 U.S. at 827-28, 96 S.Ct; *255at 1965-66. Put otherwise, Brown presumed that the passage of Section 717 not only created new rights but also affirmed specific limits on the government’s waiver of sovereign immunity.
In this case, however, Congress could hardly assume that sovereign immunity barred appellants’ Takings Clause claims. The Constitution itself provides for the requisite waiver of sovereign immunity. Indeed, any congressional attempt to quash such claims might itself be unconstitutional. Here congressional statements as to the absence of alternative remedies merely constituted a recognition of the power of the military necessity defense, not a specific limit to a waiver of sovereign immunity.
In sum, where Congress speaks of a lack of alternative remedies in circumstances where it could not constitutionally limit a waiver of sovereign immunity, such congressional observations do not imply that any newly created rights must be exclusive of all other remedies. We therefore do not believe that in passing the Claims Act Congress sought to preclude appellants’ Takings Clause claims.
B. Finality and Discharge
Even though the Act did not provide for an exclusive remedy, it did contain a provision suggesting that if an evacuee brought a claim under its provisions he would be barred from bringing subsequent claims concerning the evacuation and internment programs. Thus Section 1984(d) of the American-Japanese Evacuation Claims Act, 50 U.S.C.App. § 1981 et seq. (1982), reads as follows:
[T]he payment of an award shall be final and conclusive for all purposes, notwithstanding any other provision of law to the contrary, and shall be a full discharge of the United States * * * with respect to all claims arising out of the same subject matter.
The plain language of Section 1984(d) bars all suits brought under the Takings Clause once an evacuee has received an award under the Act.69 The Claims Act must therefore be read to force claimants to choose between attempting to receive the “bounty” provided by Congress under the Act or exercising their constitutional rights under the Fifth Amendment.
We are not unmindful of the hard choice to which Congress put the evacuees. By forcing them to choose between a ready administrative remedy and a costly lawsuit, Congress effectively forced the evacuees to settle for half a loaf rather than risk a fight for what the Constitution declares to be theirs by right. In so doing Congress acted on the outer perimeter of its authority. It did not, however, exceed its authority.70 Nor are we unaware of the manner in which the Solicitor General’s alleged wrongful concealment narrowed the evacuees’ legal choices at that time. Nonetheless, it is apparent that the congressional offer was made in good faith and that the United States is not estopped from raising Section 1984(d).71 We therefore re*256luctantly conclude that petitions for reconsideration of this harsh policy of finality are properly addressed to Congress and not to this court.
VII. Declaratory Relief
We reject appellants’ independent declaratory claim. Appellants argue that there is the danger they may again be visited by racially motivated illegal government actions. Reply brief of appellants at 21. Even assuming, arguendo, that there were a substantial probability of such an unfortunate event, appellants would still not have met their burden under Article III. Our case law holds that the mere fear of future governmental action contingent upon future discretionary decisions by political officials does not provide a live case or controversy. See Halkin v. Helms, 690 F.2d 977, 1009 (D.C.Cir.1982). Appellants also maintain that a declaratory judgment will remedy “present and ongoing psychic damage.” Brief of appellants at 56. Such psychic damage, however, standing alone, does not provide the requisite injury in fact. Cf. Laird v. Tatum, 408 U.S. 1, 13-14, 92 S.Ct. 2318, 2325-26, 33 L.Ed.2d 154 (1972) (subjective effect on First Amendment rights does not provide requisite case or controversy for declaratory relief).
VIII. Conclusion
The United States cannot be presumed to be amenable to suit. Fortunately, the Founders provided that the right to obtain just compensation for the taking of one’s property should remain inviolate. In so doing, they no doubt assumed that the normal statutes of limitations would apply. But they also most certainly assumed that the leaders of this Republic would act truthfully. In the main, history has proven the Founders correct. We have also learned, however, that extraordinary injustice can provoke extraordinary acts of concealment. Where such concealment is alleged it ill behooves the government of a free people to evade an honest accounting. Should such concealment be proven here, those individuals who have not received awards under the Claims Act should be free to press this cause to its conclusion.
Affirmed in part and reversed and remanded in part.
. Although appellants had moved for class certification, a decision on this motion was postponed pending resolution of appellee’s motion to dismiss. In their motion appellants defined the class as the approximately 120,000 citizens and permanent residents, and representatives of such persons no longer living, who were subjected to the evacuation and internment program. See Hohri v. United States, 586 F.Supp. 769, 772 n. 1 (D.D.C.1984) (Hohri). Because the District Court granted appellee’s motion to dismiss, it never ruled upon the class certification motion.
. Congress subsequently authorized the arrest, fine, and imprisonment of anyone violating an order issued pursuant to Executive Order 9066. *232See Pub.L. No. 503, 56 Stat. 173, 77th Cong., 2d Sess. (1942).
. Executive Order 9066 did not establish means of administering the evacuation. This defect was cured when the President issued Executive Order 9102, 7 Fed.Reg. 2165 (March 20, 1942), creating the War Relocation Authority (WRA).
. The government did not take title to the evacuee’s property. It offered to take custody of such property, Civilian Exclusion Order No. 5, April 1, 1942, JA 114, or to facilitate its sale. Press Release, March 10, 1942, JA 136.
. The military did begin to conduct an individualized "loyalty review” program, providing for the release of individuals of established loyalty, in February 1943. But this program provided only slow, piecemeal release. It was not until Ex parte Endo, 323 U.S. 283, 65 S.Ct. 208, 89 L.Ed. 243 (1944), that the Supreme Court held it unlawful to continue to detain an internee in a Relocation Center after she had received clearance for an indefinite leave under the loyalty review program. Prior to that decision the WRA refused to release even those internees whose loyalty had been duly certified if the internee desired to live in an area that had not been approved by the WRA. Id. at 293, 65 S.Ct. at 214.
. The Commission was established by Pub.L. No. 96-317, 94 Stat. 964, 96th Cong., 2d Sess. (1980). It was charged with issuing a comprehensive report on the internment program. Its report was issued in late 1982. See Report of the Commission on Wartime Relocation and Internment of Civilians, Personal Justice Denied (1982) (Personal Justice Denied).
. By May 26, 1944 Secretary of War Stimson was proposing an end to the exclusion. The new Commanding General of the Western Command, C.H. Bonsteel, wrote there was no longer a military necessity for exclusion as of July 3, 1944. Personal Justice Denied at 228-229.
. The government noted the prevalence of dual citizenship among Japanese-Americans, their practice of Shintoism (which entails emperor worship), Japanese language schools on the West Coast, the links between West Coast Japanese organizations and Japan, the large number of Japanese aliens within the community, and a significant number (about 10,000) of Japanese-Americans who had been sent to Japan for their education. Hirabayashi, Brief for the United States at 11.
. The Court found congressional ratification of the exclusion program in Pub.L. No. 503, 56 Stat. 173, 77th Cong., 2d Sess. (1942), providing for criminal penalties for violation of orders issued pursuant to Executive Order 9066. See Hirabayashi v. United States, 320 U.S. 81, 91, 63 S.Ct. 1375, 1381, 87 L.Ed. 1774 (1943). But it is important to recognize that to the extent that the Hirabayashi Court based its opinion on deference to a congressional judgment, see id. at 90-91, 63 S.Ct. at 1380-81, it was only deferring to a congressional decision to defer to the military on the validity of the exclusion program. Congress did not make an independent factual analysis. Although hearings were held before the Select Committee Investigating National Defense Mitigation (Tolan Committee), none of the witnesses were members of the intelligence community. See H.R.Rep. No. 1911, 77th Cong., 2d Sess. 4 & n. 2 (1942). Moreover, the Committee expressly declared that it based its endorsement of the evacuation program on the need to defer to the judgment of the military authorities, not on its own analysis of the facts. See id. at 13, 15. See also H.R.Rep. No. 2124, 77th Cong., 2d Sess. 11 (1942) (reiterating that "[wjith respect to the question of the evacuation of the Japanese population * * *, the decision of the military must be final”); Personal Justice Denied at 98 (concluding that the Tolan Committee merely “assumed that Secretary [of the Navy] Knox knew what he was talking about and that the President was acting on informed opinion”). Thus to the extent that the Court was misled as to the soundness of the military judgment it was similarly misled as to the soundness of the congressional judgment to defer to the military.
. Lt. Commander Ringle first compiled his conclusions in K. Ringle, Report on the Japanese Question (Jan. 26, 1942) (Ringle Report), JA 91-100, which he submitted to the Chief of Naval Intelligence. He subsequently included his conclusions in an article published anonymously in the October 1942 issue of Harpers magazine, under the title The Japanese in America, the Problem and Solution (by “An Intelligence Officer”). It appears that although Ennis did not have an actual copy of Ringle’s Report when he drafted his Memorandum, he did have a copy of the Harpers article and knew that Ringle was the author of this article. Ennis also had in his *234possession a memorandum prepared by Ringle for the WRA on the Japanese-American question. Thus all of Ringle’s critical findings, including a verbatim statement of his conclusion that the “Japanese Problem” could be solved on an individual basis, quoted in text irifra at 234, were included in the materials before Ennis at the time he drafted his memorandum. See E. Ennis, Memorandum for the Solicitor General (April 30, 1943) {Ennis I) at 1, JA 115.
. He also noted that the Americanization of the Nisei (American-born) had proceeded quite rapidly (cultural societies and Shintoism notwithstanding). Thus, as to the automatic dual citizenship imposed by Japanese law, Ringle noted that many of the Nisei had divested themselves of such dual citizenship, even though this entailed loss of property rights in Japan. Finálly, he noted that although the Kibei (American-born Japanese predominantly educated in Japan) might present a loyalty risk, their identities could be ascertained from government records and they should be dealt with as a discrete problem. See Ringle Report at 2-5, JA 102-106.
. In support of this conclusion Ringle noted, inter alia, that the number of Japanese aliens and citizens who would act as enemy agents was less than 3,500 and that the identity of these individuals was well known to U.S. intelligence (indeed, the most dangerous were already in custody). See Ringle Report at 2, JA 102.
. Indeed, Ennis went so far as to say that "to a very considerable extent the Army * * * is bound by the opinion of the Naval officers in Japanese matters." Ennis I at 3, JA 117.
. Ennis was quite explicit on this point in his memorandum to the Solicitor:
[I]n view of the fact that the Department of Justice is now representing the Army in the Supreme Court of the United States and is arguing that a partial, selective evacuation was impracticable, we must consider most carefully what our obligation to the Court is in view of the fact that the responsible Intelligence agency regarded a selective evacuation as not only sufficient but preferable * * *. Thus, in one of the crucial points of the case the Government is forced to argue that individual, selective evacuation would have been impractical and insufficient when we have positive knowledge that the only Intelligence agency responsible for advising Gen. DeWitt gave him advice directly to the contrary.
Ennis I at 3, JA 117.
. The brief did cite to the Harpers article. Although that article was signed "An Intelli*235gence Officer,” there was no way that the Court could have verified this fact. Moreover, the Justice Department did not endorse all of the conclusions in that article. It cited the article only for the proposition that Japanese-Americans educated in Japan would probably be loyal to Japan. Hirabayaski, Brief for the United States at 29 n. 46.
. See Personal Justice Denied at 52-60. Of the professional intelligence services, Naval Intelligence and the FBI shared the job of monitoring the Japanese-American situation on the West Coast. Ringle’s views reflected the opinion of Naval Intelligence. FBI Director Hoover expressed his view in a memorandum to the Attorney General dated February 2, 1944. Id. at 55 & nn. 33, 35. Hoover believed that the Japanese did not rely primarily on Japanese-Americans for their espionage work. Only the San Diego and Seattle FBI field offices supported the concept of a mass evacuation, but as the CWRIC observed, "Hoover’s own opinion, and thus the Bureau’s, was that the case to justify mass evacuation for security reasons had not been made.” Id. at 55.
. Even disclosure of the Ennis memoranda would not have altered the result. For Ennis did not assert that there was no basis for the government’s position; he argued only that one Naval Intelligence report, "binding” on the Army, contradicted the government’s position. Ennis I at 3, JA 117.
. Of course, it is possible that the War Department and Justice Department might not have had access to the full range of intelligence reports that the CWRIC was able to uncover. Assuming, arguendo, that the government was simply unaware of its own intelligence reports in 1943, the Justice Department would be open to charges of gross negligence in failing to inquire whether the Ringle report was contradicted by other intelligence data. Thus, if only by a decision to remain ignorant, the government appears to have concealed the fact that there was no military necessity for the mass evacuation when it argued Hirabayaski to the Supreme Court.
. Although the Court did refer to the fact that "investigations made subsequent to the exclusion" had "confirmed” that there were "members of the group who retained loyalties to Japan,” the Court clearly was referring to statements made by internees in response to loyalty questionnaires and to requests for repatriation. Korematsu v. United States, 323 U.S. 214, 218-19, 65 S.Ct. 193, 194-95, 89 L.Ed. 194 (1944).
. Indeed, the majority opinion freely cited to the Final Report itself, see 323 U.S. at 219 n. 2, 65 S.Ct. at 195 n. 2, notwithstanding the footnote in the government's brief.
. We are aware that in Korematsu the Court changed its verbal formulation and stated that racial classifications are "immediately suspect" and therefore subject to the “most rigid scruti*237ny.” 323 U.S. at 216, 65 S.Ct. at 194. In its next breath, however, the Court reaffirmed Hirabayashi’s "conclusion” that the Court could not second-guess the "finding of the military authorities that it was impossible to bring about an immediate segregation of the disloyal from the loyal” which necessarily resulted in affirming "the validity of the curfew order as applying to the whole group." Id. at 219, 65 S.Ct. at 195. Thus, although subsequent cases certainly gave substance to the notion of "rigid scrutiny” and "suspect" classifications, in Korematsu itself these phrases accompanied a highly deferential standard of review. For example, at no point did the Court suggest that it would evaluate the exclusion decision to determine whether it was narrowly tailored to effectuate the military goals asserted. Nor was this point lost on the Justices at the time. Justice Jackson, in a stinging dissent, forcefully suggested that the “scrutiny” practiced by the Korematsu majority was "rigid” in theory but deferential in fact when he said:
How does the Court know that these orders have a reasonable basis in necessity? No evidence whatever on that subject has been taken by this or any other court. There is sharp controversy as to the credibility of the DeWitt report. So the Court, having no real evidence before it, has no choice but to accept General DeWitt's own unsworn, self-serving statement, untested by any cross-examination, that what he did was reasonable. And thus it will always be when courts try to look into the reasonableness of a military order.
See id. at 245, 65 S.Ct. at 207 (Jackson, J. dissenting).
That Justice Jackson could find the credibility of DeWitt’s report to be a matter of "sharp” controversy indicates his willingness to evaluate evidence provided by parties other than the political branches of the government. That the majority, by contrast, was less impressed with such evidence underscores the importance of the Justice Department’s failure to disclose Ringle’s analysis.
. As in Hirabayashi, however, deference to Congress did not signify deference to an independent analysis of the critical issue in the case: the practicality of segregating the loyal from the disloyal. In addition to reiterating its deference to the 1942 Act, see 323 U.S. at 217, 65 S.Ct. at 194, the Korematsu Court merely cited congressional findings that there were in fact disloyal members of the Japanese-American community. See id. at 219 & n. 2, 65 S.Ct. at 195 & n. 2.
. See, e.g., 50 U.S.C.App. § 1982(b)(5) (denying compensation for loss of anticipated profits). Similar limits were interpolated through subsequent adjudications. See, e.g., Claim of Mary Sogawa, 1 Adjudications of the Attorney General 126 (1950) (denying compensation for expenses of the evacuation); Claim of George M. Kawaguchi, 1 Adjudications of the Attorney General 14, 19-20 (1950) (limiting compensation to purchase price, implicitly denying any interest increment).
, The Senate Report merely adopted the House Report’s statement of the "facts and circumstances” justifying the Claims Act. See S.Rep. No. 1740, 80th Cong., 2d Sess. 2 (1948). In addition, nearly identical language was included in the predecessor bill to H.R. 3999, H.R. 6780. See H.R.Rep. No. 2679, 79th Cong., 2d Sess. 4 (1946).
. The floor debate surrounding passage of the Act also suggests that Congress believed that Korematsu had absolved the United States of civil liability. Although there was general discussion of the need to do "justice,” see, e.g., 93 Cong.Rec. 9872 (1947) (remarks of Rep. Walter), there was no suggestion that the Japanese-Americans had suffered a legally cognizable wrong. On the contrary, at least two Representatives insisted, without rebuttal, that military necessity had absolved the United States of all liability. See 93 Cong.Rec. 9872-9873 (remarks of Representatives Goff and Gwynne, affirming the legality of the evacuation). At no point was it suggested that the evacuees could gain compensation through the courts.
. The Sogawa principle takes on particular importance in light of congressional silence on this matter throughout the amendment process attending the Claims Act. On two separate occasions Congress had the opportunity to reverse the view expressed in Sogawa. It did not do so. Ordinarily, mere congressional inaction does not shed light on the intent of Congress. See Aaron v. SEC, 446 U.S. 680, 694 n. 11, 100 S.Ct. 1945, 1954 n. 11, 64 L.Ed.2d 611 (1980). But where the matter has been subject to subsequent congressional attention then congressional acquiesence may be considered among other relevant factors. Cf. Bob Jones University v. United States, 461 U.S. 574, 600, 103 S.Ct. 2017, 2033, 76 L.Ed.2d 157 (1983). Congressional inaction in this instance may therefore be properly considered as supplementing more direct evidence that the legislature believed that Korematsu absolved the United States of civil liability from claims arising from the evacuation.
. The dissent correctly warns against the danger of forum shopping in those instances where an attorney adds frivolous FTCA claims to Takings Clause claims in order to obtain appellate jurisdiction in the regional Courts of Appeals. If we believed appellants’ FTCA claims were in fact frivolous we would certainly agree with the dissent’s conclusion that appeal must lie in the Federal Circuit alone. Cf. Doe v. U.S. Dep’t of Justice, 753 F.2d 1092, 1101-02 (D.C.Cir.1985) (refusing to find that the Federal Circuit had exclusive jurisdiction where the plaintiff’s characterization of her claim as falling under § 1346(a)(2) was "frivolous”). But, as discussed infra, appellants’ tort claims are not defective because they are substantively farfetched. Appellants have alleged serious wrongs traditionally compensable at common law. Appellants’ tort claims áre defective because they failed to appreciate the unyielding ("jurisdictional”) character of the filing requirements imposed by the FTCA and mistakenly assumed that the FTCA merely codified the more flexible exhaustion doctrine. Such codifications, however, are hardly unknown. See WATCH v. FCC, 712 F.2d 677, 681-82 (D.C.Cir.1983) (finding that the filing requirement of § 405 of the Communications Act merely codified the judge-made exhaustion doctrine, thereby incorporating its various equitable exceptions). Appellants’ failure to grasp in full the distinction between filing requirements that are nonwaivable and those that are subject to waiver on equitable grounds hardly renders their tort claims frivolous.
Any suggestion that the lawyers here indulged in forum shopping is without warrant. Cf. dissent at 259. No deliberate shopping occurred in this and other recent cases presenting a question as to the interpretation of the newly adopted § 1295(a)(2) — cases such as those cited by the dissent at 260. Rather, the parties, including the government, rarely even adverted to the section. The cases thus revealed the parties’ oversight or confusion regarding § 1295(a)(2), not their deliberate attempt to steer the case to a favored forum.
. Nor does recent case law of this circuit or the Federal Circuit contradict our analysis. Although Atari, Inc. v. JS & A Group, 747 F.2d 1422, 1437 n. 13 (Fed.Cir.1984), took an expansive view of the exclusive jurisdiction of the Federal Circuit, that case concerned the problem of pendent claims in general. Here we do not consider the problem of pendent claims as a generic matter, but only those claims specifically designated by statute as falling within the "except" clause. Nor does Professional Managers Ass'n v. United States, 761 F.2d 740 (D.C.Cir. 1985), prevent us from taking appellate jurisdiction in this case. In Professional Managers this court rejected the liberal construction of § 1295(a)(2) adopted by the Seventh Circuit in SquiUacote v. United States, 747 F.2d 432 (7th Cir.1984) (refusing to transfer a case to the Federal Circuit where that would be inefficient and unfair to the litigants), cert. denied, — U.S. -, 105 S.Ct. 2021, 85 L.Ed.2d 302 (1985). Here we do not base our decision on policy concerns for fairness or efficiency. On the contrary, we take jurisdiction on the basis of our reading of the plain meaning of the statutory language.
. The “other" § 1338(a) claims to which this clause refers are patent claims. Thus under § 1295(a)(1) the Federal Circuit has exclusive appellate jurisdiction over mixed patent and copyright/trademark claims.
. The legislative history of the Federal Courts Improvement Act, 28 U.S.C. § 1295 (1982), is not to the contrary. Both the House and Senate reports indicate that § 1295(a)(2) reflects two conflicting policies. On the one hand, Congress sought to centralize the adjudication of claims in which the United States was a defendant. See S.Rep. No. 275, 97th Cong., 1st Sess. 3-4 (1981); H.R.Rep. No. 312, 97th Cong., 1st Sess. 42 (1981). On the other hand, Congress did not want to centralize adjudication of cases involving tort claims, which would often tend to turn on issues of state law. In such cases Congress preferred adjudication by the regional Courts of Appeals. See S.Rep. No. 275 at 20; H.R.Rep. No. 312 at 42.
. As indicated infra, we affirm the District Court’s dismissal of appellants’ FTCA claims. Consequently, on remand the case will no longer fit within the "except” clause, original jurisdiction being based solely on § 1346(a)(2). Thus all subsequent appeals of this case will have to be brought in the Federal Circuit, pursuant to the general rule expressed in § 1295(a)(2).
Despite the dissent's unsupported suggestion to the contrary, see dissent at 259, our holding on the statute of limitations constitutes the "law of the case.” Our decision that we have jurisdiction over this appeal is subject-to reversal only by a superior court. Having determined that we do have authority to decide the instant appeal, we are obliged to instruct the District Court on the inquiry it is to pursue on remand. Thus, because we must “actually decide" the statute of limitations issue, our instruction sets the "law of the case.” To invalidate this instruction on later review, the Federal Circuit must find both "clear error” and “manifest injustice” in our disposition of the uncommon tolling question that this case presents. Laffey v. Northwest Airlines, 740 F.2d 1071, 1082 & n. 18 (D.C.Cir. 1984), cert. denied, — U.S. -, 105 S.Ct. 939, 83 L.Ed.2d 951 (1985).
. Under this provision the District Courts have concurrent jurisdiction with the Court of Claims for actions against the United States not exceeding $10,000. This provision, often referred to as the "Little Tucker Act,” see, e.g., Panduit Corp. v. All States Plastic Mfg. Co., 744F.2d 1564, 1575 n. 15 (Fed.Cir.1984), should be distinguished from 28 U.S.C. § 1491 (1982) which provides for jurisdiction in the Court of Claims for all claims against the United States regardless of the dollar amount.
. Appellants also assert claims for declaratory relief and cite 5 U.S.C. § 702 (1982) as a waiver of sovereign immunity for such claims. Because we find no case or controversy adequate to sustain appellants' declaratory claims, see infra at 256, we do not consider the effect of sovereign immunity on such claims.
. Thus the Tucker Act reads, in pertinent part: The District Courts shall have original jurisdiction * * * of * * * any civil action or claim against the United States, not exceeding $10,000 in amount, founded either upon the Constitution, or any Act of Congress, or any regulation of an executive department, or upon any express or implied contract with the United States, or for liquidated or unliquidated damages in cases not sounding in tort[.]
28 U.S.C. § 1346(a)(2) (1982).
. Nor can it seriously be contended that the failure of the government to take title to appellants’ property bars their claims under the Takings Clause. See United States v. General Motors Corp., 323 U.S. 373, 378, 65 S.Ct. 357, 359, 89 L.Ed. 311 (1945). Indeed, the fact that the government forced appellants to give up actual possession and control of their property suggests that it has committed a taking per se. See Loretto v. Teleprompter Manhattan CATV Corp., 458 U.S. 419, 431, 102 S.Ct. 3164, 3173, 73 L.Ed.2d 868 (1982) (dicta, collecting cases).
. The cases cited by appellee do not provide the support alleged. Thus National Board of YMCA’s v. United States, 395 U.S. 85, 89-90, 89 S.Ct. 1511, 1514-15, 23 L.Ed.2d 117 (1969), merely stands for the proposition that there is no taking where the government incidentally harms plaintiffs property while trying to protect that property from rioters. The same is true of Monarch Ins. Co. of Ohio v. District of Columbia, 353 F.Supp. 1249, 1255 (D.D.C.1973), aff’d, 497 F.2d 684 (D.C.Cir.) (order), cert. denied, 419 U.S. 1021, 95 S.Ct. 497, 42 L.Ed.2d 295 (1974).
. It would also appear that the historical findings of the CWRIC, see Personal Justice Denied at 18, support appellants’ allegations on this point. Therefore, it would seem that the extra-pleading evidence of which the District Court took notice, see 586 F.Supp. at 772 n. 2, supports the District Court’s conclusion on this issue.
. Regulations also required management of Indian assets "so as to obtain the greatest revenue for the Indians * * *"United States v. Mitchell, 463 U.S. 206, 224, 103 S.Ct. 2961, 2972, 77 L.Ed.2d 580 (1983) (quoting U.S. Office of Indi*244an Affairs, Regulations and Instructions for Officers in Charge of Forests on Indian Reservations 4 (1911)).
. United States v. Mitchell, supra note 38, did state that "a fiduciary relationship necessarily arises when the Government assumes such elaborate control over forests and property belonging to Indians.” 463 U.S. at 225, 103 S.Ct. at 2972. We do not read this alternative holding, however, as articulating a broad rule in favor of finding fiduciary relationships by implication whenever the government assumes pervasive control over a group’s property. Read in context, the Court created only a narrow exception — for Indian tribes — to the requirement that the government must expressly state its intent to manage the would-be beneficiaries’ property as a trustee.
We base our narrow reading of Mitchell on the Court's reliance on Navajo Tribe of Indians v. United States, 224 Ct.Cl. 171, 624 F.2d 981, 987 (Ct.Cl.1980). Navajo Tribe also appears to limit its finding of an "implicit” trust to dealings between the United States and Indian tribes. This reading of Navajo is supported by that opinion’s reliance on Seminole Nation v. United States, 316 U.S. 286, 296-97, 62 S.Ct. 1049, 1054-55, 86 L.Ed. 1480 (1942), in which the Court noted that Congress and the judiciary had previously made numerous statements explicitly assuming fiduciary obligations. Thus Mitchell only stands for the narrow principle that statutes and regulations governing the relations between the United States and Indian tribes may be presumed to contain an implicit assumption of fiduciary obligations, given a history of explicit statements to that effect.
. Including the right to substantive due process, travel, and privacy. See Complaint at 65 ¶ 124, JA 71.
. See Clark v. Library of Congress, 750 F.2d 89, 103 n. 31 (D.C.Cir.1984) (noting that "[t]he courts have uniformly held that jurisdiction under the ‘founded upon the constitution’ grant of the Tucker Act is limited to claims under the 'takings clause’ of the Fifth Amendment”); Lombard v. United States, 690 F.2d 215, 227 (D.C.Cir. 1982) (finding sovereign immunity a bar to First, Fifth, Ninth, and Tenth Amendment claims when such claims could be construed to run against the government itself), cert. denied, 462 U.S. 1118, 103 S.Ct. 3086, 77 L.Ed.2d 1347 (1983); Jalil v. Campbell, 590 F.2d 1120, 1122 (D.C.Cir.1978) (per curiam) (no right to compensation under the equal protection clause).
. See, e.g., Radin v. United States, 699 F.2d 681, 685 n. 8 (4th Cir. 1983); Jaffee v. United States, 592 F.2d 712 (3d Cir.) (en banc), cert. denied, 441 U.S. 961, 99 S.Ct. 2406, 60 L.Ed.2d 1066 (1979); Duarte v. United States, 532 F.2d 850, 852 (2d Cir.1976).
. Appellants’ allegations that these constitutional violations provide the predicate for claims based on the Civil Rights Acts, 42 U.S.C. §§ 1981, 1983, 1985-1986 (1982), is similarly without merit. These statutes, by their terms, do not apply to actions against the United States. See Timmons v. United States, 672 F.2d 1373, 1380 (11th Cir. 1982) (§ 1981 does not waive sovereign immunity); Unimex, Inc. v. HUD, 594 F.2d 1060, 1061 (5th Cir.1979) (none of the Civil Rights Acts waive sovereign immunity); Monarch Ins. Co. of Ohio v. District of Columbia, supra note 36, 353 F.Supp. at 1252 (§ 1983 does not waive sovereign immunity).
. Appellants also argue that their constitutional torts are cognizable under the Federal Tort Claims Act. Although we doubt the validity of this argument, see Gray v. Bell, 712 F.2d 490 (D.C.Cir. 1983), cert. denied, 465 U.S. 1100, 104 S.Ct. 1593, 80 L.Ed.2d 125 (1984), we need not reach that issue here. Appellants’ failure to comply with nonwaivable filing requirements bars all claims under the FTCA whatever their substantive legal basis.
. Specifically appellants allege assault, battery, false arrest and imprisonment, abuse of process, malicious prosecution, and negligent damage to their persons and property. The FTCA bars all claims for intentional torts that arose before 1974. See 28 U.S.C. § 2680(h) (1982). Because appellants, however, have failed to comply with mandatory filing requirements we need not reach the thorny question of when appellants’ claims "arose.”
. Compare Keene Corp. v. United States, 700 F.2d 836, 841 (2d Cir.1983) (refusing to waive the FTCA filing requirement on equitable grounds), cert. denied, 464 U.S. 864, 104 S.Ct. 195, 78 L.Ed.2d 171 (1984); Lunsford v. United States, 570 F.2d 221, 224 (8th Cir.1977) (same); *246Blain v. United States, 552 F.2d 289, 291 (9th Cir.1977) (same); Best Bearings Co. v. United States, 463 F.2d 1177, 1179 (7th Cir.1972) (same); Bialowas v. United States, 443 F.2d 1047, 1049 (3d Cir.1971) (same), with Athlone Industries, Inc. v. CPSC, 707 F.2d 1485, 1488 (D.C.Cir.1983) (exhaustion doctrine is subject to equitable waiver); Natural Resources Defense Council, Inc. v. Train, 510 F.2d 692, 703 (D.C. Cir.1975) (statutory notice requirements at issue found subject to equitable waiver).
Nor do cases concerning waiver of the filing requirement in suits initially brought in state court and subsequently removed pursuant to 28 U.S.C. § 2679(d) (1982) aid ajppellants' cause. Such cases involve suits initially brought in state court on the theory that the government was not a party and only subsequently removed once it was determined that the employee was acting in his official capacity. Under such circumstances it would be nonsensical to impose a mandatory filing requirement. Consequently, it is not surprising that at least one circuit has found that § 2675(a) does not apply to such cases. See Kelley v. United States, 568 F.2d 259 (2d Cir.), cert. denied, 439 U.S. 830, 99 S.Ct. 106, 58 L.Ed.2d 124 (1978). In our case, however, appellants sued the United States from the outset. At no point did appellants purport to be suing an employee of the United States acting in his individual capacity. Section 2675(a) therefore applies and cannot be waived.
. For an analysis of congressional action in the 1980’s, see infra at 253-54.
. Nor can this court "stay" these proceedings and allow appellants to now comply with § 2675(a). Even if this were a proper course of action it would not aid appellants for their attention to § 2675(a) comes too late. Under 28 U.S.C. § 2401(b) (1982) a claimant must file with the appropriate agency within two years of the time a claim “accrues.” Even assuming that appellants’ claims did not "accrue” until early 1983, the statute has now run. See Schuler v. United States, 628 F.2d 199, 201 (D.C.Cir.1980) (en banc) (per curiam).
. It is the settled law of this circuit that the rule of Bailey v. Glover extends to cases where the underlying cause of action was not based on fraud. See Hobson v. Wilson, 737 F.2d 1, 33 (D.C.Cir.1984), cert. denied, — U.S. -, 105 S.Ct. 1843, 85 L.Ed.2d 142 (1985). According to Hobson, the only operative difference between cases where the underlying cause of action sounds in fraud and a case such as that at bar is that in the former the cause of action itself is self-concealing while in the latter the defendant must perform a subsequent act of concealment before the plaintiff can allege fraudulent concealment. But cf. McCoy v. Wesley Hospital & Nursing Training School, 188 Kan. 325, 362 P.2d 841, 847 (1961) (restricting the rule of Bailey v. Glover to cases where the underlying cause of action sounds in fraud).
. Our research reveals only two District Court opinions and two Court of Appeals opinions suggesting that fraudulent concealment does not toll a statute of limitations governing claims against the United States. None are on point. See Lien v. Beehner, 453 F.Supp. 604, 606 (N.D. N.Y.1978) (finding that the equitable considerations raised by wrongful concealment did not toll the statute of limitations in a Federal Tort Claims Act case but not addressing the question of congressional intent in passing the applicable statute of limitations); Hammond v. United. States, 388 F.Supp. 928, 934 (E.D.N.Y. 1.975) (finding that fraudulent concealment did not toll the statute of limitations but relying on the fact that at that time the FTCA expressly exempted the United States from liability for the fraudulent torts of its employees). As for the Court of Appeals opinions, see Richter v. United States, 551 F.2d 1177 (9th Cir.1977) (merely citing Hammond and providing no further analysis); KSLA-TV, Inc. v. Radio Corp. of America, 732 F.2d 441, 443 (5th Cir.1984) (per curiam ) (finding that fraudulent concealment did not toll a Louisiana statute of peremption).
. There is no indication that Congress considered the question of fraudulent concealment and then failed to address the issue in the legislation itself. See Congressional Globe, 37th Cong., 3d Sess. 415-416 (Jan. 21, 1863).
. Nor does it contradict Kendall v. United States, 107 U.S. (17 Otto) 123, 125, 2 S.Ct. 277, 278, 27 L.Ed. 437 (1883), in which the Court held that the 1863 Act’s ennumeration of "disabilities” that might toll the statute of limitations were not to be enlarged. Although appellants have alleged that their psychological condition during the post-war period should affect our tolling analysis, see brief of appellants at 31, we have refrained from adopting their suggestion. Our opinion does not turn on any alleged "disability” of the appellants, but rather on the conduct of the appellee.
. The doctrine of tolling for fraudulent concealment may originally have been of equitable origin, but according to Bailey v. Glover, 88 U.S. (21 Wall.) 342, 348-49, 22 L.Ed. 636 (1875), it had been assimilated into legal doctrine by the late 19th century and did not merely apply to bills in equity. In this connection it should be noted that we reject appellants’ suggestion that the government is equitably estopped from raising the statute of limitations defense. Brief of appellants at 38. The statute of limitations is jurisdictional. See Soriano v. United States, 352 U.S. 270, 77 S.Ct. 269, 1 L.Ed.2d 306 (1957). Thus, even if appellee’s actions were so egregious as to constitute one of those rare circumstances where estoppel of the government might be appropriate, see Heckler v. Community Health Services of Crawford County, Inc., 467 U.S. 51,-, 104 S.Ct. 2218, 2224, 81 L.Ed.2d 42 (1984), this court would still have an obligation to consider the issue on its own motion. See Insurance Corp. of Ireland, Ltd. v. Compagnie des Bauxites de Guinee, 456 U.S. 694, 702, 102 S.Ct. 2099, 2104, 72 L.Ed.2d 492 (1982).
. This split may be more apparent than real. As noted in Campbell v. Upjohn, 676 F.2d 1122, 1128 (6th Cir. 1982), the effect of all tolling rules is usually to nullify the effect of fraudulent concealment on the reasonably diligent plaintiff. Thus the "actual discovery” cases appear to have generally concerned situations where concealment had been so effective that there was no reason for a diligent plaintiff to have entered into any inquiries as to a possible cause of action.
. This is not to suggest that knowledge of injury and injurer is without legal significance. Awareness of these facts plainly puts a plaintiff on notice to conduct further inquiries into the nature of his claim. But to be on notice of an obligation to inquire is not the same thing as to have notice of the factual basis of one’s claims. See Hobson v. Wilson, supra note 49, 737 F.2d at 35.
. Our analysis should not be taken to contradict United States v. Kubrick, 444 U.S. Ill, 100 S.Ct. 352, 62 L.Ed.2d 259 (1979). Kubrick simply did not address the question of when fraudulent concealment will toll the statute of limitations. Rather, Kubrick concerned the question of when a cause of action “accrues” in a case where there have been no allegations of fraudulent concealment. Indeed in Kubrick the defendant’s failure to concede facts pertinent to the question of causation was deemed to be of little importance given that the plaintiff could have discovered the relevant information by asking any competent doctor. Id. at 122, 100 S.Ct. at 359. Thus the holding in Kubrick, that a cause of action “accrues" under 28 U.S.C. § 2401(b) (1982) when a victim of medical malpractice is aware of his injury and not when he he had reason to know all facts pertinent to his cause of action, id. at 125, 100 S.Ct. at 360, does not contradict our analysis here. Our research reveals only one Court of Appeals to have suggested, even in dicta, that Kubrick’s analysis of what a plaintiff must know to start the running of the statute of limitations in the absence of fraudulent concealment might apply to cases that do turn on fraudulent concealment. See Premium Management, Inc. v. Walker, 648 F.2d 778 (1st Cir.1981). We believe logic to be on the side of those Courts of Appeals that have rejected this extension of Kubrick. See Arvayo v. United States, 766 F.2d 1416, 1422 (10th Cir. 1985) (applying Kubrick but stressing that there was no issue of concealment); Barrett v. United States, 689 F.2d 324, 328-330 (2d Cir. 1982) (refusing to extend the Kubrick rule where the extent of fraudulent concealment was put in issue).
Moreover, not only does our analysis not conflict with the holding of Kubrick, but we believe it to be in accord with the underlying rationale governing that case. The Kubrick Court based its holding on the view that a victim of medical malpractice has some duty to make further inquiries about his condition once he is aware of his injury. 444 U.S. at 118, 122-23, 100 S.Ct. at 357, 359-60. Thus Kubrick need merely have asked some doctor other than the one that had treated him about his condition and he would have quickly obtained all the information he needed to state a claim. See id. at 123 n.10, 100 S.Ct. at 360 n. 10. As noted below, we would require appellants, even though the victim of fraudulent concealment, to conduct the sort of inquiries mandated by the Kubrick Court.
. Lest our view be misconstrued, we would stress that the statute of limitations is not tolled whenever a defendant has concealed facts material to any legal issue of significance in a case. We do not provide for tolling simply because a plaintiffs ability to mount a successful case has been impaired in some degree. Instead, we provide for tolling only when concealment has so impaired the plaintiffs case that he is not *250able to survive a threshold motion to dismiss for failure to tender a claim that would advance beyond the pleading stage.
. Specifically, the District Court examined the CWRIC report and certain intelligence and Justice Department documents.
. According to the District Court, 586 F.Supp. at 788, the Ringle, Hoover, and Fly memoranda were first cited and discussed in M. Grodzins, Americans Betrayed: Politics and the Japanese Evacuation 188-189, 291 & n. 50 (1949). See also 586 F.Supp. at 788 n.26 (collecting other works analyzing the evacuation program).
. Appellant's dismiss these references as mere "secondary" sources. Reply brief of appellants at 6 n.6. Appellants, however, confuse the rules governing admissible evidence at trial and the *251rules determining when they were on notice of the factual basis of their claims. At the very least, the Grodzins book should have alerted appellants to the need to conduct further inquiries into the factual basis of their claims. And even assuming that the government had refused to disclose the Ringle, Hoover, and Fly memoranda before 1974, after 1974 and the passage of the Freedom of Information Act appellants would have experienced little difficulty in obtaining these documents.
. Thus the JACL brief included statements by the Attorney General and the Secretary of War in 1941-42 that there were no appreciable fifth column activities by Japanese-Americans, see Korematsu, Brief for JACL at 82; statements by President Roosevelt that there had never been a serious threat of invasion of the West Coast, id. at 87; and references to Ringle’s then anonymous article (signed "An Intelligence Officer”) stating that mass evacuations were not necessary, id. at 107-08.
. Suppression of Ringle’s analysis, without more, would not have tolled the statute. Ringle’s report did provide specific facts supporting the practicality of individualized action. Ringle Report at 1-3, JA 91-93. But even Ringle’s report could not provide anything more than incremental evidence against the government’s case. Ringle did not purport to have access to all intelligence data on the issue. The Court therefore most probably would have assumed that the military had discounted this report in light of conflicting data received from other sources.
Suppression of the Fly and Hoover memoranda, and of the factual weakness of the Final Report, in the Korematsu brief would have had an even more attenuated effect on a putative Takings Clause claim. As already noted, the controversy over the Korematsu brief primarily focused on evidence tending to confirm ex post the initial military judgment, evidence of actual acts of espionage. The Korematsu and Hirabayashi decisions, however, focused on the reason*252ableness of the evacuation program when viewed ex ante. In both decisions the Court relied on the military judgment that it was impossible to segregate the loyal from the potentially disloyal in an efficient manner. Thus, although discrediting some aspect of the Final Report might have undercut the presumption of deference to some limited degree, it likely would not have changed the outcome in Korematsu or of a Takings Clause claim.
. We do not suggest that appellants’ allegation has been established as a matter of fact. We merely review the grant of a motion on the pleadings as supplemented by historical documents. Should the District Court, on remand, find that the government did have countervailing intelligence data, contrary to the finding of the CWRIC report, it would be free to find that the statute of limitations was never tolled in this case.
. We need not, and do not, pass on whether the concealment at issue violated any ethical obligation of the Solicitor General to the Court. As noted in Hobson v. Wilson, supra note 49, 737 F.2d at 42, it is not the law of this circuit that the concealment itself must be "wrongful." Moreover, the Hobson court held that even if, arguendo, concealment itself must be wrongful, the "wrongfulness” of a concealment can consist of the fact that the underlying government action — in this case a taking without just compensation — was wrongful.
. Indeed, the injuries suffered during the evacuation were probably sufficient to put appellants on inquiry notice.
. But cf. Richards v. Mileski, 662 F.2d 65, 71 (D.C.Cir.1981) (a reasonably diligent plaintiff is not necessarily expected to exercise his rights under the Freedom of Information Act).
. Appellee argues that an earlier statement by the Executive had the effect of ending the presumption of deference to the decision of the war-making branches. See Presidential Proclamation 4417, 12 Compilation of Presidential Documents 245 (1976). In that Proclamation President Ford finally repealed Executive Order 9066. In so doing he did state that Executive Order 9066 had been a "mistake." Id. He did not suggest, however, that it had been a legal error. The Proclamation is therefore fairly read merely to state that although the military had the legal authority to act as it did, the course chosen was morally mistaken. Moreover, although President Ford did affirm the loyalty of the Japanese-Americans, id. at 246, he said nothing about the critical issue of whether it would have been practical to segregate the loyal from the disloyal in an efficient manner.
. Of course, it is possible to read the Act as merely stating that Congress no longer need be satisfied with the previous explanations for the internment and that Congress, with its power to create commissions with staffs and subpoena power, would now look into this issue anew. This argument, however, only suggests that Public Law 96-317, standing alone, could not produce a legal victory for any former evacuee who brought a claim. But this court has never suggested that claimants must have all of the evidence brought before them “on a silver platter” before the statute begins to run. Brief of appellee at 37.
. It is true that Congress believed that the evacuees had no alternative remedies at the time it passed the Claims Act. In enacting § 1984(d), however, Congress was guarding against future contingencies. Our analysis of the legislative history of the Claims Act thus does not undermine our reliance on the language of § 1984(d).
. We do not believe that § 1984(d) constituted an unconstitutional condition on the exercise of the evacuees’ rights under the statute. Although as a general rule the government may not require individuals to waive constitutional rights as a precondition of receiving a bounty, see Speiser v. Randall, 357 U.S. 513, 518, 78 S.Ct. 1332, 1338, 2 L.Ed.2d 1460 (1958); Frost Trucking Co. v. Railroad Comm’n, 271 U.S. 583, 593, 46 S.Ct. 605, 607, 70 L.Ed. 1101 (1926), such conditions can be imposed where the right waived is rationally related to the benefit conferred. See, e.g., Stephenson v. Binford, 287 U.S. 251, 275, 53 S.Ct. 181, 188, 77 L.Ed. 288 (1932). We are not prepared to say that the waiver of subsequent suits was so unrelated to the provision of an inexpensive and convenient procedure as to render § 1984(d) constitutionally infirm.
. Although the Supreme Court has yet to determine whether it is ever possible to estop the United States, the Court has stated that, at a minimum, affirmative misconduct must be alleged and such misconduct must be directly responsible for a detrimental change of posi*256tion. See Heckler v. Community Health Services of Crawford County, Inc., supra note 53, 467 U.S. at-, 104 S.Ct. at 2224, 81 L.Ed.2d 42. In ornease the government’s relevant alleged affirmative misconduct occurred in 1942. Although the Solicitor General’s action was plainly a contributing factor to the Supreme Court's decision, it also appears that Congress made an independent good faith decision in 1948 to accept the facts as stated in Korematsu. Given this intervening good faith decision by Congress, appellants have failed to allege an unbroken chain of causation between the government’s misconduct in 1942 and the terms of the Claims Act in 1948. On these facts, estoppel will not lie against the United States.