SOSA V. ALVAREZ-MACHAIN (03-339) 542 U.S. 692 (2004)
331 F.3d 604, reversed.
[ Souter ]
[ Scalia ]
[ Opinion of Ginsburg ]
[ Opinion of Breyer ]
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version

Opinion of the Court

NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press.


Nos. 03—339 and 03—485


[June 29, 2004]

Justice Souter delivered the opinion of the Court.

The two issues are whether respondent Alvarez-Machain’s allegation that the Drug Enforcement Administration instigated his abduction from Mexico for criminal trial in the United States supports a claim against the Government under the Federal Tort Claims Act (FTCA or Act), 28 U.S.C. § 1346(b)(1), §§2671—2680, and whether he may recover under the Alien Tort Statute (ATS), 28 U.S.C. § 1350. We hold that he is not entitled to a remedy under either statute.


We have considered the underlying facts before, United States v. Alvarez-Machain, 504 U.S. 655 (1992). In 1985, an agent of the Drug Enforcement Administration (DEA), Enrique Camarena-Salazar, was captured on assignment in Mexico and taken to a house in Guadalajara, where he was tortured over the course of a 2-day interrogation, then murdered. Based in part on eyewitness testimony, DEA officials in the United States came to believe that respondent Humberto Alvarez-Machain (Alvarez), a Mexican physician, was present at the house and acted to prolong the agent’s life in order to extend the interrogation and torture. Id., at 657.

In 1990, a federal grand jury indicted Alvarez for the torture and murder of Camarena-Salazar, and the United States District Court for the Central District of California issued a warrant for his arrest. 331 F.3d 604, 609 (CA9 2003) (en banc). The DEA asked the Mexican Government for help in getting Alvarez into the United States, but when the requests and negotiations proved fruitless, the DEA approved a plan to hire Mexican nationals to seize Alvarez and bring him to the United States for trial. As so planned, a group of Mexicans, including petitioner Jose Francisco Sosa, abducted Alvarez from his house, held him overnight in a motel, and brought him by private plane to El Paso, Texas, where he was arrested by federal officers. Ibid.

Once in American custody, Alvarez moved to dismiss the indictment on the ground that his seizure was “outrageous governmental conduct,” Alvarez-Machain, 504 U.S., at 658, and violated the extradition treaty between the United States and Mexico. The District Court agreed, the Ninth Circuit affirmed, and we reversed, id., at 670, holding that the fact of Alvarez’s forcible seizure did not affect the jurisdiction of a federal court. The case was tried in 1992, and ended at the close of the Government’s case, when the District Court granted Alvarez’s motion for a judgment of acquittal.

In 1993, after returning to Mexico, Alvarez began the civil action before us here. He sued Sosa, Mexican citizen and DEA operative Antonio Garate-Bustamante, five unnamed Mexican civilians, the United States, and four DEA agents. 331 F.3d, at 610. So far as it matters here, Alvarez sought damages from the United States under the FTCA, alleging false arrest, and from Sosa under the ATS, for a violation of the law of nations. The former statute authorizes suit “for … personal injury … caused by the negligent or wrongful act or omission of any employee of the Government while acting within the scope of his office or employment.” 28 U.S.C. § 1346(b)(1). The latter provides in its entirety that “[t]he 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.” §1350.

The District Court granted the Government’s motion to dismiss the FTCA claim, but awarded summary judgment and $25,000 in damages to Alvarez on the ATS claim. A three-judge panel of the Ninth Circuit then affirmed the ATS judgment, but reversed the dismissal of the FTCA claim. 266 F.3d 1045 (2001).

A divided en banc court came to the same conclusion. 331 F.3d, at 641. As for the ATS claim, the court called on its own precedent, “that [the ATS] not only provides federal courts with subject matter jurisdiction, but also creates a cause of action for an alleged violation of the law of nations.” Id., at 612. The Circuit then relied upon what it called the “clear and universally recognized norm prohibiting arbitrary arrest and detention,” id., at 620, to support the conclusion that Alvarez’s arrest amounted to a tort in violation of international law. On the FTCA claim, the Ninth Circuit held that, because “the DEA had no authority to effect Alvarez’s arrest and detention in Mexico,” id., at 608, the United States was liable to him under California law for the tort of false arrest, id., at 640—641.

We granted certiorari in these companion cases to clarify the scope of both the FTCA and the ATS. 540 U.S. 1045 (2003). We now reverse in each.


The Government seeks reversal of the judgment of liability under the FTCA on two principal grounds. It argues that the arrest could not have been tortious, because it was authorized by 21 U.S.C. § 878 setting out the arrest authority of the DEA, and it says that in any event the liability asserted here falls within the FTCA exception to waiver of sovereign immunity for claims “arising in a foreign country,” 28 U.S.C. § 2680(k). We think the exception applies and decide on that ground.


The FTCA “was designed primarily to remove the sovereign immunity of the United States from suits in tort and, with certain specific exceptions, to render the Government liable in tort as a private individual would be under like circumstances.” Richards v. United States, 369 U.S. 1, 6 (1962); see also 28 U.S.C. § 2674. The Act accordingly gives federal district courts jurisdiction over claims against the United States for injury “caused by the negligent or wrongful act or omission of any employee of the Government while acting within the scope of his office or employment, under circumstances where the United States, if a private person, would be liable to the claimant in accordance with the law of the place where the act or omission occurred.” §1346(b)(1). But the Act also limits its waiver of sovereign immunity in a number of ways. See §2680 (no waiver as to, e.g., “[a]ny claim arising out of the loss, miscarriage, or negligent transmission of letters or postal matter,” “[a]ny claim for damages caused by the imposition or establishment of a quarantine by the United States,” or “[a]ny claim arising from the activities of the Panama Canal Company”).

Here the significant limitation on the waiver of immunity is the Act’s exception for “[a]ny claim arising in a foreign country,” §2680(k), a provision that on its face seems plainly applicable to the facts of this case. In the Ninth Circuit’s view, once Alvarez was within the borders of the United States, his detention was not tortious, see 331 F.3d, at 636—637; the appellate court suggested that the Government’s liability to Alvarez rested solely upon a false arrest claim. Id., at 640—641. Alvarez’s arrest, however, was said to be “false,” and thus tortious, only because, and only to the extent that, it took place and endured in Mexico.1 The actions in Mexico are thus most naturally understood as the kernel of a “claim arising in a foreign country,” and barred from suit under the exception to the waiver of immunity.

Notwithstanding the straightforward language of the foreign country exception, the Ninth Circuit allowed the action to proceed under what has come to be known as the “headquarters doctrine.” Some Courts of Appeals, reasoning that “[t]he entire scheme of the FTCA focuses on the place where the negligent or wrongful act or omission of the government employee occurred,” Sami v. United States, 617 F.2d 755, 761 (CADC 1979), have concluded that the foreign country exception does not exempt the United States from suit “for acts or omissions occurring here which have their operative effect in another country.” Id., at 762 (refusing to apply §2680(k) where a communiqué sent from the United States by a federal law enforcement officer resulted in plaintiff’s wrongful detention in Germany).2 Headquarters claims “typically involve allegations of negligent guidance in an office within the United States of employees who cause damage while in a foreign country, or of activities which take place within a foreign country.” Cominotto v. United States, 802 F.2d 1127, 1130 (CA9 1986). In such instances, these courts have concluded that §2680(k) does not bar suit.

The reasoning of the Ninth Circuit here was that, since Alvarez’s abduction in Mexico was the direct result of wrongful acts of planning and direction by DEA agents located in California, “Alvarez’s abduction fits the headquarters doctrine like a glove.” 331 F.3d, at 638.

“Working out of DEA offices in Los Angeles, [DEA agents] made the decision to kidnap Alvarez and … gave [their Mexican intermediary] precise instructions on whom to recruit, how to seize Alvarez, and how he should be treated during the trip to the United States. DEA officials in Washington, D. C., approved the details of the operation. After Alvarez was abducted according to plan, DEA agents supervised his transportation into the United States, telling the arrest team where to land the plane and obtaining clearance in El Paso for landing. The United States, and California in particular, served as command central for the operation carried out in Mexico.” Id., at 638—639.

Thus, the Ninth Circuit held that Alvarez’s claim did not “aris[e] in” a foreign country.

The potential effect of this sort of headquarters analysis flashes the yellow caution light. “[I]t will virtually always be possible to assert that the negligent activity that injured the plaintiff [abroad] was the consequence of faulty training, selection or supervision–or even less than that, lack of careful training, selection or supervision–in the United States.” Beattie v. United States, 756 F.2d 91, 119 (CADC 1984) (Scalia, J., dissenting). Legal malpractice claims, Knisley v. United States, 817 F. Supp. 680, 691—
693 (SD Ohio 1993), allegations of negligent medical care, Newborn v. United States, 238 F. Supp. 2d 145, 148—149 (DC 2002), and even slip-and-fall cases, Eaglin v. United States, Dept. of Army, 794 F.2d 981, 983—984 (CA5 1986), can all be repackaged as headquarters claims based on a failure to train, a failure to warn, the offering of bad advice, or the adoption of a negligent policy. If we were to approve the headquarters exception to the foreign country exception, the “ ‘headquarters claim’ [would] become a standard part of FTCA litigation” in cases potentially implicating the foreign country exception. Beattie, supra, at 119 (Scalia, J., dissenting). The headquarters doctrine threatens to swallow the foreign country exception whole, certainly at the pleadings stage.

The need for skepticism is borne out by two considerations. One of them is pertinent to cases like this one, where harm was arguably caused both by individual action in a foreign country as well as by planning in the United States; the other is suggested simply because the harm occurred on foreign soil.


Although not every headquarters case is rested on an explicit analysis of proximate causation, this notion of cause is necessary to connect the domestic breach of duty (at headquarters) with the action in the foreign country (in a case like this) producing the foreign harm or injury. It is necessary, in other words, to conclude that the act or omission at home headquarters was sufficiently close to the ultimate injury, and sufficiently important in producing it, to make it reasonable to follow liability back to the headquarters behavior. Only in this way could the behavior at headquarters properly be seen as the act or omission on which all FTCA liability must rest under §2675. See, e.g., Cominotto, supra, at 1130 (“[A] headquarters claim exists where negligent acts in the United States proximately cause harm in a foreign country”); Eaglin, supra, at 983 (noting that headquarters cases require “a plausible proximate nexus or connection between acts or omissions in the United States and the resulting damage or injury in a foreign country”).

Recognizing this connection of proximate cause between domestic behavior and foreign harm or injury is not, however, sufficient of itself to bar application of the foreign country exception to a claim resting on that same foreign consequence. Proximate cause is causation substantial enough and close enough to the harm to be recognized by law, but a given proximate cause need not be, and frequently is not, the exclusive proximate cause of harm. See, e.g., 57A Am. Jur. 2d §529 (2004) (discussing proper jury instructions in cases involving multiple proximate causes); Beattie, supra, at 121 (Scalia, J., dissenting) (“[I]n the ordinary case there may be several points along the chain of causality” pertinent to the enquiry). Here, for example, assuming that the direction by DEA officials in California was a proximate cause of the abduction, the actions of Sosa and others in Mexico were just as surely proximate causes, as well. Thus, understanding that California planning was a legal cause of the harm in no way eliminates the conclusion that the claim here arose from harm proximately caused by acts in Mexico. At most, recognition of additional domestic causation under the headquarters doctrine leaves an open question whether the exception applies to the claim.


Not only does domestic proximate causation under the headquarters doctrine fail to eliminate application of the foreign country exception, but there is good reason to think that Congress understood a claim “arising in” a foreign country in such a way as to bar application of the headquarters doctrine. There is good reason, that is, to conclude that Congress understood a claim “arising in a foreign country” to be a claim for injury or harm occurring in a foreign country. 28 U.S.C. § 2680(k). This sense of “arising in” was the common usage in state borrowing statutes contemporary with the Act, which operated to determine which State’s statute of limitations should apply in cases involving transjurisdictional facts. When the FTCA was passed, the general rule, as set out in various state statutes, was that “a cause of action arising in another jurisdiction, which is barred by the laws of that jurisdiction, will [also] be barred in the domestic courts.” 41 A. L. R. 4th 1025, 1029, §2 (1985). These borrowing statutes were typically restricted by express terms to situations where a cause of action was time barred in the State “where [the] cause of action arose, or accrued, or originated.” 75 A. L. R. 203, 211 (1931) (emphasis in original). Critically for present purposes, these variations on the theme of “arising in” were interpreted in tort cases in just the same way that we read the FTCA today. A commentator noted in 1962 that, for the purposes of these borrowing statutes, “[t]he courts unanimously hold that a cause of action sounding in tort arises in the jurisdiction where the last act necessary to establish liability occurred”; i.e., “the jurisdiction in which injury was received.” Ester, Borrowing Statutes of Limitation and Conflict of Laws, 15 U. Fla. L. Rev. 33, 47.

There is, moreover, specific reason to believe that using “arising in” as referring to place of harm was central to the object of the foreign country exception. Any tort action in a court of the United States based on the acts of a Government employee causing harm outside the State of the district court in which the action is filed requires a determination of the source of the substantive law that will govern liability. When the FTCA was passed, the dominant principle in choice of law analysis for tort cases was lex loci delicti: courts generally applied the law of the place where the injury occurred. See Richards v. United States, 369 U.S., at 11—12 (“The general conflict-of-laws rule, followed by a vast majority of the States, is to apply the law of the place of injury to the substantive rights of the parties” (footnote omitted)); see also Restatement (First) of Conflict of Laws §379 (1934) (defendant’s liability determined by “the law of the place of wrong”);3 id., §377, Note 1 (place of wrong for torts involving bodily harm is “the place where the harmful force takes effect upon the body” (emphasis in original)); ibid. (same principle for torts of fraud and torts involving harm to property).4 For a plaintiff injured in a foreign country, then, the presumptive choice in American courts under the traditional rule would have been to apply foreign law to determine the tortfeasor’s liability. See, e.g., Day & Zimmermann, Inc. v. Challoner, 423 U.S. 3 (1975) (per curiam) (noting that Texas would apply Cambodian law to wrongful-death action involving explosion in Cambodia of an artillery round manufactured in United States); Thomas v. FMC Corp., 610 F. Supp. 912 (MD Ala. 1985) (applying German law to determine American manufacturer’s liability for negligently designing and manufacturing a Howitzer that killed decedent in Germany); Quandt v. Beech Aircraft Corp., 317 F. Supp. 1009 (Del. 1970) (noting that Italian law applies to allegations of negligent manufacture in Kansas that resulted in an airplane crash in Italy); Manos v. Trans World Airlines, 295 F. Supp. 1170 (ND Ill. 1969) (applying Italian law to determine American corporation’s liability for negligent manufacture of a plane that crashed in Italy); see also, e.g., Dallas v. Whitney, 118 W. Va. 106, 188 S. E. 766 (1936) (Ohio law applied where blasting operations on a West Virginia highway caused property damage in Ohio); Cameron v. Vandergriff, 53 Ark. 381, 13 S. W. 1092 (1890) (Arkansas law applied where a blasting of a rock in Indian territory inflicted injury on plaintiff in Arkansas).

The application of foreign substantive law exemplified in these cases was, however, what Congress intended to avoid by the foreign country exception. In 1942, the House Committee on the Judiciary considered an early draft of the FTCA that would have exempted all claims “arising in a foreign country in behalf of an alien.” H. R. 5373, 77th Cong., 2d Sess., §303(12). The bill was then revised, at the suggestion of the Attorney General, to omit the last five words. In explaining the amendment to the House Committee on the Judiciary, Assistant Attorney General Shea said that

“[c]laims arising in a foreign country have been exempted from this bill, H. R. 6463, whether or not the claimant is an alien. Since liability is to be determined by the law of the situs of the wrongful act or omission it is wise to restrict the bill to claims arising in this country. This seems desirable because the law of the particular State is being applied. Otherwise, it will lead I think to a good deal of difficulty.” Hearings on H. R. 5373 et al. before the House Committee on the Judiciary, 77th Cong., 2d Sess., 35 (1942).

The amended version, which was enacted into law and constitutes the current text of the foreign country exception, 28 U.S.C. § 2680(k), thus codified Congress’s “unwilling[ness] to subject the United States to liabilities depending upon the laws of a foreign power.” United States v. Spelar, 338 U.S. 217, 221 (1949). See also Sami v. United States, 617 F.2d, at 762 (noting Spelar’s explanation but attempting to recast the object behind the foreign country exception); Leaf v. United States, 588 F.2d 733, 736, n. 3 (CA9 1978).

The object being to avoid application of substantive foreign law, Congress evidently used the modifier “arising in a foreign country” to refer to claims based on foreign harm or injury, the fact that would trigger application of foreign law to determine liability. That object, addressed by the quoted phrase, would obviously have been thwarted, however, by applying the headquarters doctrine, for that doctrine would have displaced the exception by recasting claims of foreign injury as claims not arising in a foreign country because some planning or negligence at domestic headquarters was their cause.5 And that, in turn, would have resulted in applying foreign law of the place of injury, in accordance with the choice-of-law rule of the headquarters jurisdiction.

Nor, as a practical matter, can it be said that the headquarters doctrine has outgrown its tension with the exception. It is true that the traditional approach to choice of substantive tort law has lost favor, Simson, The Choice-of-Law Revolution in the United States: Notes on Rereading Von Mehren, 36 Cornell Int’l L. J. 125, 125 (2003) (“The traditional methodology of place of wrong … has receded in importance, and new approaches and concepts such as governmental interest analysis, most significant relationship, and better rule of law have taken center stage” (footnotes omitted)).6 But a good many States still employ essentially the same choice of law analysis in tort cases that the First Restatement exemplified. Symeonides, Choice of Law in the American Courts, 51 Am. J. Comp. L. 1, 4—5 (2003) (“Ten states continue to adhere to the traditional method in tort conflicts”); see, e.g., Raskin v. Allison, 30 Kan. App. 2d 1240, 1242, 1241, 57 P.3d 30, 32 (2002) (under “traditional choice of law principles largely reflected in the original Restatement,” Mexican law applied to boating accident in Mexican waters because “the injuries were sustained in Mexican waters”).

Equally to the point is that in at least some cases that the Court of Appeals’s approach would treat as arising at headquarters, not the foreign country, even the later methodologies of choice point to the application of foreign law. The Second Restatement itself, encouraging the general shift towards using flexible balancing analysis to inform choice of law,7 includes a default rule for tort cases rooted in the traditional approach: “[i]n an action for a personal injury, the local law of the state where the injury occurred determines the rights and liabilities of the parties, unless … some other state has a more significant relationship … to the occurrence and the parties.” Restatement (Second) of Conflict of Laws §146 (1969); see also id., Comment e (“On occasion, conduct and personal injury will occur in different states. In such instances, the local law of the state of injury will usually be applied to determine most issues involving the tort”). In practice, then, the new dispensation frequently leads to the traditional application of the law of the jurisdiction of injury. See, e.g., Dorman v. Emerson Elec. Co., 23 F.3d 1354 (CA8 1994) (applying Canadian law where negligent saw design in Missouri caused injury in Canada); Bing v. Halstead, 495 F. Supp. 517 (SDNY 1980) (applying Costa Rican law where letter written and mailed in Arizona caused mental distress in Costa Rica); McKinnon v. F. H. Morgan & Co., 170 Vt. 422, 750 A. 2d 1026 (2000) (applying Canadian law where a defective bicycle sold in Vermont caused injuries in Quebec).

In sum, current flexibility in choice of law methodology gives no assurance against applying foreign substantive law if federal courts follow headquarters doctrine to assume jurisdiction over tort claims against the Government for foreign harm. Based on the experience just noted, the expectation is that application of the headquarters doctrine would in fact result in a substantial number of cases applying the very foreign law the foreign country exception was meant to avoid.8

Before concluding that headquarters analysis should have no part in applying the foreign country exception, however, a word is needed to answer an argument for selective application of headquarters doctrine, that it ought to be permitted when a State’s choice of law approach would not apply the foreign law of place of injury. See In re “Agent Orange” Product Liability Litigation, 580 F. Supp. 1242, 1254 (EDNY 1984) (noting that the purpose of the exception did not apply to the litigation at hand because foreign law was not implicated). The point would be well taken, of course, if Congress had written the exception to apply when foreign law would be applied. But that is not what Congress said. Its provision of an exception when a claim arises in a foreign country was written at a time when the phrase “arising in” was used in state statutes to express the position that a claim arises where the harm occurs; and the odds are that Congress meant simply this when it used the “arising in” language.9 Finally, even if it were not a stretch to equate “arising in a foreign country” with “implicating foreign law,” the result of accepting headquarters analysis for foreign injury cases in which no application of foreign law would ensue would be a scheme of federal jurisdiction that would vary from State to State, benefiting or penalizing plaintiffs accordingly. The idea that Congress would have intended any such jurisdictional variety is too implausible to drive the analysis to the point of grafting even a selective headquarters exception onto the foreign country exception itself. We therefore hold that the FTCA’s foreign country exception bars all claims based on any injury suffered in a foreign country, regardless of where the tortious act or omission occurred.


Alvarez has also brought an action under the ATS against petitioner, Sosa, who argues (as does the United States supporting him) that there is no relief under the ATS because the statute does no more than vest federal courts with jurisdiction, neither creating nor authorizing the courts to recognize any particular right of action without further congressional action. Although we agree the statute is in terms only jurisdictional, we think that at the time of enactment the jurisdiction enabled federal courts to hear claims in a very limited category defined by the law of nations and recognized at common law. We do not believe, however, that the limited, implicit sanction to entertain the handful of international law cum common law claims understood in 1789 should be taken as authority to recognize the right of action asserted by Alvarez here.


Judge Friendly called the ATS a “legal Lohengrin,” IIT v. Vencap, Ltd., 519 F.2d 1001, 1015 (CA2 1975); “no one seems to know whence it came,” ibid., and for over 170 years after its enactment it provided jurisdiction in only one case. The first Congress passed it as part of the Judiciary Act of 1789, in providing that the new federal district courts “shall also have cognizance, concurrent with the courts of the several States, or the circuit courts, as the case may be, of all causes where an alien sues for a tort only in violation of the law of nations or a treaty of the United States.” Act of Sept. 24, 1789, ch. 20, §9(b), 1 Stat. 79.10

The parties and amici here advance radically different historical interpretations of this terse provision. Alvarez says that the ATS was intended not simply as a jurisdictional grant, but as authority for the creation of a new cause of action for torts in violation of international law. We think that reading is implausible. As enacted in 1789, the ATS gave the district courts “cognizance” of certain causes of action, and the term bespoke a grant of jurisdiction, not power to mold substantive law. See, e.g., The Federalist No. 81, pp. 447, 451 (J. Cooke ed. 1961) (A. Hamilton) (using “jurisdiction” interchangeably with “cognizance”). The fact that the ATS was placed in §9 of the Judiciary Act, a statute otherwise exclusively concerned with federal-court jurisdiction, is itself support for its strictly jurisdictional nature. Nor would the distinction between jurisdiction and cause of action have been elided by the drafters of the Act or those who voted on it. As Fisher Ames put it, “there is a substantial difference between the jurisdiction of courts and rules of decision.” 1 Annals of Cong. 807 (Gales ed. 1834). It is unsurprising, then, that an authority on the historical origins of the ATS has written that “section 1350 clearly does not create a statutory cause of action,” and that the contrary suggestion is “simply frivolous.” Casto, The Federal Courts’ Protective Jurisdiction Over Torts Committed in Violation of the Law of Nations, 18 Conn. L. Rev. 467, 479, 480 (1986) (hereinafter Casto, Law of Nations); Cf. Dodge, The Constitutionality of the Alien Tort Statute: Some Observations on Text and Context, 42 Va. J. Int’l L. 687, 689 (2002). In sum, we think the statute was intended as jurisdictional in the sense of addressing the power of the courts to entertain cases concerned with a certain subject.

But holding the ATS jurisdictional raises a new question, this one about the interaction between the ATS at the time of its enactment and the ambient law of the era. Sosa would have it that the ATS was stillborn because there could be no claim for relief without a further statute expressly authorizing adoption of causes of action. Amici professors of federal jurisdiction and legal history take a different tack, that federal courts could entertain claims once the jurisdictional grant was on the books, because torts in violation of the law of nations would have been recognized within the common law of the time. Brief for Vikram Amar et al. as Amici Curiae. We think history and practice give the edge to this latter position.


“When the United States declared their independence, they were bound to receive the law of nations, in its modern state of purity and refinement.” Ware v. Hylton, 3 Dall. 199, 281 (1796) (Wilson, J.). In the years of the early Republic, this law of nations comprised two principal elements, the first covering the general norms governing the behavior of national states with each other: “the science which teaches the rights subsisting between nations or states, and the obligations correspondent to those rights,” E. de Vattel, The Law of Nations, Preliminaries §3 (J. Chitty et al. transl. and ed. 1883) (hereinafter Vattel) (footnote omitted), or “that code of public instruction which defines the rights and prescribes the duties of nations, in their intercourse with each other,” 1 James Kent Commentaries *1. This aspect of the law of nations thus occupied the executive and legislative domains, not the judicial. See 4 W. Blackstone, Commentaries on the Laws of England 68 (1769) (hereinafter Commentaries) (“[O]ffenses against” the law of nations are “principally incident to whole states or nations”).

The law of nations included a second, more pedestrian element, however, that did fall within the judicial sphere, as a body of judge-made law regulating the conduct of individuals situated outside domestic boundaries and consequently carrying an international savor. To Blackstone, the law of nations in this sense was implicated “in mercantile questions, such as bills of exchange and the like; in all marine causes, relating to freight, average, demurrage, insurances, bottomry … ; [and] in all disputes relating to prizes, to shipwrecks, to hostages, and ransom bills.” Id., at 67. The law merchant emerged from the customary practices of international traders and admiralty required its own transnational regulation. And it was the law of nations in this sense that our precursors spoke about when the Court explained the status of coast fishing vessels in wartime grew from “ancient usage among civilized nations, beginning centuries ago, and gradually ripening into a rule of international law … .” The Paquete Habana, 175 U.S. 677, 686 (1900).

There was, finally, a sphere in which these rules binding individuals for the benefit of other individuals overlapped with the norms of state relationships. Blackstone referred to it when he mentioned three specific offenses against the law of nations addressed by the criminal law of England: violation of safe conducts, infringement of the rights of ambassadors, and piracy. 4 Commentaries 68. An assault against an ambassador, for example, impinged upon the sovereignty of the foreign nation and if not adequately redressed could rise to an issue of war. See Vattel 463—464. It was this narrow set of violations of the law of nations, admitting of a judicial remedy and at the same time threatening serious consequences in international affairs, that was probably on minds of the men who drafted the ATS with its reference to tort.


Before there was any ATS, a distinctly American preoccupation with these hybrid international norms had taken shape owing to the distribution of political power from independence through the period of confederation. The Continental Congress was hamstrung by its inability to “cause infractions of treaties, or of the law of nations to be punished,” J. Madison, Journal of the Constitutional Convention 60 (E. Scott ed. 1893), and in 1781 the Congress implored the States to vindicate rights under the law of nations. In words that echo Blackstone, the congressional resolution called upon state legislatures to “provide expeditious, exemplary, and adequate punishment” for “the violation of safe conducts or passports, … of hostility against such as are in amity, … with the United States, … infractions of the immunities of ambassadors and other public ministers … [and] “infractions of treaties and conventions to which the United States are a party.” 21 Journals of the Continental Congress 1136—1137 (G. Hunt ed. 1912) (hereinafter Journals of the Continental Congress). The resolution recommended that the States “authorise suits … for damages by the party injured, and for compensation to the United States for damage sustained by them from an injury done to a foreign power by a citizen.” Id., at 1137; cf. Vattel 463—464 (“Whoever offends … a public minister … should be punished … , and … the state should, at the expense of the delinquent, give full satisfaction to the sovereign who has been offended in the person of his minister”). Apparently only one State acted upon the recommendation, see First Laws of the State of Connecticut 82, 83 (J. Cushing ed. 1982) (1784 compilation; exact date of Act unknown), but Congress had done what it could to signal a commitment to enforce the law of nations.

Appreciation of the Continental Congress’s incapacity to deal with this class of cases was intensified by the so-called Marbois incident of May 1784, in which a French adventurer, Longchamps, verbally and physically assaulted the Secretary of the French Legion in Philadelphia. See Respublica v. De Longchamps, 1 Dall. 111 (O. T. Phila. 1784).11 Congress called again for state legislation addressing such matters, and concern over the inadequate vindication of the law of nations persisted through the time of the constitutional convention. See 1 Records of the Federal Convention of 1787, p. 25 (M. Farrand ed. 1911) (speech of J. Randolph). During the Convention itself, in fact, a New York City constable produced a reprise of the Marbois affair and Secretary Jay reported to Congress on the Dutch Ambassador’s protest, with the explanation that “the federal government does not appear … to be vested with any judicial Powers competent to the Cognizance and Judgment of such Cases.” Casto, Law of Nations 494, and n. 152.

The Framers responded by vesting the Supreme Court with original jurisdiction over “all Cases affecting Ambassadors, other public ministers and Consuls.” U.S. Const., Art. III, §2, and the First Congress followed through. The Judiciary Act reinforced this Court’s original jurisdiction over suits brought by diplomats, see 1 Stat. 80, ch. 20, §13, created alienage jurisdiction, §11 and, of course, included the ATS, §9. See generally Randall, Federal Jurisdiction over International Law Claims: Inquiries into the Alien Tort Statute, 18 N. Y. U. J. Int’l L. & Pol. 1, 15—21 (1985) (discussing foreign affairs implications of the Judiciary Act); W. Casto, The Supreme Court in the Early Republic 27—53 (1995).


Although Congress modified the draft of what became the Judiciary Act, see generally Warren, New Light on the History of the Federal Judiciary Act of 1789, 37 Harv. L. Rev. 49 (1923), it made hardly any changes to the provisions on aliens, including what became the ATS, see Casto, Law of Nations 498. There is no record of congressional discussion about private actions that might be subject to the jurisdictional provision, or about any need for further legislation to create private remedies; there is no record even of debate on the section. Given the poverty of drafting history, modern commentators have necessarily concentrated on the text, remarking on the innovative use of the word “tort,” see, e.g., Sweeney, A Tort only in Violation of the Law of Nations, 18 Hastings Int’l & Comp. L. Rev. 445 (1995) (arguing that “tort” refers to the law of prize), and the statute’s mixture of terms expansive (“all suits”), see, e.g., Casto, Law of Nations 500, and restrictive (“for a tort only”), see, e.g., Randall, supra, at 28—31 (limiting suits to torts, as opposed to commercial actions, especially by British plaintiffs).12 The historical scholarship has also placed the ATS within the competition between federalist and antifederalist forces over the national role in foreign relations. Id., at 22—23 (nonexclusiveness of federal jurisdiction under the ATS may reflect compromise). But despite considerable scholarly attention, it is fair to say that a consensus understanding of what Congress intended has proven elusive.

Still, the history does tend to support two propositions. First, there is every reason to suppose that the First Congress did not pass the ATS as a jurisdictional convenience to be placed on the shelf for use by a future Congress or state legislature that might, some day, authorize the creation of causes of action or itself decide to make some element of the law of nations actionable for the benefit of foreigners. The anxieties of the preconstitutional period cannot be ignored easily enough to think that the statute was not meant to have a practical effect. Consider that the principal draftsman of the ATS was apparently Oliver Ellsworth,13 previously a member of the Continental Congress that had passed the 1781 resolution and a member of the Connecticut Legislature that made good on that congressional request. See generally W. Brown, The Life of Oliver Ellsworth (1905). Consider, too, that the First Congress was attentive enough to the law of nations to recognize certain offenses expressly as criminal, including the three mentioned by Blackstone. See An Act for the Punishment of Certain Crimes Against the United States, §8, 1 Stat. 113—114 (murder or robbery, or other capital crimes, punishable as piracy if committed on the high seas), and §28, id., at 118 (violation of safe conducts and assaults against ambassadors punished by imprisonment and fines described as “infract[ions of] the law of nations”). It would have been passing strange for Ellsworth and this very Congress to vest federal courts expressly with jurisdiction to entertain civil causes brought by aliens alleging violations of the law of nations, but to no effect whatever until the Congress should take further action. There is too much in the historical record to believe that Congress would have enacted the ATS only to leave it lying fallow indefinitely.

The second inference to be drawn from the history is that Congress intended the ATS to furnish jurisdiction for a relatively modest set of actions alleging violations of the law of nations. Uppermost in the legislative mind appears to have been offenses against ambassadors, see id., at 118; violations of safe conduct were probably understood to be actionable, ibid., and individual actions arising out of prize captures and piracy may well have also been contemplated. Id., at 113—114. But the common law appears to have understood only those three of the hybrid variety as definite and actionable, or at any rate, to have assumed only a very limited set of claims. As Blackstone had put it, “offences against this law [of nations] are principally incident to whole states or nations,” and not individuals seeking relief in court. 4 Commentaries 68.


The sparse contemporaneous cases and legal materials referring to the ATS tend to confirm both inferences, that some, but few, torts in violation of the law of nations were understood to be within the common law. In Bolchos v. Darrel, 3 F. Cas. 810 (No. 1,607) (S. C. 1795), the District Court’s doubt about admiralty jurisdiction over a suit for damages brought by a French privateer against the mortgagee of a British slave ship was assuaged by assuming that the ATS was a jurisdictional basis for the court’s action. Nor is Moxon v. The Fanny, 17 F. Cas. 942 (No. 9, 895) (Pa. 1793), to the contrary, a case in which the owners of a British ship sought damages for its seizure in United States waters by a French privateer. The District Court said in dictum that the ATS was not the proper vehicle for suit because “[i]t cannot be called a suit for a tort only, when the property, as well as damages for the supposed trespass, are sought for.” Id., at 948. But the judge gave no intimation that further legislation would have been needed to give the District Court jurisdiction over a suit limited to damages.

Then there was the 1795 opinion of Attorney General William Bradford, who was asked whether criminal prosecution was available against Americans who had taken part in the French plunder of a British slave colony in Sierra Leone. 1 Op. Atty. Gen. 57. Bradford was uncertain, but he made it clear that a federal court was open for the prosecution of a tort action growing out of the episode:

“But there can be no doubt that the company or individuals who have been injured by these acts of hostility have a remedy by a civil suit in the courts of the United States; jurisdiction being expressly given to these courts in all cases where an alien sues for a tort only, in violation of the laws of nations, or a treaty of the United States … .” Id., at 59.

Although it is conceivable that Bradford (who had prosecuted in the Marbois incident, see Casto, Law of Nations 503, n. 201) assumed that there had been a violation of a treaty, 1 Op. Atty. Gen., at 58, that is certainly not obvious, and it appears likely that Bradford understood the ATS to provide jurisdiction over what must have amounted to common law causes of action.


Against these indications that the ATS was meant to underwrite litigation of a narrow set of common law actions derived from the law of nations, Sosa raises two main objections. First, he claims that this conclusion makes no sense in view of the Continental Congress’s 1781 recommendation to state legislatures to pass laws authorizing such suits. Sosa thinks state legislation would have been “absurd,” Reply Brief for Petitioner Sosa 5, if common law remedies had been available. Second, Sosa juxtaposes Blackstone’s treatise mentioning violations of the law of nations as occasions for criminal remedies, against the statute’s innovative reference to “tort,” as evidence that there was no familiar set of legal actions for exercise of jurisdiction under the ATS. Neither argument is

The notion that it would have been absurd for the Continental Congress to recommend that States pass positive law to duplicate remedies already available at common law rests on a misunderstanding of the relationship between common law and positive law in the late 18th century, when positive law was frequently relied upon to reinforce and give standard expression to the “brooding omnipresence”14 of the common law then thought discoverable by reason. As Blackstone clarified the relation between positive law and the law of nations, “those acts of parliament, which have from time to time been made to enforce this universal law, or to facilitate the execution of [its] decisions, are not to be considered as introductive of any new rule, but merely as declaratory of the old fundamental constitutions of the kingdom; without which it must cease to be a part of the civilized world.” 4 Commentaries 67. Indeed, Sosa’s argument is undermined by the 1781 resolution on which he principally relies. Notwithstanding the undisputed fact (per Blackstone) that the common law afforded criminal law remedies for violations of the law of nations, the Continental Congress encouraged state legislatures to pass criminal statutes to the same effect, and the first Congress did the same, supra, at 23.15

Nor are we convinced by Sosa’s argument that legislation conferring a right of action is needed because Blackstone treated international law offenses under the rubric of “public wrongs,” whereas the ATS uses a word, “tort,” that was relatively uncommon in the legal vernacular of the day. It is true that Blackstone did refer to what he deemed the three principal offenses against the law of nations in the course of discussing criminal sanctions, observing that it was in the interest of sovereigns “to animadvert upon them with a becoming severity, that the peace of the world may be maintained,” 4 Commentaries 68.16 But Vattel explicitly linked the criminal sanction for offenses against ambassadors with the requirement that the state, “at the expense of the delinquent, give full satisfaction to the sovereign who has been offended in the person of his minister.” Vattel 463—464. Cf. Stephens, Individuals Enforcing International Law: The Comparative and Historical Context, 52 DePaul L. Rev. 433, 444 (2002) (observing that a “mixed approach to international law violations, encompassing both criminal prosecution … and compensation to those injured through a civil suit, would have been familiar to the founding generation”). The 1781 resolution goes a step further in showing that a private remedy was thought necessary for diplomatic offenses under the law of nations. And the Attorney General’s Letter of 1795, as well as the two early federal precedents discussing the ATS, point to a prevalent assumption that Congress did not intend the ATS to sit on the shelf until some future time when it might enact further legislation.

In sum, although the ATS 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.


We think it is correct, then, to assume that the First Congress understood that the district courts would recognize private causes of action for certain torts in violation of the law of nations, though we have found no basis to suspect Congress had any examples in mind beyond those torts corresponding to Blackstone’s three primary offenses: violation of safe conducts, infringement of the rights of ambassadors, and piracy. We assume, too, that no development in the two centuries from the enactment of §1350 to the birth of the modern line of cases beginning with Filartiga v. Pena-Irala, 630 F.2d 876 (CA2 1980), has categorically precluded federal courts from recognizing a claim under the law of nations as an element of common law; Congress has not in any relevant way amended §1350 or limited civil common law power by another statute. Still, there are good reasons for a restrained conception of the discretion a federal court should exercise in considering a new cause of action of this kind. Accordingly, we think courts should require any claim based on the present-day law of nations to rest on a norm of international character accepted by the civilized world and defined with a specificity comparable to the features of the 18th-century paradigms we have recognized. This requirement is fatal to Alvarez’s claim.


A series of reasons argue for judicial caution when considering the kinds of individual claims that might implement the jurisdiction conferred by the early statute. First, the prevailing conception of the common law has changed since 1789 in a way that counsels restraint in judicially applying internationally generated norms. When §1350 was enacted, the accepted conception was of the common law as “a transcendental body of law outside of any particular State but obligatory within it unless and until changed by statute.” Black and White Taxicab & Transfer Co. v. Brown and Yellow Taxicab & Transfer Co., 276 U.S. 518, 533 (1928) (Holmes, J., dissenting). Now, however, in most cases where a court is asked to state or formulate a common law principle in a new context, there is a general understanding that the law is not so much found or discovered as it is either made or created. Holmes explained famously in 1881 that

“in substance the growth of the law is legislative … [because t]he very considerations which judges most rarely mention, and always with an apology, are the secret root from which the law draws all the juices of life. I mean, of course, considerations of what is expedient for the community concerned.” The Common Law 31—32 (Howe ed. 1963).

One need not accept the Holmesian view as far as its ultimate implications to acknowledge that a judge deciding in reliance on an international norm will find a substantial element of discretionary judgment in the decision.

Second, along with, and in part driven by, that conceptual development in understanding common law has come an equally significant rethinking of the role of the federal courts in making it. Erie R. Co. v. Tompkins, 304 U.S. 64 (1938), was the watershed in which we denied the existence of any federal “general” common law, id., at 78, which largely withdrew to havens of specialty, some of them defined by express congressional authorization to devise a body of law directly, e.g., Textile Workers v. Lincoln Mills of Ala., 353 U.S. 448 (1957) (interpretation of collective-bargaining agreements); Fed. Rule Evid. 501 (evidentiary privileges in federal-question cases). Elsewhere, this Court has thought it was in order to create federal common law rules in interstitial areas of particular federal interest. E.g., United States v. Kimbell Foods, Inc., 440 U.S. 715, 726—727 (1979).17 And although we have even assumed competence to make judicial rules of decision of particular importance to foreign relations, such as the act of state doctrine, see Banco Nacional de Cuba v. Sabbatino, 376 U.S. 398, 427 (1964), the general practice has been to look for legislative guidance before exercising innovative authority over substantive law. It would be remarkable
to take a more aggressive role in exercising a jurisdiction that remained largely in shadow for much of the prior two centuries.

Third, this Court has recently and repeatedly said that a decision to create a private right of action is one better left to legislative judgment in the great majority of cases. Correctional Services Corp. v. Malesko, 534 U.S. 61, 68 (2001); Alexander v. Sandoval, 532 U.S. 275, 286—287 (2001). The creation of a private right of action raises issues beyond the mere consideration whether underlying primary conduct should be allowed or not, entailing, for example, a decision to permit enforcement without the check imposed by prosecutorial discretion. Accordingly, even when Congress has made it clear by statute that a rule applies to purely domestic conduct, we are reluctant to infer intent to provide a private cause of action where the statute does not supply one expressly. While the absence of congressional action addressing private rights of action under an international norm is more equivocal than its failure to provide such a right when it creates a statute, the possible collateral consequences of making international rules privately actionable argue for judicial caution.

Fourth, the subject of those collateral consequences is itself a reason for a high bar to new private causes of action for violating international law, for the potential implications for the foreign relations of the United States of recognizing such causes should make courts particularly wary of impinging on the discretion of the Legislative and Executive Branches in managing foreign affairs. It is one thing for American courts to enforce constitutional limits on our own State and Federal Governments’ power, but quite another to 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 to hold that a foreign government or its agent has transgressed those limits. Cf. Sabbatino, supra, at 431—432. Yet modern international law is very much concerned with just such questions, and apt to stimulate calls for vindicating private interests in §1350 cases. Since many attempts by federal courts to craft remedies for the violation of new norms of international law would raise risks of adverse foreign policy consequences, they should be undertaken, if at all, with great caution. Cf. Tel-Oren v. Libyan Arab Republic, 726 F.2d 774, 813 (CADC 1984) (Bork, J., concurring) (expressing doubt that §1350 should be read to require “our courts [to] sit in judgment of the conduct of foreign officials in their own countries with respect to their own citizens”).

The fifth reason is particularly important in light of the first four. We have no congressional mandate to seek out and define new and debatable violations of the law of nations, and modern indications of congressional understanding of the judicial role in the field have not affirmatively encouraged greater judicial creativity. It is true that a clear mandate appears in the Torture Victim Protection Act of 1991, 106 Stat. 73, providing authority that “establish[es] an unambiguous and modern basis for” federal claims of torture and extrajudicial killing, H. R. Rep. No. 102—367, pt. 1, p. 3 (1991). But that affirmative authority is confined to specific subject matter, and although the legislative history includes the remark that §1350 should “remain intact to permit suits based on other norms that already exist or may ripen in the future into rules of customary international law,” id., at 4, Congress as a body has done nothing to promote such suits. Several times, indeed, the Senate has expressly declined to give the federal courts the task of interpreting and applying international human rights law, as when its ratification of the International Covenant on Civil and Political Rights declared that the substantive provisions of the document were not self-executing. 138 Cong. Rec. 8071 (1992).


These reasons argue for great caution in adapting the law of nations to private rights. Justice Scalia, post, p. 1 (opinion concurring in part and concurring in judgment) concludes that caution is too hospitable, and a word is in order to summarize where we have come so far and to focus our difference with him on whether some norms of today’s law of nations may ever be recognized legitimately by federal courts in the absence of congressional action beyond §1350. All Members of the Court agree that §1350 is only jurisdictional. We also agree, or at least Justice Scalia does not dispute, post, at 2, 7, that the jurisdiction was originally understood to be available to enforce a small number of international norms that a federal court could properly recognize as within the common law enforceable without further statutory authority. Justice Scalia concludes, however, that two subsequent developments should be understood to preclude federal courts from recognizing any further international norms as judicially enforceable today, absent further congressional action. As described before, we now tend to understand common law not as a discoverable reflection of universal reason but, in a positivistic way, as a product of human choice. And we now adhere to a conception of limited judicial power first expressed in reorienting federal diversity jurisdiction, see Erie R. Co. v. Tompkins, 304 U.S. 64 (1938), that federal courts have no authority to derive “general” common law.

Whereas Justice Scalia sees these developments as sufficient to close the door to further independent judicial recognition of actionable international norms, other considerations persuade us that the judicial power should be exercised on the understanding that the door is still ajar subject to vigilant doorkeeping, and thus open to a narrow class of international norms today. Erie did not in terms bar any judicial recognition of new substantive rules, no matter what the circumstances, and post-Erie understanding has identified limited enclaves in which federal courts may derive some substantive law in a common law way. For two centuries we have affirmed that the domestic law of the United States recognizes the law of nations. See, e.g., Sabbatino, 376 U.S., at 423 (“[I]t is, of course, true that United States courts apply international law as a part of our own in appropriate circumstances”);18 The Paquete Habana, 175 U.S., at 700 (“International law is part of our law, and must be ascertained and administered by the courts of justice of appropriate jurisdiction, as often as questions of right depending upon it are duly presented for their determination”); The Nereide, 9 Cranch 388, 423 (1815) (Marshall, C. J.) (“[T]he Court is bound by the law of nations which is a part of the law of the land”); see also Texas Industries, Inc. v. Radcliff Materials, Inc., 451 U.S. 630, 641 (1981) (recognizing that “international disputes implicating … our relations with foreign nations” are one of the “narrow areas” in which “federal common law” continues to exist). It would take some explaining to say now that federal courts must avert their gaze entirely from any international norm intended to protect individuals.

We think an attempt to justify such a position would be particularly unconvincing in light of what we know about congressional understanding bearing on this issue lying at the intersection of the judicial and legislative powers. The First Congress, which reflected the understanding of the framing generation and included some of the Framers, assumed that federal courts could properly identify some international norms as enforceable in the exercise of §1350 jurisdiction. We think it would be unreasonable to assume that the First Congress would have expected federal courts to lose all capacity to recognize enforceable international norms simply because the common law might lose some metaphysical cachet on the road to modern realism. Later Congresses seem to have shared our view. The position we take today has been assumed by some federal courts for 24 years, ever since the Second Circuit decided Filartiga v. Pena-Irala, 630 F.2d 876 (CA2 1980), and for practical purposes the point of today’s disagreement has been focused since the exchange between Judge Edwards and Judge Bork in Tel-Oren v. Libyan Arab Republic, 726 F.2d 774 (CADC 1984), Congress, however, has not only expressed no disagreement with our view of the proper exercise of the judicial power, but has responded to its most notable instance by enacting legislation supplementing the judicial determination in some detail. See supra, at 34 (discussing the Torture Victim Protection Act).

While we agree with Justice Scalia to the point that we would welcome any congressional guidance in exercising jurisdiction with such obvious potential to affect foreign relations, nothing Congress has done is a reason for us to shut the door to the law of nations entirely. It is enough to say that Congress may do that at any time (explicitly, or implicitly by treaties or statutes that occupy the field) just as it may modify or cancel any judicial decision so far as it rests on recognizing an international norm as such.19


We must still, however, derive a standard or set of standards for assessing the particular claim Alvarez raises, and for this case it suffices to look to the historical antecedents. Whatever the ultimate criteria for accepting a cause of action subject to jurisdiction under §1350, we are persuaded that federal courts should not recognize private claims under federal common law for violations of any international law norm with less definite content and acceptance among civilized nations than the historical paradigms familiar when §1350 was enacted. See, e.g., United States v. Smith, 5 Wheat. 153, 163—180, n. a (1820) (illustrating the specificity with which the law of nations defined piracy). This limit upon judicial recognition is generally consistent with the reasoning of many of the courts and judges who faced the issue before it reached this Court. See Filartiga, supra, at 890 (“[F]or purposes of civil liability, the torturer has become–like the pirate and slave trader before him–hostis humani generis, an enemy of all mankind”); Tel-Oren, supra, at 781 (Edwards, J., concurring) (suggesting that the “limits of section 1350’s reach” be defined by “a handful of heinous actions–each of which violates definable, universal and obligatory norms”); see also In re Estate of Marcos Human Rights Litigation, 25 F.3d 1467, 1475 (CA9 1994) (“Actionable violations of international law must be of a norm that is specific, universal, and obligatory”). And the determination whether a norm is sufficiently definite to support a cause of action20 should (and, indeed, inevitably must) involve an element of judgment about the practical consequences of making that cause available to litigants in the federal courts.21

Thus, Alvarez’s detention claim must be gauged against the current state of international law, looking to those sources we have long, albeit cautiously, recognized.

“[W]here there is no treaty, and no controlling executive or legislative act or judicial decision, resort must be had to the customs and usages of civilized nations; and, as evidence of these, to the works of jurists and commentators, who by years of labor, research and experience, have made themselves peculiarly well acquainted with the subjects of which they treat. Such works are resorted to by judicial tribunals, not for the speculations of their authors concerning what the law ought to be, but for trustworthy evidence of what the law really is.” The Paquete Habana, 175 U.S., at 700.

To begin with, Alvarez cites two well-known international agreements that, despite their moral authority, have little utility under the standard set out in this opinion. He says that his abduction by Sosa was an “arbitrary arrest” within the meaning of the Universal Declaration of Human Rights (Declaration), G. A. Res. 217A (III), U. N. Doc. A/810 (1948). And he traces the rule against arbitrary arrest not only to the Declaration, but also to article nine of the International Covenant on Civil and Political Rights (Covenant), Dec. 19, 1996, 999 U. N. T. S. 171,22 to which the United States is a party, and to various other conventions to which it is not. But the Declaration does not of its own force impose obligations as a matter of international law. See Humphrey, The UN Charter and the Universal Declaration of Human Rights, in The International Protection of Human Rights 39, 50 (E. Luard ed. 1967) (quoting Eleanor Roosevelt calling the Declaration “ ‘a statement of principles … setting up a common standard of achievement for all peoples and all nations’ ” and “ ‘not a treaty or international agreement … impos[ing] legal obligations’ ”).23 And, although the Covenant does bind the United States as a matter of international law, the United States ratified the Covenant on the express understanding that it was not self-executing and so did not itself create obligations enforceable in the federal courts. See supra, at 33. Accordingly, Alvarez cannot say that the Declaration and Covenant themselves establish the relevant and applicable rule of international law. He instead attempts to show that prohibition of arbitrary arrest has attained the status of binding customary international law.

Here, it is useful to examine Alvarez’s complaint in greater detail. As he presently argues it, the claim does not rest on the cross-border feature of his abduction.24 Although the District Court granted relief in part on finding a violation of international law in taking Alvarez across the border from Mexico to the United States, the Court of Appeals rejected that ground of liability for failure to identify a norm of requisite force prohibiting a forcible abduction across a border. Instead, it relied on the conclusion that the law of the United States did not authorize Alvarez’s arrest, because the DEA lacked extraterritorial authority under 21 U.S.C. § 878 and because Federal Rule of Criminal Procedure 4(d)(2) limited the warrant for Alvarez’s arrest to “the jurisdiction of the United States.”25 It is this position that Alvarez takes now: that his arrest was arbitrary and as such forbidden by international law not because it infringed the prerogatives of Mexico, but because no applicable law authorized it.26

Alvarez thus invokes a general prohibition of “arbitrary” detention defined as officially sanctioned action exceeding positive authorization to detain under the domestic law of some government, regardless of the circumstances. Whether or not this is an accurate reading of the Covenant, Alvarez cites little authority that a rule so broad has the status of a binding customary norm today.27 He certainly cites nothing to justify the federal courts in taking his broad rule as the predicate for a federal lawsuit, for its implications would be breathtaking. His rule would support a cause of action in federal court for any arrest, anywhere in the world, unauthorized by the law of the jurisdiction in which it took place, and would create a cause of action for any seizure of an alien in violation of the Fourth Amendment, supplanting the actions under Rev. Stat. §1979, 42 U.S.C. § 1983 and Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S. 388 (1971), that now provide damages remedies for such violations. It would create an action in federal court for arrests by state officers who simply exceed their authority; and for the violation of any limit that the law of any country might place on the authority of its own officers to arrest. And all of this assumes that Alvarez could establish that Sosa was acting on behalf of a government when he made the arrest, for otherwise he would need a rule broader still.

Alvarez’s failure to marshal support for his proposed rule is underscored by the Restatement (Third) of Foreign Relations Law of the United States (1987), which says in its discussion of customary international human rights law that a “state violates international law if, as a matter of state policy, it practices, encourages, or condones … prolonged arbitrary detention.” Id., §702. Although the Restatement does not explain its requirements of a “state policy” and of “prolonged” detention, the implication is clear. Any credible invocation of a principle against arbitrary detention that the civilized world accepts as binding customary international law requires a factual basis beyond relatively brief detention in excess of positive authority. Even the Restatement’s limits are only the beginning of the enquiry, because although it is easy to say that some policies of prolonged arbitrary detentions are so bad that those who enforce them become enemies of the human race, it may be harder to say which policies cross that line with the certainty afforded by Blackstone’s three common law offenses. In any event, the label would never fit the reckless policeman who botches his warrant, even though that same officer might pay damages under municipal law. E.g., Groh v. Ramirez, 540 U.S. ___ (2004).28

Whatever may be said for the broad principle Alvarez advances, in the present, imperfect world, it expresses an aspiration that exceeds any binding customary rule having the specificity we require.29 Creating a private cause of action to further that aspiration would go beyond any residual common law discretion we think it appropriate to exercise.30 It is enough to hold that a single illegal detention of less than a day, followed by the transfer of custody to lawful authorities and a prompt arraignment, violates no norm of customary international law so well defined as to support the creation of a federal remedy.

* * *

The judgment of the Court of Appeals is



1. In the Ninth Circuit’s view, it was critical that “DEA agents had no authority under federal law to execute an extraterritorial arrest of a suspect indicted in federal court in Los Angeles.” 331 F.3d, at 640. Once Alvarez arrived in the United States, “the actions of domestic law enforcement set in motion a supervening prosecutorial mechanism which met all of the procedural requisites of federal due process.” Id., at 637.

2. See also Couzado v. United States, 105 F.3d 1389, 1396 (CA11 1997) (“ ‘[A] claim is not barred by section 2680(k) where the tortious conduct occurs in the United States, but the injury is sustained in a foreign country’ ”) (quoting Donahue v. United States Dept. of Justice, 751 F. Supp. 45, 47 (SDNY 1990)); Martinez v. Lamagno, No. 93—1573, 1994 WL 159771, *2, judgt. order reported at 23 F.3d 402 (CA4 1994) (per curiam) (unpublished opinion) (“A headquarters claim exists where negligent acts in the United States proximately cause harm in a foreign country”), rev’d on other grounds, 515 U.S. 417 (1995); Leaf v. United States, 588 F.2d 733, 736 (CA9 1978) (“A claim ‘arises’, as that term is used in … 2680(k), where the acts or omissions that proximately cause the loss take place”); cf. Eaglin v. United States, Dept. of Army, 794 F.2d 981, 983 (CA5 1986) (assuming, arguendo, that headquarters doctrine is valid).

3. See also Restatement (Second) of Conflict of Laws 412 (1969) (“The original Restatement stated that, with minor exceptions, all substantive questions relating to the existence of a tort claim are governed by the local law of the ‘place of wrong.’ This was described … as ‘the state where the last event necessary to make an actor liable for an alleged tort takes place.’ Since a tort is the product of wrongful conduct and of resulting injury and since the injury follows the conduct, the state of the ‘last event’ is the state where the injury occurred”).

4. The FTCA was passed with precisely these kinds of garden-variety torts in mind. See S. Rep. No. 1400, 79th Cong., 2d Sess., p. 31 (1946) (“With the expansion of governmental activities in recent years, it becomes especially important to grant to private individuals the right to sue the Government in respect to such torts as negligence in the operation of vehicles”); see generally Feres v. United States, 340 U.S. 135, 139—140 (1950) (Congress was principally concerned with making the Government liable for ordinary torts that “would have been actionable if inflicted by an individual or a corporation”).

5. The application of foreign law might nonetheless have been avoided in headquarters cases if courts had been instructed to apply the substantive tort law of the State where the federal act or omission occurred, regardless of where the ultimate harm transpired. But in Richards v. United States, 369 U.S. 1 (1962), we held that the Act requires “the whole law (including choice-of-law rules) … of the State where the [allegedly tortious federal] act or omission occurred,” id., at 3, 11. Given the dominant American choice-of-law approach at the time the Act was passed, that would have resulted in the application of foreign law in virtually any case where the plaintiff suffered injury overseas.

6. See also Rydstrom, Modern Status of Rule that Substantive Rights of Parties to a Tort Action are Governed by the Law of the Place of the Wrong, 29 A. L. R. 3d 603, *2 (1970) (“Many courts are now abandoning the orthodox rule that the substantive rights of the parties are governed by the law of the place of the wrong”). We express no opinion on the relative merits of the various approaches to choice questions; our discussion of the subject is intended only to indicate how, as a positive matter, transjurisdictional cases are likely to be treated today.

7. Under the Second Restatement, tort liability is determined “by the local law of the state which … has the most significant relationship to the occurrence and the parties,” taking into account “the place where the injury occurred,” “the place where the conduct causing the injury occurred,” “the domicil, residence, nationality, place of incorporation and place of business of the parties,” and “the place where the relationship, if any, between the parties is centered.” Restatement (Second) of Conflict of Laws §145 (1969).

8. The courts that have applied the headquarters doctrine, believing it to be intimated by our emphasis, in Richards v. United States, supra, on the place of the occurrence of the negligent act, have acknowledged the possibility that foreign law may govern FTCA claims as a function of Richards’s further holding that the whole law of the pertinent State (including its choice-of-law provisions) is to be applied. See, e.g., Leaf, 588 F.2d, at 736, n. 3. Some courts have attempted to defuse the resulting tension with the object behind the foreign country exception. See, e.g., Sami v. United States 617 F.2d 755, 763 (CADC 1979) (believing that norm against application of foreign law when contrary to forum policy is sufficient to overcome possible conflict). We think that these attempts to resolve the tension give short shrift to the clear congressional mandate embodied by the foreign country exception. Cf. Shapiro, Choice of Law Under the Federal Tort Claims Act: Richards and Renvoi Revisited, 70 N. C. L. Rev. 641, 659—660 (1992) (noting that the Richards rule that the totality of a State’s law is to be consulted may undermine the object behind the foreign country exception).

9. It is difficult to reconcile the Government’s contrary reading with the fact that two of the Act’s other exceptions specifically reference an “act or omission.” See 28 U.S.C. § 2680(a) (exempting United States from liability for “[a]ny claim based upon an act or omission of an employee of the Government, exercising due care, in the execution of a statute or regulation”); §2680(e) (“Any claim arising out of an act or omission of any employee of the Government in administering [certain portions of the Trading with the Enemy Act of 1917]”). The Government’s request that we read that phrase into the foreign country exception, when it is clear that Congress knew how to specify “act or omission” when it wanted to, runs afoul of the usual rule that “when the legislature uses certain language in one part of the statute and different language in another, the court assumes different meanings were intended.” 2A N. Singer, Statutes and Statutory Construction §46:06, p. 194 (6th ed. 2000).

10. The statute has been slightly modified on a number of occasions since its original enactment. It now reads in its entirety: “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.

11. The French minister plenipotentiary lodged a formal protest with the Continental Congress, 27 Journals of the Continental Congress 478, and threatened to leave Pennsylvania “unless the decision on Longchamps Case should give them full satisfaction.” Letter from Samuel Hardy to Gov. Benjamin Harrison of Virginia, June 24, 1784, in 7 Letters of Members of the Continental Congress 558, 559 (E. Burnett ed. 1934). Longchamps was prosecuted for a criminal violation of the law of nations in state court. The Congress could only pass resolutions, one approving the state-court proceedings, 27 Journals of the Continental Congress 503, another directing the Secretary of Foreign Affairs to apologize and to “explain to Mr. De Marbois the difficulties that may arise … from the nature of a federal union,” 28 Journals of the Continental Congress 314, and to explain to the representative of Louis XVI that “many allowances are to be made for” the young Nation. Ibid.

12. The restriction may have served the different purpose of putting foreigners on notice that they would no longer be able to prosecute their own criminal cases in federal court. Compare, e.g., 3 Commentaries 160 (victims could start prosecutions) with the Judiciary Act §35. (creating the office of the district attorney). Cf. 1 Op. Atty. Gen. 41, 42 (1794) (British consul could not himself initiate criminal prosecution, but could provide evidence to the grand jury.)

13. The ATS appears in Ellsworth’s handwriting in the original version of the bill in the National Archives. Casto, Law of Nations 498, n. 169.

14. See Southern Pacific Co. v. Jensen, 244 U.S. 205, 222 (1917) (Holmes, J., dissenting).

15. Being consistent with the prevailing understanding of international law, the 1781 resolution is sensibly understood as an act of international politics, for the recommendation was part of a program to assure the world that the new Republic would observe the law of nations. On the same day it made its recommendation to state legislatures, the Continental Congress received a confidential report, detailing negotiations between American representatives and Versailles. 21 Journals of the Continental Congress 1137—1140. The King was concerned about the British capture of the ship Marquis de la Fayette on its way to Boston, id., at 1139, and he “expresse[d] a desire that the plan for the appointment of consuls should be digested and adopted, as the Court of France wished to make it the basis of some commercial arrangements between France and the United States.” Id., at 1140. The congressional resolution would not have been all that Louis XVI wished for, but it was calculated to assure foreign powers that Congress at least intended their concerns to be addressed in the way they would have chosen. As a French legal treatise well known to early American lawyers, see Helmholz, Use of the Civil Law in Post-Revolutionary American Jurisprudence, 66 Tulane L. Rev. 1649 (1992), put it, “the laws ought to be written, to the end that the writing may fix the sense of the law, and determine the mind to conceive a just idea of that which is established by the law, and that it not be left free for every one to frame the law as he himself is pleased to understand it … .” 1 J. Domat, The Civil Law in its Natural Order 108, (W. Strahan transl. and L. Cushing ed. 1861). A congressional statement that common law was up to the task at hand might well have fallen short of impressing a continental readership.

16. Petitioner says animadversion is “an archaic reference to the imposition of punishment.” Reply Brief for Petitioner Sosa 4 (emphasis in original). That claim is somewhat exaggerated, however. To animadvert carried the broader implication of “turn[ing] the attention officially or judicially, tak[ing] legal cognizance of anything deserving of chastisement or censure; hence, to proceed by way of punishment or censure.” 1 Oxford English Dictionary 474 (2d ed. 1989). Blackstone in fact used the term in the context of property rights and damages. Of a man who is disturbed in his enjoyment of “qualified property” “the law will animadvert hereon as an injury.” 2 Commentaries 395. See also 9 Papers of James Madison 349 (R. Rutland ed. 1975) (“As yet foreign powers have not been rigorous in animadverting on us” for violations of the law of nations).

17. See generally R. Fallon, D. Meltzer, D. Shapiro, Hart and Wechsler’s The Federal Courts and the Federal System ch. 7 (5th ed. 2003); Friendly, In Praise of Erie–and of the New Federal Common Law, 39 N. Y. U. L. Rev. 383, 405—422 (1964).

18. Sabbatino itself did not directly apply international law, see 376 U.S., at 421—423, but neither did it question the application of that law in appropriate cases, and it further endorsed the reasoning of a noted commentator who had argued that Erie should not preclude the continued application of international law in federal courts. 376 U.S., at 425 (citing Jessup, The Doctrine of Erie Railroad v. Tompkins Applied to International Law, 33 Am. J. Int’l L. 740 (1939)).

19. Our position does not, as Justice Scalia suggests, imply that every grant of jurisdiction to a federal court carries with it an opportunity to develop common law (so that the grant of federal-question jurisdiction would be equally as good for our purposes as §1350), see post, at 8, n. 1. Section 1350 was enacted on the congressional understanding that courts would exercise jurisdiction by entertaining some common law claims derived from the law of nations; and we know of no reason to think that federal-question jurisdiction was extended subject to any comparable congressional assumption. Further, our holding today is consistent with the division of responsibilities between federal and state courts after Erie, see supra, at 32, 35—36, as a more expansive common law power related to 28 U.S.C. § 1331 might not be.

20. A related consideration is whether international law extends the scope of liability for a violation of a given norm to the perpetrator being sued, if the defendant is a private actor such as a corporation or individual. Compare Tel-Oren v. Libyan Arab Republic, 726 F.2d 774, 791—795 (CADC 1984) (Edwards, J., concurring) (insufficient consensus in 1984 that torture by private actors violates international law), with Kadic v. Karadzic, 70 F.3d 232, 239—241 (CA2 1995) (sufficient consensus in 1995 that genocide by private actors violates international law).

21. This requirement of clear definition is not meant to be the only principle limiting the availability of relief in the federal courts for violations of customary international law, though it disposes of this case. For example, the European Commission argues as amicus curiae that basic principles of international law require that before asserting a claim in a foreign forum, the claimant must have exhausted any remedies available in the domestic legal system, and perhaps in other fora such as international claims tribunals. See Brief for European Commission as Amicus Curiae 24, n. 54 (citing I. Brownlie, Principles of Public International Law 472—481 (6th ed. 2003)); cf. Torture Victim Protection Act of 1991, §2(b), 106 Stat. 73 (exhaustion requirement). We would certainly consider this requirement in an appropriate case. Another possible limitation that we need not apply here is a policy of case-specific deference to the political branches. For example, there are now pending in federal district court several class actions seeking damages from various corporations alleged to have participated in, or abetted, the regime of apartheid that formerly controlled South Africa. See In re South African Apartheid Litigation, 238 F. Supp. 2d 1379 (JPML 2002) (granting a motion to transfer the cases to the Southern District of New York). The Government of South Africa has said that these cases interfere with the policy embodied by its Truth and Reconciliation Commission, which “deliberately avoided a ‘victors’ justice’ approach to the crimes of apartheid and chose instead one based on confession and absolution, informed by the principles of reconciliation, reconstruction, reparation and goodwill.” Declaration of Penuell Mpapa Maduna, Minister of Justice and Constitutional Development, Republic of South Africa, reprinted in App. to Brief for Government of Commonwealth of Australia et al. as Amici Curiae 7a, ¶3.2.1 (emphasis deleted). The United States has agreed. See Letter of William H. Taft IV, Legal Adviser, Dept. of State, to Shannen W. Coffin, Deputy Asst. Atty. Gen., Oct. 27, 2003, reprinted in id., at 2a. In such cases, there is a strong argument that federal courts should give serious weight to the Executive Branch’s view of the case’s impact on foreign policy. Cf. Republic of Austria v. Altmann, 541 U.S. ___, ___ (2004) (slip op., at 23—24) (discussing the State Department’s use of statements of interest in cases involving the Foreign Sovereign Immunities Act of 1976, 28 U.S.C. § 1602 et seq.).

22. Article nine provides that “[n]o one shall be subjected to arbitrary arrest or detention,” that “[n]o one shall be deprived of his liberty except on such grounds and in accordance with such procedure as are established by law,” and that “[a]nyone who has been the victim of unlawful arrest or detention shall have an enforceable right to compensation.” 999 U. N. T. S., at 175—176.

23. It has nevertheless had substantial indirect effect on international law. See Brownlie, supra, at 535 (calling the Declaration a “good example of an informal prescription given legal significance by the actions of authoritative decision-makers”).

24. Alvarez’s brief contains one footnote seeking to incorporate by reference his arguments on cross-border abductions before the Court of Appeals. Brief for Respondent Alvarez-Machain 47, n. 46. That is not enough to raise the question fairly, and we do not consider it.

25. The Rule has since been moved and amended and now provides that a warrant may also be executed “anywhere else a federal statute authorizes an arrest.” Fed. Rule Crim. Proc. 4(c)(2).

26. We have no occasion to decide whether Alvarez is right that 21 U.S.C. § 878 did not authorize the arrest.

27. Specifically, he relies on a survey of national constitutions, Bassiouni, Human Rights in the Context of Criminal Justice: Identifying International Procedural Protections and Equivalent Protections in National Constitutions, 3 Duke J. Comp. & Int’l L. 235, 260—261 (1993); a case from the International Court of Justice, United States v. Iran, 1980 I. C. J. 3, 42; and some authority drawn from the federal courts, see Brief for Respondent Alvarez-Machain 49, n. 50. None of these suffice. The Bassiouni survey does show that many nations recognize a norm against arbitrary detention, but that consensus is at a high level of generality. The Iran case, in which the United States sought relief for the taking of its diplomatic and consular staff as hostages, involved a different set of international norms and mentioned the problem of arbitrary detention only in passing; the detention in that case was, moreover, far longer and harsher than Alvarez’s. See 1980 I. C. J., at 42, ¶91 (“detention of [United States] staff by a group of armed militants” lasted “many months”). And the authority from the federal courts, to the extent it supports Alvarez’s position, reflects a more assertive view of federal judicial discretion over claims based on customary international law than the position we take today.

28. In this case, Sosa might well have been liable under Mexican law. Alvarez asserted such a claim, but the District Court concluded that the applicable law was the law of California, and that under California law Sosa had been privileged to make a citizen’s arrest in Mexico. Whether this was correct is not now before us, though we discern tension between the court’s simultaneous conclusions that the detention so lacked any legal basis as to violate international law, yet was privileged by state law against ordinary tort recovery.

29. It is not that violations of a rule logically foreclose the existence of that rule as international law. Cf. Filartiga v. Pena-Irala, 630 F.2d 876, 884, n. 15 (CA2 1980) (“The fact that the prohibition of torture is often honored in the breach does not diminish its binding effect as a norm of international law”). Nevertheless, that a rule as stated is as far from full realization as the one Alvarez urges is evidence against its status as binding law; and an even clearer point against the creation by judges of a private cause of action to enforce the aspiration behind the rule claimed.

30. Alvarez also cites, Brief for Respondent Alvarez-Machain 49—50, a finding by a United Nations working group that his detention was arbitrary under the Declaration, the Covenant, and customary international law. See Report of the United Nations Working Group on Arbitrary Detention, U. N. Doc. E/CN.4/1994/27, pp. 139—140 (Dec. 17, 1993). That finding is not addressed, however, to our demanding standard of definition, which must be met to raise even the possibility of a private cause of action. If Alvarez wishes to seek compensation on the basis of the working group’s finding, he must address his request to Congress.