Personal Jurisdiction and the Montreal Convention

SK/SZS SAS Ireland Airbus 320 EI-SIB by Riik@mctr (CC BY-SA 2.0)

I recently discussed the Fifth Circuit’s remarkably unremarkable personal jurisdiction analysis in a case involving a Montreal Convention claim. Before reaching the constitutional personal jurisdiction analysis, however, the panel in Hardy v. Scandinavian Airlines System first rejected the plaintiff’s argument that the Montreal Convention itself established personal jurisdiction over the defendant airline, either directly or as incorporated in the airline’s contract of carriage.

Though not an expert on the Montreal Convention, I was puzzled by aspects of the court’s analysis on this question. This post explains how U.S. negotiators believed they had expanded jurisdiction under the Montreal Convention to reach such personal injury claims and how changes in personal jurisdiction doctrine may have undercut those hopes—as well as possible routes for addressing this gap going forward.

The Montreal Convention’s “Fifth Jurisdiction”

The Montreal Convention of 1999 governs airline liability regarding international carriage of persons, baggage, and cargo and is intended to unify and simplify the patchwork of prior treaties and protocols governing such liability. It has been ratified by the United States, 135 other states, and the European Union.

Article 33 addresses “Jurisdiction.” Article 33(1) adopts the four bases for jurisdiction employed by its predecessor, the Warsaw Convention:

An action for damages must be brought, at the option of the plaintiff, in the territory of one of the States Parties, either before the court [1] of the domicile of the carrier or [2] of its principal place of business, or [3] where it has a place of business through which the contract has been made or [4] before the court at the place of destination.

Article 33(2), however, is new to the Montreal Convention. It provides:

In respect of damage resulting from the death or injury of a passenger, an action may be brought before one of the courts mentioned in paragraph 1 of this Article, or in the territory of a State Party in which at the time of the accident the passenger has his or her principal and permanent residence and to or from which the carrier operates services for the carriage of passengers by air, either on its own aircraft, or on another carrier’s aircraft pursuant to a commercial agreement, and in which that carrier conducts its business of carriage of passengers by air from premises leased or owned by the carrier itself or by another carrier with which it has a commercial agreement.

As Hans Huggler explained in a recent article, U.S. negotiators pushed hard to include this “fifth jurisdiction” in the Montreal Convention to cure a problem colloquially known as the “wandering American.” The Warsaw Convention did not permit U.S. courts to hear cases brought by injured Americans against foreign airlines if the U.S. plaintiff bought the ticket while abroad (or if someone like an employer had booked the ticket outside the United States) and the final destination was outside of the United States, even if the flight in question departed from a U.S. airport.

Article 33(2) strikes a careful balance between enabling individuals to sue for personal injury claims in their home courts and protecting airlines from having to defend cases in countries where they have no business presence. Only if an airline has a meaningful business presence in a country can a personal injury claimant file suit there on the basis of the Convention’s “fifth jurisdiction.”

Characterizing Article 33 from a U.S. Perspective

What cases Article 33 permits state parties to hear is a separate question from what cases U.S. courts are authorized to hear as a matter of domestic law. The Montreal Convention is a self-executing treaty, and it preempts state law. Because Article 33 delimits the range of claims that can be brought in U.S. courts, U.S. courts have consistently held that Article 33 (and its analog under the Warsaw Convention) defines the scope of their subject-matter jurisdiction over such claims via 28 U.S.C. § 1331’s grant of federal question jurisdiction.

This is the approach the Second Circuit took, for example, in National Union Fire Insurance Co. of Pittsburgh v. UPS Supply Chain Solutions, Inc. (2023). The Second Circuit interpreted Article 33(1) as addressing “treaty jurisdiction,” which for domestic purposes is a limit on subject-matter jurisdiction. While 28 U.S.C. § 1331 grants district courts subject-matter jurisdiction over claims based on treaties, the Second Circuit explained, Article 33 effectively restricts subject-matter jurisdiction to a subset of potential Montreal Convention claims.

The plaintiff in Hardy argued that Article 33 should be read as a grant of personal jurisdiction. The Fifth Circuit rejected that argument, but on a different basis than the Second Circuit in National Union Fire Insurance. The Fifth Circuit instead interpreted Article 33 as a venue provision because its language “precisely mirrors” the language of domestic venue statutes. It then reasoned that “a venue prescription sans authorization of service of process does not independently create personal jurisdiction.” Because Article 33 does not itself provide for service of process, the court concluded, Article 33 does not authorize the exercise of personal jurisdiction.

This reasoning gives me pause because it turns on reading the language of a multilateral treaty based on how domestic statutes are written and interpreted. That the Article “precisely mirrors” language in some U.S. statutes seems like happenstance, and the failure of a provision in a multilateral treaty to meet the language requirements that U.S. courts expect of certain U.S. statutes seems both unsurprising and immaterial.

Nonetheless, as I described in my prior post, the Hardy court went on to find that personal jurisdiction was separately authorized by Federal Rule of Civil Procedure 4(k)(2) and that the exercise of such jurisdiction in Hardy would not offend the Constitution.

Why It Matters

While Hardy was ultimately able to uphold personal jurisdiction under Rule 4(k)(2), its approach suggests that the Montreal Convention did not ameliorate the “wandering American” problem after all. Applying specific jurisdiction analysis in light of the airline’s contacts with the United States as a whole, the Hardy court reasoned that the plaintiff’s claim arose out of those U.S. contacts: not only did the plaintiff purchase her ticket in the United States, but she boarded the plane in the United States as well.

But that analysis might have led to a different conclusion if the ticket had been bought by the plaintiff’s daughter in Sweden (whom she was traveling to visit), or if once in Sweden the plaintiff had purchased a ticket to fly to Paris on AirFrance and had been injured on that flight. In both those scenarios, Article 33(2) would permit U.S. courts to hear the plaintiff’s claim against the foreign airline based on the airline’s significant operations within the United States. But it is unlikely that a U.S. court would have found specific jurisdiction over such a claim (at least for the second scenario), even taking into account the airline’s U.S. contacts.

What has changed since 1999, when the Montreal Convention was adopted, is the scope of general jurisdiction. At the time U.S. negotiators secured the inclusion of Article 33(2), a foreign airline doing continuous and systematic business in the United States was subject to personal jurisdiction on any and all claims against it; personal jurisdiction was thus not a limit on claims otherwise permitted under Article 33(2). But the Supreme Court’s decision in Daimler AG v. Bauman (2014) effectively ended general jurisdiction over foreign corporations, including foreign airlines.  During the same period, the Supreme Court has directed judges to rigorously police specific jurisdiction’s requirement that there be a relationship between the defendant, the forum, and the dispute (see, for example, Bristol-Myers Squibb Co. v. Superior Court of California (2017)). That combination of limiting both general and specific jurisdiction could leave many Article 33(2) cases on the table for want of personal jurisdiction.

Possible Alternatives

There are several ways that U.S. courts could salvage the Montreal Convention’s “fifth jurisdiction” on behalf of U.S. plaintiffs, though some are more appealing than others.

First, the courts could decide that the due process analysis under the Fifth Amendment does differ from that of the Fourteenth Amendment, such that federal courts applying Rule 4(k)(2) would not be limited by the specific and general jurisdiction analyses that the Supreme Court has grafted onto International Shoe Co. v. Washington (1945). That option is foreclosed in the Fifth Circuit by an en banc opinion (and I think the Fifth Circuit was correct to be skeptical of such arguments, which threaten to destabilize personal jurisdiction doctrine based on shaky historical claims). But it is a question the Supreme Court could decide this Term in a pair of cases regarding personal jurisdiction under the Foreign Sovereign Immunity Act.

Second, the Supreme Court could instead shave some of those analytical barnacles from International Shoe, an approach I have argued for but for which I am not holding my breath. The core insights of International Shoe remain powerful, and if that decision were applied on its own terms, it could arguably reach the “wandering American” cases. The distinction between general and specific jurisdiction, after all, was not articulated by scholars until the 1960s and was not embraced by the Supreme Court until the 1980s. The continuum of contacts and disputes that Justice Stone described in International Shoe would catch some of the cases that are falling between general and specific jurisdiction as currently articulated by the Roberts Court.

Third, courts could use the jurisdictional door that the Supreme Court recently opened in Mallory v. Norfolk Southern Railway (2023), which held constitutional Pennsylvania’s statutory scheme that treats companies’ registration to do business in the state as an acceptance of general jurisdiction in the state’s courts. As Huggler points out, Pennsylvania’s statutory scheme has already been applied successfully in an Article 33(2) case via Rule 4(k)(1)(A).

The plaintiff in Hardy made a similar argument in light of FAA regulations, which require incorporating at least some aspects of the Convention into the airline’s contract of carriage, but the Fifth Circuit was quick to reject that approach. I am a little unclear about the Fifth Circuit’s reasoning on this point. It may be that the FAA regulations only require foreign airlines to incorporate some of the Montreal Convention’s terms, not including the Convention’s jurisdictional provisions. (That to me would be the strongest argument against finding consent to jurisdiction.) It seems, though, that the Fifth Circuit relied on its interpretation of Article 33 as a venue provision to reason that incorporating a venue provision into a contract does not constitute a waiver of personal jurisdiction. (I find that explanation less persuasive because I am not convinced that Article 33 should be analogized to a domestic venue statute.)

Somewhat surprisingly, nowhere did the Fifth Circuit cite Mallory v. Norfolk Southern Railway (2023). I am a little nervous myself about how broadly Mallory’s endorsement of consent-by-statute can be read. But here such reasoning seems defensible because there is necessarily a link between the airline’s conduct that would be interpreted as consent (e.g., registering to operate flights within U.S. airspace and/or selling tickets to U.S. residents) and the scope of jurisdiction to which it would be consenting (e.g., personal injury claims arising from its flight operations—admittedly worldwide, but limited to those brought by U.S. permanent residents).

The simplest answer would be for Congress to pass a statute authorizing nationwide service of process for Montreal Convention claims and stating that foreign airlines consent to personal jurisdiction for claims within the scope of Article 33 when they operate any flights within the United States (either directly or, as recognized by Article 33, via a commercial agreement with another airline). But given that the Montreal Convention is a self-executing treaty, could it be possible to interpret the U.S. ratification of the convention as doing precisely that, with Rule 4(k)(2) establishing authorization for service of process that would be constitutionally sufficient in light of Mallory?

Conclusion: A Mess of the Supreme Court’s Making

I continue to think that the Supreme Court used overly strong language in Bristol-Myers Squibb when it rejected a “sliding scale” approach to specific jurisdiction. If it had not, it might still be possible to assert specific jurisdiction in “wandering American” cases even without statutory or regulatory requirements based on the significance of the foreign airline’s U.S. contacts and the (admittedly indirect) relationship between those contacts and the U.S. plaintiff’s injuries abroad. But as matters currently stand, U.S. passengers should be aware that they may still not be able to sue at home for injuries they sustain on flights between foreign countries, even if they fly on airlines with significant operations in the United States.