concurring in judgment only:
Although I agree with my colleagues that the plaintiffs’ suit for damages against the United States should be dismissed, I reach that conclusion by a different route. I believe that Congress did intend that the statutory provisions depriving federal district courts of jurisdiction over the underlying tort suit would be contingent on the validity of the espousal agreement. I also conclude that the plaintiffs’ central challenge to the espousal agreement is suitable for judicial resolution and does not present a political question. In my view, however, the espousal arrangement at issue here is an entirely valid international agreement. I therefore agree that the restrictions on federal jurisdiction should be enforced and the suit should be dismissed.
I. Are the Jurisdiction-Stripping Provisions Dependent on the Validity of the Espousal?
A. Constitutional Concerns
The majority regards the jurisdiction-stripping provisions of the Compact of Free Association Act as separate and apart from the espousal agreement. In my colleagues’ view, the Congress has plenary power to strip the federal district courts of jurisdiction, at least over nonconstitutional claims, whatever the legal status of the espousal. This argument presumes that Congress could constitutionally have withdrawn the government’s consent to suit on these claims even if it had provided no alternative compensation mechanism at all. I find that assumption quite problematic.
The majority’s position rests in large part on the fact that the plaintiffs’ underlying claim for relief sounds in tort. While much controversy surrounds the extent of Congress’ obligation to provide a judicial forum for the vindication of constitutional claims, see, e.g., Bartlett v. Bowen, 816 F.2d 695 (D.C.Cir.1987); see also Webster v. Doe, — U.S. —, 108 S.Ct. 2047, 2053, 100 L.Ed.2d 632 (1988) (noting “the ‘serious constitutional question’ that would arise if a federal statute were construed to deny any judicial forum for a colorable constitutional claim”), there has been next to no discussion of jurisdiction-stripping practices where nonconstitutional claims are concerned.1 The Federal Tort Claims Act is of course a fairly recent innovation, and no one suggests that it is constitutionally compelled. The Act contains a number of exceptions, and Congress could clearly enact new ones. The government in essence portrays the Compact Act as simply one more exception to the FTCA: “Since there manifestly was no constitutional requirement that Congress waive the sovereign immunity of the United States for tort claims, there is likewise no constitutional impediment to the reassertion of immunity to bar the further exercise of court jurisdiction over these claims.” Brief for United States at 22-23. Therefore, the government argues, Congress could have removed these claims from the jurisdiction of the *386federal courts even without creating an alternative compensation system.
Some Supreme Court opinions contain expansive language regarding the power of Congress to reassert the sovereign immunity of the United States. See, e.g., Lynch v. United States, 292 U.S. 571, 581, 54 S.Ct. 840, 844, 78 L.Ed. 1434 (1934) (“The rule that the United States may not be sued without its consent is all embracing.”); Maricopa County v. Valley National Bank, 318 U.S. 357, 362, 63 S.Ct. 587, 589, 87 L.Ed. 834 (1943) (citing Lynch for the proposition that “the power to withdraw the privilege of suing the United States or its instrumentalities knows no limitations”). Despite this sweeping language, it seems to me that there must be some limits on Congress’ ability to reassert sovereign immunity, even over nonconstitutional claims.
The government’s position is in essence that, since the United States need not ever consent to be sued for damages, it has absolute authority to define the circumstances under which suit will be permitted. This contention, however, cannot be accommodated with equal protection doctrine generally, which clearly holds that even government “privileges” may not be allocated in a discriminatory manner.2 This exception is particularly relevant in a case involving access to the courts.3 Certainly Congress could have repealed the FTCA entirely, and it could have provided that the FTCA would not apply to suits arising out of U.S. nuclear testing. I cannot believe, though, that Congress could amend the Act to provide that blacks cannot sue the United States for damages. A slightly different constitutional problem would be posed by a statute providing that a particular individual cannot invoke the FTCA. Such a statute would appear to deny equal protection and would also implicate values protected by the bill of attainder clause. The same sorts of problems, it appears to me, would arise here if the jurisdiction-stripping provisions of the Compact Act stood alone.
The jurisdiction-stripping language (considered in isolation) does not define a neutral class of claims (e.g., all claims arising from nuclear testing) for which consent to suit has been withdrawn. Nor does it distinguish between citizen and alien,4 or between domestic and extraterritorial claims5 — distinctions which, while not without their drawbacks, would probably in my view withstand constitutional scrutiny. Rather, the Act singles out a narrow group of claimants, and a narrow class of claims, which are in all relevant respects (save, of course, for the existence of the Compact and the espousal, which for the sake of this discussion we must ignore) indistinguishable from other claims and claimants for which a federal cause of action remains. It seems likely, moreover, that the great majority of these plaintiffs are Asians;6 and *387Asians have, through most of our country’s history, been subjected to repeated racial discrimination. Against this backdrop, I cannot accept my colleagues’ blithe assumption that withdrawal of jurisdiction implicates no constitutional concerns.
The dearth of record information concerning the plaintiffs’ racial characteristics —or the existence of ethnic variations within the Micronesian community — does not (as my colleagues suggest, see Majority opinion (Maj. op.) at 378-79) eliminate the danger of invidious discrimination. After all, national origin is itself a suspect basis for government classification. See Hernandez v. Texas, 347 U.S. 475, 479, 74 S.Ct. 667, 671, 98 L.Ed. 866 (1954). The Compact, which withdraws federal jurisdiction over “all claims, past, present and future, of the Government, citizens and nationals of the Marshall Islands which are based upon ... the Nuclear Testing Program,” clearly differentiates upon this basis. For purposes of equal protection review, the crucial question is not whether a precisely definable racial or ethnic class is involved, but whether “the group is one that is a recognizable, distinct class, singled out for different treatment under the laws, as written or as applied.” Castaneda v. Partida, 430 U.S. 482, 494, 97 S.Ct. 1272, 1280, 51 L.Ed.2d 498 (1977). It seems quite plain to me that the Marshall Islanders do comprise such a group.
My colleagues argue that no suspect classification based on national origin exists because an espousal agreement by its very nature can settle only the claims of a particular country’s nationals. See Maj. op. at 379. I agree, but I do not see how this argument can be reconciled with the rest of the majority’s analysis. I readily concede that, within the context of the Compact as a whole, there is an obvious, legitimate, nondiscriminatory reason for singling out this group of potential plaintiffs. The problem lies with the majority’s insistence on reviewing, and approving, the jurisdiction-stripping provision in isolation. My colleagues contend that the espousal and the withdrawal of jurisdiction are sev-erable provisions, and they would uphold the latter without examination of the former. Yet in rebutting the inference of discrimination that a selective withdrawal of jurisdiction would create, they point out that the withdrawal of jurisdiction is only a part of a complex international agreement.
If the jurisdiction-stripping provision of this bill stood alone, it would raise grave constitutional concerns. The class of claims over which federal jurisdiction has been withdrawn has not been defined by neutral criteria. Rather, the statute has identified a very narrow set of claimants, defined on the basis of nationality, most of whom would appear to belong to a racial group which, by traditional equal protection standards, constitutes a suspect class. I certainly do not suggest that the espousal of these claims is an act of racial discrimination; my point is simply that, if the jurisdiction-stripping provisions were reviewed in isolation, the racial composition of the plaintiff class would surely be a relevant concern. I am not prepared to state categorically that the Constitution would forbid Congress to withdraw its consent to suit on these claims without providing an alternative means of redress. I do believe, though, that the constitutional doubts concerning such a statute are strong enough that the Compact of Free Association Act should be construed, if possible, so as to avoid these difficulties. In my view, the petitioners’ construction of the statute — under which the government’s withdrawal of consent to suit is conditioned upon the validity of the espousal — alleviates these constitutional doubts and is fully consistent with both the language and history of the Act.
B. Statutory Language
Section 103(g)(2) of the CFAA states:
It is the explicit understanding and intent of Congress that the jurisdictional limitations set forth in Article XII of such Agreement are enacted solely and exclu*388sively to accomplish the objective of Article X of such Agreement and only as a clarification of the effect of Article X, and are not to be construed or implemented separately from Article X.
Pub.L. No. 99-239, 99 Stat. 1770, 1782 (1986) (App. I at 101). This language is hardly free from ambiguity. On the one hand, the Senate clearly rejected other wording, originally contained in the House bill, which would have stated unequivocally that the jurisdiction-stripping provisions would apply only if the espousal were held to be valid.7 On the other hand, the Congress did not simply eliminate all references to the interdependence of the jurisdictional and espousal provisions — as would seem to be the most logical course of action if the withdrawal of jurisdiction were intended to stand alone. Instead House and Senate leaders accepted compromise language which gave complete satisfaction to no one. It is therefore not too surprising that the words of the statute fail to resolve this question unambiguously.
What is most striking, in my view, is the failure of either the majority or the government to suggest a plausible alternative reading of § 103(g)(2). The majority suggests that “[t]he language of Section 103(g)(2) upon which plaintiffs rely simply makes it plain that the deprivation of jurist diction applies not to all claims by the Marshall Islanders against the United States, but only those described in Articles X and XI of the Section 177 Agreement. Presumably, any other claim, under the Federal Tort Claims Act or other authority, could proceed.” Maj. op. at 375-76. With respect, this seems to me most unlikely. Article XII provides that “[a]ll claims described in Articles X and XI of this Agreement shall be terminated. No court of the United States shall have jurisdiction to entertain such claims, and any such claims pending in the courts of the United States shall be dismissed.” Even in the absence of § 103(g)(2), I do not see how this language could possibly suggest that consent to suit was withdrawn as to claims other than those described in Articles X and XI.8 The majority’s interpretation of § 103(g)(2) would render that section entirely superfluous. The same may be said of the government’s contention that § 103(g)(2) “simply confirms that [Articles X and XII] are to be implemented together, as already required by the terms of the Agreement itself.” Brief for United States at 16. It is hard to see what legal force can attach to a requirement that two provisions must be “implemented together,” unless the applicability of one is contingent upon the lawfulness of the other.
Section 103(g)(2) is assuredly not the most straight-forward way of saying that the withdrawal of federal jurisdiction was contingent upon judicial validation of the espousal provisions. But I fail to see what else § 103(g)(2) could mean. Rather than assume that this section is a nullity — that it either says nothing at all or else repeats what is already clear from other portions of the statute — I would read this language to provide that a valid espousal is a neces*389sary precondition for the abrogation of judicial relief. My preference for this reading is of course supported by my view that withdrawal of jurisdiction, standing alone, would raise a substantial constitutional question.
C. Legislative History
The legislative history offers only slight assistance in resolving the interpretive questions at issue here. Unlike the majority, however, I believe that this history supports rather than undermines the plaintiffs position.
The plaintiffs rely heavily on the following statement of Representative Seiberling:
[Section 103(g)(2) ] is intended to make it clear that court-stripping provisions of article XII of the section 177 agreement have no independent force or effect and their sole function is to implement the provisions of article X. Thus, if article X is valid, espousal stands; and if article X is invalid, claims covered by the espousal provisions will remain justiciable in U.S. courts, regardless of article XII.
131 Cong.Rec. H11,829 (daily ed. Dec. 11, 1985) (App. II at 609).9 This statement is of somewhat doubtful authority: as the government points out, it was not spoken on the floor during debate but was inserted into the text subsequently under permission to extend remarks. See Brief for United States at 19. Representative Seiberling did, however, state during floor debate that “the provisions of the compact regarding espousal of the Marshall Islands’ claims are subject to judicial review like any act of Congress,” 131 Cong.Rec. H11,-838 (daily ed. Dec. 11, 1985) (App. II at 618), and that statement went uncontradicted.10
The statements of a single Representative constitute pretty scanty evidence, but the government has offered little in response. The government and the majority rely heavily on statements to the effect that the Compact was intended to provide a comprehensive settlement of tort claims stemming from American nuclear testing in Micronesia. Representative Solarz, a sponsor of the Act, stated that “the compact settles all nuclear claims resulting from our nuclear weapons testing program.” 131 Cong.Rec. H11,836 (daily ed. Dec. 11, 1985) (App. II at 616). The section-by-section analysis of the Compact Act stated that § 103(g) “reiterates the provisions of Section 177 of the Compact which provides that there is full and final settlement of all nuclear effects claims.” 131 Cong.Rec. S17,651 (daily ed. Dec. 13, 1985) (App. II at 621). Both the government and the majority view these statements as evincing a congressional intention to shield the espousal provisions from judicial review. See Brief for United States at 20; Maj. op. at 377.
In my view this is an unconvincing argument. If Congress enacted legislation dealing with a knotty banking problem (for example), and if the relevant committee reports described the statute as a “final settlement” or “comprehensive solution” to the problem, no one would suppose that Congress thereby intended to preclude judicial review. In the present case, a Congressperson who believed that the espousal provisions were judicially reviewable, but that they would ultimately be upheld in court, might very easily state that the Compact would constitute a “full and final settlement.” In fact, Representative Lago-marsino — a supporter of the Compact Act on whose views the government relies, see Brief for United States at 20-21 — expressed essentially this view:
*390If the section 177 agreement is challenged in the courts, it is the view of the Department of Justice and the Department of State, whose representatives testified on the section 177 agreement during our hearings on the compact, that the United States will prevail and that the constitutionality of this agreement will be affirmed.
132 Cong.Rec. 520 (January 23, 1986) (extension of remarks) (App. II at 624). In short, I see no conflict between Representative Solarz’s view that “the compact settles all nuclear claims” and Representative Seiberling’s assurance that the espousal provisions “are subject to judicial review like any act of Congress.” It is surely the norm that Congress, when it passes legislation, will expect both that the new statute will be challenged in court, and that the law will survive judicial review and thereafter fulfill its stated purpose.
Finally, I think that the plaintiffs’ position is supported by the logic of the situation. This is not a case in which Congress first decided to strip the courts of jurisdiction and then looked around to see what might be done to lessen the hardship of the Marshall Islanders. Rather, the espousal and jurisdiction-stripping provisions appear to have been regarded throughout the process as integrally connected. Given this background, it is hard to imagine that Congress would have wished the jurisdiction-stripping provisions to take effect even if the espousal provisions were invalidated.11
The majority concludes that the withdrawal of jurisdiction over these claims should be viewed as standing alone, and should be affirmed on this basis. In light of the serious constitutional difficulties that would attend such a selective denial of judicial relief, I would hesitate to interpret the statute in that way if another plausible reading is available. In my view, both the language and the history of the Compact Act support the plaintiffs’ contention that their suit for damages should be dismissed only if the espousal is found to be valid.
II. Does the Validity of the Espousal Present a Political Question?
Judge Sentelle also states that “even if we err in our interpretation of [the Compact] Act ... the District Court was without jurisdiction over this matter of international relations by reason of the political question doctrine.” Sentelle opinion (Sen-telle op.) at 379. My colleague does not, however, follow through with the implications of that statement. For if Judge Sen-telle has erred in his interpretation of the Compact Act, then Congress has in fact provided by statute that the validity of the jurisdiction-stripping provisions is to be contingent on a judicial assessment of the espousal. The political question doctrine can furnish a true alternative rationale only if my colleague is willing to say that judicial intervention in this area is constitutionally forbidden even if Congress has invited the courts to play a role. If we assume that the Compact Act provides for judicial review of the espousal, then judicial review is not, in my view, an untoward intrusion into the political branches’ negotiation of international agreements. Rather, the courts, by reviewing the espousal, would (as plaintiffs’ counsel put it at oral argument) simply be enforcing the agreement according to its terms.
Leaving that point aside, I believe that Judge Sentelle’s political question analysis is deeply flawed. My colleague appears to assert two distinct bases for finding this to be a political question: (1) that the issues presented are unsuitable for judicial resolution; and (2) that the political branches’ conduct of foreign policy would be undermined if the court were to review the espousal. The first argument rests on a misunderstanding of the plaintiffs’ claims; the *391second rests on a misconception of the political question doctrine.
A. Suitability of the Issues for Judicial Resolution
If the plaintiffs were attacking the espousal on the ground that the purported government of the Marshall Islands was not truly sovereign, then I would agree that a political question had been posed. Plaintiffs, however, have expressly disavowed such a challenge. See Brief for Appellants at 12. As I discuss at greater length infra, I also believe that any challenge to the adequacy of the settlement is nonjusticiable. Plaintiffs’ central contention, however, is that the espousal is invalid because international law forbids the espousal of claims held by persons who were not nationals of the espousing state at the time the claims arose. That issue seems to me entirely suitable for judicial resolution.
To accept the plaintiffs’ argument, we must accept three distinct propositions: (1) that the withdrawal of jurisdiction is contingent on the validity of the espousal; (2) that the espousal is invalid as a matter of domestic law if it contravenes international law; and (3) that under international law a state may espouse only the claims of individuals who were its nationals when the claims arose. My colleague may disagree with each of these propositions, but I cannot fathom how any one of them could be deemed unfit for judicial resolution. Each is a pure question of law which requires no foreign policy expertise and implicates no uniquely political concerns.
B. Interference With the Political Branches
My colleague also argues that this issue raises a political question because, by undermining the relationship between the governments of the United States and the Marshall Islands, it may interfere with the political branches’ conduct of foreign relations. This argument, in my view, sweeps much too broadly. In Baker v. Carr, 369 U.S. 186, 211, 82 S.Ct. 691, 706-07, 7 L.Ed. 2d 663 (1962), the Supreme Court cautioned that “it is error to suppose that every case or controversy which touches foreign relations lies beyond judicial cognizance.” More recently, the Court reaffirmed, in a case with a potentially substantial impact on foreign relations, that “the courts have the authority to construe treaties and executive agreements, and it goes without saying that interpreting congressional legislation is a recurring and accepted task for the federal courts_ [Ujnder the Constitution, one of the Judiciary’s characteristic roles is to interpret statutes, and we cannot shirk this responsibility merely because our decision may have significant political overtones.” Japan Whaling Association v. American Cetacean Society, 478 U.S. 221, 230, 106 S.Ct. 2860, 2866, 92 L.Ed.2d 166 (1986).
There exists, in fact, a long line of Supreme Court cases resolving on the merits legal issues with highly significant implications for foreign affairs. In Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 590-91, 72 S.Ct. 863, 868-69, 96 L.Ed. 1153 (1952), the Court invalidated President Truman’s seizure, pursuant to his authority as Commander-in-Chief during the Korean War, of private steel mills to avert a nationwide strike which the President asserted would “immediately jeopardize and imperil our national defense and the defense of those joined with us in resisting aggression, and would add to the continuing danger of our soldiers, sailors, and airmen engaged in combat.” In United States v. Nixon, 418 U.S. 683, 94 S.Ct. 3090, 41 L.Ed.2d 1039 (1974), the Court enforced a subpoena of Presidential documents, although a foreseeable result of its decision was a diminution of Presidential prestige and authority, with obvious ramifications for American foreign policy. In Japan Whaling Association, supra, the Court considered and rejected the government’s contention that the political question doctrine barred judicial resolution of a challenge to the Commerce Secretary’s decision not to certify Japan for its violation of whale-harvesting rules. The closest parallel to the present case would appear to be Dames & Moore v. Regan, 453 U.S. 654, 101 S.Ct. 2972, 69 L.Ed.2d 918 (1981), which involved a challenge to the Iran claims *392settlement. A decision adverse to the government would plainly have had far-reaching effects on U.S. foreign policy: it would have infuriated Iran by abrogating the settlement, and it would have demonstrably tied the President’s hands in dealing with similar crises in the future. The Court nevertheless resolved the dispute on the merits; I believe that the present case is also justiciable.
The central principle, cited but (in my view) misconstrued by my colleague, is that “[t]he doctrine of which we treat is one of ‘political questions,’ not one of ‘political cases.’ ” Baker v. Carr, 369 U.S. at 217, 82 S.Ct. at 710. I read this passage as a reminder that our focus should be on the particular issue presented for our consideration, not the ancillary effects which our decision may have on political actors. If we are called upon to resolve a question (such as the identity of a foreign state’s legitimate ruler) which courts are ill-qualified to decide, then invocation of the political question doctrine is appropriate. But when the issue is well-suited for judicial resolution (as is surely the case with the purely legal questions presented here), then we should not decline to adjudicate simply because the impact of our ruling cannot be cabined within a wholly apolitical realm.
III. Is the Espousal Valid?
Having concluded that the jurisdiction-stripping provisions of the Compact Act are contingent upon the validity of the espousal, and that the plaintiffs’ central challenge to the espousal is justiciable, I proceed to the merits of the plaintiffs’ challenge. Plaintiffs assert three distinct attacks on the espousal provisions of the Compact Act.
A. Adequacy of the Settlement
The plaintiffs assert that the espousal should be invalidated because the amount of the settlement — $150 million — is inadequate in comparison to the injuries that they have suffered. They invoke concepts of trusteeship and fiduciary duty in arguing that the United States Government owes a continuing obligation to the residents of the Marshall Islands. In my view, however, this court is precluded from inquiring into the adequacy of the settlement.12
Whatever the prior relationship between the United States Government and the people of the Marshall Islands, under the Compact it is the Marshall Islands Government which must protect the interests of its residents. Once we acknowledge the power of the Marshall Islands Government to espouse these claims, then any dispute as to the adequacy of the settlement must be recognized as in substance a dispute between the plaintiffs and their own government. Such a dispute would seem to lie beyond the purview of this court, both because this court lacks the authority to inquire into the adequacy of another government’s representation of its own people, and because the United States Government cannot in any event be held responsible for another government’s failings. Nor is this transformed into a claim against the United States by the contention that the United States Government has “coerced” a favorable settlement. The American Government, after all, typically bargains from a position of strength; and I see no basis on which this court could scrutinize international agreements and inquire into the ade*393quacy of the consideration given to foreign states.
Moreover, any inquiry into the adequacy of the settlement figure would require, in essence, that the district court try the lawsuit. The adequacy of the settlement, after all, depends both on the extent of the plaintiffs’ injuries — a very difficult determination in itself13 — and on the likelihood of their collecting a judgment in the face of the government’s formidable defenses.14 The whole point of the espousal, however, was to achieve expeditious settlement of these claims and avoid protracted litigation. I believe that Congress did intend for the withdrawal of federal jurisdiction over the tort claims to be contingent on the validity of the espousal. But it would seem to me a bizarre result if we could uphold the espousal only after completing the sort of extended inquiry which the settlement was designed to prevent.
B. Takings Claim
The plaintiffs also contend that the espousal is invalid because it constitutes a taking of property without just compensation, and thus violates the fifth amendment.15 This claim fails on two grounds. First, it ignores the Supreme Court’s admonition that “consent to sue the United States is a privilege accorded; not the grant of a property right protected by the Fifth Amendment.” Lynch v. United States, 292 U.S. 571, 581, 54 S.Ct. 840, 844, 78 L.Ed. 1434 (1934). Moreover, the fifth amendment does not prohibit all takings, but only takings for which “just compensation” has not been paid. The compensation provided by the international agreement is presumptively just; for reasons noted earlier, this court is precluded from inquiring into the adequacy of the settlement.
C. International Law Claim
The plaintiffs’ principal claim is that the espousal provisions violate international law because the individuals whose claims were settled were not nationals of the Marshall Islands at the time the injuries occurred.16 This line of argument requires the plaintiffs to establish both that the espousal is invalid as a matter of domestic law if it violates international law, and that international law in fact forbids this sort of espousal agreement. The plaintiffs would appear to face a formidable obstacle in this court’s recent decision in Committee of U.S. Citizens Living in Nicaragua v. Reagan, 859 F.2d 929 (D.C.Cir.1988). The court there considered the question whether a federal statute could be invalidated on the ground that it violates customary international law.17 The panel concluded that *394“the law in this court remains clear: no enactment of Congress can be challenged on the ground that it violates customary international law.” 859 F.2d at 939. It therefore might appear that, even if the withdrawal of jurisdiction is contingent on the validity of the espousal, the espousal cannot be deemed invalid on the basis of its alleged incompatibility with international law. In response, the plaintiffs offer two closely related arguments in support of their contention that international legal norms should be controlling here.
The first argument is based on congressional intent. Plaintiffs appear to concede that Congress may, if it wishes, pass a statute which contravenes customary international law, and that federal courts in such a circumstance are obliged to give effect to the statute. But in plaintiffs’ view Congress, by negotiating an international agreement in accordance with international legal processes, and by using the technical term “espousal,” has signaled its intention that the Compact Act should be reviewed on the basis of international law. Thus, plaintiffs do not argue that customary international law can supersede the will of Congress; they simply contend that application of international law is essential to a proper construction of the Compact Act.
In advocating this interpretation of the statute, the plaintiffs rely on the Supreme Court’s recent statement that “[i]t has been a maxim of statutory construction since the decision in Murray v. The Charming Betsy, 2 Cranch 64, 118, 2 L.Ed. 208 (1804), that ‘an act of congress ought never to be construed to violate the law of nations, if any other possible construction re-mains_’” Weinberger v. Rossi, 456 U.S. 25, 32, 102 S.Ct. 1510, 1516, 71 L.Ed.2d 715 (1982). If congressional intent regarding the applicability of international legal norms is ambiguous, plaintiffs argue, then the court should assume that international law is to govern. The court should not, in other words, in the absence of clear evidence presume that Congress would wish for domestic legal consequences to follow from an international agreement which is itself violative of international law.
In the end, though, I think it is unnecessary to determine whether international law applies, because in my "view the espousal does not violate international law. The plaintiffs are correct in stating that espousal agreements are typically limited to the settlement of claims held by persons who were nationals of the espousing country at the time the claims arose. But I think the government is correct in arguing that this is a matter of custom rather than a binding rule, and that no international legal norm forbids countries from agreeing to the espousal of claims held by individuals who do not meet the strictures of the continuous nationality rule. See, e.g., 1 G. Schwarzenberger, International Law 594 (3d ed. 1957) (App. II at 767) (“Like any other rule of international customary law, the rule on the nationality of claims may be modified or abrogated by means of treaties.”). See also 1 R. Lillich and B. Weston, International Claims: Their Settlement by Lump Sum Agreements 48-50 (1975) (App. II at 742-43).
An understanding of the reason for the general custom shows why it is of no help to the plaintiffs in this case. General acceptance of the continuous nationality rule is a principle of relations between nations. It prevents one nation from exerting undue pressure on another by espousing the claims of individuals to whom it bears only a tenuous relationship. It also removes incentives for individuals to change nationality in order to gain the assistance of a powerful state. Neither of these concerns is implicated in the present case. Rather, plaintiffs’ quarrel with the espousal is at bottom a quarrel with their own government. But the rule concerning the nationality of claims — like international law gen*395erally — was never intended to be a measure for limiting the authority of a government vis-a-vis its own nationals.18 These plaintiffs are seeking in essence to invoke a rule of international law as a basis for challenging their own government’s power to settle their claims. I do not believe, though, that international law in any way forbids the agreement reached here.
IV. Conclusion
Although I, like my colleagues, conclude that the district court lacked jurisdiction to address the plaintiffs’ claim for damages under the FTCA, the difference between my approach and theirs seems to me sufficiently important to justify a separate opinion. Despite the broad dictum in some Supreme Court decisions, I am troubled by the majority’s uncritical assumption that Congress may deprive a narrow class of plaintiffs of judicial relief, even if their claims are nonconstitutional. The interplay between the very broad congressional control over federal jurisdiction (particularly in the context of damage suits against the United States), and the fundamental requirement that access to the federal courts must be granted on a nondiscriminatory basis, would seem to me to pose a difficult constitutional question. I hesitate to address this question absent clear evidence that Congress did in fact intend to effect an unconditional withdrawal of jurisdiction.
In my view, no such evidence is present in this case. Rather, it appears to me that the jurisdiction-stripping and espousal provisions of the Compact Act were consistently thought of as parts of a package. Congress did not seriously consider enacting one without the other, and it seems artificial to review the withdrawal of jurisdiction as though it stood in isolation. I believe that Congress intended that the withdrawal of jurisdiction should be contingent on judicial approval of the espousal mechanism. However, because I find no infirmity in the espousal, I conclude that the withdrawal of federal jurisdiction was valid and that the plaintiffs’ claims for damages were properly dismissed.
. The presence of constitutional doubts concerning the power of Congress to withdraw federal jurisdiction over constitutional claims is not, however, entirely irrelevant to the present case. The Compact Act espoused, and withdrew federal jurisdiction over, constitutional as well as nonconstitutional causes of action. Under the majority’s construction of the statute, the jurisdiction-stripping provisions are to be implemented without regard to the propriety of the espousal. My colleagues thus impute to Congress the view that the federal courts may be stripped of jurisdiction over constitutional claims even if no alternative remedy is provided. Admittedly, our resolution of this tort suit does not require us to decide whether Congress actually possesses such power. But in interpreting the statute, we should not lightly presume that Congress intended to take so constitution?' ly problematic a step.
. The Supreme Court "now has rejected the concept that constitutional rights turn upon whether a governmental benefit is characterized as a ‘right’ or as a ‘privilege.’ ’’ Graham v. Richardson, 403 U.S. 365, 374, 91 S.Ct. 1848, 1853, 29 L.Ed.2d 534 (1971). The Court’s sweeping pronouncements in Lynch and Maricopa County, it might be noted, were issued in an era when the right-privilege distinction held far greater sway. See generally Van Alstyne, The Demise of the Right-Privilege Distinction in Constitutional Law, 81 Harv.L.Rev. 1439 (1968).
. Cf. Lindsey v. Normet, 405 U.S. 56, 77, 92 S.Ct. 862, 876, 31 L.Ed.2d 36 (1972) ("This Court has recognized that if a full and fair trial on the merits is provided, the Due Process Clause of the Fourteenth Amendment does not require a State to provide appellate review.... When an appeal is afforded, however, it cannot be granted to some litigants and capriciously or arbitrarily denied to others without violating the Equal Protection Clause.”).
. In my view a law providing that "citizens may sue the United States for damages, but aliens may not” would be far less constitutionally suspect than a law providing that ."aliens of British ancestry may sue the United States for damages, but aliens of Japanese ancestry may not."
. The Federal Tort Claims Act contains a foreign claims exception. See 28 U.S.C. § 2680(k).
. The record contains no precise delineation of the racial or ethnic characteristics of the plaintiff class. One House committee report states that "[t]he Trust Territory of the Pacific Islands, commonly referred to as Micronesia, consists of over 2,000 islands scattered over 3 million square miles of oceán. The population of 130,-000 live on some 100 islands and consist of six distinct ethnic groups speaking nine different languages.” H.R.Rep. No. 188, Part 1, 99th Cong., 1st Sess. 3 (1985), U.S.Code Cong. & Admin.News 1985, pp. 2746, 2748 (App. I at *387181). Although the population of Micronesia is characterized by significant diversity, “[t]he peoples of all of these islands are ethnically linked, and trace their ancestry to migrations from Asia.” S.Rep. No. 433, 94th Cong., 1st Sess. 23 (1975).
. Section 5(n)(2) of the House bill provided that: "If, notwithstanding the enactment into law of this Act, a United States court of competent jurisdiction determines that the provisions of Article X of the Agreement ... are invalid as a matter of international law or for any other reason, the provisions of Article XII of the Section 177 Agreement shall not, of themselves, prevent any court of the United States otherwise having jurisdiction over claims described in Article X and XI of the Section 177 Agreement from entertaining such claims; and the time between the effective date of the Compact and any subsequent final determination of invalidity of Article X of the Section 177 Agreement shall not be included in any calculation regarding applicable statutes of limitations or other similar limitations pertaining to the presentation of any such claims to any such court.” Reprinted in H.R.Rep. No. 188, Part 2, 99th Cong., 1st Sess. 79 (1985), U.S.Code Cong. & Admin.News 1985, p. 2746 (App. I at 411).
. Section 103(g)(1) of the Act states: "It is the intention of the Congress of the United States that the provisions of section 177 of the Compact of Free Association ... constitute a full and final settlement of all claims described in Articles X and XI of the Section 177 Agreement, and that any such claims be terminated and barred except insofar as provided for in the Section 177 Agreement.” 99 Stat. 1782 (App. I at 101). It seems to me that the statute, like the Agreement, quite clearly restricts the withdrawal of jurisdiction to the claims described in Articles X and XI.
. Representative Seiberling also stated: "The House bill also included language specifying that should article X of the section 177 subsidiary agreement be held invalid, article XII of that agreement would have no effect; while this language has been somewhat revised, the new revision is to the same effect.” 131 Cong.Rec. H11,829 (daily ed. Dec. 11, 1985) (App. II at 609).
. The Congressman went on to say that "a legacy of the unique relationship of the United States to the Marshall Islanders under the U.N. trusteeship agreement will be the pending constitutional questions with respect to their rights, questions which cannot be foreclosed from court review.” 131 Cong.Rec. H11.838 (daily ed. Dec. 11, 1985) (App. II at 618).
. The interpretive problem posed here might be characterized as a question concerning the severability of statutory provisions. To give independent effect to the jurisdiction-stripping language, we must conclude both that this provision would be constitutional in isolation, and that Congress would have wished to withdraw federal jurisdiction even if the espousal were deemed invalid. See Lynch, supra, 292 U.S. at 586, 54 S.Ct. at 846. Both of these propositions are in my view quite doubtful.
. Congress, it might be noted, has been quite inconsistent in its defense of the settlement figure. The report of the House Committee on Foreign Affairs stated that “[t]he compact is also very much in the economic interests of the United States.... At the present time, there are approximately $5 billion in nuclear claims suits pending against the U.S. Government. Without the settlement contained in the compact, these suits will proceed, and it is quite possible that adverse decisions will be handed down against the U.S. Government that will cost it, and thus the taxpayer, considerably more than $150 million.” H.R.Rep. No. 188, Part 1, 99th Cong., 1st Sess. 5 (1985) U.S.Code Cong. & Admin.News 1985, pp. 2746, 2750 (App. I at 183). Later in the same document, the Committee defended the settlement on the ground that "it is entirely possible that if there were no settlement in the compact and all the court cases proceeded to decisions, that the people on the four atolls might receive less than the $150 million we have agreed to provide as a settlement.” Id. at 9-10 (App. I at 188).
. As the government’s brief points out, "plaintiffs would be required to individually prove specific injuries caused by exposure to radiation. This burden of proof would be made more difficult here by the problems of distance, language, and illiteracy encountered in communicating with thousands of plaintiffs scattered across hundreds of miles throughout the islands." Brief for United States at 25 n. 14.
. Plaintiffs in these suits “would face a multitude of jurisdictional defenses, including the discretionary function exception (28 U.S.C. 2680(a)), the foreign claims exception (28 U.S.C. 2680(k)), the statute of limitations (28 U.S.C. 2401(b)), and the administrative claim requirement (28 U.S.C. 2675(a)).” Brief for United States at 25 n. 14.
. This takings claim should be distinguished from the takings claim at issue in People of Enewetak v. United States, 864 F.2d 134 (Fed.Cir.1988). Plaintiffs in that case argued that the damage wrought by U.S. nuclear testing was itself a taking of property in violation of the fifth amendment. The property allegedly taken was plaintiffs’ lands, homes, and businesses. Here the plaintiffs’ cause of action is a tort claim under the Federal Tort Claims Act. Plaintiffs respond to the government’s defense (lack of jurisdiction) by arguing that a withdrawal of jurisdiction would constitute an uncompensated taking; the property allegedly taken here is the plaintiffs’ cause of action in tort.
. The broadest form of the rule on nationality of claims is the continuous nationality rule, which requires that a claim be held continuously by a national of the espousing state from the time of the injury to the time of the espousal.
. The court in Nicaragua distinguished between "customary” and “peremptory" norms of international law, and left open the possibility that "peremptory” norms might supersede domestic law. "Peremptory” norms are those principles of international law deemed so fundamental that no deviation from them is permitted. The Nicaragua court cited “genocide, slavery, murder, torture, prolonged arbitrary deten*394tion, and racial discrimination” as examples of peremptory norms. 859 F.2d at 941. Plaintiffs here do not contend that the espousal violates a peremptory norm of international law. Nor do I believe that such a claim would be tenable: although the jurisdiction-stripping provisions of the Compact Act, standing alone, might raise concerns about racial bias, the espousal viewed in its entirety cannot plausibly be portrayed as an instrument of racial discrimination.
. In certain limited areas, international law does restrict the authority of governments over their own people. See, e.g., Filartiga v. Pena-Irala, 630 F.2d 876, 884 (2d Cir.1980) ("[O]fficial torture is now prohibited by the law of nations. The prohibition is clear and unambiguous, and admits of no distinction between treatment of aliens and citizens.”). These restrictions, however, are extraordinary; international law plainly imposes no generalized "duty of fair representation” on any government.