* This text was reproduced and reformatted from the text appearing at the findlaw website (visited May 21, 2007) <http://caselaw.lp.findlaw.com/data2/circs/9th/0256256opp.pdf.>

1 The plaintiffs, who appear as appellants and cross-appellees in this appeal, will be referred to as“plaintiffs” throughout.

2 A jus cogens norm ‘ ‘is a norm accepted and recognized bythe international community of states as a whole as a norm from which no derogation is permitted and which can be modified only by a subsequent norm of general international law having the same character.” Siderman de Blake v. Republic of Argentina, 965 F.2d 699, 714 (9th Cir. 1992) (quoting Vienna Convention on the Law of Treaties, art. 53, May 23,1969, 1155 U.N.T.S. 332, 8 I.L.M. 679).

3 See note 15, infra.

4 In assuring ourselves of jurisdiction we are mindful of “the rule that a federal court may not hypothesize subject-matter jurisdiction for the purpose of deciding the merits.” Hodgers-Durgin v. De La Vina, 199 F.3d 1037, 1042 n. 3 (9th Cir. 1999) (en bane). Although we would not be compelled to address subject matter jurisdiction if we were affirming the district court's dismissal on another threshold ground, sec Sinochem Int'l Co. Ltd. v. Malaysia Int'l Shipping Corp., 127 S. Ct. 1184,1191-92 (2007), we may not review merits issues, such as the act of state doctrine, without initially having juris-diction to do so. See, e.g., Siderman de Blake, 965 F.2d at 707 (“Because sovereign immunity is jurisdictional and the act of state doctrine is not, we must consider sovereign immunity before reaching the act of state doctrine.” (emphasis added) (citation and internal quotation marks omitted)); see also Republic of Austria v. Altmann, 541 U.S. 677, 700 (2004) (noting that the “act of state doctrine provides foreign states with a substantive defense on the merits” (emphasis added)).

5 We rely here on authority interpreting 28 U.S.C. § 1331's grant of federal question jurisdiction in part because in both circumstances ‘ ‘the court must assume jurisdiction to decide whether the allegations state a cause of action on which the court can grant relief’ and thus the legal question of whether plaintiffs have properly invoked a federal court's power to recognize actionable international law torts under the ATCA “must be decided after and not before the court has assumed jurisdiction over the controversy.” See Bell, 327 U.S. at 682. Making the jurisdictional showing under § 1350 the same as under § 1331 is also consistent with Sosa v. Alvarez-Machain, 542 U.S. 692 (2004), which suggests that where a federal court has recognized an international law tort under the ATCA, the suit arises under federal common law, and thus federal jurisdiction may alternatively be premised upon § 1331. See Sosa, 542 U.S. at 732 (recognizing that ATCA claims are “private claims under federal common law for violations of … international law norm[s]“); id. at 745 n. (Scalia, J.,concurring in judgment) (“[A] federal-common-law cause of action of the sort the Court reserves discretion to create would'arise under’ the laws of the United States … for purposes of statutory federal-question jurisdiction.” (original emphasisremoved)); see also Illinois v. City of Milwaukee, 406 U.S. 91, 99 (1972) (concluding that “§ 1331 jurisdiction will support claims founded upon federal common law“). Consequently, although prior to Sosa some courts required ‘ ‘a more searching preliminary review of the merits” before ATCA jurisdiction attached, see Filartiga v. Pena-Irala, 630 F.2d 876, 887-88 (2d Cir. 1980), we believe the better view after Sosa is that at least where a case is brought by an alien for a “tort only,“ the jurisdictional burden under § 1350 and § 1331 is the same. Accord Tel-Oren v. Libyan Arab Republic, 726 F.2d 774, 779 n.4 (D.C. Cir. 1984) (Edwards, J., concurring) (“As to aliens … jurisdiction under § 1331 is available at least to the extent that § 1350 applies.“).

6 Plaintiffs, the United States and an amicus who filed a briefin response to Rio Tinto's Petition for Rehearing and for Rehearing En Bane are largely in agreement with our under standing of the jurisdictional threshold in ATCA cases. See Plaintiffs’ Opposition to Petition for Panel Rehearing and for Rehearing En Bane at 12; Brief for the United States as Amicus Curiae Supporting Panel Rehearing or Rehearing En Bane at 4-5; Brief for the National Foreign Trade Council as Amicus Curiae in Support of Defendants’ Petition for Panel Rehearing and for Rehearing En Bane at 89.

7 The Supreme Court ultimately concluded that under this standard, the petitioner's claim for arbitrary arrest and detention was not cognizable under the ATCA. Sosa, 542 U.S. at 737- 38.

8 We do not address the second and third Baker factors, as Rio Tinto does not contend they are applicable.

9 As discussed infra, the act of state doctrine also involves a determination of the political repercussions of judicial action, and in that context courts have held that statements of interest, although entitled to respect, are not conclusive. See Allied Bank Int'l v. Banco Credito Agricola de Cartago, 757 F.2d 516, 521 n.2 (2d Cir. 1985) (“This estimation [of the applicability of the act of state doctrine] may be guided but not controlled by the position, if any, articulated by the executive as to the applicability vel non of the doctrine to a particular set of facts. Whether to invoke the act of state doctrine isultimately and always a judicial question. “); Environmental Tectonics v. W.S. Kirkpatrick, Inc., 847 F.2d 1052, 1062 (3d Cir. 1988) (holding that the State Department's legal conclusions “are not controlling on the courts,” but that its “factual assessment of whether fulfillment of its responsibilities will be prejudiced by the course of civil litigation is entitled to substantial respect“). The Supreme Court also recently stated in the context of assertions of foreign sovereign immunity that “should the State Department choose to express its opinion on the implications of asserting jurisdiction over particular petitioners in connection with their alleged conduct, the opinion might well be entitled to deference as the considered judgment of the Executive on a particular question of foreign policy.” Altmann, 541 U.S. at 702 (emphasis in original omitted).

10 In Ungaro-Benages, the court ultimately dismissed the claims on comity grounds. Id. at 1240.

11 The SOI adds that “[countries participating in the multilateral peace process have raised this concern” as well.

12 The D.C. Circuit recently confronted a similarly equivocalSOI and, citing our earlier opinion in this appeal, denied an ATCA defendant's petition for a writ of mandamus invoking the political question doctrine. See Doe v. Exxon Mobil Corp., 473 F.3d 345, 354-55 (D.C. Cir. 2007) (interpreting the SOI at issue “not as an unqualified opinion that this suit must be dismissed, but rather as a word of caution to the district court alerting it to the State Department's concerns“).

13 We address below the separate question of whether the act of state or international comity doctrines warrant dismissal due to a balancing of the interests of PNG and the United States.

14 We need not determine whether a refusal to honor an explicit request to dismiss would constitute sufficient “disrespect” to warrant dismissal under this factor, although we note the Second Circuit's conclusion in Kadic that it would not. Kadic, 70 F.3d at 250.We also note that after we issued our original opinion in this appeal the United States filed an amicus brief in support of Rio Tinto's Petition For Panel Rehearing and for Rehearing En Bane. The brief, which was signed both by officials from the State Department and the Department of Justice, does not take issue with our understanding of the SOI, but instead agrees that the ‘’ statement did not recommend a specific disposition of any of the legal issues presented” in this case. Brief for the United States as Amicus Curiae Supporting Panel Rehearing or Rehearing En Bane at 14 n.3. Indeed, after noting that the SOI was based on concerns in 2001 “which are different from the interests and circumstances that exist today” the government expressly declines to endorse a dismissal of this case based on the SOI. See id. (“[T]he United States is not here seeking dismissal of the litigation based on purely case-specific foreign policy concerns.” (citing Sosa, 542 U.S. at 733 n.21)).

15 The plaintiffs have submitted recent letters from members of PNG's government urging that the suit will not harm or affect the ongoing Bougainville peace process. The Chief Secretary to the Government of PNG, Joseph Kalinoe, wrote to the United States Ambassador to PNG on March 30, 2005 that ’ ‘the [PNG Government] does not see the case presently before the U.S. courts in the US affecting diplomatic and bilateral relations between our two countries nor does it see it affecting the peace process on the island of Bougainville.” And on January 8, 2005, John Momis, the Interim Bougainville Provincial Governor, wrote to the State Department's legal advisor under whose name the SOI was written, “urg[ing] theGovernment of the United States to support the Prime Minister's position to permit the case to proceed in the courts of America, and to explain that the people of Bougainville strongly desire the case to proceed in America … .” Momis’ letter includes detail about the current state of the Bougainville peace process, and about how “the litigation has not hindered or in any way adversely affected the peace negotiations.” Indeed, the letter adds that “the Sarei litigation has helped facilitate the process as it is viewed as another source of rectifying the historic injustices perpetrated against the people of Bougainville.” Finally, the letter asserts that “the only way that the litigation will impact [U.S./PNG] foreign relations is if the litigation is discontinued.“Whether these letters are properly authenticated is in dis-pute. But if they are authentic and their authors accurately describe the current state of affairs in PNG, that would seri-ously undercut the State Department's concerns expressed in its November 5, 2001 SOI — which itself depended on assessments by local government officials, including Joseph Kalinoe's predecessor as Chief Secretary to the Government of PNG. For whatever reason, the State Department has declined to update the SOI. Under these circumstances, we do not rely on the letters’ substantive representations. But the letters, by suggesting there exists today a different reality in PNG from that portrayed in the SOI, illustrate why it is inappropriate to give the SOI final and conclusive weight as establishing a political question under Baker.

16 We also recalled Baker's warning against “sweeping statements that imply all questions involving foreign relations are political ones” and its command to courts to undertake a“case-by-case analysis to determine whether the question posed lies beyond judicial cognizance.” 410 F.3d at 544-45. We characterized the dissent, which ‘ ‘would have the political question doctrine remove from our courts all matters that fall by their constitutional DNA into th[e] sphere of conduct involving foreign relations,” id. at 547 (internal quotations omitted), as setting forth “an over-inclusive approach [that] threatens to sweep all cases touching foreign relations beyond the purview of the courts — a practice warned against in Baker.” Id.

17 Rio Tinto has not appealed the nondismissal of the war crimes and violations of the laws of war claims under the act of state doctrine, but argues against it only in response to the plaintiffs' appeal as to the act of state dismissals.

18 The agreement is codified in the Bougainville Copper Agreement Act.

19 As noted above, see supra note 17, Rio Tinto has waived any appeal of the district court's failure to dismiss the war crimes and violations of the lawsofwarclaimsonactof state grounds. We note, however, that the act of state doctrine has been interpreted to apply only to legitimate acts of warfare. See, e.g., Under v. Portocarrero, 963 F.2d 332, 336 (11th Cir. 1992) (holding that “there is no foreign civil war exception to the right to sue for tortious conduct that violates the funda mental norms of the customary laws of war'’); see also Flatow v. Islamic Republic of Iran, 999 F. Supp. 1, 24 (D.D.C. 1998) (concluding that political assassinations “are not valid acts of state of the type which bar consideration of this case“). Because such conduct violates jus cogens norms, it does not constitute an official act. See, e.g., Siderman de Blake, 965 F.2d at 715-18.

20 Our circuit has not explicitly held that district court dismissals or refusals to dismiss on the ground of international comity are reviewed for abuse of discretion, although it has settled that comity decisions in general are reviewed under that standard. See, e.g., Stock West Corp. v. Taylor, 964 F.2d 912. 917-18 (9th Cir. 1992) (regarding comity owed to state courts). We join our sister circuits in clarifying that this abuse of discretion review applies to dismissals on grounds of international comity as well.

21 Once again, Rio Tinto failed to appeal the district court's comity ruling, confining its arguments against it to its response to plaintiffs’ appeal. See supra notes 17&19.

22 We also note that whether the presence of a conflict is a predicate inquiry, or simply one factor in a multipart inquiry, is academic here, as the district court did not abuse its discretion in identifying a conflict.

23 A “compensation claim” is defined to include any claim” in connection with” a mining project “which relates to or concerns” environmental harm, takings, or, more broadly “extends to any other matter” or “seeks the payment of damages, compensation or any other form of monetary relief.''

24 As with the act of state claims, Rio Tinto has failed to appeal the district court's ruling, but argues in response to the plain tiffs’ arguments that the district court inappropriately failed to dismiss the war crimes and violations of the laws of war claims on comity grounds. Even if Rio Tinto had not waived any appeal, the district court did not abuse its discretion in failing to dismiss these claims on comity grounds.

25 The ambiguity alluded to arose from Judge Bork's concurring opinion in Tel-Oren, 726 F.2d at 799, where he held that the ATCA and its reference to the law of nations did not amount to a congressional grant of a cause of action. In Congress' words, “[t]he TVPA would provide such a grant.” H.R. Rep. No. 102-367 at 4, 1992 U.S.C.C.A.N. at 86; see also 28 U.S.C. § 1350, historical and statutory notes, Torture Victim Protection, Section 2 - Establishment of civil action. The ambiguity has also been resolved by the Supreme Court's interpretation of the ATCA in Sosa:[Although the [ATCA] is a jurisdictional statute creating no new causes of action, the reasonable inference from the historical materials is that the statute was intended to have practical effect the moment it became law. The jurisdictional grant is best read as having been enacted on the understanding that the common law would provide a cause of action for the modest number of international law violations with a potential for personal liability at the time.42 U.S. at 724.

26 Compare 28 U.S.C. § 1350, historical and statutory notes, Torture Victim Protection, Section 2(a) (TVPA) (’ ‘An individual who, under actual or apparent authority, or color of law, of any foreign nation . .. subjects an individual to torture … or … extrajudicial killing shall, in a civil action, be liable for damages … .“) with 28 U.S.C. § 1350 (ATCA) (“The district courts shall have original jurisdiction of any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States.“) (emphasis added).

27 Congress also included an exhaustion requirement in another statute involving claims of human rights violations against foreign states, the Foreign Sovereign Immunities Act, 28 U.S.C. § 1602, et seq. Under the FSIA, a court shall decline to hear a case alleging specified human rights abuses against a foreign sovereign unless that sovereign is designated as a state sponsor of terrorism and the plaintiff has afforded thesovereign “a reasonable opportunity” to arbitrate the claim.” 28 U.S.C. § 1605(a)(7)(B)(i).

28 Despite the dissent's assertion to the contrary (Dissent at 4183-84), we agree that the TVPA has expanded rather than narrowed U.S. remedies for torture and extrajudicial killing overseas. First, the TVPA created remedies for U.S. citizens never available to them under the ATCA. Second, aliens had only what Congress considered an ambiguous right to bring torture claims under the ATCA because of Judge Bork's opinion in Tel-Oren. Thus, the TVPA confirmed the existence of a clear, safe-harbor cause of action for alien victims of torture and extrajudicial killings.The passage of the TVPA can be said to be a narrowing of an alien's remedies for torture in U.S. courts only if, after the TVPA, an alien can no longer bring a torture claim under the ATCA. This was the Seventh Circuit's conclusion in Ena-horo. See 408 F.3d at 886 (“It is hard to imagine that the Sosa Court would approve of common law claims based on torture and extrajudicial killing when Congress has specifically provided a cause of action for those violations … .“). Rio Tinto has not made this latter argument, and we are not endorsing such a result.31We do not read torture and extrajudicial killing out of the ATCA, as the dissent claims. (Dissent at 4186.) That issue is not squarely before us, and we note that Enahoro's resolution of the complexities that result from the apparent overlap between the TVPA and the ATCA may not be the only appropriate approach — a clear safe harbor statute need not eclipse the more general and ambiguous statute that preceded it.Lastly, the dissent's argument that “[i]f the majority is correct, Congress has made it more difficult for aliens to bring torture claims into U.S. courts because now (under TVPA) they must first exhaust their remedies, whereas previously (under ATCA) they did not” is deeply ironic. (Dissent at 4184 n.4.) This is actually one of the premises of the dissent's approach — not ours. We hold that exhaustion is not required at this time under the ATCA. It is the dissent that would import exhaustion into the ATCA (which may or may not encompass torture claims after the TVPA) and make it more difficult to bring torture or any other ATCA claim in U.S. 32 courts.

29 The dissent's invocation of the Charming Betsy interpretive doctrine in light of the ambiguity in congressional intent is not quite on point. See Murray v. Schooner Charming Betsy, 6 U.S. (2Cranch) 64, 118 (1804) (“[A]n act of Congress ought never to be construed to violate the law of nations if any other possible construction remains.“). (Dissent at 4187.) Though reading exhaustion into the ATCA would be consistent with international law norms, failing to read exhaustion into the ATCA would not violate those norms.

30 The dissent is incorrect in its assertion that we ‘ ‘infer[ ] that exhaustion, which was expressly included in TVPA, must necessarily have been left out of ATCA, which has no exhaustion requirement.” (Dissent at 4185.) All that we can conclude with any certainty from the TVPA is that it does not answer the question of what Congress intended with respect to an exhaustion requirement in the ATCA.Furthermore, our reasoning has little connection to that of Papa v. United States, 281 F.3d 1004 (9th Cir. 2002), where we read the statute of limitations from the TVPA into the ATCA. (Dissent at 4185.) The question of whether there should be a limitations period was never considered by Papa. Relying on North Star Steel Co. v. Thomas, 515 U.S. 29, 33 (1995) (” [O] ften federal statutes fail to provide any limitations period for the causes of action they create, leaving courts toborrow a period … to limit these claims.“) (cited by Papa, 281 F.3d at 1012 n.30), Papa assumed that like any other civil cause of action considered by the Supreme Court, an ATCA suit must have a limitations period. Therefore, Papa was left only with the question of how long that limitations period should be. If the ATCA is to have an imported exhaustion requirement, then the TVPA may be a good source for determining the standards of that requirement. But this does not answer the antecedent question of whether exhaustion should be imported into the ATCA in the first instance. Lastly, we acknowledge that it is hazardous to attempt to gauge congressional intent from silence or comments expressed in the context of another statute's legislative history, but we must make do with the meager legislative history available to us. Moreover, it seems no more hazardous than the dissent's preference for reading a new requirement into a law that does not explicitly provide for it. “If Congress intended to do something different, let Congress fix it.” Amalgamated Transit Union Local 1309, AEL-CIO v. Laidlaw Transit Servs., Inc., 448 F.3d 1092, 1100 (9th Cir. 2006) (Bybee, J., dissenting from denial of rehearing en bane).

31 The dissent's suggestion that “[n]othing in Sosa or ATCA indicates that this distinction [between domestic and international tribunals] matters,” (Dissent at 4198), misses the point. Sosa did not hold simply that the “ATCA was written in order to bring the law of nations into American courts” (Dissent at 4198), but that only some portions of the law of nations are brought into American courts through the ATCA. See Sosa, 542 U.S. at 720, 725. If in determining which portions of the law of nations are usable for ATCA purposes, we rely upon sound judicial discretion (as argued at length by the dissent at 4201-16), we should not be lulled into a false sense of familiarity with the term “exhaustion” just because it is the same term that we use to describe an analogous doctrine in our domestic law. Although the concepts of exhaustion may be analogous in the international and domestic spheres, they are not identical, and the international law of exhaustion has developed in part as a result of uniquely international concerns.

32 This premise is consistent with the rules of the two ad hoc international criminal tribunals, for the former Yugoslavia and Rwanda, which establish the “primacy” of the respective international criminal tribunal over its national counterpart without regard for the international rule of exhaustion. See International Criminal Tribunal for the former Yugoslavia, Rules of Procedure and Evidence, pt. 2, rules 8-13, U.N. Doc. IT/32/Rev.7 (1996), entered into force Mar. 14, 1994, as amended Jan. 8, 1996, available at http://wwwl.umn.edu/humanrts/icty/ct-rules7.html; International Criminal Tribunal for Rwanda, Rules of Procedure and Evidence, pt. 2, rules 8-13, U.N. Doc. ITR/3/REV.1 (1995), entered into force June 29, 1995, available at http://wwwl.umn.edu/humanrts/africa/RWANDAl.htm.

33 Many of the policy arguments in favor of exhaustion — including (i) the dissent's concern with “undermin[ing] local gov-ernments” (Dissent at 4203), (ii) “single-handedly derailing] diplomacy” (Dissent at 4214), (iii) the inability of courts “to make … subtle adjustments in response to national and world events” (Dissent at 4213) and (iv) the premise that “[a] lawsuit in U.S. courts will rarely be the best way to resolve supranational conflicts” (Dissent at 4204) — apply nearly as well to ATCA (and TVPA) cases where local remedies have been exhausted. Further, because of the “futility” exception to the exhaustion rule, see Foreign Relations Law Restatement § 703 cmt. d (“That [exhaustion] requirement is met if it is shown that no domestic remedies are] available or that it would be futile to U.S. 731, 739 (2001)), is the appropriate starting point for pursue them.”), many potential human rights cases brought determining the scope of our discretion with respect to the under the ATCA will be excused from satisfying the exhaus-ATCA. Given the specialized jurisprudence developing tion requirement altogether even if we were to read it into the around the ATCA, we question whether Patsy and McCarthy's statute — thus side-stepping exhaustion's purported benefits. holdings regarding judicial discretion to impose exhaustion And for those countries where evidence of the futility of of administrative remedies can be uncritically transferred to requiring exhaustion is less clear cut, ATCA1) to determine the adequacy of those countries legal curiam (“Although we find its disposition to have been in and political systems. In sum, the proffered exhaustion rule error the Sev plaintiffs and the international law context, defendants will no doubt ask U.S. courts to conduct sensitive inquiries into the internal affairs of other countries”(Dissent at 421enth Circuit felt bound to apply. [precedent], may not accomplish quite so much as the dissent predicts. even thouSh il expressed grave doubts …. This was a prudent course. It neither forced the issue by upsetting.

34 Sosa, rather than the administrative law cases relied upon by Court of Appeals took to be our settled precedents, nor buried the dissent, dissent at 4183 (citing McCarthy v. Madigan,the issue by proceeding in a summary fashion. By adhering 503 U.S. 140, 144 (1992) (quoting Patsy, 457 U.S. at 501), to its understanding of precedent, yet plainly expressing its superseded by statute as stated in Booth v. Churner, 532 doubts, it facilitated our review.)

35 cf. Eberhart v. United States, 126 S.Ct.403, 407(2005)(percuriam)(“Although we find its disposition to have been in error, The seventh circuit delt bound to apply [precedent].even though it expressed grave doubts …. This was a prudent course. It neither forced the issue by upsetting what the Court of Appeals took to be our settled precedents, nor buried the issue by proceeding in a summary fashion. By adhering to its understanding of precedent, yet plainly expressing its doubts, it facilitated our review.”)

1 This is a case of first impression in this circuit. We have not addressed the application of Sosa to exhaustion. The only issue before us in Alperin v. Vatican Bank, 410 F.3d 532 (9th Cir. 2005), cited by the majority at 4155, was whether that case should be dismissed under the political question doctrine. See id. at 538, 541 n.4 (“by agreement of the parties the district court limited its discussion to the issue of whether the Holocaust Survivors’ claims should be dismissed under the political question doctrine,” and the “viability of the Holocaust Survivors’ claims apart from the issue of the political question doctrine is not before us“). We left open the possibility of other grounds for dismissal, and opined that “[g]iven … a myriad of other procedural and jurisdictional hurdles, the Holocaust Survivors may indeed face an uphill battle in pursuing their claims.” Id. at 539.

2 Statutorily required exhaustion, by contrast, may be jurisdictional. See Weinberger v. Salfi, 422 U.S. 749, 764-67 (1975); Marathon Oil Co. v. United States, 807 F.2d 759, 768 (9th Cir. 1986); Montgomery v. Rumsfeld, 572 F.2d 250, 252 (9th Cir. 1978).

3 In its entirety, ATCA provides: “The district courts shall have original jurisdiction of any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States.” 28 U.S.C. § 1350.

4 The majority responds that remedies for torture are only narrowed if aliens can no longer bring a claim for torture under ATCA. Maj. op. at 4159 n.26. The Seventh Circuit has so held. Enahoro, 408 F.3d at 884-85, 886 (holding that torture claims are precluded now that TVPA “occupfies] the field” and finding it “hard to imagine that the Sosa Court would approve of common law claims based on torture and extrajudi-cial killing when Congress has specifically provided a cause of action for those violations and has set out how those claims must proceed“). My point is a little different: If the majority is correct, Congress has made it more difficult for aliens to bring torture claims into U.S. courts because now (under TVPA) they must first exhaust their remedies, whereas previously (under ATCA) they did not.

5 I can only speculate as to why Congress chose not to clarify ATCA when it drafted the exhaustion provisions of TVPA. It may be that because TVPA extended the cause of action to United States citizens, Congress wished to make clear that traditional exhaustion would also apply to citizens suing in their own national courts.

6 Judge Cudahy ultimately concluded that Sosa offered “little guidance” on the question and that is was “far from clear” whether exhaustion was required under ATCA. 408 F.3d at 890 (Cudahy, J., dissenting).

7 The majority argues that because Sosa urges caution in expanding the causes of action available under ATCA, judicial caution precludes recognition of exhaustion. Maj. op. at 4165 (citing Sola's directive that “[t]hese reasons argue for great caution in adapting the law of nations to private rights“). This turns the Court's reasoning on its head; the majority's conception of caution would expand, rather that restrict, the availability of claims under ATCA. It overlooks the Court's “reasons” for caution, which include the fear that courts might “consider suits under rules that would go so far as to claim a limit on the power of foreign governments over their own citizens,” and “raise risks of adverse foreign policy consequences.” 542 U.S. at 727, 728. An exhaustion require ment would help abate such fears and would serve to fortify the “high bar” Sosa established for ATCA claims. Id. at 727.

8 The exhaustion requirement predates both the Jay Treaty and ATCA. “The requirement that local remedies should be resorted to seems to have been recognized in the early history of Europe, before the modern national state had been born … .” AMERASINGHE, supra, at 22; see also A.A. Canҫado trindade, the application of the rule of exhaustion of local remedies in international law 1 (1983); A.A. Canҫado trindade, Origin and Historical Development of the Rule of Exhaustion of Local Remedies in Interna tional Law, 12 Revue belge de droit international 499, 507,512 (1976) (noting that “the practice of States throughout the eighteenth century was clearly directed towards upholding the local remedies rule“). For reasons that I explain, infra, it reads too much into ATCA to assume that because the Senate and the President approved the Jay Treaty with an express exhaustion requirement, the House and Senate, five years earlier, could not have anticipated an exhaustion requirement in ATCA. But see Maj. op. at 4158 (“the explicit exhaustion requirement in the Jay Treaty may reveal that the First Congress did not view exhaustion as an automatic rule of customary international law“).

9 Like its U.S. domestic counterpart, the rule of exhaustion Of local remedies is largely judge-made. “[P]erhaps more so than any other rule of international law,” it “undoubtedly grew up under the nurturing of professional administrators.’ Mummery, supra, at 393. The law continues to be “sheltered from the harsher political winds” that shape much of international law because its development rests in the hands of judges. Id.

10 The matter is by no means resolved. There was a trend towards the procedural view at the time of Phosphates in Morocco (Italy v. Fr.), 1938 P.C.I.J. (ser.A/B) No. 74, but the International Law Commission adopted the substantive view in the 1970s. See Maj. op. at 4167; Second Report on Diplomatic Protection, supra 35-37. The International Law Commission's current draft articles currently stop short of declaring when international responsibility attaches, but declare that ‘ ‘The responsibility of a State may not be invoked if … [t]he claim is one to which the rule of exhaustion of local remedies applies and any available and effective local remedy has not been exhausted.” Responsibility of States for Internationally Wrongful Acts, Art. 44,2001 in Official Records of the General Assembly, Fifty-sixth Session, Supplement No. 10 (A/56/10). The current Third Restatement takes no position on the issue; it has relinquished even the noncommittal stance of the Second Restatement, which declared that’ ‘the exhaustion of available remedies is primarily a procedural requirement.” Re-statement (Second) Of the foreign relations law of the united states § 206 comment d (1965) (emphasis added).

11 I believe that Sosa's rule would incorporate even procedural exhaustion, because the international community does not recognize virtually any “violation of the law of nations“ without it. Contrary to the majority's opinion, the Sosa rule is not limited to “substantive norms comparable to ‘violation of safe conducts, infringement of the rights of ambassadors, and piracy,’ “ Maj. op. at 4168 (quoting Sosa, 542 U.S. at 724), but requires recognition of “any claim based on the present-day law of nations [that] rest[s] on a norm of international character accepted by the civilized world and defined with specificity comparable to the features of the 18th-century paradigms we have recognized.” Sosa, 542 U.S. at 725.

12 Judicial inquiry into available exceptions in ATCA cases would be similar to that performed when applying the doctrine of forum non conveniens, see Menendez Rodriguez v. Pan American Life Insurance Co., 311 F.2d 429, 433 (5th Cir. 1962) (finding that political refugees could receive a fair hearing in Cuba), vacated on other grounds, 376 U.S. 779 (1964), or considering the enforceability of a forum selection clause, see McDonnell Douglas Corp. v. Islamic Republic of Iran, 758 F.2d 341, 346 (8th Cir. 1985) (refusing to consider en forcement of a forum selection clause where litigation in Iran would ‘ ‘be so gravely difficult and inconvenient that McDon nell Douglas would for all practicable purposes be deprived of its day in court“).

13 I have deliberately refrained from referring to comity between federal courts and foreign courts, because I do not wish to prejudge what form local remedies might take. As I discuss infra, at 4143-44 n.16, we have only to look at South Africa's Truth and Reconciliation Commission to see that some coun tries may be able to craft unique institutions for dealing with difficult internal matters. Not every solution requires a judicial solution. Indeed, governments have much greater flexibility in terms of process and remedies than American courts may be able to offer.

14 The majority observes that hearing exhausted cases presents the same potential for undermining local governments. Maj. op. at 4167 n.31. While this is true, it misses the point. If we respect the international norm requiring exhaustion, we will minimize the number of cases that involve potential meddling in other country's affairs. In fact, we will intervene only where governments are so inadequate as not to provide adequate remedies. With exhaustion, we get the best of both worlds. We retain a commitment to the principle of international com ity, while stepping in to render justice where required—and only where required. While adjudicating the exhaustion re quirement requires occasional inquiries into the internal affairs of other nations (in cases where the adequacy of remedies is not readily apparent), id., this infrequent inquiry is preferable to offering international intervention on behalf of every foreign plaintiff who shops a claim in American courts.

15 The majority complains about the lack of “empirical data showing improvements in the quality or accessibility of local remedies as a result of the application of the local remedies rule.” Maj. op. at 4169. A study of this scope would indeed be difficult to complete—since cases like the one before us are rare. We need not obtain proof of the exhaustion's effec-tiveness beyond a reasonable doubt before considering scholars’ arguments and case studies as part of our prudential inquiry. That said, there are convincing empirical examples, see infra, 4143 n.16. These examples fall short of quantifying “the full effects of human rights treaties,” Maj. op. at 4169, but they strongly indicate exhaustion helps foster local remedies, while preserving international litigation for those “individuals whose access to justice is blocked in their home country.” Lutz, Ellen&Sikkink, Kathryn, The Justice Cascade: The Evolution and Impact of Human Right Trials in Latin America, 2 CHI. J. INT'L L. 1,4 (2001)Google Scholar. And even if I believed that the academic empirical debate was a draw, I would defer to Congress's observation that exhaustion “can be expected to encourage the development of meaningful remedies in other countries.” H.R. REP. NO. 102-367, pt. 3, at 5, reprinted in 1992 U.S.C.C.A.N. 84, 88-89, and the State Department's decision, in the case of South Africa, that foreign litigation over reparations for apartheid interfered with that nation's domestic development. See In re S. Afr. Apartheid Litig., 346 F. Supp. 2d 538, 553 (S.D.N.Y. 2004).

16 Failure to exhaust threatens international institutions, not just local ones. Most claims brought under ATCA could have been brought in an international forum—one that most likely requires exhaustion. Allowing claims to be brought in United States courts before efforts at a local solution allows litigants to forum shop and discourages them from seeking adjudication in international bodies. See Bradley, Curtis A., The Costs of International Human Rights Litigation, 2 CHI. J. INT'L L. 457, 469 (2001)Google Scholar (arguing that alien tort actions are costly to the international system because they risk preempting or disrupting local remedies or international institutional re sponses).

17 One trend we have seen in recent years is the development of creative, indigenous legal solutions to address alleged human rights violations. [T]he transition from autocratic rule to democracy in numerous countries, beginning in South America but extending to Eastern Europe and parts of Africa, Central America, and Asia, has caused new governments to devise strategies for coming to terms with the human rights abuses of prior regimes and, in some case, guerrilla opposition groups. In most cases in which states have decided to seek accountability, they have charted their own course under domestic law, creating mechanisms tailored to their individual circumstances. This pattern has led to criminal trials, truth commissions, purging of former officials from office, and civil suits against abusers. Ratner, Steven R., New Democracies, Old Atrocities: An Inquiry In International Law, 87 GEO. L.J. 707, 714 (1999)Google Scholar. We can see evidence of these local solutions most prominently in South Africa, with the creation of the Truth and Reconciliation Commission to address the injustices of apartheid, see Schrage, supra, at 166, and most recently in Rwanda, with the development of an alternative dispute resolution method known as gacaca courts, which emphasize the admission of guilt and expression of remorse by defendants complicit in the Rwandan genocide, see Schabas, William A., Genocide Trials and Gacaca Courts, 3 J. INT'L CRIM. JUST. 879 (2005)CrossRefGoogle Scholar.