BEATTIE v. UNITED STATES

Opinion for the Court filed by Senior Circuit Judge WILKEY.

Opinion dissenting filed by Circuit Judge SCALIA.

Opinion concurring filed by Circuit Judge WALD.

WILKEY, Senior Circuit Judge:

This case arises out of the crash of an Air New Zealand aircraft into Mount Erebus, Antarctica, on 28 November 1979. All persons on board were killed. On 12 January 1983 plaintiffs filed a complaint in the United States District Court for the District of Columbia, seeking recovery against the United States for wrongful death under the Federal Tort Claims Act (FTCA).1 Plaintiffs' amended complaint ultimately alleged negligence of United States Navy Air Traffic Controllers at McMurdo Naval Air Station, Antarctica, as well as negligence in the selection, training, and supervision of the navy personnel at McMurdo Base by officials of the Department of Defense.

The United States filed a motion to dismiss under Rule 12(b)(1), (3), and (6), claiming that the District Court lacked subject matter jurisdiction, that plaintiffs had failed to state a cause of action upon which relief could be granted, and that venue was improper. The primary basis for the motion was the "foreign country" exception to the Federal Tort Claims Act.2 This exception removes from the scope of the FTCA "[a]ny claim arising in a foreign country."3

The issue before the District Court was one of first impression: Is Antarctica, a continent which is not now subject to the sovereignty of any nation, a "foreign country" within the meaning of the FTCA? By interlocutory order on 25 June 1984 the District Court denied the government's motion to dismiss, 592 F.Supp. 780, and certified this case for consideration by this Court in conformity with 28 U.S.C. § 1292(b). The government sought permission to appeal, which we granted.

To resolve the question before us, we must deal with three broad issues. The first issue is whether the District Court has subject matter jurisdiction. This issue hinges on a determination of whether Antarctica is a foreign country within the meaning of the FTCA. The second issue is whether the venue rules of the FTCA have been satisfied. The final issue involves a determination of which forum's law to apply. Our analysis of these issues leads us to affirm the interlocutory order of the District Court.

I. SUBJECT MATTER JURISDICTION

A. The Exception of Section 2680(k)

The FTCA acts as a waiver of sovereign immunity in specified types of cases. Section 2680 of the FTCA lists several exceptions to that waiver. One of those retentions of sovereign immunity is involved here: section 2680(k), which withholds FTCA jurisdiction from "[a]ny claim arising in a foreign country."4 As previously noted, the question of whether the District Court has subject matter jurisdiction depends on whether Antarctica is a foreign country within the meaning of the FTCA.

1. The Nature of Antarctica

Antarctica can properly be characterized as something of an international anomaly. It is a large continent which has never been and is not now subject to the sovereignty of any nation. Under the Antarctica Treaty of 1959 the signatory nations agreed not to exercise sovereignty in Antarctica, although their claims to sovereignty were not extinguished.5

The United States currently operates four active year-round stations, several summer camps, and numerous temporary tent cities in Antarctica.6 McMurdo Base is America's largest station, with a summer population in excess of 850 persons and a winter population of about 92. It consists of approximately 130 buildings. McMurdo Station has been assigned a zip code by the United States Postal Service.7

McMurdo Station has an airfield which supports frequent flights to and from New Zealand during the Antarctic summer. The airfield has two air traffic control facilities. One is Ice Tower; it is located adjacent to the runway, and has radio communication capability with incoming aircraft. The second facility is Mac Center, which has both radio capability to communicate with aircraft and radar capability to locate aircraft via radar returns.8

These United States activities are not set forth to demonstrate that, by virtue of extensive involvement, the United States can bring some distant land within the scope of United States sovereignty. These activities are relevant only to a fairly narrow and straightforward issue — is Antarctica a foreign country? The answer to this question is determined in part by answering the question of whether the United States treats this admittedly sovereignless land like a foreign country. The answer is that it does not.

During the pendency of the Antarctica Treaty the United States has consistently reaffirmed its position regarding Antarctica. In 1981, for example, Assistant Secretary of State James L. Malone reiterated that, while the United States does not recognize territorial sovereignty in Antarctica, it maintains its own basis to claims of sovereignty in Antarctica.9

Based on the foregoing information, and on a common sense approach to the plain language of the statute, it would appear obvious that Antarctica is not a foreign country within any ordinary meaning of that term. That sort of "plain meaning" approach formed part of the basis of the District Court's decision. As the District Court explained:

Reference to the legislative history and relevant case law illustrates that Congress did not intend the term "foreign country" to extend beyond its ordinary meaning.

2. The Legislative History

The FTCA was the product of many years of congressional drafting and redrafting. A variety of amendments were proposed to the original legislation, including several different ways to structure the foreign country exception. A look at some of the rejected language highlights the meaning which should be given to the version which was eventually passed. In 1940 language was proposed to the foreign country exception which would have delineated the geographical jurisdiction of the FTCA to approximately the area the government now contends is covered, and by a positive inclusion instead of the negative exclusion we now have. The suggested language read:

Obviously, this language, if it had been accepted, would have limited the geographic jurisdictional scope of the FTCA to what the government now says it is. In fact, even if such language had never been suggested in the legislative history, the government would be hard put now to explain why, if the intent was to confine the FTCA solely to the geographical United States, the statute did not use straightforward language such as, "the applicability of the FTCA is limited to United States territory," or, "limited to claims arising in this country." However, no such language was accepted, and the previous "foreign country" version was retained.12 We find it persuasive that Congress did not place a strict geographical limit on the scope of the FTCA, preferring simply to make an exception in the case of foreign countries.

One instructive item of legislative history was highlighted in the only Supreme Court case to interpret this section of the FTCA, United States v. Spelar.13 The Court quoted a pertinent colloquy between Assistant Attorney General Francis M. Shea, testifying to the House Committee on the Judiciary, and Congressman Robson of that Committee.

Although Mr. Shea did testify that it would be "wise to restrict the bill to claims arising in this country," the entire basis for that statement was his premise that liability is to be determined by the law of the situs of the wrongful act or omission.15

It should be noted that in examining legislative history, one is justified in placing greater reliance on the exact choice of words in the bill as enacted than in oral testimony to a committee hearing. Mr. Shea's words must be given their general intent, which is that liability based on the law of a foreign country was to be avoided.

Although the legislative history does not point decisively to any answer, the weight of the evidence is in favor of the concept that Congress did not intend to limit the application of the FTCA to the United States and its territories and possessions. It had the opportunity to do so and chose to retain only the "foreign country" limitation. Rather, the legislative will seems to be as the Supreme Court summarized it in Spelar, that "though Congress was ready to lay aside a great portion of the sovereign's ancient and unquestioned immunity from suit, it was unwilling to subject the United States to liabilities depending upon the laws of a foreign power."16

3. Cases Interpreting Section 2680(k)

The only Supreme Court case to interpret § 2680(k) is United States v. Spelar.17 That case involved a suit for wrongful death arising out of the death of a flight engineer at a United States Air Base in Newfoundland, leased by the United States from Great Britain. The Supreme Court determined that even though the accident occurred on a United States Base, that the negligent acts complained of nevertheless arose in a foreign country. The law which would have been applied to the case was the wrongful death law of Newfoundland. The holding of Spelar is based on the principle that § 2680(k)'s purpose is to prevent the United States from being subject to "liabilities depending upon the laws of a foreign power,"18 and that the mere presence of United States government activities in a foreign country will not convert that portion of the foreign country into something else.

Both parties in this proceeding cite a wealth of lower federal court cases relating to the issue of jurisdiction. Since the issue in this case is one of first impression, none of the cases can be said to be truly on point. However, these cases are instructive and stand for several helpful propositions.

B. Operative Effect Cases

Several cases have determined that a claim arises where the act or omission complained of occurs.19 Thus, a large body of the cases cited in the briefs illustrate that an FTCA claim may arise in the United States, because the negligent act or omission occurred here, even though the act or omission had its "operative effect," i.e., the injury occurred, in a foreign country.20

For example, in the case of Sami v. United States21 an Afghanistani economist was arrested by German police in Germany, at the request of the United States representative for Interpol. Mr. Sami had fled to Germany with his children as a result of a child custody dispute. He sued the United States for false arrest after the State Department told Germany that the offense was not extraditable.22

The District Court dismissed the case as one arising in a foreign country, because the last event necessary to liability had occurred in Germany. This Court responded that the FTCA focuses on the place where the negligent or wrongful act or omission of the government employee occurred. Since the instructions to make the arrest and most of the other operative facts in Sami occurred in the District of Columbia, this Court determined that § 2680(k) was not a bar, because the case actually arose in the United States.

These operative effect cases relate not so much to the definition of "foreign country," but to the meaning of "arising in." They determine that "arising in" does not necessarily refer to the situs of the injury, but to the situs of the negligence. The operative effect cases support subject matter jurisdiction over a portion of the plaintiffs' claims in this case, those which have been characterized as the "headquarters claims."23 The headquarters claims are the allegations of negligent selection, training, and supervision of the McMurdo Air Traffic Controllers by officials in Washington, D.C. Since these claims allege negligent acts or omissions by government employees which occur within the United States, and which merely had their operative effect in Antarctica, they are not claims which "arise in" a foreign country. However, the operative effect cases do not support subject matter jurisdiction over the claims of negligence by the air traffic controllers at McMurdo Base, Antarctica. These "Antarctica claims" constitute a separate ground for relief, with the negligent acts or omissions complained of occurring in Antarctica.

Thus, we accept for present purposes the validity of the headquarters claims, as we must on appeal of a Rule 12(b) motion.24 Doing so does not eviscerate the jurisdictional bar of the foreign country exception, as the government fears. For one thing, it has always been possible to allege headquarters claims in cases involving injury in a foreign land. But unless subject matter jurisdiction can be separately established for those claims truly arising in a foreign country, the parties would be left with the headquarters claims as their entire case. Since, with the exception of international anomalies such as Antarctica, those claims would be barred by § 2680(k), it is seldom worth their while to try to make a case live or die on the basis of headquarters claims standing alone.

C. The Foreign Country Cases

A variety of cases have been cited which merely decide that, despite outward appearances, the situs of the acts or omissions was a foreign country. For example, torts occurring on American embassies or military bases which are located in foreign countries are barred by the foreign country exception.25 This is also related to the idea that embassies and military bases in foreign countries often operate under agreement to apply that foreign country's laws in cases occurring there.26

A variety of other cases make subtle distinctions as to the level of sovereignty existing in some foreign land. For example, in Burna v. United States,27 an FTCA case in post World War II Japan, the court concluded that, while the applicable treaty conferred some authority over Okinawa to the United States, Okinawa was still a foreign country within the meaning of the FTCA. This and other vestiges of sovereignty in Japan led the court to conclude that the foreign country exception applied. In Cobb v. United States,28 the court's extensive analysis led it to the conclusion that Japanese tort law would be applicable. The court determined that Okinawa was in a foreign country because of the fact that the United States Military Government was not free to alter the tort law of Okinawa at will but was bound to maintain the preexisting "foreign law." As the court concluded, "[s]ince Congress was unwilling to subject the United States to liability based on that sort of law, the action was properly dismissed."29 Since Antarctica, by agreement, is not now and never has been subject to the sovereignty of any nation, the present case does not require such an analysis.

From this wealth of cases, plaintiffs have selected one case as that "most closely analogous to the present litigation,"30 the case of In Re "Agent Orange" Product Liability Litigation.31 In Agent Orange, Vietnam veterans, their spouses and children brought suit under the FTCA for alleged injuries incurred from exposure of the servicemen to the herbicide Agent Orange while they were stationed in Vietnam. The claims brought by the servicemen were dismissed pursuant to the doctrine enunciated in Feres v. United States.32 However, the independent claims of the wives and children were permitted to proceed. The government put forth several grounds for dismissal, one of which was the foreign country exception, since the alleged injuries occurred in Vietnam. In examining the foreign country exception, Judge Weinstein, "[a]t the peril of restating the obvious," stated that "the purpose of the exception, then, is to ensure that `the United States [is not subject] to liabilities dependent upon the laws of a foreign power.'"33 He went on to say that

Because of the procedural posture of the case, the judge noted that "[i]t is unclear at this point if the decisions relating to that misuse took place in the United States or Vietnam.... There is no reason to attribute those mistakes to Vietnam rather than to the United States and no policy reason to apply the `foreign claim' exception."35

The exact reason for the court's decision that the foreign country exception did not bar the case is not clear from the language of the opinion. The court appears to have been influenced by cases such as Sami,36 an operative effect case, and yet it was also clearly influenced by the apparent lack of sovereignty in South Vietnam and Cambodia. Agent Orange, like the present case, involved an undetermined mix of acts or omissions, some occurring within the United States and others in a distant land, all contributing to the harm complained of. Agent Orange supports the proposition that § 2680(k) is not a bar to jurisdiction over cases arising at least in part outside the United States, and in areas where "there is no theoretical justification for application of foreign law."37

For all of the above reasons, we conclude that the District Court has jurisdiction over this case. Congress, in drafting and passing section 2680(k), only intended to exclude foreign countries from the scope of the FTCA.

D. Analogous Statutes and Cases

Plaintiff also lists a potpourri of cases and statutes which have decided, in other contexts, that Antarctica was not a foreign country or a foreign state. By themselves, none of these cases would be persuasive, since they all involve interpretations of other statutes with other purposes. "Foreign country" is a fluid concept and Congress is capable of defining it very differently for different reasons. In addition, waiver of sovereign immunity is not to be inferred lightly. But these cases have a cumulative effect which is persuasive. They demonstrate that across a broad scheme of regulations Congress and the courts have consistently determined that Antarctica does not satisfy any definition of "foreign country."

For example, in Larry R. Martin v. Commissioner38 the Tax Court held that Antarctica was not a foreign country within the meaning of income tax regulations39 which define "foreign country" as follows:

The Tariff Act of 1930,40 the Interstate Transportation of Wagering Paraphernalia Act,41 and the State Conducted Lotteries Act42 each defined "foreign country" as "any empire, country, dominion, colony, or protectorate, or any subdivision or subdivisions thereof (other than the United States and its possessions)." The International Flight Information Manual, published by the Federal Aviation Administration, does not list Antarctica in its designations of "foreign countries" in international aviation.43 As previously noted,44 the McMurdo Station has been assigned a zip code by the United States Postal Service. United States dollars are the currency of exchange at McMurdo Base, and to obtain dollars to spend at McMurdo, New Zealanders must apply for permits to export money from New Zealand.45

As Judge Greene noted, "to the extent that there is any assertion of governmental authority in Antarctica, it appears to be predominantly that of the United States. The United States conducts all search and rescue operations in Antarctica and, significantly, it controls all air transportation."46

E. The Analogy to Outer Space

The legal status of Antarctica has been most frequently analogized to outer space.47 United States spokesmen suggested the 1959 Antarctic Treaty as a possible model for an outer space treaty during initial formulation discussions in 1965 and 1966.48 Obviously, the provisions of a treaty relating to outer space are only relevant to the present case by analogy. However, they are instructive as to the way in which the United States has acted with reference to sovereign immunity and liability for acts of its agents in a context very similar to Antarctica.

The Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, Including the Moon and Other Celestial Bodies49 [the Space Treaty], was signed at Washington, London, and Moscow on 27 January 1967, and entered into force on 10 October 1967. By the terms of the treaty, the United States has agreed to be internationally liable for its space objects and retain jurisdiction over its own objects and persons.

For example, Article VII of the Space Treaty provides:

In addition, Article VIII of the Space Treaty provides

The Space Treaty is obviously not couched in terms of tort claims. However, the basic principle is that in the sovereignless reaches of outer space, each state party to the treaty will retain jurisdiction over its own objects and persons. Like the decisions under the statutes discussed in section D above, holding that Antarctica is not a "foreign country" for various purposes, the treatment of outer space is persuasive by analogy.

II. VENUE

Venue of cases under the FTCA is governed by the provisions of 28 U.S.C. § 1402, which provides that any such "civil action on a tort claim against the United States ... may be prosecuted only in the judicial district where the plaintiff resides or wherein the act or omission complained of occurred."52 Since the plaintiffs in this case are all residents of Great Britain or New Zealand, venue must be satisfied through the place of the act or omission. This would be true even if the plaintiffs were American citizens resident in New Zealand, which brings into stark relief the state of utter lawlessness which would be created in Antarctica by adoption of the government theories.

Plaintiffs have posited two separate grounds for relief, which may be characterized as the headquarters claims and the Antarctica claims. The headquarters claims alleged negligence by officers of the United States, occurring in Washington, D.C., and in the Pentagon. On their face, the headquarters claims satisfy the venue requirements of section 1402(b). Of course, we must analyze these claims under the settled rule for assessing the propriety of dismissal under Rule 12(b) of the Federal Rules of Civil Procedure. Many potentially dispositive facts are intensely disputed by the parties. Because there has been neither factfinding by the district court nor very much if any discovery by the parties, we must accept as true all of the material allegations in the plaintiffs' complaint. Dismissal for failure to state a claim for relief is proper only when it "appears beyond doubt that the plaintiff can prove no set of facts in support of his claim which would entitle him to relief."53 All factual doubts must be resolved and all inferences made in favor of the plaintiffs.54 On that basis, it is clear that the headquarters claims satisfy the requirements of section 1402(b) by complaining of acts or omissions within the jurisdiction of the District Court for the District of Columbia.

This leaves us with the claims involving negligence in Antarctica. These claims satisfy the venue requirements under two discrete but interrelated theories.

The general rule is that venue must be established as to each separate cause of action.55 However, as one commentator has explained, the focus is on the word separate.56 Professor Moore notes that, whether the case involves a federal and nonfederal claim,57 or two federal claims,58 if they amount to only one cause of action with two grounds for relief, proper venue as to one federal ground will support adjudication of both grounds. His theory is that "there is no true problem of joinder of actions unless there be two or more causes of action; that a broad concept of cause of action eliminates or lessens the problem."59 He goes on to apply to the venue context the broad definition of cause of action from Hurn v. Oursler.60 This notion of venue being required only for separate causes of action is still used,61 even though Hurn's concept of cause of action has been replaced with respect to subject matter jurisdiction by the doctrine of pendent jurisdiction announced in United Mine Workers v. Gibbs.62

Applying that principle to the present case, it is clear that this litigation can accurately be described as a single cause of action with separate grounds for relief. Plaintiffs seek damages for "an essentially single wrong,"63 i.e., wrongful death. They allege against the same defendant two separate grounds for relief. These two grounds are identical as to parties and very similar as to the proof. There will be great commonality as to witnesses and evidence. Recognizing that there is but a single cause of action, venue under § 1402(b) is satisfied for the entire case by virtue of the headquarters claims.

In the context of subject matter jurisdiction, the concept of "cause of action" has been replaced as the touchstone for analysis by the doctrine of pendent jurisdiction. This familiar concept has also spilled over into the rules of venue. The doctrine of "pendent venue" is now well established, particularly in cases where the court has previously exercised its discretion to hear a certain claim under pendent jurisdiction. As one commentator noted, "[i]t would seem that if procedural convenience is enough to avoid the constitutional limitations on the jurisdiction of the federal court, it should suffice also to dispense with the purely statutory requirements as to venue."64 Consequently, many pendent jurisdiction cases also apply the principles of pendent jurisdiction by analogy when one or more claims arising out of a common nucleus of operative facts do not satisfy the requirements of the applicable venue statute.65

This has even been carried over into cases where subject matter jurisdiction was independently satisfied on each claim, although with a more limited application.66 This extension was foreseen by the authors of another treatise, in which they explain the rule that venue must be proper for each claim.67 They cite a Supreme Court case involving both patent and breach of contract claims.68 Jurisdiction was proper on all claims due to the complete diversity of the parties, but venue was only proper as to the patent claim. The other claims had to be dismissed. The authors note that the case was decided long before the concept of pendent jurisdiction was developed, and that the result should be different today.69

It would seem that there is no practical reason for limiting the application of pendent venue to cases where pendent jurisdiction has also been applied, and judges have recognized this by utilizing pendent venue when the case has merited it.

For example, in Laffey v. Northwest Airlines70 the District Court for the District of Columbia found that it "need not decide whether [venue] is proper" for the cause of action under the Civil Rights Act of 196471 since venue was clearly proper under the Equal Pay Act claim.72 The Equal Pay Act claim involved allegations that Northwest Airlines paid female in-flight cabin attendants less than male in-flight cabin attendants, and sought past-due and future wages. The Civil Rights Act claim involved allegations that Northwest discriminated in its employment practices against its female in-flight cabin attendants in that it used a discriminatory employment classification scheme and employment conditions including male priority in the "chain of command" and unequal treatment in regard to weight and height requirements, eyeglasses, uniform cleaning, and hotel accommodations.73 Thus, Laffey can be aptly characterized as two separate causes of action arising out of a common nucleus of operative facts. This Court has subsequently cited Laffey, with respect to the general rule that venue must be satisfied as to each cause of action, for the notion that such a rule may be undesirable "when parties and proofs for all causes of action are identical."74

These same principles were applied in the case of Zenith Radio Corp. v. Matsushita Electric Industrial Co.75 That case involved, among other things, allegations under the Anti-Dumping Act of 191676 and federal antitrust law.77 The court addressed defendant's argument that even if venue were proper under § 12 of the Clayton Act, it was not proper for the counts arising under the Anti-Dumping Act. As the court explained:

In the case of Seamon v. Upham,79 which originally was a Voting Rights Act case charging that the Texas apportionment plan in various ways diluted minority voting strength, plaintiff sought to file a supplemental complaint, adding under Federal Rule of Civil Procedure 18(a) a claim challenging the legality and constitutionality of the Texas Governor's proclamation which set a date for a special congressional election. This supplemental complaint clearly lacked venue. The court, in its examination of the doctrine of pendent venue, noted that a court exercising its discretion in a pendent venue case "should consider the same factors that it would consider in deciding whether to exercise pendent jurisdiction.... Thus it should consider whether the claims derive from a `common nucleus of operative fact,' ... and whether the exercise of jurisdiction furthers the goals of judicial economy, convenience, and fairness to the litigants."80 The court also noted that

The court went on to decline to exercise pendent venue because it determined that the supplemental complaint arose out of a completely discrete set of circumstances.82

Goggi v. Outboard Marine Corp.83 was a case where plaintiff sought to apply the concept of pendent venue to a patent claim. The court explained that, at least with regard to the special patent venue statute,84 this area of "pendent venue" has received limited application.85 The court in Goggi distinguished those cases which did apply the concept of pendent venue by noting that "the sense of these opinions was that where there is substantial identity of issues and proof, it is most economical to try the various claims at one time."86 Certainly the present case bears the same identity of parties and proofs found in Laffey and other pendent venue cases.

Whether to apply the principle of pendent venue in any given case is a discretionary decision, based on applicable policy considerations. Some of these considerations will be the same as those that support the exercise of pendent jurisdiction — judicial economy, convenience, avoidance of piecemeal litigation, and fairness to the litigants. Other considerations unique to the context of venue will apply.87 For example, the purpose of venue rules is generally considered to be "primarily a matter of convenience of litigants and witnesses."88 It is also oriented to the convenience of the court system.89

Given that all parties to this lawsuit will be properly before the district court on the headquarters claims, it does not seem too inconvenient to include the Antarctica claims in the same lawsuit. Indeed, even if the Antactica claims were sued on alone, the District of Columbia might well be the most convenient forum, since many of the witnesses will be either here in Washington, D.C. or scattered around the globe.90 Records relating to the accident investigation of the Air New Zealand flight are located in Washington, D.C.91 Much of the other evidence in the case will be either in the District of Columbia or at the Pentagon.92

Hearing the Antarctica claims also allows the district court to act in harmony with congressional intent and a rule of construction adopted by the Supreme Court, both of which are explained in Brunette Machine Works v. Kockum Industries.93 The court noted that there had been and perhaps still were occasional cases in which the federal courts have jurisdiction but there is no district in which venue is proper, but that Congress had acted to close such gaps in 1966 by amending the general venue statute. The court noted that "the development supports the view that Congress does not in general intend to create venue gaps, which take away with one hand what Congress has given by way of jurisdictional ground with the other. Thus, in construing venue statutes it is reasonable to prefer the construction that avoids having such a gap."94 In the present case, dismissal of the Antarctica claims for lack of venue creates a gap between jurisdiction and venue and works a hardship not usually found in dismissals for lack of venue, which are often done with an eye toward another court in which subject matter jurisdiction and venue can both be established. No one contends here that venue lies elsewhere. The government, consistent with its position that there is no law in Antarctica, maintains that venue exists — nowhere.

Whether viewed as a single cause of action with two grounds for relief (two claims), or as an appropriate case to apply the principle of "pendent venue" to the headquarters and Antarctica claims, this case satisfies the venue requirements of § 1402(b).

III. CHOICE OF LAW

Choice of law issues in FTCA actions are governed by 28 U.S.C. § 1346(b), which states that the district courts "shall have exclusive jurisdiction of civil actions on claims against the United States ... 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."95 This section has been construed to mean that the district court should apply as a choice of law rule the whole law of the place where the act or omission occurred.96

In this wrongful death case, some undetermined portion of the acts or omissions complained of occurred in the District of Columbia. Thus, at least as to the headquarters claims, and probably as to the entire case, application of D.C. law would mean that this "civil action" would be tried "in accordance with the law of the place where the act or omission occurred."

The Antarctica claims presents another novel issue. By directing application of the law of the place where the act or omission occurred, § 1346 leads the court to a place where this is no civil tort law to apply.97 As we have seen, this is a situation directly opposite to the fear of becoming entangled in foreign law, which produced the "foreign country" exception. Hence, for this and other reasons, contrary to the government's assertions, we do not feel that the logical conclusion to be derived therefrom is that the entire case ought to be dismissed.98 Rather, we feel that this represents a question not answered by the statute, and turn to those factors relevant to the choice of the applicable rule of law when there is no such statutory directive. Those factors, outlined in Restatement 2d, Conflict of Laws § 6, point to the District of Columbia as the forum whose law should be applied in this case.99

This is not a case such as Sami,100 where a foreign state's substantive law is supplanted by the law of the forum because it conflicts with a strong public policy of the forum.101 Rather, by traditional analysis, among the two "forums" with the greatest interest in the outcome of this litigation — Antarctica and the United States — only one has any civil law to apply. And, in fact, Antarctica has no "forum" either; fundamentally, because Antarctica is not a "country," foreign or otherwise.

These considerations are similar to those which influenced the court's decision in Agent Orange.102 In that case the erratic ebb and flow of sovereignty in Southeast Asia led the court to conclude that "there is no theoretical justification for application of foreign law."103

These conclusions are buttressed by the recognized principle of international law that a nation may exercise jurisdiction over its nationals as a legitimate exercise of the nationality principle. The United States recognizes the validity of the nationality principle.104

The United States has a strong public policy interest in the outcome of this litigation. In addition, the District Court for the District of Columbia has a strong interest in bringing to a resolution a case which has been properly filed in that court in which it has both subject matter jurisdiction and venue. In contrast to these and other strong interests of the District of Columbia forum, there is no foreign sovereign even to assert a countervailing interest in Antarctica. The proper law to be applied to the case would be District of Columbia law.

In summary, the government argument requires us to accept an analysis of the FTCA, section 2680(k), that there are two areas of the world only — first, the United States, and second, "foreign countries." There are obviously several other areas in which people operate, e.g., outer space, the high seas, and Antarctica. While the other non-United States/non-foreign country areas may be covered by some law, we have a no-man's land of law in Antarctica, unless United States law covers the actions of United States citizens — not an unfair concept — and United States law includes the Federal Tort Claims Act.

All of this attempted limitation of coverage rests on one indefensible concept — that Antarctica is a "foreign country." Such an interpretation does violence to the plain meaning of the statute and the purpose behind the "foreign country" exception. While there are theoretical procedural obstacles, these can be logically overcome once the basic concept that United States law applies at least to the actions of United States citizens in Antarctica is accepted, and the other side of the coin is visualized, i.e., that unless this concept is accepted, Antarctica is an area without any law whatsoever.

The decision of the District Court is Affirmed and the case is Remanded for further proceedings consistent with this opinion.

So Ordered.

SCALIA, Circuit Judge, dissenting:

It is conceivable (though barely) that a court might have to conclude that a particularly ill-drafted statute created federal causes of action over which no federal court had venue, and directed the courts to apply the law of a place that has no law. What makes the majority opinion in this case unique is that it does not reach these strange conclusions by sheer compulsion, but adopts them as the preferred interpretation of a statute that could readily be construed otherwise. One is at a loss to explain the result except as a manifestation of the territorial imperative that impels this court to extend its writ to foreign lands when the United States or its officers are sued. Cf. Ramirez v. Weinberger, 745 F.2d 1500 (D.C.Cir.1984) (en banc). I dissent because in my view it is clear beyond doubt that the Federal Tort Claims Act does not apply to a claim arising in Antarctica; and because I think that this case presents only a single claim, which arose there.

I. BACKGROUND

Although, as subsequent discussion will show, I do not share the majority's assessment of the importance of the legal and operational details of the United States' presence in Antarctica (other than the undisputed fact that the United States asserts no legislative jurisdiction there), it is nonetheless of some value to be clear on what we are talking about. It is not accurate to characterize Antarctica as a "large continent which has never been and is not now subject to the sovereignty of any nation." Maj. op. at 93. It is more properly described, in the words of one leading book on the area, as "an entire continent of disputed territory." F. AUBURN, ANTARCTIC LAW AND POLITICS 1 (1982) (emphasis added) [hereinafter cited as F. AUBURN, ANTARCTICA]. Seven countries — Argentina, Australia, Chile, France, New Zealand, Norway and the United Kingdom — have made formal claims (some of which are conflicting) to portions of the continent that total about 85 percent of its expanse. Boczek, The Soviet Union and the Antarctic Regime, 78 AM. J. INT'L L. 834, 840 (1984). Other nations, including the United States and the Soviet Union, have substantial bases for claims which they have not yet chosen to assert. See id. at 841; F. AUBURN, ANTARCTICA, at 48-83.

It is also wrong to conclude that "[u]nder the Antarctica Treaty of 1959 the signatory nations agreed not to exercise sovereignty in Antarctica," Maj. op. at 93. Article IV(1) of that treaty merely provides that the treaty does not prejudice any nation's claim or basis for a claim to Antarctica; and Article IV(2) merely provides that while the treaty is in force no activities shall constitute a basis for asserting a claim in Antarctica and no new claim shall be asserted.1 Claims asserted prior to the effective date of the treaty, or not yet asserted but based upon pre-treaty activities, are unaffected, as is the right of all nations to implement past-asserted claims by the exercise of sovereignty. See The Antarctic Treaty: Hearings Before the Senate Comm. on Foreign Relations, 86th Cong., 2d Sess. 39, 61-62 (1960) (testimony of Herman Phleger of the Department of State, Head of the United States Delegation to the Antarctic Conference) [hereinafter cited as Treaty Hearings]. Thus, Article VIII of the treaty, which provides for exclusive jurisdiction of each contracting party over its nationals who are designated observers and scientific personnel, specifically states that this provision is "without prejudice to the respective positions of the Contracting Parties relating to jurisdiction over all other persons in Antarctica," and further provides that "the Contracting Parties concerned in any case of dispute with regard to the exercise of jurisdiction in Antarctica shall immediately consult together with a view to reaching a mutually acceptable solution." 12 U.S.T. at 797-98. All seven countries claiming territory in Antarctica have "either specifically enacted legislation governing criminal conduct in Antarctica or appear to consider their domestic criminal legislation applicable to areas they claim." Bilder, Control of Criminal Conduct in Antarctica, 52 VA.L.REV. 231, 260 (1966).

McMurdo Station is located on Ross Island, as is Mount Erebus, into which Flight 901 crashed. The island forms part of an area known as "the Ross Dependency," encompassing land from 160° East longitude to 150° West longitude, over which New Zealand has a longstanding claim based upon a 1923 Order in Council by the United Kingdom to the Governor of New Zealand. Joyner, Antarctica and the Law of the Sea: Rethinking the Current Legal Dilemmas, 18 SAN DIEGO L.REV. 415, 417 n. 8 (1981); F. AUBURN, THE ROSS DEPENDENCY 5 (1972) [hereinafter cited as F. AUBURN, ROSS DEPENDENCY]. See generally Richardson, New Zealand's Claims in the Antarctic, 33 NEW ZEALAND L.J. 38 (1957). New Zealand's claim is recognized by Australia and the United Kingdom, F. AUBURN, ANTARCTICA, at 29,2 and according to one commentator by Argentina, Chile, France and Norway as well, P. QUIGG, A. POLE APART: THE EMERGING ISSUE OF ANTARCTICA 113 (1983). Also, if "sector" claims3 are recognized in Antarctica, New Zealand might be able to assert such a claim over the area at issue here as the country which, longitudinally, is immediately above it. New Zealand has explicitly extended its domestic criminal law to the Ross Dependency, New Zealand Antarctic Act of 1960, reprinted in 11 POLAR RECORD 303 (1962), and by exchange of notes has granted the United States air and naval facilities located there exemption from taxation, customs duties, radio station licensing, and criminal jurisdiction in minor cases, see Agreement on Operations in Antarctica, Dec. 24, 1958, United States-New Zealand, 9 U.S.T. 1502, T.I.A.S. No. 4151, renewed indefinitely, Oct. 18, 1960, 11 U.S.T. 2205, T.I.A.S. No. 4591.

The United States refuses to recognize territorial sovereignty in Antarctica; at the same time, however, it maintains a basis for asserting claims of its own. Law of the Sea Negotiations: Hearings Before the Subcomm. on Arms Control, Oceans, International Operations and Environment of the Senate Comm. on Foreign Relations, 97th Cong., 1st Sess. 24-25 (1981) (statement of Assistant Secretary of State James L. Malone). It is not clear how much basis for a claim it has to the area in which McMurdo Station is located. See generally F. AUBURN, ANTARCTICA, at 66-74.4 Of course the determination by the Executive Branch not to recognize claims of territorial sovereignty in Antarctica is binding on this Court, see United States v. Pink, 315 U.S. 203, 229-30, 62 S.Ct. 552, 565-66, 86 L.Ed. 796 (1942), but we need not blind ourselves to the internationally disputed character of the land in question.

With regard to operational as opposed to legal considerations, the majority opinion again exaggerates the position of the United States in Antarctica. It quotes the District Court's statement that "`[t]he United States conducts all search and rescue operations in Antarctica and, significantly, it controls all air transportation.'" Maj. op. at 99 (quoting Beattie v. United States, 592 F.Supp. 780, 783 (D.D.C.1984) (footnotes omitted)). One must be dubious about an assertion that the United States could so dominate an area containing roughly one-tenth of the world's land mass, at the bottom of the opposite hemisphere, in which numerous other countries conduct extensive operations. The majority offers as authority for this remarkable assertion only the District Court, which in turn offered only the following footnote:

592 F.Supp. at 783 n. 21. Professor Auburn was clearly not advancing the proposition that the United States "controls all air transportation" in Antarctica. Indeed, he was not even implying such control in the small portion of Antarctica known as the Ross Dependency. Rather, he was exemplifying the point that New Zealand relies upon the United States for logistical support of its operations in the Ross Dependency. That does not even suggest that the United States is the only nation with flight capacity in that limited region. In fact, both New Zealand and Australia have flown regular support operations for the United States to McMurdo Base. See F. AUBURN, ANTARCTICA, at 68; U.S. Antarctic Program: Hearings Before the Subcomm. on Science, Research, and Technology of the House Comm. on Science and Technology, 96th Cong., 1st Sess. 80 (1979). Moreover, it is clear that the United States does not "control all air transportation" in the sense of granting and denying permission for, and directing the path of, flights by other states. "Tourist flights over Antarctica depend on McMurdo Station for weather reports, but whether or not planes fly, on what route, and at what altitude are entirely at the discretion of their captains." P. QUIGG, supra, at 101; see also Brief for Appellees at 3 (emphasis added) (asserting only that McMurdo Station was "advised of schedules and the flight plans of the anticipated sightseeing tours").

The majority again relies upon the District Court for an even more inspiring, but also even more inaccurate, statement: "`to the extent that there is any assertion of governmental authority in Antarctica, it appears to be predominantly that of the United States.'" Maj. op. at 99 (quoting 592 F.Supp. at 783). In support of this conclusion, the District Court offered the statement on rescue operations and air transportation quoted above, and the following:

592 F.Supp. at 783 (footnotes omitted). Such formidable exercises of American power at McMurdo Station hardly carry the point. There are more than thirty stations in Antarctica, of which only four are American. See F. AUBURN, ANTARCTICA, at xvi-xvii, xx (maps). Moreover, numerous countries conduct Antarctic operations, and while the United States may have the largest program, those of other countries are significant. The Soviet Union alone has "developed an Antarctic program rivaling that of the United States." Note, Thaw in International Law? Rights in Antarctica under the Law of Common Spaces, 87 YALE L.J. 804, 814 (1978) (footnote omitted).

The majority notes — presumably to suggest that agencies of our government do not consider Antarctica "foreign" — that "[t]he International Flight Information Manual, published by the Federal Aviation Administration [FAA], does not list Antarctica in its designations of `foreign countries' in international aviation." Maj. op. at 99 (footnote omitted). The omission is utterly without significance, however, since the manual lists only airfields which are open to the public and which "business and private aviators" can fly to directly. See generally FAA, International Flight Information Manual, vol. 31 at 1 (Apr. 1983). Virtually all American military airfields overseas (e.g., McMurdo Naval Air Station, Antarctica) are not listed since special permission is required to use them. See id. at 8. The airfields of other nations in Antarctica apparently operate under similar restrictions. See F. AUBURN, ANTARCTICA, at 279; P. QUIGG, supra, at 103.

Again presumably as evidence that that distant land is not really "foreign," the majority thinks the assignment of a postal zip code to McMurdo Station (96692) important enough to bear mention twice in its opinion. See Maj. op. at 93, 99. The zip code listing for "McMurdo Station, Antarctica" is part of a list of military post offices outside the continental United States that includes, inter alia, Berlin, Germany (09742); Rio de Janeiro, Brazil (34030); and Zama, Japan (96343). McMurdo will be found immediately after the listing for Christchurch, New Zealand and just before the listing for Singapore. U.S. Postal Service, NATIONAL FIVE DIGIT ZIP CODE AND POST OFFICE DIRECTORY cxxv (1983). The instructions at the top of this list read: "A higher rate of postage will be charged if the foreign country is used in conjunction with the MPO's [military post office's] address." Id. at cxxiv. Thus, a letter sent to McMurdo Station, Antarctica 96692 (as opposed to simply McMurdo Station 96692), pays the higher "foreign country" mailing rate. In any case, I personally feel that as a display of sovereign epistolary power, the assignment of a zip code is topped by New Zealand's seizure of the philatelic monopoly; it has issued Ross Dependency postage stamps, in more than one denomination, see F. AUBURN, ROSS DEPENDENCY, at 61.

With this clarification of background, I proceed to discussion of the substantive issues in this case.

II. ANTARCTICA IS A FOREIGN COUNTRY

The Statutory Text

The statutory language upon which this case turns is 28 U.S.C. § 2680(k) (1982), which provides that the Federal Tort Claims Act's (FTCA's) waiver of sovereign immunity does not apply to "[a]ny claim arising in a foreign country." The essence of appellees' position, accepted by the majority, is that the word "country" in this text means "sovereign state" — so that if a land such as Antarctica, subject (in the view of this government) to no national jurisdiction, is involved, the exception does not apply.

The first dictionary meaning of "country" is not "sovereign state" but simply "region," Webster's Ninth New Collegiate Dictionary 298 (1983), see also id. at 483 (first definition of "foreign" is "situated outside a place or country; esp. situated outside one's own country") — and it so happens that the term was used precisely in that customary sense in the congressional discussions of the Antarctica Treaty. See, e.g., Treaty Hearings, at 61 (statement of Sen. Aiken) ("If this agreement is approved, Antarctica becomes a country without a government"). It must be acknowledged, however, that the term can be used in the sense appellees would derive — when one speaks, for example, of the "democratic countries of the world."5 It is appropriate, then, to look for clarification elsewhere in the statute — which readily discloses that the former meaning must be intended, since otherwise two crucial portions of the legislative scheme make no sense and a third conflicts with the Antarctica Treaty.

The Venue Provision

Section 1402(b) of 28 U.S.C. (1982) provides:

Our recent decision in Reuber v. United States, 750 F.2d 1039 (D.C.Cir.1984), notes the importance of this provision:

Id. at 1048 & n. 11.

If, as the appellees assert, "foreign country" means only "foreign state," and sovereign immunity has therefore been waived with respect to torts occurring in stateless regions such as Antarctica, then the venue provision for those regions makes no sense, since the "only ... judicial district" in which suit has been allowed will usually not exist — i.e., will exist only if the plaintiff happens to reside in the United States.

The majority quotes Brunette Machine Works, Ltd. v. Kockum Industries, Inc., 406 U.S. 706, 710 n. 8, 92 S.Ct. 1936, 1939 n. 8, 32 L.Ed.2d 428 (1972), to the effect that "`Congress does not in general intend to create venue gaps which take away with one hand what Congress has given by way of jurisdictional grant with the other.'" Maj. op. at 104. Quite so. In Brunette the jurisdiction of the federal courts was clear, and the consequence of applying the enunciated principle was to give broad interpretation to an ambiguous venue provision — to avoid taking away what one hand clearly had given. Here, by contrast, the limited character of the venue provision is entirely clear, and the consequence of the principle is to give narrow interpretation to the ambiguous jurisdictional provision — to avoid giving what one hand has clearly taken away.

Nor can this gap in the venue provision plausibly be explained as a product of legislative oversight — a detail beyond the normal capacity or inclination of Congress to address. For in other legislation, including specifically legislation dealing with Antarctica, Congress has anticipated and prevented venue gaps by including catch-all provisions. In the Antarctic Conservation Act of 1978, 16 U.S.C. § 2404(e) (1982), it provided for review of decisions involving permits to take Antarctic plants or animals "in the United States district court wherein the applicant for a permit resides, or has his principal place of business, or in the United States District Court for the District of Columbia." In the Public Vessels Act, 46 U.S.C. § 782 (1982), it provided that "in case none of such parties reside or have an office for the transaction of business in the United States, and such vessel or cargo be outside the territorial waters of the United States, then [venue is proper] in any district court of the United States." In the Tucker Act, 28 U.S.C. § 2502 (1982), it provided that "[c]itizens or subjects of any foreign government which accords to citizens of the United States the right to prosecute claims against their government in its courts may sue the United States in the Claims Court if the subject matter of the suit is otherwise within such court's jurisdiction."

In short, the venue provision of the FTCA, which assumes the availability of a court only for those torts that occur within a federal judicial district, makes it clear beyond doubt that "foreign country" in the exclusion provision has its ordinary meaning of any region outside the United States.

The Choice-of-Law Provision

The choice-of-law provision of the FTCA reads in relevant part as follows:

28 U.S.C. § 1346(b) (1982) (emphasis added). The Supreme Court has made clear that this language means precisely what it says:

Richards v. United States, 369 U.S. 1, 9, 82 S.Ct. 585, 591, 7 L.Ed.2d 492 (1962) (footnote omitted); see also Sami v. United States, 617 F.2d 755, 761 (D.C.Cir.1979) ("The entire scheme of the FTCA focuses on the place where the negligent or wrongful act or omission of the government employee occurred"). The majority finds that "§ 1346(b) leads the court to a place where there is no civil tort law to apply." Maj. op. at 104-105 (footnote omitted). I think, to the contrary, that it leads the court to the same place the venue provision does — to the conclusion that the FTCA no more envisions suits for torts occurring in stateless foreign regions than it does suits for torts occurring in foreign sovereignties. If "foreign country" is given its usual meaning, there will always be "civil tort law to apply" just as there will always be a district court with venue.

The Antarctica Treaty

The interpretation of "foreign country" the majority has adopted not only sets various sections of the FTCA itself at war with one another, but also creates a clear conflict between that legislation and the Antarctica Treaty. In the treaty, the United States agreed not to assert for the time being United States territorial jurisdiction over Antarctica. (The exercise of such jurisdiction presupposes a claim; the United States has asserted no claim; and the treaty provides that during its effectiveness no new claims shall be asserted. See discussion at pages 106-107, supra; Bilder, supra, 52 VA.L.REV. at 270.) I agree with the majority that the United States may permit itself to be sued for its acts anywhere in the world, pursuant to the law of the place where those acts occurred, through the exercise of a basis of jurisdiction other than territorial sovereignty, analogous to "the recognized principle of international law that a nation may exercise jurisdiction over its nationals," Maj. op. at 105. The FTCA, however, does more than merely permit suit against the United States. Section 2679 of 28 U.S.C. (1982), entitled "Exclusiveness of remedy," and enacted subsequent to the treaty, provides as follows:

If the Act is interpreted as the majority suggests, this provision would bar a New Zealander from suing an employee of the United States (even if that employee happened to be a New Zealand national) for a tort committed by that employee in Antarctica; plaintiff's substantive rights would be limited to his suit against the United States in United States courts. There is no conceivable basis for this assertion of legislative jurisdiction except the principle of territorial sovereignty, which assertion conflicts with our treaty obligations. It does not matter, of course, that this exclusivity provision is limited to a type of tort not immediately at issue in this case (injury or damage caused by motor vehicles). The point is that the rationale of the majority's decision — holding Antarctica not to be a "foreign country" for purposes of the FTCA — renders this provision in violation of the treaty.

Since the treaty predates this provision, which was added to the FTCA in 1961, Pub.L. No. 87-258, § 1, 75 Stat. 539 (Sept. 21, 1961), the result of the conflict would be that our assertion of territorial jurisdiction in Antarctica would stand, despite its violation of treaty commitments. It is contrary to sound principles of statutory construction, however, to interpret an ambiguous provision (in this case the "foreign country" exception) in a fashion that produces such consequences.

Whitney v. Robertson, 124 U.S. 190, 194, 8 S.Ct. 456, 458, 31 L.Ed. 386 (1888). It should not be thought, incidentally, that the treaty provisions assure that our exercise of territorial jurisdiction will not be a matter of concern to other nations. See generally Comment, Quick, Before It Melts: Toward a Resolution of the Jurisdictional Morass in Antarctica, 10 CORNELL INT'L L.J. 173, 185-86 n. 57 (1976) (noting that jurisdictional assertions even during the pendency of the Treaty could be used to strengthen territorial claims). Of course it would be possible to say that this conflicting exclusive-remedy provision of the FTCA was simply not meant to have the same scope as the other provisions of the Act, and does not apply outside the territorial United States — just as the majority has found it possible to create venue and choice-of-law solutions out of whole cloth. But when these imaginative exercises can be avoided by giving the phrase "foreign country" its ordinary meaning, to perform them is to rewrite the statute rather than interpret it.

Accepted Principles of Construction

One would therefore logically conclude, even without the benefit of any special principles of statutory construction, that "foreign country" in the present text means any location outside the territorial jurisdiction of the United States. In fact, however, several such principles of construction are applicable, all of which reinforce the same conclusion and render it in my view inevitable.

First, there is "the `canon of construction which teaches that legislation of Congress, unless a contrary intent appears, is meant to apply only within the territorial jurisdiction of the United States....'" United States v. Spelar, 338 U.S. 217, 222, 70 S.Ct. 10, 13, 94 L.Ed. 3 (1949) (quoting Foley Brothers, Inc. v. Filardo, 336 U.S. 281, 285, 69 S.Ct. 575, 577, 93 L.Ed. 680 (1949)). See also Natural Resources Defense Council, Inc. v. NRC, 647 F.2d 1345, 1357 n. 54 (D.C.Cir.1981); RESTATEMENT (SECOND) OF FOREIGN RELATIONS LAW OF THE UNITED STATES § 38 (1965). In other words, even without the "foreign country" exclusion it would be improper to interpret the FTCA as applying to acts or events in Antarctica. It is perverse to give the explicit exclusion of foreign claims the consequence of expanding the Act. Rather, as the Supreme Court observed in Spelar, the presumption against extra-territorial application "is doubly fortified by the language of [the FTCA] and the legislative purpose underlying it." 338 U.S. at 222, 70 S.Ct. at 13.

Second, the Supreme Court has set forth a guide to the interpretation of the FTCA in particular. "[T]he effect of the Act," it has said, "was `to waive immunity from recognized causes of action and ... not to visit the Government with novel and unprecedented liabilities.'" Stencel Aero Engineering Corp. v. United States, 431 U.S. 666, 671, 97 S.Ct. 2054, 2058, 52 L.Ed.2d 665 (1977) (quoting Feres v. United States, 340 U.S. 135, 142, 71 S.Ct. 153, 157, 95 L.Ed. 152 (1950)).6 To my knowledge, not only has no state ever permitted itself to be sued for alleged torts committed in a stateless land such as Antarctica, but no modern state has ever entertained in its courts a tort suit between private parties who were not both its citizens relating to activities in such a land.7

Finally, even if applying the FTCA to Antarctica in the present case represents no more than an exercise of non-territorial jurisdiction, it violates a broad principle of statutory construction — of which the canon of presumed territorial application discussed earlier is merely one manifestation — which avoids giving an ambiguous provision content that is likely to cause what the Supreme Court has described in another context as "embarrassment to the Executive Branch in handling foreign relations," Banco Nacional de Cuba v. Sabbatino, 376 U.S. 398, 412, 84 S.Ct. 923, 932, 11 L.Ed.2d 804 (1964). Appellees' statement that "[t]he [Antarctic] Treaty itself acknowledges each treaty country's jurisdiction over its own nationals," Brief for Appellees at 27, is incorrect. The treaty does not speak to jurisdiction over anyone except certain designated scientists and observers. See Bilder, supra, 52 VA.L.REV. at 237-38. The failure to provide nationality jurisdiction resulted not from oversight but from the fears of several nations that it would undermine their claims. Id. at 238 n. 21; F. AUBURN, ANTARCTICA, at 184. The United States is undoubtedly free to apply that principle (or an analogous principle) in Antarctica, but I would leave that decision to the political branches rather than make it judicially, by interpreting an ambiguous statute to produce that result. Such a course is simply insensitive to the delicate diplomatic modus vivendi that our Executive has worked out among nations with claims or potential claims to this important part of the world.

Legislative History

The evidence of the statute itself is so overwhelmingly clear as to require no resort to the questionable assistance of legislative history. But even there the conclusion is only strengthened rather than called into doubt. The weakness of the majority's case on this score is demonstrated by what it chooses to discuss first: an amendment proposed in 1940 that would have specifically limited the scope of the FTCA to "damage or injury occurring within the geographical limits of the United States, Alaska, Hawaii, Puerto Rico [and] the Canal Zone." Tort Claims Against the United States: Hearings on S. 2690 Before a Subcomm. of the Sen. Comm. on the Judiciary, 76th Cong., 3d Sess. 65 (1940) [hereinafter cited as 1940 Senate Hearings]. The majority notes that "no such language was accepted, and the previous `foreign country' version was retained." Maj. op. at 95. What it neglects to note is that there is no reason to believe the "proposed amendment" was ever considered. It was in fact not even "proposed" in the ordinary sense of having been offered by a Member of Congress as a bill or an amendment to a bill. Rather it was nothing more than part of an alternative version of the bill offered to a Senate subcommittee for inclusion in the record by one J.J. Keegan, a member of the United States Employees' Compensation Commission. It was never even put to a vote of the subcommittee — and indeed subsequent hearings are devoid of any discussion of Keegan's proposals in general and of this proposed change in particular. Understandably so, since Keegan began his testimony in support of six pages of proposed stylistic and substantive changes with the following statement:

1940 Senate Hearings at 53. One does not have to be overly impressed with the virtues of a law degree to conclude that the subcommittee decided to leave what was quintessentially a legal matter (how to waive the long-standing sovereign immunity of the government) to "very able lawyers" and simply ignored the suggestions of Mr. Keegan. In addition, even if a subcommittee of Congress did consider this snippet of Mr. Keegan's presentation, its rejection would prove nothing. Congress may have thought it preferable to express the Act's limitation to United States territory by simply excluding torts in "foreign countries," instead of reciting a list of "nonforeign countries." See Western Coal Traffic League v. United States, 677 F.2d 915, 924 (D.C.Cir.), cert. denied, 459 U.S. 1086, 103 S.Ct. 568, 74 L.Ed.2d 931 (1982) ("amendments to a bill's language are frequently latent with ambiguity: they may either evidence a substantive change in legislative design or simply a better means for expressing a provision in the original bill"). In fact, we do not even know whether Mr. Keegan himself thought there was any difference in substance between the two formulations. And finally, to cap it all, Mr. Keegan's proposal not only explicitly spelled out the geographical limits but also made the criterion whether the "damage or injury occurr[ed]" within those limits, 1940 Senate Hearings at 65, — which is of course not the same as whether the "claim ar[ose]" there, see Richards v. United States, 369 U.S. 1, 9-10, 82 S.Ct. 585, 590-591, 7 L.Ed.2d 492 (1962). If Congress was aware of this proposal, and if it made a conscious rejection, it seems to me a better guess that this significant feature was the reason.

One must admire the thoroughness of appellees' research that could have unearthed Mr. Keegan and this portion of his extensive suggestions. But that this is the closest thing to a friend they have been able to find at the ever-congenial banquet of legislative history (in the case of the FTCA, a banquet with separate sittings in a number of years before it was finally adopted in 1946) is perhaps the strongest indication of how bereft of support their position must be. I may add that the willingness of the majority to rely upon such an irrelevancy is perhaps the strongest indication of the wisdom of the English courts in refusing to attend these feasts.

If it were necessary to consult legislative history in order to determine whether that interpretation of the statute should be adopted which leads to completeness and consistency or that which leads to inadequacy and confusion; then the relevant portion is the testimony by Assistant Attorney General Francis M. Shea, explaining to the House Committee on the Judiciary the revised version of the bill proposed by the Attorney General and (in the aspect at issue here) actually adopted by the Congress:

Tort Claims: Hearings on H.R. 5373 and H.R. 6463 Before the House Comm. on the Judiciary, 77th Cong., 2d Sess. 35 (1942) (emphasis added). Besides the fact that it is common practice to give deference to the contemporaneous interpretation of the department responsible for drafting a statutory provision, see Vermont Yankee Nuclear Power Corp. v. NRC, 435 U.S. 519, 546, 98 S.Ct. 1197, 1213, 55 L.Ed.2d 460 (1978), Assistant Attorney General Shea's testimony happens to be the only portion of the legislative history relied upon by the Supreme Court in the one case it has decided dealing with this aspect of the FTCA. See Spelar, 338 U.S. at 220-21, 70 S.Ct. at 11-12. It can hardly be thought, of course, that the "good deal of difficulty" which Mr. Shea envisioned from deciding a case under a law other than that of this country would be at all reduced when that other law is nonexistent. It may be that, for reasons I discuss in the immediately following paragraph of this opinion dealing with similar statements in judicial opinions, Mr. Shea's testimony is ultimately inconclusive. But to the extent that this sole authoritative portion of the legislative history proves anything, both its language ("restrict the bill to claims arising in this country") and its rationale ("otherwise, ... a good deal of difficulty") are only compatible with the interpretation the majority has rejected.

Prior Case-Law

Most of the decided cases involving § 2680(k) are not really helpful to the question before us, since none of them involved a situation such as this, in which not only does United States law not apply, but also the law of no foreign jurisdiction applies as well. To be sure, dicta uttered without contemplating this issue can be quoted on both sides. The majority, for example, quotes the statement in Spelar that the purpose of § 2680(k) is to prevent the United States from being subject to "liabilities depending upon the laws of a foreign power," 338 U.S. at 221, 70 S.Ct. at 12. And I could quote language from the same case to the effect that "[b]y the exclusion of claims `arising in a foreign country,' the coverage of the Federal Tort Claims Act was geared to the sovereignty of the United States," and that "[t]he amended version [i.e., the version amended pursuant to the Justice Department's proposals] identified the coverage of the Act with the scope of United States sovereignty." 338 U.S. at 219, 220-21, 70 S.Ct. at 11, 12.8 But to be honest about it, none of these dicta proves anything. In the context of that case (as in the context of almost all the other cases) the phrase "liabilities depending upon the laws of a foreign power" could have been elliptical, leaving out the implied conclusion "and therefore not depending upon the laws of the United States"; just as the phrase "geared to the sovereignty of the United States" could have been elliptical, leaving out the implied conclusion "and therefore depending upon the laws of a foreign power." There is in fact no way of telling whether the nonapplicability of United States law or the applicability of foreign law was crucial to the Court's thinking — and indeed it seems highly unlikely that either the Spelar court or most of the other courts cited by the majority had any intent whatever on the point.9

A few cases, however, contain holdings which, on the assumptions adopted by the rendering courts, are inconsistent with the majority's thesis that a country is "foreign" only if foreign law applies. Meredith v. United States, 330 F.2d 9 (9th Cir.), cert. denied, 379 U.S. 867, 85 S.Ct. 137, 13 L.Ed.2d 70 (1964), involved an FTCA claim based upon a negligent act allegedly occurring in the United States embassy in Bangkok, Thailand. The court held the claim to be barred by § 2680(k) without definitively concluding that foreign law was applicable. Pertinent portions of Judge Browning's opinion are as follows:

330 F.2d at 10-11 (emphasis added). This holding, arrived at without a conclusive determination that United States substantive law was not applicable, is incompatible with the majority's analysis.

So also is the holding of the Fourth Circuit in Burna v. United States, 240 F.2d 720 (1957), which involved an FTCA claim arising from an alleged negligent act on Okinawa while the United States had, by treaty, "the right to exercise any and all powers of administration, legislation and jurisdiction" over that island. Id. at 721. Here again the court did not determine whether Japanese or United States law was applicable to the tort. It said:

240 F.2d at 721-23.

Neither the analysis of these cases, nor their holdings (since they did not find foreign law applicable), can be squared with the majority's decision here. To my knowledge, the interpretation I propose neither creates a circuit conflict nor requires the positing of nonexistent findings of law to avoid one.

III. THE "HEADQUARTERS" ALLEGATIONS DO NOT ELIMINATE APPLICATION OF § 2680(k)

I believe the foregoing discussion establishes that any claim arising in Antarctica arises in a "foreign country" for purposes of § 2680(k), and is therefore not maintainable against the United States. The question I turn to in this section is whether the appellants' allegation of negligence in the training, selection and supervision of the air traffic controllers sent to Antarctica — which negligence allegedly occurred in the United States — either (1) suffices to establish that the entire claim arises in the United States, or (2) constitutes a separate claim that arises in the United States.10 I turn to the second part of this question first.

Only a Single Claim is at Issue

The majority acknowledges that all elements of the present complaint present "a single cause of action," but contends that the cause of action contains "two claims." Maj. op. at 104; see also Wald op. at 131.11 This strange usage accords neither with common understanding nor with what internal and external evidence establishes as the meaning of "claim" in the FTCA. The term "claim" in § 2680(k) is at least as broad as, if not broader than, the concept "cause of action."

Shortly before the adoption of the FTCA, Congress had revised the Federal Rules of Civil Procedure, substituting the word "claim" for the phrase "cause of action" wherever it appeared. The intent was to embrace the broadest meaning of the earlier phrase. As Judge Swan observed three years before the FTCA was enacted:

Original Ballet Russe, Ltd. v. Ballet Theatre, Inc., 133 F.2d 187, 189 (2d Cir.1943) (footnote omitted). Or as the Supreme Court later described the development:

United Mine Workers of America v. Gibbs, 383 U.S. 715, 724, 86 S.Ct. 1130, 1138, 16 L.Ed.2d 218 (1966). In a case cited by United Mine Workers as exemplifying the (if anything more limited) meaning of the earlier phrase, the Supreme Court had specifically said: "The mere multiplication of grounds of negligence alleged as causing the same injury does not result in multiplying the causes of action." Baltimore S.S. Co. v. Phillips, 274 U.S. 316, 321, 47 S.Ct. 600, 602, 71 L.Ed. 1069 (1927). See also Rhodes v. Jones, 351 F.2d 884, 886-87 (8th Cir.1965), cert. denied, 383 U.S. 919, 86 S.Ct. 914, 15 L.Ed.2d 673 (1966) (citation omitted) ("Under the Federal Rules of Civil Procedure, the word `claim' denotes the same thing [as `cause of action'], i.e., `the aggregate of operative facts which give rise to a right enforceable in the courts'").

The lawyers who drafted § 2680(k) were unquestionably aware of this general meaning of the newly prominent word "claim," and unquestionably used it in that sense not only in § 2680(k), but elsewhere in the FTCA. Section 2672, for example, provides that acceptance of an administrative award "shall constitute a complete release of any claim against the United States ... by reason of the same subject matter" (emphasis added). And § 2675(a) provides that suit cannot be instituted "upon a claim" unless the plaintiff "shall have first presented the claim to the appropriate Federal agency and his claim shall have been finally denied." If the word is interpreted as the majority would have it ("a single cause of action with two grounds for relief (two claims)," Maj. op. at 104), then it is highly questionable whether the headquarters "claim" is properly before us, since the notice provision is jurisdictional, see Bialowas v. United States, 443 F.2d 1047, 1049 (3d Cir.1971), and since the headquarters allegations were not even made in court until the amended complaint. Compare Complaint, ¶ 28(a)-(d), with Amended Complaint, ¶ 28(a)-(e).

Practical considerations relating to § 2680(k) in particular reinforce this meaning of the statutory language. Since suits against the government on the basis of actions of its employees are ex hypothesi based upon actions in the course of those employees' official duties, see 28 U.S.C. § 1346(b), it will virtually always be possible to assert that the negligent activity that injured the plaintiff 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. And regard for malpractice liability, if nothing else, will induce plaintiffs' lawyers regularly to include such an assertion in their complaints — for who can tell, at least until full discovery, where it is that proper training or supervision did not occur? Thus, the so-called "headquarters claim" will become a standard part of FTCA litigation involving the activities of federal employees abroad,12 and few such suits will be able to be dismissed at the jurisdictional threshold. I think it unlikely that the exclusion of "claims arising in a foreign country" was meant to operate in this fashion. To the contrary, it seems to me that the intent of Congress was to insulate the government from claims for injuries caused by its employees abroad, whether or not those employees' actions can in turn be attributed to actions or inactions by other employees stateside.

No one would contend that a single injury caused by a private employer, through various "separate acts or omissions" by various "separate sets of actors" in his employ, would produce a multiplicity of "claims." See, e.g., RESTATEMENT (SECOND) OF JUDGMENTS § 24 (1982). There is no reason for treating the government's liability differently, or for attributing a novel meaning to the word "claim" in the FTCA. The authorities the majority and concurrence rely upon for the notion that there are really two § 2680(k) "claims" involved here represent narrow exceptions to the ordinary meaning of the term, unrelated to the policies at issue here. Thus, comment c to the RESTATEMENT (SECOND) OF JUDGMENTS § 26 (1982) states:

(Emphasis added.) Comment c gives as an example the ability to sue in federal court under a federal statute on a claim that has already been rejected under state law theories in the state courts. In my view (and evidently in the RESTATEMENT's) it is more precise to describe this situation as involving two separate "phases of a claim" rather than two separate "claims." But even if the latter terminology is used, there is no basis for making such an exception from the general rule in the circumstances of this case.13

The Claim Arose in Antarctica

Since, then, only a single "claim" is involved, one would think that the next question would be in what place that single "claim" arose. It is at this point that Judge Wald's expressed willingness to assume "arguendo" that only a single claim is involved, Wald op. at 132, turns out to be a promise kept to the ear but broken to the heart. It is easy to assume a single claim, so long as one attaches no legal consequences to the concept — which is perhaps why the concurrence regards the matter of one versus more claims to be "no more than a semantic quibble," id. at 132. Specifically, Judge Wald would find it possible to hold that a unitary claim both arises in a foreign country and does not arise in a foreign country; and that the choice of law governing liability under the unitary claim is determined by the rules of Antarctica yet also determined by the rules of the District of Columbia.

There is a semantic quibble afoot here — but it does not pertain to whether a unitary claim is a unitary claim. Rather, it pertains to whether it makes any difference for Judge Wald to achieve the result she desires by frankly admitting (no arguendo about it) that she is fragmenting a unitary claim; or rather by calling it a unitary claim but depriving that concept of all reasonable consequences. It seems to me that when a statute refers to a place at which a unitary claim arose, it refers to a unitary place. It will not do to reject this approach as "inappropriate" because "it results in completely denying plaintiffs a forum for pursuing their claim," Wald op. at 133. Since the whole object of the provision we are construing is to deprive plaintiffs of a forum, it is hardly a useful or even a rational approach to reject ordinary meanings because they produce such a result.

The claim fragmentation or deprivation of the legal consequences of claim unity that the concurrence engages in (whichever of the two characterizations one chooses) has important consequences not merely with regard to determining whether a "claim arose" in a foreign country for purposes of § 2680(k); and not only, as we shall see, for purposes of the choice-of-law applicable to the claim under § 1346(b); but also for purposes of virtually all of the exemptions in § 2680. In one line of cases, for example, which is now the subject of a petition for certiorari that has been granted by the Supreme Court, plaintiffs have attempted to apply Judge Wald's technique to the exemption contained in § 2680(h), which excludes "any claim arising out of" various specified intentional torts. There also, one would assume that the unitary claim either arises out of the intentional tort or does not. Plaintiffs have sought to avoid the bar against intentional tort suits, however, just as the concurrence here would avoid the bar against foreign claims, by splitting the claim or denying the consequence of its unitary nature. Specifically, they have alleged the equivalent of "headquarters negligence" in the government's training, selection, or supervision of the employee who committed the intentional tort. I will discuss these cases in more detail below, see pages 124-125, infra, but mention them here merely to note the broad consequences of the approach the concurring opinion adopts.

I proceed, then, to discussion of the question whether the locus of the present unitary claim is the United States or Antarctica. The issue can be precised somewhat further, in accordance with this court's analysis in Sami v. United States, 617 F.2d at 762 n. 7:

Pursuant to this analysis, and assuming, as discussed above, that the single claim cannot be fragmented, we are faced with the necessity of selecting an "act or omission" to determine the issue of FTCA jurisdiction.

Since this is a suit for negligence, the first selection that comes to mind is that "act or omission" (in the chain of events producing the injury) which was blameworthy. But it is readily apparent that this solution cannot work. While it might fortuitously provide an answer in some circumstances, in the ordinary case there may be several points along the chain of causality at which genuine "fault" on the part of a government employee occurred; or perhaps no point at which the unquestioned fault can be located — as where negligence is established through the doctrine of res ipsa loquitur. In the present case, for example, the controllers may have been erroneously trained, but they may also have been inattentive to their duties at the time of the crash. In such circumstances the assignment of fault provides no basis for determining where the unitary claim arose. Moreover, establishing the particular location or locations of fault is often realistically impossible when the fault consists of an "omission" rather than an "act." For instance, if the real fault in the present case was that the controllers received no training whatever, where is it that a blameworthy federal employee failed to provide them training? In addition, determining where the claim arose on the basis of where government employee "fault" occurred would have the same harmful practical effect as permitting fragmentation of the single claim: "Headquarters fault" would always be pleaded, and the jurisdictional issue would rarely be resoluble at the outset. Finally and most importantly, resolving the issue on the basis of employee fault would produce results incompatible with plausible congressional intent. If an air crash in this country were caused by a military air traffic controller's failure to comply with an applicable objective standard of behavior, it is impossible to imagine that the United States could plead as a complete defense that the controller was simply acting in accordance with negligent training received at the Rhein-Main military base in the Federal Republic of Germany.

I think, therefore, that one must look to some test other than the allocation of individual employee fault. It is, after all, not liability of individual employees we are assessing in these cases, but liability of the United States. The relevant "act or omission" is the act or omission of the United States that caused the harm. In a case where it is claimed that one or more elements of negligence by federal employees affected the plaintiff through the failure of a particular federal action, or of a particular condition that should have been remedied by federal action, to comply with requisite standards, the place of the relevant "act or omission" is the place where that operative noncompliance occurs — regardless of whether any specific "blame" can be attributed to any particular federal employee at that point. Thus, if it is claimed that an accident was caused by a government car making a left-hand turn without signalling, it would not matter whether the government-employee driver was negligent in failing to activate the turn signal, or whether he had been negligently trained to turn without signalling at some other location, or whether he in fact activated the signal which did not work because of negligent servicing by government employees at yet a third location; the "act or omission" occurred where the alleged breach of standard that injured the plaintiff took place.14

The rule that a claim "arises" for purposes of § 2680(k) where there occurs the alleged violation of standard (attributable to government action or inaction) nearest to the injury serves to further one of the primary purposes of the provision — to avoid the application of foreign law against the United States. For whatever law may be applied to other elements of the claim, the crucial question of standard-violation in the government action immediately producing the injury is almost certain to be governed by the law of the place where that alleged violation occurred. Suppose a traffic accident is caused in a foreign country by a federal employee, acting within the scope of his duties, who makes a right turn at a red light. Suppose further that such action complies with the traffic rules of the District of Columbia, where such employee was selected and trained, but does not comply with the law of the foreign country. Even if negligence in the selection and training is asserted, surely under anyone's choice-of-law theory the crucial question, central to recovery, of whether the employee failed to comply with an objective standard of conduct would be governed by the law of the place where that alleged failure occurred. By making the same criterion the test of whether the claim arises in a foreign country, we most effectively achieve § 2680(k)'s purpose of avoiding the application of foreign law.

Persuasively analogous authority rejecting a "place of fault" criterion and adopting essentially the test I have set forth above is to be found in a Supreme Court decision determining whether a particular tort suit came within the constitutional grant of federal power over "Cases of admiralty and maritime Jurisdiction," U.S. CONST. art. III, § 2. In The Admiral Peoples, 295 U.S. 649, 55 S.Ct. 885, 79 L.Ed. 1633 (1935), the libelant had been injured while disembarking from a vessel, by falling from a gangplank leading from the vessel to the dock. The shipowner asserted that admiralty jurisdiction did not exist because, among other reasons, the alleged acts of negligence consisted of improper construction of the gangplank, and improper placing of the gangplank against the ship — acts which occurred on shore. After acknowledging that "[w]here the cause of action arises upon the land, the state law is applicable," 295 U.S. at 651, 55 S.Ct. at 886, the Court disposed of the issue as follows:

Id. at 652, 55 S.Ct. at 886 (emphasis added). Thus, the place of standard-violation, rather than the place where the negligence occurred, governed.

Analogous support can also be found in Richards v. United States, supra, which involved the issue of where the "act or omission occurred" not for foreign-tort immunity purposes, but rather for purposes of determining which state's law was applicable under the choice-of-law provision of the FTCA, 28 U.S.C. § 1346(b). There the plaintiffs sought damages for deaths in an airline crash allegedly attributable to the FAA's failure to assure adequate overhaul practices at American Airlines' depot in Tulsa, Oklahoma. The Supreme Court's opinion simply assumed (as apparently did the parties) that the place where the "act or omission occurred" was Tulsa. It nowhere appears in the opinion, however, that any individual federal employee was actually "negligent" or even was actually present at that particular location; and I think it most unlikely that the governing law would have changed if the reason for noninspection at Tulsa was a decision in Washington not to send FAA employees to Oklahoma. Rather, it seems to me that the "omission" within the meaning of the statute occurred there because it was there that the unlawful and injury-producing condition that a federal employee should have acted to remedy existed — even though the genuinely operative decision not to remedy it may have occurred all the way up the chain of command and causality, in Washington.

In a recent case involving FTCA liability of the United States for an "omission," the Ninth Circuit adopted (again for choice-of-law rather than "foreign claim" purposes) an analysis substantially similar to that which I have suggested (and, incidentally, incompatible with the majority's choice-of-law approach):

Ducey v. United States, 713 F.2d 504, 508 n. 2 (9th Cir.1983).15

Finally, closely analogous support can be found in those cases, alluded to above, which have rebuffed the attempt to convert an intentional tort (excluded from suit under the FTCA by § 2680(h)) into a negligent tort by reliance upon a claim of "headquarters negligence." Some of these opinions apply reasoning quite similar to that I have set forth above, as the following examples indicate:

Collins v. United States, 259 F.Supp. 363, 364 (E.D.Pa.1966).

Hughes v. Sullivan, 514 F.Supp. 667, 670 (E.D.Va.1980), aff'd "for reasons adequately stated by the District Court" sub nom. Hughes v. United States, 662 F.2d 219, 220 (4th Cir.1981). See also Wine v. United States, 705 F.2d 366, 367 (10th Cir.1983); Naisbitt v. United States, 611 F.2d 1350, 1353-56 (10th Cir.), cert. denied, 449 U.S. 885, 101 S.Ct. 240, 66 L.Ed.2d 111 (1980); Gale v. United States, 525 F.Supp. 260 (D.S.C.1981); Taylor v. United States, 513 F.Supp. 647, 649-53 (D.S.C.1981); Pennington v. United States, 406 F.Supp. 850, 851-53 (E.D.N.Y.1976); Davidson v. Kane, 337 F.Supp. 922, 923 (E.D.Va.1972). The Court of Appeals for the Third Circuit has taken a contrary view in two cases, Shearer v. United States, 723 F.2d 1102 (3d Cir.1983), cert. granted, 469 U.S. 929, 105 S.Ct. 321, 83 L.Ed.2d 259 (1984), and Gibson v. United States, 457 F.2d 1391 (3d Cir.1972).16 In the former of these, as noted earlier, the Supreme Court has accepted certiorari. If the approach adopted by Shearer is rejected by the Supreme Court for purposes of § 2680(h), I think it clear that it must also be rejected for purposes of § 2680(k) — and if adopted for the one, then also adopted for the other. While relying for the present upon the weight of authority regarding the § 2680(h) issue in support of my position here, I am hopeful that the Supreme Court's disposition of Shearer will address the "headquarters claim" point, and thereby conclusively resolve the issue between the concurrence and me.

Judge Wald believes that cannot be, since "[i]n contrast [to § 2680(k) cases] the rationale for not holding the government liable on the basis of headquarters negligence in § 2680(h) cases is that the intentional, criminal act of the employee is the sole proximate cause of the harm and supersedes the headquarters negligence." Wald op. at 138, citing RESTATEMENT (SECOND) OF TORTS § 442(B) (1965). Even in theory, that cannot possibly be so. Any general tort rationale of "superseding cause" would surely be sought in the applicable state law, and not in the text of the FTCA. Moreover, if that were the principle explaining the § 2680(h) "headquarters" cases, it would apply as well to the intentional, criminal acts of nongovernment employees; but as correctly stated in one of the cases:

Pennington v. United States, 406 F.Supp. at 851, citing Panella v. United States, 216 F.2d 622 (2d Cir.1954) (Harlan, J.). Of course in order to recover a plaintiff must establish legally "proximate" cause even if the bar of sovereign immunity does not apply, and opinions in some suits against the government involving intentional torts choose to go off on this nonjurisdictional ground, see, e.g., United States v. Shively, 345 F.2d 294 (5th Cir.1965), cert. denied, 382 U.S. 883, 86 S.Ct. 177, 15 L.Ed.2d 124 (1965) — a phenomenon which some of the § 2680(h) cases fully acknowledge, e.g., Naisbitt, supra, 611 F.2d at 1355. But none of the § 2680(h) cases I have cited above nonsuiting the plaintiffs are "proximate cause" cases in this ordinary tort-law sense, and there is simply no basis for the concurrence's appeal to this doctrine as a means of deflecting in advance the Supreme Court's action in Shearer.17

The concurrence brings ordinary tort principals of "proximate cause" into the debate even more generally, asserting that my insistence upon selecting a single jurisdiction in which the claim arose is incompatible with those principles. Wald op. at 135-136. That is not so. I have no quarrel with such cases as Ingham v. Eastern Air Lines, Inc., 373 F.2d 227, 237 (2d Cir.1967); the RESTATEMENT (SECOND) OF TORTS; and Dean Prosser; all cited by Judge Wald for the proposition that "where there are multiple, proximate concurrent or contributing causes, the courts do not arbitrarily choose the most proximate as the sole cause for purposes of liability with respect to FTCA claims." Wald op. at 135. I would not do so either, but that says nothing about the question whether, in the case of entirely sequential and dependent negligence such as we have here, the "most proximate" governmentally connected cause — or what perhaps should be called, in order to avoid confusion with the general tort doctrine, the "nearest cause" — should be chosen for purposes of determining where the unitary claim arose. For example, in the Ingham case that Judge Wald discusses, see Wald op. at 135, involving negligence by both pilots and air traffic controllers, I would permit either or both acts of negligence to be the basis of recovery. They are both "proximate causes" in the sense that they are not so remote as to be beyond the legally recognized chain of causality. The only issue is whether suit would be barred — not by reason of the "supervening cause" theory dismissed in Ingham but by reason of § 2680(k) — if the government negligence nearest to the injury occurred abroad. In other words, for purposes of determining where the unitary claim arises in the case of a chain of dependent causes, I would genuinely refer to the proximate governmental cause — using the word "proximate" in its proper superlative sense of "nearest," as opposed to its usage for purposes of determining which acts in the chain can be a legal basis for liability, which latter usage admits of multiple "proximate" causes.

Judge Wald asserts that my analysis is contrary to the holding of the Ninth Circuit in Leaf v. United States, 588 F.2d 733 (9th Cir.1978). Wald op. at 133-135 & n. 3. Not so. In that case, the plaintiff brought an action against Bean, allegedly an informant for the Drug Enforcement Agency, and against the United States, for the loss in Mexico of a plane that had been leased from the plaintiff for the purpose (undisclosed to him) of conducting an undercover drug-trafficking operation. Bean thought the flight back from Mexico was to be a "practice run," but armed Mexicans overloaded the plane with marijuana, as a result of which it crashed on takeoff. The plaintiff asserted that this was the result of alleged acts of negligence by Bean in both Mexico and the United States, and by other alleged government agents in the United States. The district court had granted summary judgment for the government, based on its finding "that Bean was not an agent of the United States," that "the acts of nongovernment actors in Mexico were the sole proximate cause of the injury," and that the claim therefore arose in a foreign country. 588 F.2d at 735. The nub of the Ninth Circuit's reasoning in reversing is shown by the following excerpts from its opinion:

Id. at 736. On my analysis, if Bean was a government agent and his violation of a standard of care in Mexico was the "nearest governmental cause" in the sense I have defined, dismissal was proper. But if Bean was not a government agent and there was proximate causality (in the tort sense) traceable to the violation of a standard of care by a government agent only in the United States; or if Bean was a government agent, but it was in the United States that his last operative violation of a standard of care occurred; then dismissal was improper. Since on the disputed facts the plaintiff had to be given the benefit of the latter suppositions, the reversal was correct.18 I think, in other words, that the Leaf court meant by "proximate cause" precisely what I mean by the last operative act of negligence. But the case is in any event in no way contrary to the analysis I have set forth.

Judge Wald also relies on Pelphrey v. United States, 674 F.2d 243 (4th Cir.1982). There a district court had rendered summary judgment against the plaintiffs on the grounds that the foreign country allegations were barred by § 2680(k) and that the headquarters allegations had no factual support. The court of appeals affirmed on the same grounds. If courts were fastidious about the impropriety of reaching merits issues without first establishing jurisdiction, it could be argued that this opinion implicitly holds that the occurrence of the "most proximate" negligence abroad did not dispose of the whole case. But courts are not so fastidious. See, e.g., Doe v. United States Dep't of Justice, 753 F.2d 1092, 1101 (D.C.Cir.1985). It is obvious that the Fourth Circuit in Pelphrey did precisely what this court would probably have done — decided the case the simple way without even considering the complex issue here under discussion. Neither in what it says nor in what it does can the opinion provide any support for the concurrence.

Finally, I cannot allow to pass unchallenged the concurrence's assertion that my approach is "calculated ... to insure a result of non-suit in complex cases where non-suit may very well not be the result most attuned to congressional intent." Wald op. at 143. My approach happens to result in a non-suit in this case, but in other cases would avoid a non-suit which the concurrence's approach would produce. For example, to modify Judge Wald's hypothetical, see Wald op. at 142, if a defective backup system was installed in a government plane in Mexico and was negligently activated in the United States, my approach would allow an FTCA suit on the basis of both the backup system and its negligent activation while the majority's approach seems "calculated ... to insure a ... non-suit" with respect to the backup system.19 Thus, if Judge Wald can write "Congress [did not] intend by the foreign country exception to sanction the government's exportation to foreign countries of poorly trained, unsupervised, or inept personnel with free license to injure unsuspecting residents of those countries," Wald op. at 143; I suppose I could reply that Congress did not intend to sanction the government's importation from foreign countries of poorly trained, unsupervised, or inept personnel with, under the majority's approach, free license to injure unsuspecting residents of this country so long as they follow instructions received overseas. But of course all such rhetoric is misdirected when one is dealing with a statutory provision which, in one manner or another, is unquestionably intended to "sanction" (if one chooses to put it that way) governmental negligence.

IV. EVEN ASSUMING THAT ANTARCTICA IS NOT A FOREIGN COUNTRY, THE CASE SHOULD HAVE BEEN DISMISSED

What I have said above shows that this suit presents a single claim arising in Antarctica, which, since Antarctica is a foreign country, must be dismissed for lack of jurisdiction under § 2680(k).

Even if I thought that Antarctica is not a foreign country, however, I would disagree with the majority's disposition of this case, and with large portions of the majority's and the concurrence's analysis. I must set forth that disagreement at least briefly, since what the majority has said with regard to an FTCA claim straddling two (in its view) nonforeign jurisdictions is of great practical significance.

First, as my analogous use of Richards v. United States and Ducey v. United States in the immediately preceding discussion suggests, I think it no more valid to split a unitary claim such as this for purposes of the choice-of-law provision of the FTCA, 28 U.S.C. § 1346(b), than for purposes of the foreign claim exclusion. Neither of those choice-of-law cases (nor any other I am aware of in the 39-year history of the FTCA) even considered such a course — and in Ducey in particular the plaintiffs had specifically raised a "headquarters claim." Thus, it seems to me that what really confronts the majority (on its erroneous assumption that Antarctica is not foreign) is not the clear applicability of D.C. law to the so-called "headquarters claim" and a supposed choice-of-law problem for the "other" claim, but rather a problem similar to that which I discussed in the immediately preceding section: which law applies to the entire suit?

To fragment a unitary claim for choice-of-law purposes as the majority would do creates needless and insoluble conflicts regarding applicable rules. Assume, for example, an FTCA suit by a resident of Virginia, alleging that a federal employee, also a resident of Virginia, while driving in the course of his duties a vehicle registered and garaged in that state, was involved in an automobile accident in West Virginia, in which the plaintiff was injured. The cause of the accident is alleged to be excessive speed of the federally driven vehicle, attributable to (1) the negligence of the driver, (2) negligent training of the driver at a federal training center in Maryland, and (3) negligent selection of the driver by federal supervisors in the District of Columbia. The problem which the majority's fragmentation of this unitary claim poses is that the three separate issues of standard of care will be governed by the whole law (including the conflicts law) of three different jurisdictions; so that issues common to liability under all three "claims" of negligence may be resolved differently depending upon which "claim" is at issue. Thus, for example, the court might have to find that since, on the "negligent driving claim," the law of West Virginia, where that negligence occurred, would apply an "interest analysis" choice-of-law approach, the law of Virginia would govern the issues of whether contributory negligence precludes recovery and whether the doctrine of "last clear chance" applies; while at the same time finding that on the "negligent training claim" the law of Maryland (where that negligence occurred) would use a "place of the accident" approach, and apply West Virginia's law to those same issues; and finding that on the "negligent selection claim" the law of the District of Columbia (where that negligence occurred), would adopt a "best law" (so-called) principle and apply its own contributory negligence and "last clear chance" rules. See generally E. SCOLES & P. HAY, CONFLICT OF LAWS 550-602 (1984). Assuming that Virginia resolves both of these common issues to the plaintiff's advantage, West Virginia both of them to the defendant's advantage, and the District of Columbia one to the advantage of each, the process of deciding the case is not only absurdly complex but produces a disreputable result. Indeed, it will produce no result at all when all three alleged elements of governmental negligence are found to have caused the accident — unless the jury is given the impossible task of assigning a percentage of causation to each of three acts of negligence that are interdependent and concurrently operative. This is not, as Judge Wald suggests, merely a routine instance of the "veritable jungle" which modern choice-of-law rules have produced, Wald op. at 139. It is vegetation of an entirely new strain. None of the respectable or even disrespectable (if there can any longer be such a thing) theories of choice of law, and no case I am aware of, would apply the whole law of different jurisdictions to the various elements of a single cause of action. It is inconceivable that Congress positively intended such contortion.20

Judge Wald suggests a more "flexible" approach to choice-of-law. I have no difficulty with the flexibility involved in permitting a court to consider first whether the choice-of-law rules of each connected state refer to the same law. If that is so, it is obviously unnecessary to decide which jurisdiction § 1346(b) points to. See the four cases discussed in the paragraph at page 140 of the concurrence. Judge Wald believes, however, that "where a true conflict of law problem is encountered after looking to the whole law of each place where an act or omission occurred, courts may either apply the law of each place to the acts occurring in that place or by some formula choose one place such as the place where the most significant act or omission occurred," Wald op. at 141 (emphasis added). That degree of flexibility is the issue here. In my view the court must "choose one place." It is interesting that the only court of appeals decision cited by the concurrence in support of its choice-of-law analysis in fact supports mine. Bowen v. United States, 570 F.2d 1311 (7th Cir.1978), discussed by the concurrence at pages 140-141, did feel the necessity of selecting a single jurisdiction. To be sure, it did not, as the concurrence notes, adopt the test for selection of jurisdiction I have proposed. But there is no reason why it should have done so, since the case did not involve acts of sequential and dependent causality — i.e., the negligence of one actor that ultimately has its effect only through the negligence of another — but rather three independently operative negligent acts.21 I need not decide what act I would select in such a situation; indeed, it could even be argued that the necessity of selecting a single jurisdiction is not as great when independent causality is at issue. The latter consideration would distinguish Kohn v. United States, 591 F.Supp. 568 (E.D.N.Y.1984), which also involved independently operative acts. In any case, to the extent that Bowen has any relevance to the present case, it supports my single-jurisdiction approach rather than the approach of the concurrence.

My other disagreements with the majority's analysis (accepting its assumption that Antarctica is not foreign) are of lesser consequence, since they relate to the relatively rare situation — created only by the majority's resolution of the Antarctica issue — in which the choice-of-law provision of the FTCA refers to a jurisdiction that has no law. Were I to create for myself the necessity of dealing with such a situation, I would not conclude, as does the majority, that I was confronted with a "question" of what law to apply — "a question not answered by the statute," Maj. op. at 105, and therefore requiring the court to devise an alternate choice-of-law rule. To the contrary, it seems to me no question is posed at all. According to the statute, we have jurisdiction only "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." 28 U.S.C. § 1346(b).22 It is clear that under the circumstances of this case, a private person would not be liable to the claimant in accordance with the law of Antarctica. It is therefore clear that we have no jurisdiction. Converting this plain answer into a "question" is not only incompatible with the rule that a waiver of sovereign immunity must be "unequivocally expressed," United States v. Mitchell, 445 U.S. 535, 538, 100 S.Ct. 1349, 1351, 63 L.Ed.2d 607 (1980), but distorts the intent of the statute by making the United States liable where a private person would not be.

I suppose it must be regarded as fortunate that the majority's decision to replace the choice-of-law rule of the statute with those of the RESTATEMENT (SECOND) OF CONFLICT OF LAWS, Maj. op. at 105, led to the District of Columbia in the present case. But one must fear that the circumstances of the next Antarctica case (or perhaps a revision of the RESTATEMENT) will lead next time to the substantive law of the Soviet Union. If that happens, one wonders whether the consequence will be to convert Antarctica (for purposes of that case) into a "foreign country" (since foreign law would then be applicable) with the result that the suit will be dismissed; or rather to set the court off in search of another nonjurisdiction to replace RESTATEMENT (SECOND), which will perhaps once again lead to the United States substantive law. I am tempted to confess that a decision which produces such endlessly interesting ramifications cannot be all bad.

* * * * *

Even insofar as the majority's holding regarding the nonforeign status of Antarctica is concerned, this is not necessarily the "one-of-a-kind" case Judge Wald believes and I would like to hope. Antarctica is a dangerous land. "[A]n uncomfortably large number of the buildings and facilities at McMurdo are named for those who died here, and crosses stand silhouetted on the black hills." M. Parfit, The Last Continent, SMITHSONIAN, Oct. 1984, at 48, 56. Fifty-one aircraft of this nation alone have been lost in Antarctica since 1946. Id. at 57. Tourism in Antarctica is particularly hazardous, and "[s]ome scientists believe that [it] should be banned," P. QUIGG, supra, at 103; see also id. at 101 (noting accidents of Antarctic tourist ship). The fact that what few activities take place in Antarctica are governmental activities or are heavily dependent upon governmental activities for their support, suggests that this may not be the last multi-million dollar tort suit against the world's deepest of pockets, arising from events in that desolate region, to which we will be asked to apply the local law of the District of Columbia.

But not even the hope of "one-of-a-kind" can be held forth for those portions of the decision permitting dismemberment of a unitary and assertedly nonforeign FTCA claim. That analysis will unquestionably cause confusion in determining the choice of law for domestic FTCA tort claims in the future. On all counts, I dissent.

WALD, Circuit Judge, concurring:

When Congress enacted the Federal Tort Claims Act (FTCA), it obviously did not consider the legal ramifications of torts committed by civilian personnel in Antarctica and, of course, made no specific provision for such an action. Cf. Antarctica Conservation Act of 1978, 16 U.S.C. § 2404(e) (including special venue provision). As a result,

Dalehite v. United States, 346 U.S. 15, 49, 73 S.Ct. 956, 975, 97 L.Ed. 1427 (1953) (Jackson, J., dissenting) (quoting Cardozo, The Growth of the Law, at 102).

Upon weighing the relevant factors, I find Judge Wilkey's and Judge Greene's judgment call as to whether this one-of-a-kind case ought to fall within the "foreign country" exception to the FTCA more persuasive than Judge Scalia's counter-position. As Justice Frankfurter stated in United States v. Spelar, 338 U.S. 217, 70 S.Ct. 10, 94 L.Ed. 3 (1949), the only Supreme Court case directly interpreting the FTCA foreign country exception:

Id. at 223, 70 S.Ct. at 13 (Frankfurter, J., concurring). I fully concur in Judge Wilkey's opinion that Antarctica is not a "foreign country" within the meaning of the FTCA, 28 U.S.C. § 2680(k), and see no cause to elaborate further upon his reasoning. I write separately to address the novel and, I believe, unjustified conclusions about choice of law and viability of FTCA actions that the dissent draws from its conceptualization of the cause of action in the present case as a "unitary claim."1 The dissent would have this court apply a mechanical yardstick for determining where "the unitary act or omission" occurred in every FTCA action to the end that no cause of action would be recognized in this case or in many other multi-layered negligence cases because some of the alleged negligent activities occurred in a foreign country. I find this attempt to impose one monolithic formula for deciding where a FTCA claim arose when the claim is based upon multiple acts or omissions not only in direct conflict with precedent but potentially mischievous in its consequences as well.

As Judge Wilkey states, see Maj.Op. at 101, this is essentially a single cause of action based upon a single claim of injury — wrongful death — resulting from the plane crash. Nonetheless, there are two alleged grounds of negligence: (1) the "headquarters claims" based upon the negligent selection, training, and supervision of the air traffic controllers by government personnel in Washington, D.C.; and (2) the "Antarctica claims" based upon the negligent acts or omissions of the air traffic controllers in Antarctica. Thus there are two separate allegations of negligence committed by two separate sets of actors at different points in time which either singly or in concert allegedly caused the plane crash.

Judge Scalia, in my view, has misread his FTCA map and thus has taken an unnecessary and unenlightening detour through the Federal Rules of Civil Procedure and the Restatement (Second) of Judgments in order to prove that there is only one FTCA claim in this case. See Dissent at 118-120. While there is no simple test for determining when claims are separate, I tend to agree with him that in this instance the plaintiffs would have run afoul of the rule against splitting claims had they decided to bring separate or successive actions. See Baltimore S.S. Co. v. Phillips, 274 U.S. 316, 321-22, 47 S.Ct. 600, 602-03, 71 L.Ed. 1069 (1927) ("The injured respondent was bound to set forth in his first action for damages every ground of negligence ... upon which he relied, and cannot be permitted ... to rely upon them by piecemeal in successive actions to recover for the same wrong and injury."); cf. Tolson v. United States, 732 F.2d 998 (D.C.Cir.1984) (Where FTCA claim was predicated on three alleged grounds of negligence — the acts of a prison guard, negligent supervision, and negligent retention — dismissal of the claim based on the acts of the prison guard under Fed.R.Civ.P. 54(b) was not a final appealable judgment since this was not a separate claim and relief could still be granted upon one of the two remaining grounds.). Nevertheless, whether plaintiffs in this case stated two claims for relief or two grounds for relief is no more than a semantic quibble2 and certainly not the real sticking point between Judge Scalia and myself. Thus, to join issue at the outset, I accept arguendo that this is a single cause of action or a single claim for wrongful death alleging two grounds for relief based upon two negligent acts or omissions.

Where I part company with Judge Scalia is with respect to the unwarranted conclusions he draws from the assumption that there is but one wrongful death claim here. For Judge Scalia relies on this assumption to launch a new and restrictive theory of FTCA liability: if there is a unitary claim, even though it is based on multiple acts of negligence, then the court must choose only one act or omission for purposes of determining the place where the entire claim arose. To locate that one special place where the one governing act or omission occurred, the dissent formulates a standard which can be restated simply as follows: The place of the relevant act or omission is the place where the most proximately operative failure to comply with standards occurs — regardless of whether any specific blame can be attributed to any particular federal employee at that point. See Dissent at 122. Application of this standard in the present case, of course, precludes relief altogether because the most proximate act or omission occurred in the Antarctica which Judge Scalia argues is a foreign country, or, even if it is not, provides no law under which to determine FTCA liability. Judge Scalia's standard, or indeed the requirement that there be only one place where a multicausative "claim" arose, is neither grounded in the language of the FTCA nor consistent with existing precedent.

Indeed to reach his chosen destination, Judge Scalia has to put aside his FTCA map and rely on sign posts which he himself has erected. As this court has recognized "[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 (D.C.Cir.1979) (citing § 1346(b)). With respect to the foreign country exemption, section 2680(k), we have said: "What must be in a foreign country under the exemption is, we think, not a `claim arising' but `an act or omission of an employee of the government.'" Id. at 726 n. 7. Similarly, the operative effect cases, see Maj.Op. at 96, have soundly rejected the approach of locating the situs of the relevant acts or omissions at the place where the acts or omissions had their operative effect or came to fruition (i.e., where the claim arose). For jurisdictional and choice of law purposes, the FTCA focuses on the place where the acts or omissions upon which the claim is based occurred. I find nothing in the FTCA to indicate that a single claim cannot be based upon multiple acts or omissions or that one governing act or omission must always be identified. As the Supreme Court pointed out in Richards v. United States, 369 U.S. 1, 9, 82 S.Ct. 585, 591, 7 L.Ed.2d 492 (1962), Congress apparently focused on cases where all the relevant events including the injury occurred in one jurisdiction and never explicitly considered cases where the relevant events occurred in more than one place. Merely asserting that there is a single claim here does not compel any particular answer to the very practical question left completely open by the FTCA: How should courts decide jurisdiction and choice of law questions when there are multiple acts or omissions occurring in different places?

To answer this question, the court need not pursue a singleminded, literalistic search for the one place where the one governing act or omission occurred. In fact, such an approach is particularly inappropriate where, as in this case, it results in completely denying plaintiffs a forum for pursuing their claim. As the Supreme Court has admonished:

United States v. Aetna Surety Co., 338 U.S. 366, 383, 70 S.Ct. 207, 216, 94 L.Ed. 171 (1949). Indeed when confronted with complex multistate or multinational tort actions, courts have for the most part adopted a more flexible approach which takes into account those acts or omissions — be they single or multiple — having operative significance in the case.

With respect to the present action, it is true that any negligence on the part of government personnel in Washington, D.C. in the selection, training or supervision of the air traffic controllers will be manifest only through the actions of those air traffic controllers. Nonetheless, there are still two separate allegations of negligence committed by separate sets of actors: one based on acts or omissions occurring in Washington, D.C., and one based on acts or omissions occurring in the Antarctica. The primary cause of the plane crash may have been inattentiveness on the part of the air traffic controllers or the primary cause may have been inadequate training in the use of the radar equipment. The mere fact that the alleged inadequate operation of the radar equipment by the air traffic controllers occurred in Antarctica does not negate the fact that the core, contributing act or omission of providing inadequate training which led to the plane crash may have been committed by personnel in Washington, D.C. I see no rationale in the statute or elsewhere in our jurisprudence for disregarding separate negligent acts or omissions — albeit connected in a chain of causation — committed by separate sets of actors in determining liability for the ultimately resulting harm. Indeed, to do so, in my view, conflicts with both existing precedent in analogous cases and traditional principles of tort law, and potentially leads to undesirable results for claimants with no offsetting benefits for judicial administration.

I. ANALOGOUS SECTION 2680(k) CASES

Decisions by other courts in analogous cases involving alleged acts or omissions occurring both in the United States and a foreign country do not support Judge Scalia's position that choice of law and even jurisdiction of the whole claim in section 2680(k) cases must be determined by a "last clear chance" theory of where the "one relevant act or omission" occurred.

In Leaf v. United States, 588 F.2d 733 (9th Cir.1978), the plaintiffs, owners of a plane damaged in Mexico, appealed from a summary judgment granted in favor of the defendant United States. The claim of governmental negligence was rooted in actions of an alleged agent of the Drug Enforcement Agency (DEA), Bean, who leased and used plaintiffs' airplane in a drug smuggling operation, and the DEA's failure to supervise and control Bean. These alleged negligent acts occurred in California and Arizona. The plane, however, was damaged in Mexico as a result of other negligent acts occurring in Mexico: overloaded with marijuana, it crashed on take-off and was subsequently sunk in a reservoir to prevent its discovery by police. The district court held that Bean was not a government agent and that the acts of non-government actors in Mexico were the sole proximate cause of the loss since their intervening conduct broke the chain of causation between alleged negligent acts in the United States and ultimate injury in Mexico. Thus, since the relevant acts or omissions occurred in Mexico, the FTCA claim was barred by the foreign country exception. Noting that "[a] claim `arises,' as the term is used in Sections 1346(b) and 2680(k), where the acts or omissions that proximately cause the loss take place," the Ninth Circuit reversed the district court, stating:

Id. at 736 (emphasis added). Yet, under the dissent's monolithic approach toward locating the place where the last government omission occurred, this case would have to be dismissed on jurisdictional grounds.3

Similarly, in Pelphrey v. United States, 674 F.2d 243 (4th Cir.1982), the plaintiff brought an FTCA claim against the United States for damages arising from a radical mastectomy performed at the Navy Regional Medical Center (NRMC) in the Philippines. The plaintiff's claim alleged wrongful acts of the surgeon which occurred in the Philippines and wrongful "headquarter" acts of the Navy which occurred in the United States. With regard to the latter, the plaintiff specifically alleged that the Navy had been negligent in the selection of the doctor and in the administration of NRMC with respect to providing adequate supervision, staff, and equipment. The court affirmed the district court's dismissal with respect to the alleged negligent acts of the surgeon in the Philippines since these were barred by the foreign country exemption. Since this was the last proximate act of negligence under Judge Scalia's approach, the inquiry should have ended here. Instead, the court went on to consider the allegations of negligent selection, supervision, and administration which occurred in the United States. The court ultimately concluded that the district court's grant of summary judgment was proper but only "[b]ecause Ms. Pelphrey failed to present any evidence, in the form of affidavits, depositions, or otherwise, to refute the government's affidavits demonstrating that the Navy had complied with the standards regarding the qualifications of Dr. Street and accreditation of NRMC...." Id. at 248.

Finally in Bryson v. United States, 463 F.Supp. 908 (E.D.Pa.1978), the court rejected the government's motion to dismiss for lack of jurisdiction in an action alleging that the government's negligence was a proximate cause of a serviceman's death where he was killed by a fellow, intoxicated serviceman whom he attempted to help out of the men's room in a barracks in Germany. Although the most proximate alleged claims of negligence occurred in Germany, i.e., providing intoxicating beverages and failing to guard against excessive intoxication and its consequences, and were thus barred by the foreign country exception, the court held:

Id. at 912. Again the dissent's approach would have precluded jurisdiction in this case. I suggest that these cases show Judge Scalia's FTCA jurisprudential notion about the need for identifying a single act or omission to be without ancestry in FTCA law.4

II. TRADITIONAL TORT PRINCIPLES RELATING TO PROXIMATE CAUSE

Traditional tort law principles as well suggest that, where there are multiple, proximate concurrent or contributing causes, the courts should not arbitrarily choose the most proximate as the sole cause for purposes of determining liability with respect to FTCA claims but should look instead to the various acts or omissions which proximately caused the harm. The Second Circuit, for example, in Ingham v. Eastern Air Lines, Inc., 373 F.2d 227 (2d Cir.), cert. denied, 389 U.S. 931, 88 S.Ct. 295, 19 L.Ed.2d 292 (1967), rejected the government's contention that it was relieved of liability predicated upon air traffic controllers' failure to provide up-to-date weather information because the crew of the plane subsequently negligently failed to execute a "missed approach maneuver." Although Ingham, unlike the section 2680(k) cases discussed above, does not involve a "headquarters claim" in the United States followed by subsequent acts of negligence in a foreign country, it clearly illustrates the application of well-accepted proximate cause principles. Finding the acts of the air traffic controllers a proximate and concurrent cause of the accident, the court stated:

Id. at 237 n. 11.

Similarly, the Sixth Circuit, in a case involving the parachute deaths of sixteen sky divers, made clear that concurrent or subsequent negligence did not exonerate the government. See Freeman v. United States, 509 F.2d 626 (6th Cir.1975). The court found that the air traffic controller's negligent act caused the pilot to believe that he was over the parachute jump target. The consequence of this act continued during the entire sequence of events so that the subsequent negligence of the pilot and jump master did not alone cause the drowning of the jumpers. The subsequent acts of negligence were not independent intervening causes but instead produced the injury in cooperation with the original wrong.

In general, case law recognizes that the United States can be held liable in tort in airplane crash cases if any negligent act of a government employee was a proximate cause of the injury. See also Delta Air Lines, Inc. v. United States, 561 F.2d 381, 389, 394 (1st Cir.1977), cert. denied, 434 U.S. 1064, 98 S.Ct. 1238, 55 L.Ed.2d 764 (1978); Hartz v. United States, 387 F.2d 870, 873-74 (5th Cir.1968); Thinguldstad v. United States, 343 F.Supp. 551, 553 (S.D.Ohio 1972). Under traditional proximate cause doctrine, any substantial factor in causing the harm is a proximate cause to which liability attaches, not just the last proximate act which could have prevented the harm. The Restatement (Second) of Torts § 442 B (1965) states:

See also id. at §§ 431, 433, 443, 447 (establishing that defendant is liable if negligent conduct is a substantial factor causing the harm); W. Prosser, Handbook of the Law of Torts § 41, at 240 (4th ed. 1971) ("If the defendant's conduct was a substantial factor in causing the plaintiff's injury, it follows that he will not be absolved from liability merely because other causes have contributed to the result....").

Although Judge Scalia has "no quarrel" with the cited authorities on proximate cause he apparently fails to understand the relevance of the authorities. See Dissent at 125. If Judge Scalia can assert that, "[t]he lawyers who drafted § 2680(k) were unquestionably aware of this general meaning of the newly prominent word `claim,' and unquestionably used it in that sense ... in the FTCA," see Dissent at 118, then I believe that they also were unquestionably aware of general principles of tort law. And, even if they had no glimmer of the concept of proximate cause, the FTCA specifically incorporates state and common law tort principles. As the Supreme Court noted in Richards:

369 U.S. at 6-7, 82 S.Ct. at 589-590. Thus, it is fair to assume that the words "act or omission" in the FTCA refer to acts or omissions which have operational significance in the case according to traditional tort law principles, i.e., acts or omissions which proximately caused the harm. While the FTCA does not explicitly address cases involving multiple acts or omissions occurring in different places, I submit that the "reasonable judge," when faced with the question of how to apply the FTCA in such a case, would look to FTCA cases and traditional principles of tort law to find an answer rather than creating an arbitrary rule of her own completely lacking any foundation in relevant law or precedent. Under well-established principles of proximate cause, when there are multiple acts or omissions, liability attaches to each act or omission which is a substantial factor in causing the harm, not just the most proximate or last act or omission. Thus, Judge Scalia's standard which determines, and in this case precludes, liability solely on the basis of the most proximate act or omission, regardless of its relative significance, in the chain of events culminating in the ultimate harm is totally out of sync with traditional tort principles which attach liability to any act or omission which is a substantial factor in causing the harm.

Judge Scalia, however, asserts that analogous support for his standard can be found in cases construing the intentional tort exemption of the FTCA. 28 U.S.C. § 2680(h) (specifically excluding claims arising out of an assault and battery committed by a government employee). In the cited cases, the plaintiffs allege that an injury or death was caused by the intentional, and usually criminal, act of a government employee, and that the government was somehow negligent in failing to prevent the harm (e.g., by discharging the employee, supervising the employee, or warning the victim). Since liability predicated on intentional torts of government employees is barred by section 2680(h), the issue is whether the government may nonetheless be held liable in these cases based upon the alleged headquarter negligence. The answer to this question has tended to turn on the specific facts of the case. The Third Circuit recently addressed the question and concluded:

Shearer v. United States, 723 F.2d 1102, 1106-07 (3d Cir.1983), cert. granted, ___ U.S. ___, 105 S.Ct. 321, 83 L.Ed.2d 259 (1984). The court then went on to hold that:

Id. at 1108. The Supreme Court has granted certiorari on Shearer, which Judge Scalia construes as a signal that the Supreme Court will vindicate his view that claims should not be allowed to go forward on the basis of allegations of headquarter negligence where the final act resulting in harm comes within one of the FTCA exemptions. The Supreme Court's ultimate disposition of Shearer is purely a matter of speculation especially since the case also involved a claim that government liability was precluded under the Feres doctrine. See Feres v. United States, 340 U.S. 135, 71 S.Ct. 153, 95 L.Ed. 152 (1950) (barring claims where injury is "incident to military service").5

In any event, the ultimate disposition of Shearer will not illuminate the appropriate path of analysis for this court to take in the present case. Judge Scalia, by adamantly refusing to acknowledge the relevance of traditional proximate cause principles in construing the FTCA and cases thereunder, has misconstrued the import of the section 2680(h) cases with respect to the present action. In the present case, if the government did negligently select and negligently train the air traffic controllers then the natural and foreseeable consequence is the negligent performance of those air traffic controllers. The government's negligent selection and training is a, if not the most, substantial factor causing the harm and thus is a proximate cause of the harm. In contrast, the rationale for not holding the government liable on the basis of headquarter negligence in section 2680(h) cases is that the intentional, criminal act of the employee is the sole proximate cause of the harm and supersedes the headquarter negligence. A claim involving a superseding intentional act is entirely different than a multilayer negligence claim.

As a general matter in tort law, the intervening intentional or criminal acts of third parties will break the chain of causation. See Restatement (Second) of Torts § 442B, supra p. 136. (Intervening force does not generally relieve original negligent actor of liability "except where the harm is intentionally caused by a third person and is not within the scope of the risk created by the actor's conduct."); cf. Romero v. National Rifle Ass'n of America, 749 F.2d 77, 79 (D.C.Cir.1984) (Scalia, J.) ("when the intervening act involves criminal, rather than negligent, conduct by a third party, the ability to anticipate (or foresee) the intervention with the normally required degree of specificity is not enough," a duty to guard against the harm must exist). The dissent in Shearer stated:

Shearer, 723 F.2d at 1111 n. 5 (Garth, J., dissenting) (citations omitted). Later the dissent distinguished Loritts v. United States, 489 F.Supp. 1030, 1032 (D.Mass.1980), where an FTCA action based on the rape of a member of a choral group performing at West Point by a cadet was held not barred by section 2680(h) because the "attack was ... not an intervening cause but rather, was a foreseeable result of the defendant's breach" of its voluntary undertaking to protect the female choral group. The dissent states:

Shearer, 723 F.2d at 1112 n. 6 (Garth, J., dissenting).

Thus even if the dissent's view in Shearer were to prevail in the Supreme Court it would still not provide support for Judge Scalia's view that liability should be precluded in this multilayer negligence case. The air traffic controllers' alleged negligence is not necessarily an independent superseding act constituting the sole proximate cause of the plane crash. If the plaintiffs' can prove their allegations of negligent selection, training, and supervision then the negligent performance of the air traffic controllers in Antarctica is a natural and foreseeable result of the government's headquarter negligence making that headquarter negligence a proximate cause of the plane crash.

III. CHOICE OF LAW

Pursuant to section 1346(b), the liability of the United States is determined "in accordance with the law of the place where the act or omission occurred." Once again, however, the statute says nothing about what law governs when there are multiple acts or omissions occurring in more than one state or country. Judge Scalia gives a hypothetical example to illustrate the "parade of horribles" which results if his standard of locating the one place, where the unitary act or omission occurred, is not adopted6 at least for choice of law purposes. See Dissent at 127-28. He suggests that applying the whole law of each jurisdiction where an act or omission occurred in a multi-layered negligence case would result in a whole new dimension of chaos in choice of law. This is nothing more than a restatement of the well-recognized complexity of the choice of law problem in multistate or multinational tort actions. See, e.g., Forsyth v. Cessna Aircraft Co., 520 F.2d 608, 609 (9th Cir.1975) (referring to "the wilderness in which courts sometimes find themselves when searching for solutions to problems arising under the judicial nightmare known as Conflict of Laws."); In re Paris Air Crash, 399 F.Supp. 732, 739 (C.D.Cal.1975) ("The law on `choice of law' in the various states and in the federal courts is a veritable jungle...."). Even if this court felt the need to construct a single approach to the problems of FTCA claims involving acts or omissions occurring in multiple places, and chose to do it in the context of a case as factually unique as the present one, I caution against a rule which is patently inconsistent with traditional tort principles and existing precedent.7

The only Supreme Court case addressing the choice of law problem under FTCA is Richards. There the Court held only that the FTCA requires application of the whole law of the place where the act or omission occurred, including its choice of law rule. Thus, the question here is not which substantive law will apply but which choice of law rule will apply. In Richards, the Court, after observing that Congress apparently did not consider choice of law problems in enacting the FTCA, stated:

369 U.S. at 13-14, 82 S.Ct. at 593. Yet, Judge Scalia would read into section 1346(b) his own choice-of-law rule despite the Supreme Court's explicit admonition that we should not assume Congress has intended rules independent of controlling choice-of-law principles. As far as I can tell, despite Judge Scalia's unsupported assertion to the contrary, see Dissent at 128, in cases involving multiple acts or omissions occurring in different places, the courts have traditionally resolved the choice of law problems by applying the whole law of the place of each act or omission.

Looking to the whole law of the place of each act or omission will often result in no true conflict either because each place uses the same choice of law rule thus pointing to one body of governing substantive law or because, although the choice of law rules point to different places, the substantive law of each place is the same. In fact, in the majority of cases involving multistate activity and addressing the choice-of-law question the problem has resolved itself in one of these ways. For example, in In re Silver Bridge Disaster Litigation, 381 F.Supp. 931 (S.D.W.Va.1974), the court found that the acts or omissions occurred in possibly two or three locations and, thus, pursuant to Richards, the court should look to the law of West Virginia, Ohio, and the District of Columbia. The court concluded, however, that, because the substantive legal principles of each of the jurisdictions were not in conflict, the choice-of-law problem could be largely disregarded. In Insurance Company of North America v. United States, 527 F.Supp. 962 (E.D.Ark.1981), where an airplane allegedly crashed due to the negligence of air traffic controllers in both Memphis, Tennessee and Blythville, Arkansas, the court looked to the whole law of both Arkansas and Tennessee. Since each state followed the lex loci delicti rule in negligence cases and the plane crashed in Tennessee, Tennessee law applied. See also Suchomajcz v. United States, 465 F.Supp. 474 (E.D.Pa.1979) (Though unclear from the complaint whether the negligent acts or omissions occurred in New York or Pennsylvania because Pennsylvania and New York have identical choice of law rules, under either jurisdiction Pennsylvania law applies.); Kantlehner v. United States, 279 F.Supp. 122 (E.D.N.Y.1967) (where acts or omissions may have occurred in any combination of nine states and seven states' conflict rules pointed to Maryland law and the remaining two states' rules pointed to New York, summary judgment could still be granted on all but one issue since the result would be the same under either Maryland or New York substantive law).

In these cases, the need to actually decide the choice-of-law issue was obviated. This may often turn out to be the case, but of course not always. At times, there will be true conflicts. In Bowen v. United States, 570 F.2d 1311 (7th Cir.1978), the pilot of a private aircraft brought an FTCA action against the United States alleging negligence on the part of air traffic control personnel at airports in three states for failure to warn of icing conditions in the vicinity of his final landing destination.8 The district court had found all three states' choice of law rules would point to the same substantive law. However, the Seventh Circuit on appeal noted that one of the states, Arkansas, had changed its choice of law rule and would now apply its own law which would create a true conflict. Noting that Richards shed no light on the question of which law to apply when acts or omissions occur in more than one state and cases where the need to decide the issue had been unnecessary, the Seventh Circuit stated:

Id. at 1318. The court found that Indiana law would apply under either approach and that the pilot's contributory negligence barred the action. Thus, while the Seventh Circuit, in cases involving a true conflict of law problem, would apparently choose one governing act or omission for choice of law purposes, it would choose not the last act as Judge Scalia suggests but the act or omission having the most significant causal effect.

On the other hand, in Kohn v. United States, 591 F.Supp. 568 (E.D.N.Y.1984), the family of a serviceman shot by a fellow soldier brought a FTCA action against the United States for infliction of emotional distress caused by negligent acts of the Army occurring in Kentucky and New York with respect to the disposition of the decedent's body and notification of the family. The court concluded that it would "look to the choice-of-law principles of each state, New York or Kentucky, where an alleged tortious act or omission took place in order to determine the substantive law to apply to each claimed tort." Id. at 572. Looking to those choice-of-law rules, the court concluded that Kentucky law should apply to acts occurring in Kentucky and New York law to acts occurring in New York and resolved the case accordingly.

The foregoing examples demonstrate that a flexible approach which takes account of all the acts or omissions having operative significance in a case, is the preferred approach and the one which has been followed by courts faced with complex causation cases. In many cases, no true conflict of law problem will be encountered. In those cases, where a true conflict of law problem is encountered after looking to the whole law of each place where an act or omission occurred, courts may either apply the law of each place to the acts occurring in that place or by some formula choose one place such as the place where the most significant act or omission occurred. I am confident our district court judges are up to the task.

In the present case, as well, it is unnecessary for us to decide how we would choose to resolve a case involving a true conflict. Alleged negligent acts or omissions occurred in the District of Columbia and Antarctica, thus, the court must look to the whole law of each place. In this novel case, Antarctica has no law, consequently there is no conflict of law problem. In my view, the majority's application of the whole law of the District of Columbia, under these circumstances, is clearly more in accord with FTCA precedent than Judge Scalia's approach which would coincidentally and to my view arbitrarily result in dismissing the action altogether at the outset.

The District of Columbia has adopted the "interest balancing" approach to choice of law problems which assesses the significant contacts. See Hitchcock v. Hitchcock, 665 F.2d 354, 359-60 (D.C.Cir.1981). In the present case, I think there is sufficient basis for this

In re Air Crash Disaster Near Saigon, 476 F.Supp. 521, 527 (D.D.C.1979) (citing In re Paris Air Crash, 399 F.Supp. 732, 745 (C.D.Cal.1975)). There is no conflict of law problem here or overwhelming policy interest of another forum in applying its law. Significant acts or omissions allegedly occurred in the District of Columbia making application of its law appropriate. Cf. Hitchcock, 665 F.2d at 359-69 (Although the vaccine resulting in the paralysis of a diplomat's wife was administered by a nurse in Virginia and that nurse failed to give the patient appropriate warnings and risk information, the locus of the negligent acts or omissions was in Washington, D.C. where the ultimate supervisory decisions were made and thus the D.C. choice of law rule applied.); Gelley v. Astra Pharmaceutical Products, Inc., 610 F.2d 558, 560 (8th Cir.1979) (applying D.C. choice of law rule where the bulk of the FDA activity addressed in the complaint occurred in the District of Columbia).

Finally, Judge Scalia suggests that considerations of judicial administrability of the FTCA's "act or omission" language commends the adoption of his standard for determining where the one governing act or omission occurred. Specifically, he asserts that courts will find it easier to identify the last proximate act which should have occurred to prevent the harm than to identify the most "blameworthy" or significant act causing the harm. Even if I accepted the underlying assumption that one unitary act or omission must always be identified in FTCA cases, which I do not, Judge Scalia's argument is still not persuasive. It may be just as difficult, if not more difficult to pinpoint the most proximate failure or omission which could have prevented the harm as to pinpoint the most significant or direct proximate cause of the harm. See, e.g., supra note 7. Moreover, such an approach — if applied in an undeviating way as Judge Scalia seems to anticipate — will result in mischievous and, in my view, unattractive outcomes. Take, for example, the following hypothetical situation: An emergency back-up system is improperly installed in a military transport plane here in the United States which will cause the plane to explode if the emergency back-up system is activated. An Air Force pilot flying the transport plane over Mexico carrying a number of civilians, family members of servicemen, negligently decides to conduct an unauthorized in-flight test of the back-up system, or even negligently decides that circumstances warrant activation of the system, and the plane explodes. The pilot's negligent act occurred in Mexico and thus an FTCA action is precluded by the foreign country exception. Clearly, however, but for the negligent installation of the back-up system in the United States the plane would not have exploded. The negligent installation was the primary cause of the plane exploding. I cannot conscience a standard which would immunize this act from being actionable or make whether it is actionable dependent on whether the pilot's decision to activate the system was negligent.9 Arguably, Judge Scalia's standard superficially may appear more reasonable and relatively innocuous when confined to cases where the negligent acts are interrelated such as in the present case;10 however, as the foregoing hypothetical illustrates, the standard is patently unacceptable when generalized to other fact patterns. Moreover, even in cases like the present one, where separate but dependent acts are involved, it may very well be that the local actor is merely executing to the tee orders received from a distant headquarters, or that the headquarters has failed to furnish needed equipment despite pleas from the local actors. Clearly in such cases, the gravamen of the negligence is located at the headquarters and should not be insulated from liability on the basis of an arbitrary choice of law rule.

IV. CONCLUSION

Nothing in the FTCA suggests that when several acts or omissions contribute to an injury, only one can be selected as the determinant of whether FTCA liability lies under either section 2680(k) or section 1346(b) and that one must be the most proximately operative failure to comply with standards. This is a judicial gloss, albeit a creative one, calculated by the dissent to insure a result of non-suit in complex cases where non-suit may very well not be the result most attuned to congressional intent. Although clearly a more easily administrable standard, Congress rejected use of the place of injury in FTCA — a standard which the dissent's most proximate operative failure standard closely approximates. See Richards v. United States, 369 U.S. at 10, 82 S.Ct. at 591. Unlike Judge Scalia, I do not believe that "the intent of Congress [in the FTCA] was to insulate the government from claims for injuries caused by its employees abroad, whether or not those employees' actions can in turn be attributed to actions or inactions by other employees stateside." See Dissent at 119. I simply do not believe that Congress intended by the foreign country exception to sanction the government's exportation to foreign countries of poorly trained, unsupervised, or inept personnel with free license to injure unsuspecting residents of those countries, as well as any Americans travelling abroad. Indeed, the extraordinary lengths to which the dissent must go to preclude these plaintiffs from proceeding with their FTCA claim against the government surely suggests the opposite.