Balancing Nationalities in International Investment Law

I. Introduction

How can you tell where someone is “from”? Historically, an individual’s national identity was singular, starting with formal nationality. One’s national center of gravity was readily determinable. In most cases individuals lived and died in the country in which they were born, in which case no question could even have been presented. Even in cases involving migration, there would come a time when the shift from country of origin to country of resettlement would have been clear. That often coincided with the act of naturalization, which typically extinguished nationality of birth. Even where an individual somewhat anomalously held two nationalities, there would often have been one nationality with which the individual was more clearly aligned.

Not so today. Determining one’s primary national identification is no longer an easy thing in many cases. This is partly a function of increased mobility; it is no longer difficult, or even particularly expensive, to spread one’s time across countries on a continuing basis. One can own property and have family in more than one country. One can obviously identify culturally with more than one nation. One can even vote in multiple countries. Much of this is consequentially enabled by the growing acceptance of dual citizenship. One can openly identify as a formal member of more than one country in a way that was disfavored in the past. One can be from here and there. In almost no context is there is a need to determine which is more important.

International investment law, however, has not caught up to this reality. In an increasing number of international arbitrations, tribunals are shutting the doors on dual national claimants under the doctrine of “dominant and effective nationality.” The test finds roots in nineteenth century decisions of bilateral arbitral commissions faced with claims by individuals holding nationality in each of the two states party to a claims procedure. It was once the case that dual nationals were barred in all cases from filing claims against either country of nationality. In its earlier guises, an effective nationality test was deployed to allow claims that would have otherwise been precluded, to the end of allowing claims against a state of nationality to which ties were merely nominal or clearly subordinated. Today, the test is being put to work largely to bar claims by dual citizens who have substantial organic ties to each state of citizenship. “Effectiveness” is typically clear. Tribunals are left attempting to determine which nationality is “dominant.”

This Article in Part I first traces the origins of the “dominant and effective nationality” test in arbitrations from the late nineteenth through the twentieth century. The test was given an ostensible boost in the 1955 decision of the International Court of Justice in Nottebohm,1 easily the most famous international ruling relating to nationality. Although the case did not implicate a dual national, Nottebohm elaborated a sociologically-oriented analysis for determining the effect of formal nationality under international law. Several rulings of the post-war Italian-United States Conciliation Commission applied the test, most notably in the Mergé ruling,2 the first major arbitral decision to measure national attachments on a relative basis. The Iran-United States Claims Tribunal likewise adopted dominant nationality to determine jurisdiction in a raft of cases beginning with its 1984 decision in the A/18 Case.3 Although the approach in those cases was not without detractors, there emerged some agreement that “dominant and effective nationality” had come to represent customary international law.4 But into the beginning of the 2000s there were few cases in which the principle was tested.

That has changed. Part III considers a recent wave of arbitral decisions considering dual nationals under international investment agreements (IIAs). As part of state efforts to facilitate the global mobility of capital, bi- and multilateral trade and investment treaties enable individual investors to sustain claims against host states. At the same time, the incidence of dual citizenship has grown dramatically as more states come to accept the status.5 The result has been a raft of cases in investor-state dispute settlement (ISDS) involving dual nationals. Some investment agreements have expressly adopted the dominant nationality test. Arbitral tribunals have incorporated the test into other regimes that are silent on the treatment of dual national claims. The result has been a growing number of recent decisions applying the test. Most notable among these decisions is a Dominican Republic-Central America Free Trade Agreement (DR-CAFTA) tribunal’s 2019 decision in Ballantine v. Dominican Republic.6 In that case, U.S.-born spouses with significant continuing ties to the United States were barred from asserting claims against the Dominican Republic, where they had moved and naturalized, the tribunal finding that their Dominican nationality had become dominant in the context of their investments in that country. Meanwhile other tribunals have rejected the dominant nationality test, often on the grounds that treaties silent on the treatment of dual nationals are not intended to render them ineligible as claimants. The differential approaches have created a high level of uncertainty in the practice.7

Some commentators have advocated for applying the dominant and effective test as the default standard for assessing jurisdiction in claims involving dual nationals.8 Part IV trains a critical perspective on these rulings and on the incorporation of the dominant nationality test into ISDS regimes. The test is unworkable in cases like Ballantine, where the claimants’ national identities defied meaningful comparative assessment. In this respect, the dominant nationality test is rooted in antiquated conceptions of nationality as essentially singular where contemporary sociological conditions now allow for fluid and non-zero-sum national associations. Decisions applying the test have hewed to Nottebohm’s framing of nationality, which has been largely discredited in the scholarly literature.9 Moreover, the dominant nationality test and its application in cases like Ballantinewill have unintended consequences. Although the dominant nationality test ostensibly takes aim at so-called citizenship shopping, it will incentivize other strategic behaviors.10 It may, in the end, revive an imperial era practice in which investors from the Global North carefully nurture their homeland citizenship even while they establish themselves permanently as non-citizens, alongside their investments, in states of the Global South. In short, the dominant nationality approach is of another era and should be abandoned.

For better or worse, citizenship’s place in the world has been transformed.11 International investment law has been generally slow to absorb change, siloed from scholarship outside the perimeter of specialized arbitration journals.12 The nature of international arbitration, moreover, systemically inclines it to putative doctrinal regularity. In contrast to judicial counterparts, arbitrators are ill-equipped to engage in policy-inflected reasoning.13 Here as in other areas tribunals should come to incorporate elements of global social meanings into their decision-making presumptions. And while academic commentary engages abstract theoretical critiques of the international investment law,14 important questions involving the application of international investment treaties have been left unaddressed.15 This Article takes up one such question.

II. Dealing With Dual National Claimants: Early Cases

As a historical matter, dual nationality was highly disfavored.16 However, because of the near-complete discretion afforded states respecting nationality practice, dual nationality arose from the interplay of different state approaches to allocating nationality.17 In an age of migration, dual nationality was persistent notwithstanding state efforts to suppress it. This persistence coincided with the rise of arbitral mechanisms to resolve disputes between states relating to the treatment of transnational capital.18 Arbitral tribunals were called upon to determine jurisdictional boundaries in cases involving individuals holding nationality in each of two states party to arbitral agreements. Early decisions denied claims where an individual asserted the nationality of a state to which they had attenuated ties against a state in which they had nationality and in which they resided. Later cases articulated a doctrine of “effective” nationality.

The nineteenth-century decisions in Drummond’s Case19 and Alexander’s Case20 held that an individual holding two nationalities could not claim against either, consistent with the traditional international law dictum that a sovereign could do as it pleased with respect to any of its subjects, in other words, not be answerable under international law for any such conduct. In Drummond’s Case, the France-born grandson of an English duke who had emigrated to France almost a century prior sought to recover as an English subject for the confiscation of an estate in France, under an 1814 treaty between England and France resolving property claims arising in the years following the French Revolution.21 The treaty commission denied Drummond’s claim on the ground that the claimant was not in fact a British subject. The Privy Council, sitting on the appeal, found that Drummond did have British nationality, but denied the claim nonetheless:

[T]hough formally and literally, by the law of Great Britain, he was a British subject, the question is, whether he was a British subject within the meaning of the treaty. He might be a British subject and might also be a French subject; and if he were a French subject, then no act done towards him by the Government of France could be considered an illegal act, within the meaning of the treaty, which could only mean to provide indemnity where the act done towards the British subject was illegal by reason of the law of nations, or of some treaty subsisting between France and Great Britain. But neither the law of nations, nor any treaty bound the French Government to, act otherwise than as it thought fit Towards its own subjects.22

Finding Drummond to have French nationality under French law, the Privy Council upheld the Commission’s denial of the claim.23

Alexander’s Case similarly involved an individual making a claim against the country in which he had resided on the basis of nationality in a state in which he had neither been born or resided. Alexander was born in Kentucky to a British immigrant.24 His horse farm was damaged by Union soldiers during the Civil War, and he made a claim as a British subject against the U.S. under the Treaty of Washington, which set the terms of Civil War claims.25 The claims commission denied jurisdiction. “The practice of nations in such cases,” Commissioner James Somerville Frazer observed in oral comments,

is believed to be for their sovereign to leave the person who has embarrassed himself by assuming a double allegiance to the protection which he may find provided for him by the municipal laws of that other sovereign to whom he thus also owes allegiance. To treat his grievances against that other sovereign as subjects of international concern would be to claim a jurisdiction paramount to that of the other nation of which he is also a subject. Complications would inevitably result, for no government would recognize the right of another to interfere thus in behalf of one whom it regarded as a subject of its own.26

These cases were consistent with traditional understandings that no claim could be brought under international law by or on behalf of individuals against a state of nationality, a rule grounded in sovereignty and the complete discretion afforded states over their own nationals. Writing in 1915, Edwin Borchard concluded that “[t]he principle generally followed has been that a person having dual nationality cannot make one of the countries to which he owes allegiance a defendant before an international tribunal.”27 Some tribunals and other authorities continued to hold that under no circumstances could a person sustain a claim against a state of nationality, even when that nationality was ineffective, under what was known as the doctrine of “non-responsibility.”28 The approach was incorporated in the 1930 Hague Convention on Nationality.29

But on their facts, both Drummond’s Case and Alexander’s Case were also consistent with the subsequent articulation of a rule of effective nationality, under which an individual could not make a claim as a national of a state in which he had nominal nationality. Rather, he could only make a claim as a national of a state whose nationality he held in more meaningful terms.30 As even Borchard allowed, there were other cases, in which “international tribunals have endeavored to resolve the conflict of nationalities by applying various criteria to determine which of the two nationalities could more properly be attributed to the claimant.”31 Borchard elided the general rule of non-responsibility by supposing that attribution comprised an implied “election” of one nationality over another, in other words, a constructive choice on the claimant’s part.32 Borchard highlighted domicile as a key criterion. Tribunals also privileged nationality jus soli (by place of birth) over citizenship jus sanguinis (by descent).33

In most of these cases, there was no real question as to which of two nationalities was the “real” one. In Lebret, a woman holding both French and U.S. nationality who had lived in the U.S. for 40 years and centered her entire life there brought a claim under a U.S.-French treaty settling Civil War damages.34 The French commissioner on the claim conceded that her attachment to that country was “too platonic” to justify France demanding compensation for the losses she had sustained.35 In Hammer and Brissot, claims against Venezuela of two women holding Venezuelan and U.S. nationality were rejected where they had acquired nationality automatically by marriage to American men; neither had ever resided in the United States.36 Similarly, in the 1912 decision in Canevaro, the Permanent Court of Arbitration denied a claim where the individual suffering loss sought to claim against Peru as an Italian; he held nationality in both states.37 The individual was an Italian national through descent and held Peruvian citizenship on the basis of territorial birth. The claimant “on several occasions conducted himself as a Peruvian citizen, both by standing as a candidate for the Senate, where none are admitted except Peruvian citizens and where the individual went on to defend his election.”38 There was no evidence, on the other hand, that he had ever held himself out as an Italian national. In such circumstances, Peru had “the right to consider him as a Peruvian citizen and to deny his status as an Italian claimant.”39

In most of these cases, jurisdiction was rejected after finding nationality in the respondent state to be nominal, a result that could also have been accomplished under the bright-line “non-responsibility” approach against dual national claims.40 The Montfort decision supplies an exception.41 The widow claimant in that case held both French and German nationality, the latter through the naturalization of her husband. (A number of these cases implicated the once pervasive practice under which a woman’s nationality automatically followed that of her husband, precisely because the acquisition of nationality in these cases was involuntary.42 ) She lodged a claim against Germany in a mixed arbitral tribunal established to consider damages incurred during the First World War. The claim against Germany was sustained notwithstanding her German nationality. The tribunal drew from a non-state precedent involving the Institute of International Law in which an Austrian had been extended German citizenship by reason of his appointment to the faculty of a German university.43 The body found him to be Austrian for purposes of holding office in the Institute. “We think that it is natural to consider him,” concluded the Institute, “as being only the national of the State to which he is bound by the law and the fact of which he is the national and in whose territory he resides and in whose service he is. It is, so to speak, the active nationality which must be envisaged and not the somewhat theoretical nationality which may subsist alongside it.”44 Noting that the widow was born and had always lived in France and had held herself out as a German only to collect interest on a bank account, the tribunal found that “there is reason to consider the applicant as French.”45

These were not close cases once it was allowed that the dual national could sustain a claim in the first place. There was invariably one meaningful nationality and one nominal one. That reflected the nature of national attachments in the eighteenth and nineteenth centuries. Where dual nationality persisted, notwithstanding strong norms against it,46 there was typically one nationality that centered an individual’s sociological attachment.47 This may have been not the least because dual nationality was such a disfavored status. Dual nationality persisted through the mismatch of state nationality practices, not because people were looking for it; indeed it was a status to be avoided.48 Individuals who held it would have done nothing to advertise the condition. Moreover, material circumstances at the time would have made more balanced connections between each of two countries of nationality difficult. Expensive, slow transportation and balky channels of communication obstructed possibilities for a more sociologically distributed binational existence, even were individuals so inclined.

That context enabled an approach assessing the reality of ties to each of two states of nationality, indeed it almost demanded it. To have allowed nominal nationality to defeat jurisdiction in claims processes would have been arbitrary in obvious ways. The “effective nationality” test had yet to be formulated as such, and some observers denied its prevalence,49 but the drift of arbitral proceedings was pointing in that direction.50

III. Dominant Nationality Ascendant

From there, the test of effective nationality evolved to include a relative assessment of ties to more than one state of nationality, an inquiry into which nationality was “dominant” even if both were “effective.” The approach crystallized with three notable cases in the latter half of the twentieth century: the Nottebohm ruling of the International Court of Justice,51 the Mergé case out of the Italian-United States Conciliation Commission,52 and the A/18 decision of the Iran-United States Claims Tribunal.53 These cases grafted onto the notion of effective or actual nationality the more relativistic test under which an individual’s nationalities would be weighed in terms of sociological connection, hence the supplemental, or clarifying, element of “dominant” along with “effective” nationality.

A. Nottebohm

The Nottebohm decision notoriously did not involve a case of dual nationality at all,54 and yet remains the touchstone for the doctrine of dominant and effective nationality, almost surely because it is the only major decision of the International Court of Justice considering the nationality of individuals in any permutation. In that case, Liechtenstein looked to espouse a claim on behalf of its national, Frederic Nottebohm, who had recently naturalized as a citizen notwithstanding his lack of preexisting ties to the country. Nottebohm had for more than 30 years resided in Guatemala as a German national; as the Second World War loomed he shed his German nationality for nationality in the neutral Liechtenstein, apparently to avoid enemy alien status in Guatemala.55 When Guatemala persisted with the expropriation of his property, Liechtenstein asserted a claim on his behalf. At the time of expropriation and espousal of the claim, Nottebohm held nationality only in Liechtenstein, having automatically lost his German nationality upon naturalization in another state and having never naturalized in Guatemala.56

The ICJ rejected Liechtenstein’s capacity to sustain the claim on the grounds that Nottebohm lacked ties to Liechtenstein sufficient to support the claim. Along the way, the Court defined nationality as “a legal bond having as its basis a social fact of attachment, a genuine connection of existence, interests and sentiments, together with the existence of reciprocal rights and duties.”57 This famous articulation gave rise to the doctrine of “genuine link,” under which nationality will be given international effect only where supported by sociological connection.58 The Court accordingly examined whether Nottebohm’s connection with Liechtenstein was

sufficiently close, so preponderant in relation to any connection which may have existed between him and any other State, that it is possible to regard the nationality conferred upon him as real and effective, as the exact juridical expression of a social fact of a connection which existed previously or came into existence thereafter.59

Finding on the contrary Nottebohm’s actual connections to Liechtenstein to be “extremely tenuous,”60 with the aim of securing protection “but not of becoming wedded to its traditions, its interests, its way of life,”61 the Court denied the efficacy of the nationality as interposed against Guatemala.

The Nottebohm decision has been the target of withering criticism on its facts, as has the “genuine link” theorem which it bore.62 Less controversial were its dicta relating to the treatment of dual nationals in the context of international claims (dicta because Nottebohm himself was not a dual national). The Court observed that in cases involving dual nationals, international arbitrators had given “their preference to the real and effective nationality, that which accorded with the facts, that based on stronger factual ties between the person concerned and one of the States whose nationality is involved,” considering first the place of habitual residence but also other factors including “family ties, his participation in public life, attachment shown by him for a given country and inculcated in his children, etc.”63 Nottebohmdrew on these cases by way of supporting its examination of relative ties: were his ties to Liechtenstein closer than his ties to any other state?64 The ICJ was treating Nottebohm as if he were a dual national, a citizen of Germany and/or Guatemala as well as Liechtenstein. As such, the decision was in line with prior arbitral decisions. Moreover, like those decisions, the comparison produced an easy answer. Nottebohm had deep ties to both Germany, his county of birth and ethnicity, and to Guatemala, where he had been long resident. His ties to Liechtenstein, by contrast, were readily characterized as something other than “real and effective.”65

B. Mergé

With its lofty language, Nottebohm supplies the cornerstone for “romantic” conceptions of nationality.66 But it was the lesser-known Mergé case that fully consolidated the dominant and effective test for dual national claims.67 Mergé involved a claim brought against Italy by a U.S.-born citizen who had acquired Italian nationality by virtue of her marriage to an Italian diplomat. She had lived with her husband in Italy from 1933 until 1937; from 1937 until 1946 she accompanied him on his assignment to the Italian embassy in Tokyo, travelling there on her Italian passport. After living in the U.S. for nine months after the war, she resumed her residence in Italy until the date of the tribunal’s decision in 1955. She had renewed her U.S. passport on several occasions and had registered as a U.S. citizen with U.S. consular authorities in both Tokyo and Rome.68

The Italian-U.S. Conciliation Commission rejected her claim as an American against Italy. The tribunal did so with a subtle twist on prior approaches. Working off earlier decisions, the Hague Convention,69 and Nottebohm, and rejecting the doctrine of non-responsibility, the tribunal accepted the “principle of effective, in the sense of dominant, nationality.”70 Elaborating, the tribunal concluded that “effective nationality does not mean only the existence of a real bond, but means also the prevalence of that nationality over the other, by virtue of facts which exist in the case.”71 On that basis, the commission concluded that Mergé could “in no way be considered to be dominantly a United States national . . . because the family did not have its habitual residence in the United States and the interests and the permanent professional life of the head of the family were not established there.”72

The Mergé case marked an important evolution in the treatment of dual national claims, adding the “dominant” prong to the existing test of effectiveness. Earlier decisions had applied the “real and effective” test to consider whether one of two nationalities was (to reverse the inquiry) not real and ineffective, in other words, nominal. That is what had made these cases relatively easy to decide, mostly to the end of rejecting claims that were founded on nationalities that were formal and nothing more. “Effective” in these terms was not the same as “less meaningful,” the tribunal’s assertion of equivalence notwithstanding. Mergé evolved the test to apply where an individual had two nationalities, both of which were real and effective in the sense of being something more than nominal, instituting a comparative analysis to determine which of two nationalities reflected the more significant, “prevailing” connection. This was necessary to decide the case, insofar as Mergé’s U.S. citizenship was not happenstance or technical—she had lived in the U.S. for twenty-four years after her birth, during which time she had presumably been “wedded to its traditions, its interests, its way of life,” in Nottebohm’s terms.73 As of the coming into force of the peace treaty with Italy, at which date nationality was fixed for purposes of claims against Italy, Mergé had only recently returned to Italy from sojourns in Japan and the U.S. The decision leaned heavily on the claimant’s sometimes habitual residence in Rome, and perhaps the facts demonstrated that her Italian nationality, at the relevant times, was the more important of the two. But it was a much closer call than earlier cases.

Other cases decided by the Italian-U.S. Conciliation Commission further demonstrated the challenges of applying the test. In Salvoni, a U.S.-born claimant acquired Italian nationality by virtue of her marriage to an Italian national.74 The couple resided in the U.S. for ten years before relocating to Italy in 1937. Notwithstanding “numerous letters showing her attachments to America and her desire to return thereto, as well as affidavits filed with the American Consular offices in Italy,” her claim was denied on the grounds of dominant Italian nationality, in light of the fact that “if she wanted to return to the United States before or after the war she would have had ample opportunity to do so.”75 In Mazzonis, a U.S.-born claimant who acquired Italian nationality upon marriage to an Italian man was not “considered to have been dominantly a United States national” when she resided with her husband in Italy, notwithstanding her return to the U.S. after his death in 1948.76 Spaulding involved a U.S.-born claimant who acquired Italian nationality upon marriage to an Italian man in 1938. In 1943, she established permanent residence in Switzerland and then returned to the U.S. in 1945. The commission found it “obvious that her Italian citizenship was dominant in that she remained in Italy and Switzerland with her husband, the head of her household, practically continuously from the date of her marriage until she returned to the United States in 1945.”77 In Ruspoli-Droutskoy, a wealthy American-born woman who had married an Italian prince in 1901 essentially led a binational life after his death in 1909. “It is evident that Mrs. Ruspoli did a great deal of travelling in her lifetime and that a considerable portion of her travels, after her marriage to Mr. Ruspoli, was made between Italy and the United States.”78 She spent the entirety of both world wars in the U.S. and consistently traveled on U.S. passports. The claims commission found her dominantly American.

The factual backgrounds of these claimants brought into relief an emerging population of individuals who might accurately be described as transnational. The cases were enabled in the first place by a tolerance of dual nationality in cases involving U.S. citizen women who married foreign men. Formerly, American women who married foreign men automatically lost their U.S. citizenship, while typically securing the husband’s nationality.79 The 1922 Cable Act allowed women to retain their U.S. citizenship in most cases, while most other states (Italy included80 ) continued to automatically naturalize foreign women upon marriage.81 Under the former regime, the American-born woman would have lacked the capacity to press an international claim as an American, having lost her citizenship through the marriage to a foreign man. Now she could. But these mixed-nationality marriages and circular migration made difficult any pinning of the individual’s “real” nationality in many cases.

C. A/18

This difficulty was brought into further relief in the context of the late twentieth century Iran-United States Claims Tribunal, established in the wake of the 1979–81 embassy hostage crisis.82 In the A/18 Case, the tribunal adopted the dominant and effective test as “the applicable rule of international law” in the face of the failure of the underlying treaty to specify the treatment of dual national claims.83 Mergé had paved the way for balancing nationalities where a claimant had effective ties to each state party to the claims procedure. More so than Mergé and the Italian-U.S. Conciliation Commission cases, these claims involved complicated narratives of claimants’ engagements and experiences with Iran and the U.S., often together with meaningful connections to other states. The Iran-U.S. tribunal decided dozens of cases involving dual nationals in which it applied the dominant and effective nationality test.84 In line with Nottebohm and Mergé, relevant factors in its inquiry were topped by place of habitual residence, along with “center of interests, family ties, participation in public life and other evidence of attachment.”85

These cases set a “relevant period” for the inquiry between when the claim arose (typically a taking of property by the new revolutionary regime) and the signing of the Iran-U.S. claims agreement in early 1981.86 The tribunal was thus not focusing on the full life of the claimant in considering the question, though it did recount full life histories in all cases involving dual nationals.87   The Diba case supplies an illustrative example. Diba was born in Iran and moved to the U.S. for college in 1950.88 He married an American and had two U.S.-born children, naturalizing in 1959. After getting divorced, he returned to Iran in 1967, where he married an Iranian national. Notwithstanding reestablishing residence in Iran, Diba claimed that he continued to spend 50% of his time in the U.S., 30% in Iran, and 20% in Europe.89 The tribunal nevertheless found his Iranian nationality to be dominant during the relevant period (between late 1979 and early 1981), while conceding that Diba’s U.S. nationality may have been dominant for an earlier period. Other cases involved Iranian-born nationals who spent substantial periods of time in the U.S. only to return in the 1970s for professional reasons. Most were deemed dominantly Iranian for purposes of their claims notwithstanding continuing meaningful connections to the U.S.90

Married women were once again disadvantaged under the test. The claimant in Perry-Rohani, for example, was born in the U.S.91 When she was twelve, she moved with her parents to Switzerland, where her father worked for a U.S. corporation. She returned to the U.S. for college, after which she married an Iranian national, by virtue of which she became an Iranian national herself. They moved to Tehran in 1975, where her husband owned an engineering consulting firm, departing in the summer of 1978 as the revolution took hold in Iran. The tribunal found her a dominant Iranian national for the period between the period the claim arose in August 1979 and the claims agreement of early 1981. It stressed that she had not “cultivated or had any direct contacts with American nationals” in Iran, choosing instead “the company and friendship of the Iranian friends of her husband.”92 “It can be concluded therefore that during these years in Iran her exposure to American society and culture was virtually nonexistent.”93 But the claimant was obviously American in a deeply meaningful sense.

These cases reflected growing transnational mobility. Pre-revolutionary Iran and the U.S. had an exceptionally close relationship, especially in the years leading up to the revolution.94 Close economic ties resulted in a large community of Iranians in the U.S. with continuing ties to Iran.95 Many Iranians came to the U.S. for college and graduate school,96 establishing significant ties to the U.S. in early adulthood while retaining ties to Iran. There was significant circular—cyclical, really—migration.97 There were many binational marriages.

Critically for these purposes, dual nationality itself was becoming a more common status. The U.S. naturalization process had never by itself required the actual severing of original citizenship as a condition of naturalization; it depended on the citizenship laws of other countries to terminate nationality upon naturalization in the United States.98 Iranian nationality was at the time anomalous in providing for the retention of nationality upon naturalization in other states. Indeed, it was almost impossible to lose Iranian nationality, even by those born to Iranian fathers outside and never resident in Iran.99 Meanwhile, U.S. citizenship practice was otherwise becoming more tolerant, if not accepting, of dual citizenship. A series of U.S. Supreme Court cases had constrained the capacity of the U.S. government to terminate U.S. citizenship against an individual’s will.100 As a result, there were growing possibilities to maintain citizenship in more than one country, accompanied by active ties to each. This explains the large number of dual national claims in the Iran-U.S. process; it also explains the magnified complication of deciding which of two citizenships was the stronger. As in Mergé, these were cases in which individual claimants had substantial connections to both countries, in contrast to cases in which the connection to one nationality was merely nominal.101

Application of the dominant and effective test by the Iran claims tribunal garnered a mixed reception. David Bederman lamented the “very confused picture” that resulted from the cases.102 Bederman noted that “globetrotters”—those whose connections involved third countries as well as the United States and Iran—fared particularly badly.103 Many individuals who would clearly have self-identified as “American” did not qualify as such before the tribunal. This seemed particularly true of U.S.-born individuals who moved to Iran with Iranian spouses, as in Perry-Rohani.104 Although the tribunal took into account the full course of claimants’ lives, the focus on the short period between 1978 and 1981 gave rise to a kind of snapshot approach, pointing to dominant Iranian nationality even where the claimants were clearly binational.105 In the end, some observers questioned the utility of comparing national attachments.106

Critiques notwithstanding, the Iran-U.S. claims process proved a high-profile adoption of the dominant nationality test; in an era of few institutionalized arbitral processes implicating individual claimants, there were few competing voices on the question. Though rejected by the United Nations Compensation Commission (UNCC) established to address small claims arising out of Iraq’s 1991 invasion of Kuwait,107 the test was subsequently applied by the Eritrea-Ethiopia Claims Commission.108 Notably, it was accepted by the International Law Commission (ILC) in its draft articles on diplomatic protection, Article 7 of which provided, “A State of nationality may not exercise diplomatic protection in respect of a person against a State of which that person is also a national unless the nationality of the former State is predominant.”109 The ILC rooted its position in Mergé, abjuring Nottebohm’s focus on whether nationality should be considered “effective” for international purposes.110 “A tribunal considering this question,” the commentary continued, “is required to balance the strengths of competing nationalities.”111 The ILC position was broadly accepted as restating the prevailing understanding of customary international law in the context of diplomatic protection.112

IV. Dominant Nationality Dominant?

Enter the world of international investment agreements (IIAs), a dense web of agreements aimed at facilitating and protecting cross-border investment flows. At recent count, there are more than 2,500 such agreements in force, consisting of more than 2,200 bilateral investment agreements (BITs) and around 400 so-called “treaties with investment provisions” (TIPs), often denominated as free-trade agreements.113 The world of IIAs is entirely a post-World War II phenomenon.114 IIAs afford substantive protections to transborder investors, notably the guarantee of “fair and equitable treatment.”115 To an unprecedented degree, IIAs enable individual investors to make claims directly against states hosting investment. The resulting regime of investor-state dispute settlement (ISDS) has resulted in more than 1,300 investment treaty cases, all of which have been brought before arbitral bodies.116 An increasing number involve individuals who have nationality in both state parties to an investment treaty.

A few IIAs expressly bar dual nationals,117 and one major forum for dispute resolution, the International Centre for Settlement of Investment Disputes (ICSID), has since its founding in 1966 expressly precluded recognition of claims by dual nationals against either state of nationality.118 Otherwise, dual nationals face no categorical bar to sustaining claims under IIAs, and in this new and important context the dominant and effective test continues to enjoy significant traction.

Adoption of the dominant nationality test has followed two paths. First, the dominant nationality test has been expressly incorporated in some IIAs. With respect to claims brought under those agreements, tribunals put the test to work as a matter of treaty law, the relevant agreement serving as an enabling act under the international law doctrine of lex specialis.119 The U.S. has been particularly inclined to advocate the standard, including it in its model for bilateral investment treaties.120 Second, with respect to IIAs that are silent on the question of dual nationality, a number of tribunals have read the dominant nationality test into treaty terms as a principle of customary international law, as did the tribunals considering the Italy and Iran claims. Meanwhile, other tribunals have rejected the dominant nationality approach, admitting dual national claims without regard to the relative strength of national attachments. The variable practice demands further systematization.

A. Ballantine

The 2019 decision in Ballantine is perhaps the highest-profile arbitration to apply the dominant and effective standard.121 It implicated an intensive personal investment of both money and on-the-ground engagement by two native-born Americans in the Dominican Republic. The tribunal ultimately found the pair to have shifted their primary national affiliation to the situs of the investment, barring their claim to damages from the Dominican Republic under the Dominican Republic-Central America Free Trade Agreement (CAFTA-DR).122 The case highlights the difficulties of fixing dominant nationality in the age of globalization.

The Ballantines developed an upscale eco-friendly community marketed to expatriates in the mountains of the Dominican Republic. The couple began to purchase land tracts in 2003.123 A first phase was successfully executed by 2009, with infrastructure for almost a hundred home sites, though there were tensions with communities in surrounding areas.124 A second phase did not go smoothly, however, plagued by permitting issues with Dominican environmental regulators. Boundaries of a new national park were drawn to extend into the planned development area, and Dominican Republic authorities ultimately denied approvals for the project.125 In 2014, the Ballantines filed a claim under CAFTA-DR for $35 million in damages, alleging a breach of so-called national treatment and most-favored nation obligations (essentially a discrimination claim).126

The Ballantines asserted jurisdiction as covered investors under the agreement. CAFTA-DR required that claimants be an “investor of a Party that is a party to an investment dispute with another Party;” provided, however, that “a natural person who is a dual national shall be deemed to be a national of the State of his or her dominant and effective nationality.”127 The claimants argued that their U.S. citizenship was dominant and effective.128 Respondent Dominican Republic argued that, on the contrary, their Dominican citizenship was dominant and that as a result the tribunal was required to deny jurisdiction over the claim.129

The Ballantines’ involvement in the Dominican Republic began with a year spent in the country as Christian missionaries doing humanitarian work in 2000.130 They took up residence in the Dominican Republic after their initial land purchases in 2004, acquired permanent residence in 2006, and naturalized in 2010.131 They retained their U.S. citizenship and traveled frequently to the U.S., where they always maintained at least one residence.132 They maintained bank accounts and other investments in the U.S.; used their U.S. passports to secure admission to any country other than the Dominican Republic; voted in U.S. federal elections; and filed U.S. tax returns.133 In several years, Lisa Ballantine was physically present more often in the U.S. than in the Dominican Republic.134 Their four children attended an American school and then all attended college in the U.S. They also paid taxes and voted in the Dominican Republic and undertook various declarations, formal and informal, of attachment to the country. As of the time of the filing of the claim, they had lived in the Dominican Republic for eight years.

For the determination of which nationality was dominant, the tribunal turned to customary international law in the absence of prior rulings on the dual nationality provision in the specific context of the CAFTA-DR.135 From Nottebohm, the tribunal drew a sociological conception of nationality, from Mergé an application of the relativistic assessment of the strength of multiple nationalities. The tribunal elaborated “dominant nationality” to imply:

the notion of strength and precedence of one thing over another and that closeness between an individual and a State can indicate such attributes. In turn, closeness with a State and the strength of a nationality bond, could be the result of several factors in play such as the time spent by the individual in that country, family and personal attachments, language, education, work, economic or financial attachments, i.e. a cluster of elements that make up the life of an individual and that define several connections to a particular State. We understand “dominance” as referring to the degree or magnitude in which such connections are stronger than the connections that could have also been built by the individual in relation to another State that has also bestowed its nationality.136

As in Mergé and other cases, the tribunal centered the country of habitual residence as a central factor, “but not the only one.”137 The tribunal also highlighted “the circumstances in which the second nationality was acquired, the individual's personal attachment to a particular country, and the center of the person’s economic, social and family life” as driving the analysis.138 The tribunal pegged the time of the investment and the time of the filing of the claim as the temporal frame for the inquiry, undertaking a “holistic assessment . . . to examine how, at that particular time, the connections to both States could be characterized in terms of dominance and effectiveness.”139 The tribunal considered aspects of the investment itself to be relevant to the inquiry, “for example, the conduct of a particular State towards the investor, how the investor presented himself or herself, or the reason underlying the investor’s decision to apply for naturalization.”140

Applying these factors, the tribunal found the Ballantines’ Dominican citizenship to be dominant.141 While acknowledging that both claimants spent significant periods of time back in the U.S., the tribunal observed that the test could not be reduced to “a mathematical ‘day counting’ exercise.”142 The tribunal placed significant weight on the couple’s permanent residence in the Dominican Republic. With respect to personal attachment, the tribunal, on the one hand, highlighted statements from Lisa Ballantine about how difficult it had been to leave the Dominican Republic after the project’s termination, feeling a rejection by a country she had come to love. On the other hand, it accepted that the claimants had limited social and cultural connections to Dominican society. Conceding that personal attachment was “difficult to measure objectively,” it concluded that attachment to both countries “seem to be of equal force.”143 By contrast, the tribunal found the Ballantines to be more strongly connected to the Dominican Republic in terms of their economic, social, and family life.144

The tribunal centered the Ballantines’ naturalization as Dominicans to be a “key component” in finding it to be their dominant nationality. As part of the Dominican naturalization application, the claimants spoke of “feel[ing] very identified with the sentiment and Dominican customs since we have had a close bond of coexistence and respect.”145 In the litigation, the Ballantines had asserted that naturalization was motivated more by way of protecting their investment for estate purposes and other purposes as “a business decision.”146

While the Ballantines were evidently attempting to diminish the significance of naturalization as instrumental and non-affective, the tribunal turned this back against them. The tribunal drew from Nottebohm to the effect that “naturalization is not a matter to be taken lightly.”147 “In this Tribunal’s view, naturalization should not be equated to the purchase of a good or service.”148 Moreover, even allowing that the “sole reason for becoming Dominican and domestic investors was the investment,” the Tribunal found troubling the fact that the Ballantines became Dominican to protect their investment and were now disavowing that status to the same end.149 Once naturalized, Michael Ballantine held himself out as Dominican in business transactions, and “the nationality was key to their business.”150 Although they used their U.S. passports to exit the DR, they used their Dominican passports to enter. “[W]hile this Tribunal cannot deny the fact that during the relevant period, the Claimants maintained connections to the U.S. and this also seems natural from the fact that they were born and lived in that country for the majority of their lives,” concluded the Tribunal, “during the relevant period the Dominican nationality was effective and took precedence over the American nationality.”151

Arbitrator Marney L. Cheek dissented from the ruling.152 She took aim at the panel’s focus on the Ballantines’ use of nationality in the context of the investment and argued for considering the full course of claimants’ lifetimes rather than focusing on the period between the time of the alleged breach and the filing of the claim. Recounting each of the Ballantines’ relationships with the two countries, the dissent conceded that there was “no question” that the focal point of their daily lives shifted to the Dominican Republic when they moved there in 2006.153 Nevertheless, she argued, a holistic consideration of their lifetimes supported a finding that their U.S. citizenship was dominant. “That Mr. Ballantine chose Dominican nationality not necessarily for love of country and allegiance, but out of economic self-interest,” she observed, “does not lead to a conclusion that his dominant and effective nationality was Dominican on the critical dates. Mr. Ballantine’s economic ties to the Dominican Republic and his narrow reasons for seeking Dominican citizenship are but two of many relevant factors to be considered in this analysis.”154 She turned the majority’s finding around to question whether the Ballantines could have established treaty jurisdiction against the U.S. “It is difficult to imagine that if the Ballantines launched this hypothetical case against the United States under DR-CAFTA, such a case could move forward based on the dominant Dominican nationality of the Claimants.”155

B. A Case Study in Disarray

Ballantine presents the most notable contemporary application of the dominant and effective test in an arbitration involving dual nationals. There have been a wave of other cases involving investor claims by dual nationals. Some implicated, as did Ballantine, underlying treaty regimes incorporating the dominant and effective standard. In Carrizosa Gelzis, for example, individuals with U.S. and Colombian dual citizenship claimed compensation from Colombia for discriminatory treatment under the two countries’ trade promotion agreement,156 which provides that dual nationals are to be deemed exclusively nationals of the state of their dominant and effective nationality.157 Colombia conceded that the claimants’ citizenship in the U.S. was “effective,” it having been acquired at birth through their U.S. citizen mother. Taking a “holistic approach,”158 the tribunal nonetheless found their Colombian citizenship to be dominant in light of (among other factors) their habitual residence there, and family, business, and political connections to Colombia relative to the U.S.159 The tribunal rejected the claimants’ argument that they identified more strongly as Americans. “For all the Tribunal knows,” it observed, “their lives behind their front doors may be the embodiment of modern American family life, but as to that, the Tribunal has only the Claimants’ word and expression of subjective feeling.”160

Okuashvili v. Georgia, pursued under a U.K.-Georgia bilateral investment treaty, presented a closer case.161 The claimant there was Georgian by birth and British by recent naturalization. He resided in the U.K. with his immediate family, splitting his time 70-30 with Georgia; he maintained business and political interests in his homeland.162 On the way to finding his British nationality dominant, the tribunal highlighted the fact that the claimant “completed the naturalization process with the understanding that it would entrain the loss of his Georgian nationality” as a matter of Georgian law.163 Even though Georgian authorities did not withdraw his citizenship, the tribunal found it a “critical consideration” that the claimant exhibited a “readiness to forfeit his Georgian citizenship for the sake of being naturalized as a British citizen.”164

Okuashvili and other recent arbitrations have involved investment agreements which are silent on the standing of dual citizen investors. The panel in Okuashvili assumed without deciding that background understandings required that a claim brought by a dual citizen be cloaked in dominant nationality.165 The panel in Heemsen v. Venezuela went further, finding that the dominant and effective test, as a rule of international law, should be assimilated into a treaty not otherwise specifying the treatment of dual nationals.166 One commentator called the Heemsen award “even bigger than that of the Ballantine decision” insofar as it read the dominant and effective standard into a treaty as a matter of international law, in other words, as a default rule.167

Finally, a trio of recent claims against Venezuela demonstrates prevailing confusion on the question. In García Armas v. Venezuela [García Armas I], a panel found that a national of a party could sustain a claim against a state of which he was also a national without further qualification under a bilateral investment treaty between Spain and Venezuela.168 The tribunal found jurisdiction over a claimant who would clearly have failed to satisfy a dominant nationality analysis. Serafin García Armas had been born in 1944 in Spain with Spanish nationality. He relocated to Venezuela in 1961, at the age of 17, and naturalized as a Venezuelan in 1972, as a result of which he lost his Spanish nationality by operation of Spanish law. He was able to recover his nationality in 2004 in the wake of changes in both Venezuelan and Spanish law. Serafin García Armas brought his claim as a Spanish national under the Spain-Venezuela bilateral investment treaty. Venezuela asserted that the tribunal lacked jurisdiction, insofar the claimant made Venezuela the “center of his personal interests, family, economic and political, making full and exclusive use of said nationality, even when he had already recovered his Spanish nationality” in 2004.169 He married a Venezuelan national with whom he had three children in Venezuela, and had held himself out as exclusively a Venezuelan national in various contexts. Venezuelan nationality was, in short, the nationality “he holds in real life.”170 But the tribunal rejected Venezuela’s argument, refusing to qualify the express treaty term establishing jurisdiction for claims brought by individuals holding nationality in one of the party states.171 The tribunal thus denied application of the dominant nationality test, allowing the claim to proceed.172

On essentially the same facts in a case involving relatives of Serafin García Armas, however, another tribunal found no jurisdiction over a dual national claim under the same investment treaty. Expressly rejecting the decision on Serafin García Armas claim, the tribunal assessing the relatives’ claims found them barred on either of alternative grounds.173 It first interpreted the treaty to incorporate the categoric ICSID preclusion on claims by dual nationals against either state of nationality.174 Though an additional rationale was unnecessary, the tribunal also rejected jurisdiction on the grounds that the claimants’ dominant nationality was that of defendant Venezuela.175

Meanwhile, the decision in Santamarta Devis v. Venezuela, yet another case governed by the Spain-Venezuela bilateral investment treaty, denied a dual national’s claim solely on the basis of dominant nationality.176 Reciting the application of the test in Nottebohm, Mergé, and by the Iran-U.S. claims tribunal in the A/18 Case, the tribunal found “the principle of dominant and effective nationality is a principle of customary international law,”177 which it then incorporated into the BIT through background rules of treaty interpretation.178 The tribunal found claimant’s Spanish nationality to be “effective,” as opposed to merely “formal”; for some decades he maintained “close ties” to Spain,179 to which he had traveled frequently, in which he owned property, and where he had voted.180 The state of habitual residence, typically key in the determination of dominant nationality,181 was “of little importance,” given that claimant had lived in a third country (the United States) since 1989.182 Indeed, the tribunal concluded that the “center of plaintiff’s personal, family, and social life appears to be the United States, where he is not a national.”183

The Santamarta Devis tribunal rejected Venezuela’s proffer of the claimant’s birth and prior residence in Venezuela as establishing the dominance of Venezuelan nationality.184 The tribunal instead stressed his continuing economic ties to Venezuela. It also highlighted the fact that he had been forced to leave the country because of security concerns, not because of a “lack of attachment to the country, nor due to his attachment to Spain,” and for that reason had been unable to return to Venezuela.185 The tribunal suggested, in effect, that he would be residing in Venezuela but for the security situation there. Notwithstanding his long absence from the country, on that reasoning the tribunal found his Venezuelan nationality to be dominant, denying jurisdiction over the claim.

These recent decisions supply further evidence of decisional disarray on the questions of whether and how the dominant nationality test applies. As the Swiss court reviewing the decision in Santamarta Devis observed, “it thus appears that serious divergences exist on this controversial problem.”186 Arbitral decisions have been unpredictable, at least, with respect to dual national claims in investor-state dispute resolution.187 With the rise of dual citizenship, the question requires some resolution.

V. The Perils of Balancing Nationalities

Ballantine and other decisions applying the dominant nationality test show the pitfalls of assessing national attachment on a relative basis in a contemporary context. The facts of Ballantine exemplify a new world in which individuals can sustain national identities in a non-zero-sum fashion; the approach is flawed on its own terms. Some commentators, concerned with strategic behavior, advocate constraints on dual national claimants.188 But applying the dominant and effective test will trigger perverse unintended consequences. In an era in which “citizenship planning” has become a standard component of the global lives of ultra-high net-worth individuals,189 the wealthy will assimilate the Ballantine decision and adjust their behavior accordingly.

A. Behind the Curve on Dual Citizenship

The trend towards limiting the standing of dual nationals is curiously out of step with the growing acceptance of dual citizenship in every other context. Dual citizenship is dramatically on the rise.190 The growing acceptance of dual citizenship is not a mere toleration. There are very few contexts in which dual citizens suffer discrimination of any kind. In many cases, dual citizenship reflects genuine experiential and affective connections to each of two (or more) states of nationality. This is most obviously the case with respect to the children of mixed national parents, who will typically sustain emotional attachments to both parents and (in turn) assimilate their parents’ national associations. In a prior era during which dual nationality was aggressively policed, children were forced in many cases to choose (through the mechanism of “election”) between the nationality of their parents, an autonomy-suppressing practice whose abandonment should be welcomed.191 Likewise, states required immigrants to relinquish their birth nationality as the price of naturalization or terminated citizenship upon naturalization in another country. The latter practice was near universal among European states during the early and mid-twentieth century. Today, many fewer states make naturalization contingent on termination of prior nationality, nor do many states automatically terminate the citizenship of those who naturalize elsewhere.192 In short, dual citizenship has been normalized. States increasingly accept the status.193 Millions of individuals now hold dual citizenship, many of them by operation of law rather than by affirmative choice.

Those pressing “dominant nationality” are concerned about individuals gaming nationality to internationalize claims that should be domestic, in other words, that investors are seeking out nationalities to enable investor-state claims for which jurisdiction would otherwise not exist.194 But there is little evidence of such strategic behavior in the cases to date.195 In many of these cases, claimants were born with dual citizenship.196 In other cases, dual citizenship was clearly founded in the migration paradigm in which an individual held citizenship in their country of origin and country of resettlement.197

It is true that some acquire dual citizenship for a mix of sentimental and instrumental purposes, or for instrumental purposes alone. Additional citizenships often come with material advantages, most notably mobility privileges in the form of visa-free travel, while posing few costs.198 Investor citizenship supplies the clearest example of purely instrumental citizenship acquisition.199 One might then pose a scenario in which a wealthy investor (as most are in investor-state disputes, which typically involve millions of dollars in damages) simply buys citizenship in a state having a bilateral investment treaty with his original state of citizenship. To date, there appears to be no such example.200 Even if such a case were to transpire, it could be addressed under the “effective” prong of the “dominant and effective” test. That was Nottebohm itself—an individual who acquired nationality (in Liechtenstein) through financial means apparently for purposes of securing Liechtenstein’s protection in a property-related dispute,201 with the ICJ asking whether the individual had a “genuine link” to the country espousing his claim. The approach could also, perhaps, take care of cases in which the investor state nationality was extremely attenuated, as after-discovered through some distant ancestor,202 although those cases would hardly implicate “gaming” on the part of putative claimants insofar as one cannot generate ancestors after the fact.

But in cases in which both nationalities are clearly effective, individuals should have standing to pursue claims under either nationality against the other. García Olmedo argues that the test “should focus on establishing whether a dual national is ‘foreign’ enough, that is, sufficiently connected to the home State, to render ‘international’ a dispute with the respondent State.”203 That seems unworkable and ill-advised. Whether someone reads as “foreign” will run along a complex mix of objective and subjective dimensions, set against amorphous parameters of background national identity (necessary to determine who is “not foreign”). In any case, that isn’t what is implied by “dominant,” which seems almost mathematical in its implication. If you are 51% of one nationality, then that is your nationality for purposes of investor disputes. In this respect, translating the nationality analysis into whether a dual national is “foreign enough” is precisely what is so unsettling about Ballantine and other cases. The Ballantines were certainly foreign compared to their native-born Dominican co-nationals. Foreign enough? Maybe, maybe not.204

Santamarta Devis presents another recent case in which determination of dominant nationality seems abstract and unmoored. The panel conceded that the claimant’s life was centered in a non-party state, finding, in effect, that his dominant (non)nationality was American, where he had lived for decades.205 The tribunal was left deciding which of the state-parties came in second, Venezuela or Spain. The tribunal focused on his continuing economic ties to Venezuela in finding Venezuela to be his dominant nationality, as least relative to Spain. The tribunal’s conclusion might be defensible on that basis, but the result—that he had “dominant” nationality in a country which he had not visited for 30 years—seems hardly self-evident. He had traveled frequently to Spain, owned property there, and voted in Spanish elections.206 The tribunal reasoned that he did not live in Spain because he did not want to live there but did not live in Venezuela because he could not live there,207 in what was, effectively, a “I left my heart in Venezuela” line of reasoning. But if Santamarta Devis had sought exile in Spain instead of the U.S. the tribunal would no doubt have found Spain to be his dominant nationality on the habitual residence factor alone, even if he did harbor a stronger continuing sentimental attachment to Venezuela.

B. Gaming Dominant Nationality

Normalizing the dominant nationality test will inject uncertainty with respect to investment disputes involving a growing number of dual citizens. This uncertainty will be most visible in investor dispute settlement cases like the ones described here. But to the extent the test is consolidated, it could less perceptibly impact investment decision-making on the ground. Some with dual citizenship may be deterred from making the kinds of investment that the investment treaties are intended to facilitate, unsure whether they will enjoy treaty protections.208 Those who have large investment exposure may be advised by attorneys and others as to possible jurisdictional arguments against their assertion of treaty rights.

More likely, entrenching the dominant nationality approach will have would-be investors adjusting their citizenships. García Olmedo and others lament the nationality shopping that they claim would result from a less restrictive approach to dual national claims, that prospective claimants will engage in nationality shopping by way of enabling treaty jurisdiction.209 On the contrary, it is the dominant nationality approach that is more likely to incentivize strategic behaviors by way of avoiding uncertainty. Investors will abjure nationality in a state in which they have or are looking to make investments for which investment treaty protections supply a material backstop. One could see investors renouncing host-state citizenship, of which there are examples among recent cases.210

In other cases, one would see investors rejecting naturalization, even where eligible, in states where they have significant investment exposure. The Ballantines surely regret that they acquired Dominican citizenship. There was no significant legal advantage to naturalization for purposes of their development project,211 which they could have pursued as mono-Americans and for which there would then have been clear CAFTA-DR jurisdiction for the ensuing dispute. Assuming they had a claim to treaty damages on the merits, they lost it all by taking on Dominican nationality. The case will stand as a cautionary message to future investors, who will be well advised to avoid citizenship acquisition for fear that the citizenship of naturalization will be deemed “dominant” and disqualifying for investor dispute settlement purposes.

Indeed, in this respect, the approach may generate a back-to-the-future moment. During the nineteenth and early twentieth centuries, it was common for Europeans who established businesses in Latin America to retain their nationality of origin long after they moved the center of their sociological gravity to their places of business in the Western hemisphere.212 Nottebohm itself supplies an example. Nottebohm had lived in Guatemala for 34 years without securing Guatemalan citizenship,213 no doubt well aware that his German citizenship would shield him from depredations of the Guatemalan government.214 When, with the advent of World War II, his German nationality became a liability, he did not look to change his nationality to Guatemala, a country to which he clearly enjoyed a “genuine link,” but rather to another European state that would (and in the end attempted to) exercise diplomatic protection on his behalf. Leaving aside his fate as an enemy alien, Nottebohm’s experience was typical of American and European investors in Latin America.215

At points, Latin American states attempted to naturalize these foreign investors on the basis of residence alone, without their consent.216 Nationality conferred on a nonconsensual basis was rejected by arbitral tribunals before which Latin American states sought to assert claimant nationality as a defense.217 Nonconsensual naturalization was vigorously and in the end successfully opposed by the U.S. and European states.218 These episodes gave rise to one of the few clear constraints on state allocation of nationality, namely, that it cannot be ascribed on a non-volitional basis.219

That rule persists today. Host states can’t force naturalization.220 And by the terms of most investment treaties, host states can’t discriminate against non-citizens; that’s the point.221 To the extent dual citizenship implicates risk of forfeiting valuable treaty rights (a kind of insurance), investors will shun naturalization. This will be the case even if they have actually shifted their attachments to the country to which they may have moved and have invested, in other words, even if their sociologically more important connections are to that state. Nationality planning in that case completely defeats the object of the dominant nationality test, which measures attachment but only where an individual holds dual citizenship. From the investor’s perspective it will be more about uncertainty than any balancing of national connections. Even if an investor retains substantial attachments to her country of origin, such that it might seem still to be dominant, the test will seem malleable enough to deter acquiring host state citizenship.

Resolution of a modern-era claim brought under the North American Free Trade Agreement (NAFTA) demonstrates this point.222 The claimant in Feldman v. Mexico was born with citizenship in the U.S. and spent his first 33 years there before relocating to Mexico in 1973, where he had resided continuously for the 27 years preceding the filing of his NAFTA complaint.223 He had been married to two Mexican citizens and had four children born and thereafter resident in Mexico, and Mexico was the center of his substantial business activity.224 He acquired the Mexican equivalent of permanent residence status in 1991, but he never naturalized to Mexican citizenship. He owned eleven properties in Mexico, none in the U.S.225 In defending against the claim, Mexico argued that the tribunal lacked jurisdiction because Feldman was in effect a Mexican national and that his Mexican nationality was dominant. Quoting Nottebohm, Mexico asserted that “it is unquestionable that Mr. Feldman has ‘stronger factual ties’ with Mexico than the United States.”226

The tribunal rejected the attempt to equate Feldman’s permanent residence in Mexico with Mexican citizenship, this notwithstanding NAFTA treaty language that bolstered the argument.227 Observing that “under general international law, citizenship rather than residence or any other geographic affiliation is the main connecting factor between a state and an individual, ” the tribunal concluded that “dual nationality problems, including the search of the ‘dominant or effective nationality’, require the existence of a double citizenship.”228 In contrast to Nottebohm, where the ICJ found no genuine link to enable espousal of the claim by Liechtenstein, “here there is no doubt about genuine and regular conferral by birth of U.S. citizenship to Claimant.”229 Had Feldman acquired Mexican citizenship his claim would surely have been rejected.230 Any tribunal applying the dominant nationality test would have found his Mexican citizenship to supersede his American. Under the Ballantineanalysis, the question would not have even been close.231 Even though his sociological attachment was centered in Mexico, the lack of citizenship enabled the treaty claim against what was in effect his home country.232

To the extent non-naturalization is a consequence of adopting the “dominant and effective” test, it will come with societal costs. Ballantine is again instructive. The Ballantines were investors who did everything right. They cared about the community in the Dominican Republic in which they were investing. They integrated themselves to a great extent. That integration was evidenced and reinforced by acquiring Dominican citizenship, a token of faith in and commitment to the country in which they were devoting their entrepreneurial energies.233 Without naturalization they would have remained, symbolically at least, as outsiders and takers. Future investors of similar disposition will be well advised not to naturalize even though naturalization might have positive externalities in the host society.234 At an aggregate level, the incentive structure could facilitate the unnecessary return of a detached and alienated foreign investor class in developing economies.

C. Dual Citizens are Not the Problem

Validating jurisdiction over dual national claims also implicates more general concerns about the ISDS regimes and a forceful critique that the system constrains state regulation and community control.235 That critique is a powerful one. The evolution of the investor-state claims regime has prompted some states to withdraw from the investor treaties;236 as globalization magnified the scope of foreign investment, the treaty regime became, for some states, endemically constraining.237 But the critique seems weak as applied to the particular issue of dual national claims.238 However more common dual nationality has become, dual nationals comprise a small slice of the claimant pool. No one appears now to argue for a categoric ban on dual national claimants, moreover; assuming it ever in fact prevailed on the ground, the doctrine of “non-responsibility” is a historical artifact.239 It is only dual national claimants with putative dominant nationality in a host state that are at issue here. Making any set of investors eligible to press claims will increase the number of cases that states must defend against, but there is nothing about dual national investors that makes their claims qualitatively aggravating, especially once one concedes that some dual nationals (those with dominant nationality in other than the host state) can appropriately invoke treaty jurisdiction.240

Indeed, barring some dual nationals from treaty jurisdiction can be framed as imposing a discriminatory levy on citizenship.241 Citizenship conveys an expressive aspect. Rules which encumber the acquisition or maintenance of citizenship should be understood to burden the exercise of individual autonomy and the associational freedom.242 The clearest cases implicate individuals of mixed national parentage, an element in some arbitral decisions applying the dominant nationality test. Claimants in these cases acquired citizenship at birth in each of their parents’ states of citizenship; in most country pairings, they are able to keep both through adulthood. These claimants can hardly be charged with manipulating nationality for purposes of treaty claims. Rather, they are being blocked from pressing a treaty right as a result of having retained citizenship in the host state. Such individuals might, in an exercise of nationality planning, shed nationality in the state in which they are looking to maintain treaty protections. But that creates an incentive to disown the citizenship of a parent by way of vindicating those protections, in tension at least with emerging norms protecting dual citizenship in that context.243

The dominant nationality rule also discriminates among citizens, disadvantaging those with another citizenship. In the CAFTA-DR context, for example, most U.S. citizens will enjoy treaty jurisdiction while those deemed to have “dominant” dual citizenship in a host state do not. The Ballantines were disadvantaged as a result of their dual citizenship. There are other contexts in which such discrimination has been condemned.244 In fact there are very few respects in which dual citizens are disadvantaged under domestic legal regimes. Dual citizens are increasingly enabled to vote even as permanent external residents.245 Some states are stuck with constitutional bars on dual citizen officeholding.246 Otherwise there are few ways in which dual citizenship disadvantages individuals, which explains in part its increasing incidence.247

Finally, there are nontrivial administrative costs associated with the dominant nationality approach. Arbitration proceedings are expensive.248 As evidenced by the recent cases described above, the question of whether a dual citizen’s host state nationality is dominant (and whether the test applies in the first place) requires extensive litigation. The Ballantine arbitration involved costs of more than $6 million, some substantial proportion of which would have been applied to the nationality issue.249 Because the dominant nationality test is multi-factorial and because the stakes are high, parties will fully develop their positions. Abandoning the dominant nationality test would eliminate these costs by adopting a bright-line rule and eliminating dual national status as a factor in international arbitration proceedings.

A normative assessment thus weighs heavily against continued application of the dominant nationality test. The test should be abandoned, and dual citizens should be enabled to secure arbitral jurisdiction on the basis of their status as nationals of any state party. To the extent that citizenship is transparently gamed under such a situation, appropriate guardrails can be established through the concept of abuse of rights rather than through the imposition of wooden assessments of sociological national attachment.250

VI. Conclusion

Citizenship isn’t what is used to be. Historically, for most, nationality was an ascribed status fully coinciding with sociological identity.251 National communities were highly segmented, as distinct as the territorial borders by which they were spatially demarcated. Citizenship was a jealous institution. Today, citizenship is modulated, and national identities are fluid. One can be from here and there. Citizenship today is no longer a zero-sum institution.

State practice has largely caught up to realities on the ground. Most countries accept plural citizenship, and many have embraced it as an instrument of state policy. Millions of individuals now enjoy formal membership in multiple states in much the same way they can maintain multiple memberships in nonstate associations. Those holding dual citizenship rarely face any disadvantage as a result of the status. This is now nothing new. Dual citizenship has become  commonplace in the post-Cold War world.

International investment law, by contrast, has lagged behind the shift, perhaps the most notable area of law to resist obvious changes on the ground. Investment law has caught on to the extent it no longer categorically rejects dual national claimants. That stance, reflected in the prior doctrine of non-responsibility, would be unsustainable today—he disconnect would be too extreme—and in fact no one continues to support non-responsibility after decades of disuse. But change comes slowly to investment law. States and arbitrators have dug in with an approach that evinces a skepticism to dual citizens in dispute settlement fora; they can’t shake the idea that dual nationality is a stratagem rather than a status. Dual national claimants have to prove that a nationality isn’t just genuine but is also “dominant”, in a test that might be repackaged as having to demonstrate one’s “go to” citizenship as something other than the state hosting an investment.

But the new citizenship doesn’t work that way. As with most law that has decoupled from context, the “dominant nationality” test is unlikely to work as intended. Precisely because citizenship is now more manipulable than in the past, investors will develop workarounds which will defeat the purpose of the test. Investors will look to reduce uncertainty with respect to treaty protections. In many cases, they will be able to accomplish this by abjuring citizenship in a state whose nationality might be deemed dominant. The disconnect between the doctrine’s pretensions and its consequences supplies an important illustration of a reversible decision-making deficit in international investment law.

• 1Nottebohm Case (Liech. v. Guat.), Second Phase, Judgment, 1955 I.C.J. 4 (Apr. 6).
• 2Mergé Case (U.S. v. It.), 1955 R.I.A.A. 236 (Ital.-U.S. Conciliation Comm’n 1955).
• 3Iran v. U.S., Case No. A/18, 5 Iran-U.S. Cl. Trib. Rep. 251 (1984) [hereinafter A/18 Case].
• 4See infra note 108 and accompanying text.
• 5See infra notes 101–03,188–95 and accompanying text.
• 6Ballantine v. Dom. Rep., PCA Case Repository No. 2016-17, Final Award (Sep. 3, 2019).
• 7This uncertainty has drawn increased attention from members of the international arbitration bar. See, e.g., Gustavo Mata Morreo, Exploring Dual Nationality in International Law: From Historical Perspectives to Investor-State Arbitration, 42 J. Int’l Arb. 369 (2025) (describing recent activity); You Kexin, The Jurisdictional Maze: Dual Nationality in International Investment Arbitration, 21 Asian Int’l Arb. J. 67 (2025) (same); Davy Karkason, Dual Nationality in Arbitration: ICSID v. UNICTRAL Rules, Transnat’l Matters (May 10, 2024), https://perma.cc/3EW3-5D7K (noting “growing debate within the international arbitration community about whether there should be clearer guidelines or rules regarding dual nationality in arbitral tribunal proceedings”); Mangesh Krishna, French Courts Keeping the Door Open for Dual Nationals’ Claims?, Kluwer Arb. Blog (Dec. 30, 2023), https://perma.cc/DUD9-VWUA (“tribunals at a crossroads,” leading to a “precarious situation” of uncertainty for dual national claimants); Pablo Mori Bregante, The Passports’ Game: Chronicle of a Foretold Death for Dual Nationals’ Claims, Kluwer Arb. Blog (Jan. 20, 2020), https://perma.cc/4VWS-HGHH.
• 8See Javier García Olmedo, Claims by Dual Nationals Under Investment Treaties: Are Investors Entitled to Sue Their Own States?, 8 J. Int’l Disp. Settlement 695 (2017); Javier García Olmedo, Recalibrating the International Investment Regime Through Narrowed Jurisdiction, 69 Int’l & Comp. L.Q. 301 (2020); Chitransh Vijayvergia, Dual Nationality of a Private Investor in Investment Treaty Arbitration: A Potential Barrier to the Exercise of Jurisdiction Ratione Personae?, 36 ICSID Rev. 150 (2021) (arguing that test “impos[es] reasonable checks” and “provide[s] much-needed clarity on the subject”); see also David J. Bederman, Nationality of Individual Claimants Before the Iran-United States Claims Tribunal, 42 Int’l & Comp. L.Q. 119 (1993) (“As principles go, the test of dominant and effective nationality seems workable.”); Zachary Douglas, The International Law of Investment Claims 321–22 (2009) (advocating dominant nationality as default rule). But seeStavros Michalopoulos & Edward Hicks, Dual Nationality Revisited: A Modern Approach to Dual Nationals in Non-ICSID Arbitrations, 35 Arb. Int’l 121, 148 (2019) (suggesting a “more liberal approach might be warranted” with respect to dual national claimants); Andrés A. Mezgravis, The Arbitrary Deprivation of Dual Nationality in Investment Arbitration, 39 Arb. Int’l 549 (2023) (arguing that test violates human rights by denying recognition of a nationality).
• 9See, e.g., Audrey Macklin, Is It Time to Retire Nottebohm?, 111 AJIL Unbound 492 (2018); Rayner Thwaites, The Life and Times of the Genuine Link, 49 Vict. Univ. Wellington L.R. 645, 657 (2018) (genuine link doctrine “a dead letter”); Robert D. Sloane, Breaking the Genuine Link: The Contemporary International Legal Regulation of Nationality, 50 Harv. Int’l L.J. 1, 5 (2009) (Nottebohm “increasingly anachronistic and misplaced today”); Annemarieke Vermeer-Künzli, Nationality and Diplomatic Protection: A Reappraisal, in The Changing Role of Nationality in International Law 77 (Alexandra Annoni & Serena Forlati eds., 2013) (Nottebohm“has been criticized rather strongly, and rightly so.”); Peter J. Spiro, Nottebohm and ‘Genuine Link’: Anatomy of a Jurisprudential Illusion, in Investment Migration in Europe and the World 103, 113 (Dimitry Kochenov et al. eds., 2025) (“If Nottebohm’s ‘genuine link’ analysis made little sense in the context of the mid-twentieth century, it makes even less sense today.”); Jan Młynarczyk, Reassessing Nottebohm in an Era of Global Mobility, LSE L.R. (Sep. 25, 2024), https://perma.cc/H2TB-KYV7 (“fundamental flaw of the Nottebohm test is its ambiguity and vagueness, making it highly impractical to apply”). But see Luke Dimitrios Spieker & Ferdinand Weber, Bonds Without Belonging? The Genuine Link in International, Union, and Nationality Law, 43 Y.B. Eur. L. 56 (2025) (offering limited defense of Nottebohm in the context of investor citizenship).
• 10See infra III.B.
• 11The last 20 years has witnessed rich commentary assessing citizenship’s institutional meanings in the wake of globalization and other world-historic developments. For recent notable treatments by legal scholars and political theorists, see, e.g., Linda Bosniak, The Citizen and the Alien: Dilemmas of Contemporary Membership (2006); Christian Joppke, Citizenship and Immigration (2010); Dimitry Kochenov, Citizenship (2019); Ayelet Shachar, The Birthright Lottery: Citizenship and Global Inequality (2009); Rainer Bauböck, Democratic Inclusion (2017); see also, e.g., The Oxford Handbook of Citizenship (Ayelet Shachar et al. eds., 2017); Leah Bassel & Engin Isen, Citizenship Struggles: 25th Anniversary Special Issue, 26 Citizenship Stud. 361 (2022).
• 12See, e.g., Julian Arato, The Private Law Critique of International Investment Law, 113 Am. J. Int’l L. 1 (2019).
• 13See, e.g., Jean-Michel Marcoux, The Concept of Sustainable Development in Investment Arbitration: A Disconnect from Investment Policymaking and International Adjudication, 38 Leiden J. Int’l L. 501 (2025) (highlighting failure of international arbitrators to assimilate concept of sustainable development).
• 14See, e.g., Steven R. Ratner, International Investment Law Through the Lens of Global Justice, 20 J. Int’l Econ. L. 747 (2017). See also Álvaro Santos, International Investment Law in the Shadow of Populism: Between Redomestication and Liberalism Re‐Embedded, 11 Pol. & Gov. 203 (2023) (sketching institutional reform possibilities in the face of “crisis” in the international investment regime); Rob Howse, International Investment Law and Arbitration, in International Law and Litigation: A Look into Procedure 363, 363 (Hélène Ruiz Fabri ed., 2019) (“[I]nvestor-state dispute settlement (ISDS) has become the most controversial form of international litigation.”); Weijia Rao, Large Corporations and Investor-State Arbitration, 65 Harv. Int’l L.J. 133, 136–37 (2023) (noting “renewed backlash” against ISDS system in the face of COVID-related claims); Joseph E. Stiglitz, Regulating Multinational Corporations: Towards Principles of Cross-Border Legal Frameworks in a Globalized World Balancing Rights with Responsibilities, 23 Am. U. Int’l L. Rev. 451 (2007).
• 15See Susan D. Franck, The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law Through Inconsistent Decisions, 73 Fordham L. Rev. 1521, 1614 (2005) (calling for more rigorous scholarly analysis of arbitral jurisprudence by way of helping to “promote the legitimacy of investment arbitration”). For a recent example of a non-investment law expert bringing specialized knowledge to bear on an investment law issue, see Rachel Brewster, Arbitrating Corruption, 101 Wash. U. L. Rev. 923 (2024) (corruption-law scholar critique of corruption defenses in international arbitration).
• 16As U.S. diplomat George Bancroft observed in 1849, for example, states should “as soon tolerate a man with two wives as a man with two countries; as soon bear with polygamy as that state of double allegiance which common sense so repudiates that it has not even coined a word to express it.” Letter from Mr. Bancroft, Envoy Extraordinary and Minister Plenipotentiary of the United States, to Lord Palmerston, Principal Secretary for Foreign Affairs, Great Britain (Jan. 26, 1849), in 53 British and Foreign State Papers 639, 643 (1868). President Theodore Roosevelt dubbed the “theory” of dual nationality to be a “self-evident absurdity.” Theodore Roosevelt, When Is an American Not an American?, reprinted in Fear God and Take Your Own Part 291 (1916).
• 17See Peter J. Spiro, At Home in Two Countries: The Past and Future of Dual Citizenship 11, 11–21 (2016) [hereinafter Spiro, At Home in Two Countries]; Peter J. Spiro, Dual Nationality and the Meaning of Citizenship, 46 Emory L.J. 1411, 1419–30 (1997) [hereinafter Spiro, Dual Nationality].
• 18For major earlier historical treatments, see John Bassett Moore, History and Digest of the International Arbitrations to Which the United States Has Been a Party(1898); Edwin M. Borchard, Diplomatic Protection of Citizens Abroad or the Law of International Claims (1915). The rise of arbitral institutions was seen as facilitating the end of armed conflict. See, e.g., Joseph B. Moore, International Arbitration, 7 Mich. L. Rev. 547 (1909); Mary Ellen O’Connell, Arbitration and the Avoidance of War: The Nineteenth-Century American Vision, in The Sword and the Scales: The United States and International Courts and Tribunals 30 (Cesare Romano ed., 2009).
• 19Drummond’s Case [1834] 12 Eng. Rep. 492 (PC).
• 20Executors of R.S.C.A. Alexander v. United States (1872), reprinted inMoore, supra note 18, at 2529.
• 21Treaty of Paris art. 20, May 30, 1814, 63 Consol. T.S. 171, 192.
• 22Drummond’s Case, 12 Eng. Rep. at 499.
• 23He was, the court observed, “a French subject, domiciled in France, with all the marks and attributes of French character. He and his family had resided in France for more than a century; and the act of violence that was done towards him, was done by the French Government in the exercise of its municipal authority over its own subjects.” Id. at 500.
• 24Moore, supra note 18, at 2529.
• 25Treaty of Washington, U.K.-U.S., May 8, 1871, 17 Stat. 863. The treaty provided for provided for the hearing of claims by “subjects of Her Britannic Majesty, upon the Government of the United States, arising out of acts committed against the persons or property of subjects of Her Britannic Majesty.” Id. art 12.
• 26Moore, supra note 18, at 2531. In another case, Commissioner Frazer suggested that the proposition that either the United States or United Kingdom had “by the treaty recognized its responsibility to the other for injuries done to those who are by its laws its own citizens or subjects [is a] construction it seems to me is utterly inadmissible. I can not possibly bring myself to believe that either government intended any such thing.” Halley v. United States (1872), reprinted in Moore, supra note 18, at 2242.
• 27Borchard, supra note 18, at 588.
• 28SeeVermeer-Künzli, supra note 9, at 81; see also, e.g., American-Turkish Claims Settlement 14–15 (Fred K. Nielsen ed., 1937) (reporting State Department agreement with Turkish position that Turkey could refuse to hear claims by naturalized U.S. citizens who retained Turkish citizenship by birth).
• 29See Convention on Certain Questions Relating to the Conflict of Nationality Laws art. 4, Apr. 12, 1930, 179 L.N.T.S. 89 [hereinafter 1930 Hague Convention] (providing that “A State may not afford diplomatic protection to one of its nationals against a State whose nationality such person also possesses.”). The convention was ratified by only 20 states. See Convention on Certain Questions Relating to the Conflict of Nationality Laws, UN Treaty Collection, “Convention on Certain Questions Relating to the Conflict of Nationality Laws: Ratifications or definitive accessions,” https://perma.cc/WQY7-278S (last visited June 11, 2025). See also Int’l Law Comm’n, Draft Articles on Diplomatic Protection with Commentaries art. 7 cmt. 2, reprinted in Report of the International Law Commission on Its Fifty-eighth Session, at 43–44, U.N. Doc. A/61/10 (2006) (describing doctrine of non-responsibility including under Hague Convention).
• 30In fact, counsel for the British Crown in Drummond’s Case made such an argument, although it was not echoed in the decision of the Privy Council. Drummond’s Case [1834] 12 Eng. Rep. 492, 496 (PC) (arguments of the King’s Advocate) (when “a treaty speaks of the subjects of any nation, it must mean those who are actually and effectually under its rule and government”). See also Laurent v. United States (U.K. v. U.S.), 29 R.I.A.A. 11, 15 (U.S.-U.K. Comm’n 1854) (“where a treaty speaks of the subjects of any nation it means those who are actually and effectually under its rule and government.”); Zvonko R. Rode, Dual Nationals and the Doctrine of Dominant Nationality, 53 Am. J. Int’l L. 139, 140 (1959) (asserting that Drummond’s Case was the “first case in international law in which the doctrine of dominant or effective nationality was invoked”).
• 31Borchard, supra note 18, at 588–89. Borchard was himself an influential proponent of the theory of non-responsibility, a view which he successfully pressed on the Hague Codification Conference with respect to the closely related question of whether one state of nationality could exercise diplomatic protection against another state of nationality. See A/18 Case, 5 Iran-U.S. Cl. Trib. Rep. 251 (noting Borchard’s influence on Hague Convention). See also William L. Griffin, International Claims of Nationals of Both the Claimant and Respondent States—The Case History of a Myth, 1 Int’l Lawyer 400 (1967) (arguing that non-responsibility doctrine had no basis in arbitral decisions).
• 32Borchard, supra note 18, at 589. Express “election” was a mechanism many states applied to birthright dual nationals who were forced to choose one or the other upon attaining majority. See, e.g., Richard W. Flournoy, Jr., Dual Nationality and Election, 30 Yale L.J. 693 (1921); Spiro, Dual Nationality, supra note 17, at 1437–39. Other commission cases took this approach, finding a claimant to have lost one nationality or the other through a course of conduct, thus no longer implicating the difficulties of dual nationality. See, e.g., Deucatte’s Case, digested in Moore, supra note 18, at 2582 (finding individual to have lost French nationality by initiating U.S. naturalization process, thus disallowing possibility of claim as French national); H.F. van Panhuys, The Role of Nationality in International Law: An Outline 75 (1959) (noting this strategy to elide questions relating to dual nationality).
• 33Borchard, supra note 18, at 589; John Westlake, International Law 230–31 (1910).
• 34Lebret Case (U.S. v. Fr.) (1884), reprinted in Moore, supra note 18, at 2488. The claim was brought under the Convention Between the United States of America and the French Republic for the Settlement of Certain Claims of the Citizens of Either Country Against the Other, Fr.-U.S., Jan. 15, 1880, 21 Stat. 673.
• 35Moore, supra note 18, at 2497 (“aussi platonique”). As the American commissioner concluded, “She is really an American citizen and nobody can believe she will ever return to France.” Id. at 2506.
• 36Hammer & Brissot (U.S. v. Venez.) (1885), reprinted in Moore, supra note 18, at 2458–59.
• 37Canevaro Claim (It. v. Peru), 11 R.I.A.A. 397 (Perm. Ct. Arb. 1912), translated in pca-cpa.org, https://perma.cc/2VR9-QMTR. 
• 38Id. at 2.
• 39Id.
• 40This is perhaps not surprising. Claims involving property damage or death would be more likely to occur in and be lodged against states in which individuals were habitually resident, with few ties to the other state of nationality under whose aegis the claim would have been putatively enabled.
• 41Montfort v. Treuhänder Hauptverwaltung, 6 Trib. Arb. Mixtes 806 (1926).
• 42See generally Candice Lewis Bredbenner, A Nationality of Her Own: Women, Marriage, and the Law of Citizenship(1998); Peter J. Spiro, A New International Law of Citizenship, 105 Am. J. Int’l L. 694 (2011).
• 43Montfort, 6 Trib. Arb. Mixtes at 809.
• 44See 10 Annuaire de L’Institut de Droit International 25 (Th. Falk ed., 1889). See also Paul Weis, Nationality and Statelessness in International Law 170–71 (2d ed. 1979) (describing the episode).
• 45Montfort, 6 Trib. Arb. Mixtes at 809.
• 46See, e,g., Executors of R.S.C.A. Alexander v. United States (1872), reprinted in Moore, supra note 18, at 2529, 2531 (“the person who has embarrassed himself by assuming a double allegiance”).
• 47These norms were embedded in a discourse centering loyalty as a core element of the relationship of the individual to the state. See, e.g., John H. Wigmore, Domicile, Double Allegiance, and World Citizenship, 21 Ill. L. Rev. 761, 764 (1927) (“A state cannot suffer a divided allegiance. Its claim on the loyalty of its people must be exclusive.”).
• 48See, e.g., Nissim Bar-Yaacov, Dual Nationality 4–5 (1961) (characterizing dual nationality as detrimental to individual well-being and as a source of “mental conflicts”).
• 49See, e.g., Salem Case (Egypt/U.S.), 2 R.I.A.A. 1161, 1187 (1932) (in dicta, observing that “[t]he principle of the so-called ‘effective nationality’ . . . does not seem to be sufficiently established in international law.”). See alsoBorchard, supra note 18; van Panhuys, supra note 32, at 80 (asserting that, though “regrettable,” the non-responsibility doctrine “must as a general rule be considered to be the law”).
• 50Weis, for instance, suggested that the 1912 decision in Canevero “accepted the so-called principle of active or effective nationality, i.e., that in cases of plural nationality a person should be considered as having the nationality which in fact he exercises.” Weis, supra note 44, at 170.
• 51Nottebohm, 1955 I.C.J. 4.
• 52Mergé, 1955 R.I.A.A. 236.
• 53A/18 Case, 5 Iran-U.S. Cl. Trib. Rep. 251.
• 54As was highlighted in a dissenting opinion by Judge Read. Nottebohm, 1955 I.C.J. at 42 (Read, J., dissenting). See also, e.g., Josef L. Kunz, The Nottebohm Judgment (Second Phase), 54 Am. J. Int’l L. 536, 556 (1960).
• 55See J. Mervyn Jones, The Nottebohm Case, 5 Int’l & Comp. L.Q. 230, 233 (1956).
• 56See L.F.E. Goldie, The Critical Date, 12 Int’l & Comp. L.Q.1251, 1269 (1963).
• 57Nottebohm, 1955 I.C.J. at 23.
• 58The decision spoke of nationality as representing a “genuine connection” between the individual and the state. Id. Early assessments used this formulation. See, e.g., Oliver Lissitzyn, Nottebohm Case (Liechtenstein v. Guatemala), 49 Am. J. Int’l L. 396, 400 (1955). By 1960, the now more familiar “genuine link” formulation was in use. See, e.g., Kunz, supra note 54, at 552.
• 59Nottebohm, 1955 I.C.J. at 24.
• 60Id. at 25.
• 61Id. at 26.
• 62See supra note 9 (listing recent academic critiques of Nottebohm).
• 63Nottebohm, 1955 I.C.J. at 22.
• 64Id. at 24 (“At the time of his naturalization does Nottebohm appear to have been more closely attached by his tradition, his establishment, his interests, his activities, his family ties, his intentions for the near future to Liechtenstein than to any other State?”).
• 65Id. This notwithstanding the fact that Nottebohm subsequently established residence in Liechtenstein and by the time of the decision had lived there for over a decade. See Goldie, supra note 56, at 1272.
• 66See Opinion of AG Tesauro in Case C-369/90 Micheletti [1992] ECR I-4239, ¶ 5 (ascribing doctrine of effective nationality to a “‘romantic period’ of international relations”); Jules Lepoutre, Romantic Times? Nationality and European Citizenship, 55 Nordisk Socialrättslig Tidskrift 55 (2024).
• 67Mergé, 1955 R.I.A.A. 236.
• 68Id. at 237.
• 69Although the 1930 Hague Convention rejected the assertion of diplomatic protection against a state of nationality, see 1930 Hague Convention, supra note 29, it adopted a dominant nationality approach with respect to the treatment of dual nationals by third states. A person with dual nationality in another state would thus be “treated as if he had only one,” that either in which he was “habitually and principally resident, or the nationality of the country with which in the circumstances he appears to be in fact most closely connected.” Id. art. 5. A proposal to bar dual nationals from bringing “a personal action through an international tribunal or commission in respect of another State of which he is also a national”—a characterization matching the current ISDS system—was rejected as implicating “a case that is so rare as to be of little interest to the majority of States.” See Weis, supra note 44, at 187–88.
• 70Mergé, 1955 R.I.A.A. at 246 (emphasis added).
• 71Id. at 247 (emphasis added).
• 72Id. at 248.
• 73Nottebohm, 1955 I.C.J. at 26.
• 74Salvoni Case (U.S. v. It.), 14 R.I.A.A. 311, 312 (Italian-U.S. Conciliation Comm’n 1957).
• 75Id. at 312–13.
• 76Mazzonis Case (U.S. v. It.), 14 R.I.A.A. 249, 250 (Italian-U.S. Conciliation Comm’n 1955).
• 77Spaulding Case (U.S. v. It.), 14 R.I.A.A. 292, 293 (Italian-U.S. Conciliation Comm’n 1956).
• 78Ruspoli-Droutskoy Case (U.S. v. It.), 14 R.I.A.A. 314, 315 (Italian-U.S. Conciliation Comm’n 1957). See also Zangrilli Case (U.S. v. It.), 14 R.I.A.A. 294, 294 (Italian-U.S. Conciliation Comm’n 1956) (U.S. nationality of naturalized U.S. citizen “deemed as prevalent” even though he had at one point returned to Italy for period of fifteen years).
• 79See generallyNancy F. Cott, Marriage and Women’s Citizenship in the United States, 1830–1934, 103 Am. Hist. Rev. 1440 (1998); Bredbenner, supra note 42.
• 80U.N. Dept. of Econ. & Soc. Affairs, Nationality of Married Women 63, 12 E/CN.6/254/Rev.l (1955) (reprinting Act No. 555 of June 13, 1912) (“A foreign woman who marries a citizen acquires Italian citizenship.”).
• 81Married Women’s Independent Nationality Act, ch. 411, 42 Stat. 1021 (1922). On the Cable Act, see, e.g., Helen Irving, Citizenship, Alienage, and the Modern Constitutional State: A Gendered History 151–61 (2015).
• 82On the establishment of the tribunal, see Charles N. Brower & Jason D. Brueschke, The Iran-United States Claims Tribunal 3–25 (1998).
• 83A/18 Case, 5 Iran-U.S. Cl. Trib. Rep. at 260. The A/18 (rejecting Iran’s assertion of non-responsibility). Of the Hague Convention’s preclusion of diplomatic protection by one state of nationality against another: “Not only is [the rule] more than 50 years old and found in a treaty to which only 20 States are parties, but great changes have occurred since then in the concept of diplomatic protection, which concept has been expanded.” A/18 Case, 5 Iran-U.S. Cl. Trib. at 260–61.
• 84See Mohsen Aghahosseini, Claims of Dual Nationals and the Development of Customary International Law: Issues Before the Iran-United States Claims Tribunal 13 (2007) (highlighting that tribunal addressed more than 130 claims brought by dual nationals).
• 85A/18 Case, 5 Iran-U.S. Cl. Trib. Rep. at 265.
• 86SeeBederman, supra note 8, at 127.
• 87See Aghahosseini, supra note 84, at 151–53.
• 88Diba Case (Diba v. Islamic Republic of Iran), 23 Iran-U.S. Cl. Trib. Rep. 268 (1989), 1989 WL 663873, (Iran-U.S. Cl. Trib. 1989), at *3. Iranians were at the leading edge of international students attending American colleges and universities; many of the tribunal’s cases involved native-born Iranians who first moved to the U.S. for education. Tianjian Lai & Jeanne Batalova, Immigrants from Iran in the United States, Migration Policy Inst. (July 15, 2021), https://perma.cc/SUA3-CF9Z.
• 89Diba Case, 23 Iran-U.S. Cl. Trib. Rep. 268, at *3.
• 90In Khosravi v. Iran, for example, the claimant spent thirteen years in the U.S. as a young man, where he went to college, married a U.S. citizen, and had children before returning to Iran five years before the revolution. 1996 WL 1171806 (Iran-U.S. Cl. Trib. 1996).
• 91Perry-Rohani Case (Perry-Rohani v. Iran), 22 Iran-U.S. Cl. Trib. Rep. 194 (1989), 1989 WL 663866, at *4.
• 92Id.
• 93Id.
• 94See, e.g., James A. Bill, The Eagle and the Lion: The Tragedy of American-Iranian Relations (1988).
• 95See Hossein Askari et al., Iran’s Migration of Skilled Labor to the United States, 10 Iranian Stud. 3 (1977).
• 96At the time the hostage crisis unfolded, there were more than 50,000 Iranian students studying in the U.S. See Lai & Batalova, supra note 88; Narenji v. Civiletti, 481 F. Supp. 1132, 1146 (D.D.C. 1979) (upholding special reporting requirement targeting Iranian students).
• 97See, e.g., Iran Economic Gains Lure Students Home, N.Y. Times, Oct. 10, 1976, at 6.
• 98This notwithstanding the oath of naturalization under which new citizens are required to “renounce and abjure absolutely and entirely all allegiance and fidelity to any foreign prince, potentate, state, or sovereignty, of whom or which the applicant was before a subject or citizen.” 8 U.S.C. § 1448. The renunciation has never been enforced given the historical refusal of other states to allow native born subjects to terminate birth nationality. See Spiro, At Home in Two Countries, supra note 17, at 75.
• 99This remains the case today. See Amy Malek, Paradoxes of Dual Nationality: Geopolitical Constraints on Multiple Citizenship in the Iranian Diaspora, 73 Middle E.J. 531, 538 (2019) (renunciation of citizenship “effectively impossible” for Iranians).
• 100See Afroyim v. Rusk, 387 U.S. 253 (1967) (rejecting government’s authority to strip citizenship against an individual’s will); Vance v. Terrazas, 444 U.S. 252 (1980) (requiring specific intent of desire to relinquish citizenship in engaging in expatriating conduct). See also Peter J. Spiro, Afroyim: Vaunting Citizenship, Presaging Transnationality, inImmigration Stories 147 (David A. Martin & Peter H. Schuck eds., 2005) (recounting history of involuntary termination of citizenship). Federal law continues to designate various expatriating acts, including naturalization in another state, but only where they are performed “with the intention of relinquishing United States nationality.” 8 U.S.C. § 1481(a).
• 101See Aghahosseini, supra note 84, at 258 (highlighting that the typical claimant was “not a person with one effective and one technical nationality . . . He is, to the contrary, a person with two internationally effective nationalities.”)
• 102Bederman, supra note 8, at 129. See also Aghahosseini, supra note 84, at 258 (noting “structural and conceptual failings”). But see Sloane, supranote 9, at 28 (finding dominant nationality test “unobjectionable” in that context).
• 103Bederman, supra note 8, at 130–31.
• 104Perry-Rohani, 22 Iran-U.S. Cl. Trib. Rep. 194.
• 105See Aghahosseini, supra note 84, at 200 (many of the dual national claimants had “more or less equal ties to the two communities.”).
• 106Id. at 193 (questioning value of comparative assessment).
• 107In setting criteria for eligibility to pursue small claims, the commission denied jurisdiction over claims “on behalf of Iraqi nationals who do not have bona fide nationality of any other state,” implicitly validating jurisdiction over those holding alternate nationalities. Letter dated 2 August 1991 from the President of the Governing Council of the UNCC to the President of the Security Council, ¶ 17, S/22885 (Aug. 2, 1991), reprinted in Claims against Iraq: United National Compensation Commission, 86 Am. J. Int’l L.113, 116 (1992) (rejecting the rule of dominant and effective nationality and the “more complex issue of the effectiveness of nationality,” in favor of an approach that could be applied “without having to resort to lengthy procedures.”) See U.N. SCOR Comp. Comm’n Governing Council, Report and Recommendations Made by the Panel of Commissioners Concerning the Sixth Installment of Claims for Departure from Iraq or Kuwait (Category “A” Claims), ¶ 29 & n.9, U.N. Doc. S/AC.26/1996/3 (Oct. 16, 1996) [hereinafter Iraq Panel Report]. See also John Crook, The United Nations Compensation Commission—A New Structure to Enforce State Responsibility, 87 Am. J. Int’l L. 144, 150 (1993) (describing approach as departure from the “conventional practice” of dominant and effective nationality). The Commission found any non-Iraqi nationality secured before the adoption of the test in August 1991 to qualify as “bona fide,” on the theory that such acquisition would not likely have been acquired in bad faith for purposes of becoming eligible to claim compensation. See Iraq Panel Report, ¶ 30. The approach applied to claims under $100,000 only.   
• 108Eritrea-Ethiopia Claims Commission, Partial Award (Loss of Property in Ethiopia Owned by Non-Residents) (Eritrea’s Claim 24), PCA Case Repository No. 2001-02 (Dec. 19, 2005), ¶¶ 8–11 (citing adoption of dominant nationality test by the Iran-U.S. Claims Tribunal).
• 109Int’l L. Comm’n, Draft Articles on Diplomatic Protection with Commentaries art. 7, reprinted in Rep. of the International Law Commission on its Fifty-eighth Session, at 26, 34, U.N. GAOR, 61st Sess., Supp. No. 10, U.N. Doc. A/61/10 (2006) [hereinafter ILC]. 
• 110It also substituted “predominant” for “dominant,” on the grounds that the word “conveys the element of relativity and indicates that the individual has stronger ties with one State rather than another.” Id. cmt. 7.4. This formulation appears not to have been adopted by other actors.
• 111Id.
• 112See Vermeer-Künzli, supra note 9, at 81 (ILC “did not introduce Article 7 as an exercise in progressive development, since it did not change existing law.”). See also Weis, supra note 44, at 185 (noting “tendency to give weight to the individual’s factual attachment in solving positive conflicts of nationality,” which had emerged as the “most striking and uniform in the practice of international tribunals.”)
• 113According to U.N. Trade and Development, as of June 2025 there are 2,221 BITs and 405 TIPs in force. See International Investment Agreements Navigator, UNCTAD Inv. Pol’y Hub, https://perma.cc/73RY-8JTZ (last visited June 9, 2025).  Almost 90% of these agreements have been negotiated since 1990. See Investment Treaties, OECD, https://perma.cc/7LET-MP5H (last visited June 9, 2025).
• 114See Investment Treaties, supra note 113.(noting that first modern bilateral investment treaty was entered into in 1959).
• 115See, e.g., Douglas, supranote 8, at 141–61; Arato, supra note 12, at 19.
• 116See Total Number of Known Investment Treaty Cases Rises to 1,332, UNCTAD Inv. Pol’y Hub (Mar. 27, 2024), https://perma.cc/53SJ-7SX5.
• 117See, e.g., Agreement for the Promotion and Protection of Investments art. I(7), Can.-Venez., June 25, 1982, Can. T.S. 1998 No. 20 [hereinafter Canada-Venezuela Investment Agreement] (covering “any natural person possessing the citizenship of Canada in accordance with its laws . . . who makes the investment in the territory of Venezuela and who does not possess the citizenship of Venezuela”).
• 118Convention on the Settlement of Investment Disputes Between States and Nationals of Other States art. 25(2)(b), Mar. 18, 1965, 575 U.N.T.S. 159 (defining jurisdiction to include “national of another contracting state” but excluding any person who had nationality of the contracting state against whom a claim is filed). No other arbitration center specifically addresses jurisdiction over dual national claims in its rules. See Kexin, supra note 7, at 74. Importantly, many IIAs allow covered investors to choose the arbitration center in which to bring a claim. Dual nationals in those cases will not be subject to the ICSID bar, at least not directly.
• 119In effect, lex specialis works to trump customary international law. See, e.g., Rep. of the Study Group of the Int’l Law Comm’n, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, ¶¶ 56–57, U.N. Doc. A/CN.4/L.682 (Apr. 13, 2006) (recognizing the lex specialis principle that a more specific treaty provision prevails over a general one).
• 120U.S. Dep’t of State, 2012 U.S. Model Bilateral Investment Treaty 4 (2012) (including dual nationals within definition of “investor of a Party” but providing that “they shall be deemed to be exclusively a national of the State of his or her dominant and effective nationality”); Off. of the U.S. Trade Representative, 2004 Model BIT 4 (2012) (same). The dominant and effective nationality proviso was incorporated in the 2020 United States-Mexico-Canada Agreement. See Agreement Between the United States of America, the United Mexican States, and Canada, art. 14.1, Nov. 30, 2018, T.I.A.S. No. 20-0701 [hereinafter USMCA]. Canada has also incorporated the standard into its model BIT-equivalent, the Foreign Investment Promotion and Protection Agreement [hereinafter FIPA]. See Govt. of Canada, 2021 Model FIPA,https://perma.cc/2DZ3-ZAX2 (last visited June 9, 2025) (using same definition as U.S. Model BIT).
• 121Ballantine, PCA Case Repository No. 2016-17, Final Award.
• 122Dominican Republic-Central America-United States Free Trade Agreement, Aug. 5, 2004, 119 Stat. 462, 43 I.L.M. 514 [hereinafter CAFTA-DR].
• 123Ballantine, PCA Case Repository No. 2016-17, ¶ 58.
• 124Id. ¶¶ 89–98 (recounting issues created by restricted road access).
• 125Id.¶ 149.
• 126Id.¶ 278.
• 127CAFTA-DR, supra note 118, art. 10.28.
• 128Ballantine, PCA Case Repository No. 2016-17, ¶ 199.  The claimants averred that they should not be treated as dual nationals at all, having only U.S. citizenship at the time they made their investment in the Dominican Republic. See id. at 54. The tribunal rejected the argument, concluding that qualifying nationality must be demonstrated at the time the alleged breach occurred and the time the claim was submitted. Id. at 146.
• 129Id. ¶ 193.
• 130Id. ¶ 562.
• 131Id. ¶¶ 179–82, 206.
• 132Id. ¶ 564.
• 133Id. ¶¶ 210–11.
• 134Id. ¶ 565.
• 135Id.¶¶ 530–32.
• 136Id. ¶ 538.
• 137Id. ¶ 549 (quoting Mergé, 1955 R.I.A.A. at 247).
• 138Id. ¶ 552.
• 139Id. ¶ 556 (emphasis in original).
• 140Id. ¶ 554.
• 141Id. ¶ 597.
• 142Id.¶ 562.
• 143Id. ¶ 573.
• 144Id. ¶ 576.
• 145Id. ¶ 578.
• 146Id. ¶ 581.
• 147Id. ¶ 579.
• 148Id. ¶ 583.
• 149Id. ¶ 584.
• 150Id. ¶ 598.
• 151Id.¶ 597.
• 152Ballantine, PCA Case Repository No. 2016–17 (Cheek, M., partially dissenting).
• 153Id. ¶ 17.
• 154Id. ¶ 21.
• 155Id. ¶ 22. In this respect, the Ballantines were effectively stateless for purposes of protected investments, as least as between the United States and the Dominican Republic. A similar critique was leveled at Nottebohm. See Kunz, supra note 54, at 543 (observing that the ICJ’s judgment made Nottebohm “for all practical purposes, a stateless person.”).
• 156Carrizosa Gelzis v. Colom., PCA Case Repository No. 2018-56, Award (May 7, 2021).
• 157United States-Colombia Trade Promotion Agreement art. 10.28, Nov. 22, 2006, 46 I.L.M. 1048 (2007).
• 158Carrizosa Gelzis, PCA Case Repository No. 2018-56, ¶ 252.
• 159Id. ¶¶ 240–47.
• 160Id. ¶ 251.
• 161Okuashvili v. Georgia, SCC Case No. 2019/058, Partial Final Award (Aug. 31, 2022).
• 162Id. ¶¶ 55–66.
• 163Id. ¶ 157.
• 164Id. ¶ 161.
• 165Id. ¶ 152 (“conscious that answering [the question] in the context of one case may have far-reaching implications for other cases and other BITs.”). See alsoJavier García Olmedo, International Decisions: Okuashvili v. Georgia, 117 Am. J. Int’l L. 681, 684 (2023) (criticizing decision on this point).
• 166Heemsen v. Venezuela, PCA Case Repository No. 2017-18, Award on Jurisdiction, ¶¶ 436–40 (Oct. 29, 2019). The tribunal cited adoption of the test by the Iran-U.S. Claims Tribunal as authority. See id. ¶¶ 439, 441. The incorporation of the dominance test decided the case, insofar as it was “not seriously contested” that the claimants’ dominant nationality was that of Venezuela, against whom the claim was made.
• 167Pablo Mori Bregante, The Passports’ Game: Chronicle of a Foretold Death for Dual Nationals’ Claims, Kluwer Arb. Blog (Jan. 20, 2020), https://perma.cc/4VWS-HGHH.
• 168García Armas v. Venezuela, PCA Case Repository No. 2013-3, Decision on Jurisdiction (Dec. 15, 2014) [hereinafter García Armas I].
• 169Id. ¶ 65.
• 170Id. ¶ 67.
• 171Id. ¶¶ 198–200. The decision on dual nationals was upheld upon review by the Paris Court of Appeal. SeeDecision on Set Aside of Merits Award, ¶¶ 138–146 (Oct. 24, 2023). See also Bahgat v. Egypt, PCA Case Repository No. 2012–07, Decision on Jurisdiction, ¶¶ 224–227 (Nov. 17, 2017) (following García Armas I).
• 172See Eva Paloma Treves, Investment Treaty Arbitration: Dual Nationals Are Now Welcome: A Way Out of ICSID’s Dual Nationality Exclusion, 49 N.Y.U. J. Int’l L. & Pol. 607 (2017) (explicating tribunal reasoning in García Armas I).
• 173García Armas v. Venezuela, PCA Case Repository No. 2016-08, Award on Jurisdiction, ¶ 729 (Dec. 13, 2019) [hereinafter García Armas II].
• 174Id. ¶ 723 (concluding that the term “investor” in the Treaty incorporates the treatment of dual nationals in the ICSID system). See also Canada-Venezuela Investment Agreement, supra note 117 and accompanying text (describing express ICSID bar on dual national claims); Dawood Rawat v. Mauritius, PCA Case Repository No. 2016–20, ¶¶ 173–179 (Apr. 6, 2018) (finding implicit incorporation of ICSID bar on dual national claims).
• 175García Armas II, PCA Case Repository No. 2016–08, ¶¶ 734–37 (claimants’ dominant nationality “without question” was Venezuelan).
• 176Santamarta Devis v. Venezuela, PCA Case Repository No. 2020-56, Award (July 26, 2023). The decision rejected the argument, credited in García Armas II, that the treaty should be interpreted to incorporate the ICSID bar. See id. ¶ 465.
• 177Id. ¶ 486.
• 178See id. ¶¶ 485, 493 (incorporating principle by operation of Article 31.3(c) of the Vienna Convention on the Law of Treaties, which affords taking into interpretive account “relevant rules of international law applicable in the relations between the parties”).
• 179See id. ¶ 501.
• 180Id.
• 181See supra note 85 and accompanying text (discussing factors in dominant nationality determination).
• 182Santamarta Devis, PCA Case Repository No. 2020-56, ¶ 503.
• 183Id. ¶ 508.
• 184Id. ¶ 504 (following Mergé, to the effect that “the fact that you were born and raised in a certain country does not necessarily imply that that country be the dominant nationality”).
• 185Id. ¶ 514 (noting the tribunal made a distinction between “stating ‘I do not wish to visit the country’ and stating ‘the insecurity . . . led me to leave Venezuela [and] prevents me from visiting the country.”).
• 186A.____ v. Venezuela, Dec. 4A_466/2023, ¶ 5.4.2 (Swiss Fed. Ct., Feb. 6, 2025); see also Mladen Stojiljkovic et al., Swiss Court Confirms Applicability of Dominant Nationality Test to Fill Gap in Spain-Venezuela BIT, Kluwer Arb. Blog (Apr. 16, 2025), https://perma.cc/3ZPV-7LUQ. The case is particularly reminiscent of Iran-U.S. tribunal decisions involving “globe trotters,” those whose most important ties were to third countries. See Bederman, supra note 8, at 130–31.
• 187See, e.g., Krishna, supra note 7; Kexin, supra note 7, at 80 (noting “significant inconsistencies” in arbitral cases involving dual nationals); Javier García Olmedo, Dual Nationals in Investment Treaty Arbitration: An Emerging Field of Inconsistent Decisions, EJIL:Talk! (July 27, 2023), https://perma.cc/4UFE-DWTH. The inconsistency is particularly lamentable where it implicates conflicting interpretations of the same treaty, as has been the case with the Venezuela-Spain agreement. See Julian Arato et al., Parsing and Managing Inconsistency in Investor-State Dispute Settlement, 21 J. World Inv. & Trade 336, 338 (2020) (“[M]ost glaring cases of unjustifiable inconsistency are cases ‘where the same investment treaty standard or same rule of customary international law was interpreted differently in the absence of justifiable ground for the distinction.’”); see also Franck, supra note 15, at 1614.
• 188See infra note 194.
• 189See, e.g., Ayelet Shachar, Unequal Access: Wealth as Barrier and Accelerator to Citizenship, 25 Citizenship Stud. 543 (2021) (describing advent of “citizenship portfolios” among the “über-rich”).
• 190See Maarten Vink et al., The International Diffusion of Expatriate Dual Citizenship, 7 Migration Stud. 362, 369 (2019).
• 191See Borchard, supra note 18 (describing historical practice of election). Under the election regime, the nationality of the father tended to be favored. See Karen Knop, Relational Nationality: On Gender and Nationality in International Law, in Citizenship Today: Global Perspectives and Practices89, 112 (2001).
• 192See Spiro, Dual Nationality, supra note 17, at 1434–35 (describing historical state practice of terminating nationality upon naturalization in another state).
• 193See, e.g., Vink et al., supra note 190, at 369 (proportion of countries accepting dual citizenship increased from 47% in 1960 to over 70% in 2017).
• 194See, e.g., Olmedo, supra note 187 (“the nationality of natural persons has also become a manipulable category, something that can be instrumentalized.”); Vijayvergia, supranote 8, at 152 (lamenting “treaty shopping” under current regime); Kexin, supra note 7, at 71 (vaunting Nottebohm as way to avoid treaty-shopping and “opportunistic legal manoeuvres”). But see Treves, supra note 171, at 615 (finding “passport shopping” concerns to be “overblown”).
• 195See, e.g., Jorun Baumgartner, Treaty Shopping in International Investment Law 246 n.32 (2016) (author of exhaustive study “not aware of any case where a natural person has sought to access treaty protections through a strategic change of nationality not involving the invocation of a (pre-existing) dual nationality.”).
• 196See García Armas I, supra note 168 and accompanying text.
• 197See Ballantine, PCA Case Repository No. 2016-17, ¶ 534 (noting that claimant’s acquisition of Dominican nationality posed the possible loss, not gain, of access to treaty protection); see also Okuashvili, PCA Case Respository No. 2019-58, (noting that claimant migrated from Georgia to UK); Baghat, PCA Case Respository No. 2012-07 (involving claimant’s migration from Egypt to Finland).
• 198See generally Yossi Harpaz, Citizenship 2.0: Dual Nationality as a Global Asset (2019) (describing case studies of individuals acquiring second citizenship for instrumental purposes); David Cook-Martín, The Scramble for Citizens: Dual Nationality and State Competition for Immigrants (2013) (same); see also Christian Brown Prener & Thomas Gammeltoft-Hansen, Citizenship as Legal Infrastructure, 25 Ger. L.J. 1290, 1304 (2024) (“With respect to transnational mobility, the particular nationality status one holds has only become more central in recent years.”).
• 199See generally Kristin Surak, The Golden Passport: Global Mobility for Millionaires (2023) (describing rise of investor citizenship industry); see also Citizenship and Residence Sales: Rethinking the Boundaries of Belonging (Dimitry V. Kochenov & Kristin Surak eds., 2023).
• 200See Jose-Miguel Bello y Villarino, If Mr Nottebohm Had a Golden Passport: A Study of the Obligations for Third Countries Under International Law Regarding Citizenships-For-Sale, 9 Cambridge J. Int’l L. 76 (2020) (suggesting limitations in such cases should they arise). The International Law Students Association’s 2024 Jessup Moot Court problem involved the refusal of a state to allow the exercise of consular assistance by another state on behalf of an individual who had secured citizenship on the basis of investment. See 2024 Jessup Compromis and Corrections and Clarifications, ¶ 43 (Sep. 15, 2023), https://perma.cc/N5ZJ-JZ5W.
• 201Nottebohm, 1955 I.C.J. at 15 (describing naturalization payment of 37,500 Swiss francs along with a “deposit as security a sum of 30,000 Swiss francs,” in addition to various continuing annual fees).
• 202Akin to the case above involving the French noble with British ancestors. Drummond’s Case[1834] 12 Eng. Rep. 492 (PC) (denying espousal of claim of British subject whose family had emigrated from Great Britain more than a century before). Cf. Christoph H. Schreuer, The ICSID Convention: A Commentary 272 (2d ed. 2009) (noting that ICSID bar on dual nationals might not apply where putative claimant holds host state citizenship on the basis of descent “over multiple generations”).
• 203García Olmedo, Recalibrating the International Investment Regime Through Narrowed Jurisdiction, supra note 8, at 325. See also Kexin, supra note 7, at 71 (arguing Nottebohm “can help prevent treaty shopping”).
• 204For similar reasons, the dominant nationality test should be abandoned in the context of diplomatic protection. See ILC, supra notes 109–10 and accompanying text (describing ILC adoption of “predominant nationality” with respect to espousal of international claims). To the extent that dominant nationality continues to be accepted in the context of diplomatic protection, it is distinguishable from the context of international investment agreements. See Douglas, supra note 8, at 323 (“The rules for the nationality of claims in the general international law of diplomatic protection do not apply to issues of nationality in investment treaty arbitration.”); see also A/18 Case, 5 Iran-U.S. Cl. Trib. Rep. at 260 (distinguishing espousal of claims by states from claims made directly to arbitral tribunals by private individuals); Baghat, PCA Case Repository No. 2012-07, ¶ 229. Activity surrounding international investor agreements in any case is now dramatically more refined than that implicating diplomatic protection; it would be strange to have doctrine relating to diplomatic protection, which dates mostly to the twentieth century and before, to be determinative of ISDS doctrine on dual nationals. The ILC’s commentary, for example, does not cite a case subsequent to the Iran-U.S. Claims Tribunal decisions of the early 1980s. See ILC, supra note 109, art. 7 cmt. See also Mezgravis, supra note 8, at 560 (asserting that diplomatic protection “has lost its relevance . . . precisely because of the emergence and apogee of the investment protection system”). Finally, application of the dominant nationality test to diplomatic protection has expanded the admissibility of claims, opposed as it was to the doctrine of non-responsibility, under which injuries to dual nationals were categorically rejected as the basis of state claims. See Vermeer-Künzli, supra note 9, at 126 (arguing against doctrine of non-responsibility). In the ISDS context, by contrast, dominant nationality is put to work to restrict jurisdiction. But see Javier García Olmedo, Redefining the Position of the Investor in the International Legal Order and the Nature of Investment Treaty Rights: A Closer Look at the Relationship between Diplomatic Protection and Investor-State Arbitration, in Non-State Actors and International Obligations: Creation, Evolution and Enforcement 158 (James Summers & Alex Gough eds., 2018) (arguing that law of diplomatic protection should inform practice relating to investor treaties).
• 205Santamarta Devis, PCA Case Repository No. 2020-56, ¶ 509.
• 206If the dominant nationality test could be reduced in the typical case, in which only two states are implicated, to a question of where a claimant has 51% or more of her attachment, the exercise in Santamarta Devis reduces to comparing attachments at some lower level, an inherently more difficult undertaking. Cf. Vermeer-Künzli, supra note 9, at 123 (“In theory it is possible that an individual is protected by a state of nationality against a state of nationality where neither of the nationalities are particularly relevant in the individual’s life (for instance, because she lives in yet another state), but where the former nationality is still predominant.”).
• 207Santamarta Devis, PCA Case Repository No. 2020-56, ¶ 516.
• 208See Arato et al., supra note 187, at 342 (“[S]ystemic uncertainty can inefficiently raise the costs of doing business for all concerned, and potentially dampen FDI flows in the long run – precisely the opposite of what investment treaties set out to accomplish.”); Jonathan Bonnitcha et al., The Political Economy of the Investment Treaty Regime 145 (2017).
• 209See García Olmedo, Recalibrating the International Investment Regime Through Narrowed Jurisdiction, supra note 8 and accompanying text.
• 210See, e.g., Pey Casado v. Republic of Chile, ICSID Case No. ARB/98/2, Award, ¶¶ 307–22 (May 8, 2008) (recognizing effectiveness of claimant’s renunciation of nationality in defendant state); see also Schreuer, supra note 202, at 274 (allowing for renunciation of host state nationality as a path to establishing ICSID jurisdiction). Alternatively, prospective claimants could shed nationality in a host state by naturalizing in another state in cases in which host state law provides for loss of citizenship upon acquisition of another. Cf. Okuashvili, PCA Case Repository No. 2019-58 (claimant believed he had lost Georgian citizenship upon UK naturalization).
• 211The Ballantines did claim some advantage for estate purposes and for business dealings. See Ballantine, PCA Case Repository No. 2016-17, ¶ 71 n.69 (Nov. 9, 2017) (naturalization “was undertaken both to respond to market conditions . . . and demonstrate commitment to their investment, as well as to facilitate estate planning”), but these averments appear to have been foregrounded as part of the Ballantines’ litigation strategy to highlight an instrumental, as opposed to sentimental, motivation for their naturalization in the Dominican Republic, precisely to counter the allegation that they had shifted their affective identity from the U.S. to the Dominican Republic. See supra note 146 and accompanying text.
• 212See, e.g., Harmodio Arias, Nationality and Naturalisation in Latin America from the Point of View of International Law, 11 J. Soc’y Comp. Legis. 126, 142 (1910–11) (while minimizing the prevalence of imposed naturalization practices, acknowledging the desire of Latin American states to “shield themselves from the exorbitant claims of some of the powerful States when one of their subjects deems himself injured while he is established in Latin America”). See also John Dugard, Introductory Note: Articles on Diplomatic Protection, UN Audiovisual Libr. Int’l L.(2006), https://perma.cc/KD64-Z8WP (“Nationals of the Western Powers who flocked to Latin America to exploit its natural resources and to participate in its industrial development frequently found themselves in disputes with the unstable and volatile governments of the region over their personal rights or property rights. They then turned to their national States for protection which sometimes took the form of arbitration and sometimes the use of force.”)
• 213See Nottebohm, 1955 I.C.J. at 25 (noting that after 34-year residence, Guatemala “was the main seat of his interests”).
• 214See, e.g., Max Paul Friedman, Nazis & Good Neighbors: The United States Campaign Against the Germans of Latin America in World War II 24 (2003).
• 215There were a sufficient number of German nationals in Latin America that as war loomed in 1939 the German Foreign Office circulated a notice to Latin American posts recognizing the incentives for German nationals to acquire another nationality and instructing them to assure such individuals that they could reacquire German nationality at a later date. See Weis, supra note 44, at 179.
• 216For example, under an 1889 decree, Brazil declared all resident foreigners would become nationals unless they made a declaration to the contrary within six months of the decree. Brazilian Decree Regulating the Naturalization of Foreigners Residing in the Republic, reprinted in 81 Brit. & Foreign State Papers 233 (1889). As late as 1949, Argentina considered enacting a law under which all foreigners resident for longer than two years could avoid naturalization only by leaving the country. See also Maximilian Koessler, “Subject,” “Citizen,” “National,
  and “Permanent Allegiance”, 56 Yale L.J. 58, 75 n.100 (1946) (referencing proposed provision wherein failure to declare the retention of nationality of origin would have resulted in automatic designation of Mexican nationality).
• 217See Weis, supra note 44, at 110 (“the acquisition of a new nationality must contain an element of voluntariness on the part of the individual acquiring it . . . it must not be conferred against the will of the individual.”).
• 218See, e.g., John Bassett Moore, A Digest of International Law 308 (1906) (noting that, in response to Brazilian law, the U.S. Secretary of State asserted that “[t]o hold that the mere residence of an individual in a foreign country was conclusive evidence of his desire and intention to become one of its citizens would . . . involve an assumption of a most violent character.”).
• 219See Koessler, supra note 216, at 74 (“A general exception to the rule of domestic domain in matters of nationality law is represented by the prohibition of compulsory naturalization, which, according to textual authority, forms part of the prevailing customary international law.”); Myres S. McDougal et al., Nationality and Human Rights: The Protection of the Individual in External Arenas, 83 Yale L.J. 900, 920 (1974) (“To impose naturalization upon individual persons against their will . . . is incompatible with the commonly accepted principles of international law.”); Peter J. Spiro, Citizenship Overreach, 38 Mich. J. Int’l L. 167, 174–76 (2017).
• 220The ICSID Convention drafters explicitly considered the risk that states might impose nationality on a claimant by way of evading jurisdiction through application of article 25 and the bar on dual national claimants in ICSID proceedings. See Schreuer, supra note 202, at 272 (“[D]uring the Convention’s drafting, the problem of compulsory granting of nationality was discussed and the opinion was expressed that this would not be a permissible way for a State to evade its obligation to submit a dispute to [ICSID],” leaving assessments in particular cases to arbitrators.); Douglas, supra note 8, at 291 (concluding that if “an individual has been conferred nationality against his will, then the tribunal is by no means bound to give effect to that breach of international law”).
• 221See, e.g., Julian Arato, Introductory Remarks, The Once and Future Law of Non-Discrimination: Revisiting Most Favored Nation and National Treatment, 2018 Proc. Am. Soc’y Int’l L.112, 112 (2018) (“principle of non-discrimination lies at the core of international economic law”).
• 222North American Free Trade Agreement, Can.-Mex.-U.S., Dec. 17, 1992, 32 I.L.M. 289 (1993) [hereinafter NAFTA].
• 223Marvin Roy Feldman Karpa v. Mex., ICSID Case No. ARB(AF)/99/1, Interim Decision on Preliminary Jurisdictional Issues, ¶ 27 (Dec. 6, 2000).
• 224Id. ¶¶ 27–28.
• 225Feldman v. Mex., ICSID Case No. ARB(AF)/99/1, Counter-Memorial on Preliminary Questions, ¶ 101 (Sep. 8, 2000) (detailing business activities).
• 226Id. ¶ 111 (quoting Nottebohm, 1955 I.C.J. 4).
• 227The NAFTA agreement defined “national” to mean “a natural person who is a citizen or permanent resident of a Party.” See NAFTA, supra note 222, art. 201. The tribunal interpreted the inclusion of permanent residents to enable claims by citizens of third countries prosecuting claims against a NAFTA party, in other words, a permanent resident of a country other than one in which a putatively protected investment is made. Thus, for instance, the tribunal suggested that a French permanent resident of the U.S., though lacking U.S. citizenship, would have standing to make a claim against Mexico pursuant to the treaty definition. See id. ¶ 35. See also Douglas, supra note 8, at 323 (using same example). This approach has been explicitly adopted in the USMCA, which supersedes NAFTA. See USMCA, supra note 120, art. 14.1 (providing that “a natural person who is a citizen of a Party and a permanent resident of another Party is deemed to be exclusively a national of the Party of which that natural person is a citizen”).
• 228Feldman, ICSID Case No. ARB(AF)/99/1, ¶¶ 30–31.
• 229Id. ¶ 32. See also Sloane, supra note 9, at 34 (distinguishing Feldman from Nottebohm on grounds that Feldman’s citizenship was conferred under normal circumstances).
• 230Under the NAFTA regime, the application of ICSID rules would have barred the claim without resort to nationality balancing. See supra note 118 and accompanying text.
• 231Feldman filed U.S. tax returns and was registered to vote in San Antonio. Feldman, ICSID Case No. ARB(AF)/99/1, ¶ 27. Unlike the Ballantines, however, he had no other apparent continuing social, economic, or cultural ties to the United States.
• 232See Sloane, supra note 9, at 35 (characterizing Mexico’s position as asserting that Feldman had “de facto” Mexican citizenship).
• 233Evidence of such commitment was—not surprisingly, given the contours of the dominant nationality test—used against them in the arbitration. See Ballantine, PCA Case Repository No. 2016-17, Dominican Republic’s Statement of Defense, ¶ 47 (May 25, 2017) (enumerating ways in which the Ballantines evidenced “[p]ersonal attachment for the Dominican Republic”).
• 234This is particularly so given the weight the Ballantine panel ascribed to the act of naturalization. The tribunal agreed with Nottebohm’s characterization of naturalization as “‘not a matter to be taken lightly . . . creat[ing] a particular bond between the individual and the State.” See Ballantine, PCA Case Repository No. 2016-17, ¶¶ 582–83 (quoting Nottebohm, 1955 I.C.J. 4). The tribunal appeared to credit the act of naturalization as evidencing a shift away birth nationality, at least where coupled with other facts evidencing attachment to the state of naturalization.
• 235See Arato, supranote 12; see also J. Benton Heath, The Anti-Reformist Stance in Investment Law, 24 J. World Inv. & Trade 564 (2023) (“[T]here is a fundamental normative problem with the special rights for special people that lie at the heart of international investment law.”); Gus Van Harten & Anil Yilmaz Vastardis, Critiques of Investment Arbitration Reform: An Introduction, 24 J. World Inv. & Trade 363, 371 (2023) (eschewing “timid and gradualist reforms” and arguing for a radical reconception of international investment law in the face of global societal disruption).
• 236NGOs are supplying roadmaps for other states looking to undertake withdrawal. See, e.g., Ladan Mehranvar & Martin D. Brauch, Breaking Free: Strategies for Governments on Terminating Investment Treaties and Removing ISDS Provisions, Columbia Center on Sustainable Investment (Oct. 2024).
• 237See, e.g., Santos, supra note 14.
• 238If the “special rights” that come with ISDS protections have become, in Ben Heath’s formulation, “something like a free (or low-cost) casino token that investors can drop in a slot machine after something goes wrong,” see Heath, supra note 235, at 578, it’s not clear why dual citizens should be denied a token they would otherwise be entitled to.
• 239See e.g., Griffin, supra note 31.
• 240In this sense, my critique here is obviously reformist, taking the institutional assumptions of ISDS as a constant and arguing for a change within those given parameters. See Heath, supranote 235, at 578–79 (defining reformist approaches to international investment law).
• 241See Mezgravis, supra note 8, at 565–66 (arguing that dominant nationality test effects an arbitrary deprivation of nationality).
• 242See Peter J. Spiro, Dual Citizenship as Human Right, 8 Int’l J. Const. L. 111, 119–23 (2010).
• 243See Spiro, At Home in Two Countries, supra note 17, at 123. The 1997 European Convention on Nationality requires parties to allow individuals to retain dual nationality acquired at birth. See European Convention on Nationality art. 14, Nov. 6, 1997, C.E.T.S. No. 166 (“A State Party shall allow . . . children having different nationalities acquired automatically at birth to retain these nationalities”).
• 244For example, discriminatory impact on dual citizens was a major reason why France rejected a counterterrorism law which would have allowed for citizenship termination for dual citizens only. Kim Willsher, Hollande Drops Plan to Revoke Citizenship of Dual-National Terrorists, Guardian (Mar. 30, 2016), https://perma.cc/X4ML-CSNL. See also Laura van Waas & Sangita Janghai, All Citizens are Created Equal, but Some are More Equal Than Others, 65 Neth. Int’l L. Rev. 413 (2018) (highlighting discriminatory aspects of citizenship deprivation against dual citizens).
• 245See, e.g., Rainer Bauböck, Stakeholder Citizenship and Transnational Political Participation: A Normative Evaluation of External Voting, 75 Fordham L. Rev. 2393 (2006). The Ballantines continued to vote in the United States even after they naturalized in the Dominican Republic and voted in a presidential election there. See Ballantine, PCA Case Repository No. 2016-17, ¶ 216.
• 246Australia presents the most notable example, where dual citizens are barred under the country’s constitution from sitting in Parliament. See, e.g., Five Australian Politicians Ousted Over Dual Citizenship, BBC (May 9, 2018) https://perma.cc/V8FD-PGYP; Rod McGuirk, How Ban on Dual Citizen Lawmakers Vexes Aussies, A.P. News (Aug. 21, 2017) https://perma.cc/6Y7X-YANG (describing opposition to the constitutional bar). Other countries restricting office-holding by dual citizens include Jamaica, Bangladesh, Malawi, Nigeria, Pakistan, and Latvia. See Spiro, At Home in Two Countries, supra note 17, at 126–27.
• 247In the United States, the only disadvantage of dual citizenship is an increased difficulty securing a federal government security clearance. See Spiro, At Home in Two Countries, supra note 17, at 111–12 (describing cases). Of course, extending investor protections to dual citizens gives them something that fellow host-state citizens will lack, resulting in a different sort of inequality. If, for example, the Ballantines had been allowed to pursue their claim, they would have enjoyed a right denied their fellow Dominican citizens, resulting in inequality. But that kind of inequality is a commonplace at the international level. For example, the Ballantines had a right to vote in U.S. elections denied to their mono-national Dominican compatriots, the sort of inequality that is normatively defensible because it is not located within the confines of the state. In other words, the inequality analysis should be undertaken within polities as if everyone held one citizenship only. See Peter J. Spiro, Political Rights and Dual Nationality, in Rights and Duties of Dual Nationals: Evolution and Prospects 135 (David A. Martin & Kay Hailbronner eds., 2003) (concluding that dual citizen voting does not violate equality norms).
• 248See Susan D. Franck, Rationalizing Costs in Investment Treaty Arbitration, 88 Wash. L. Rev. 769 (costs represent 10% of even partial awards).
• 249Ballantine, PCA Case Repository No. 2016-17, ¶¶ 499–502 (allocating costs).
• 250See Sloane, supra note 9.
• 251See, e.g., Rogers Brubaker, Citizenship and Nationhood in France and Germany34 (1992) (“Citizenship is both an instrument and an object of closure.”).