Nottebohm Under Attack (Again): Is it Time for Reconciliation?

Written by


The Nottebohm judgment has recently come under attack in the context of the European Commission’s position on investment by citizenship (CBI) schemes, also known as “golden passport” programmes. These schemes allow individuals to obtain a second citizenship in a host country in exchange for financial investments or even just a flat fee. On 20 October 2020, the Commission launched infringement proceedings against Malta and Cyprus over their CBI schemes, and, on 9 June 2021, urged those States to stop ‘“selling” EU citizenship’. In its press release, and with reliance on Nottebohm, the Commission observed that ‘the granting of EU citizenship for pre-determined payments or investments without any genuine link with the Member States concerned’ violates the principle of sincere cooperation and undermines the essence of EU citizenship (See also Commission’s Report on Investor Citizenship, at 5 and 6).

The Commission’s reference to Nottebohm has triggered intense normative critique. A number of scholars have fiercely opposed the proposition that a Member State must only grant citizenship to individuals who have genuine links with it. The most salient criticism comes from Peter Spiro and Dimitry Kochenov. Spiro labels Nottebohm as a ‘jurisprudential illusion’, arguing that ‘there may be no other judgment of an international tribunal which has had so much purchase on the imagination’ (Spiro, at 35). For his part, Kochenov considers that ‘Nottebohm is unquestionably bad law and the Commission was obliged to know this to be the case’ (Kochenov, at 23).

These quotes capture the traditional approach towards the 1955 ICJ ruling, which has been rejected by a consensus of legal scholars since its inception. This blog post respectfully challenges this conventional wisdom. I submit that Nottebohm has been as criticised as it has probably been misunderstood. The ICJ did not, as critics generally argue, depart from international law on nationality nor did it seek to create an international rule based on a genuine link requirement. A closer look at the majority’s reasoning reveals that, in fact, this decision is not ultimately about genuine links, but rather about preventing the mis(use) of nationality. Lastly, Nottebohm is currently present in other overlooked areas where nationality serves a transnational purpose, such as international investment law. These arguments will be briefly discussed in turn.

1. Nottebohm reaffirms international law in the field of nationality

Far from being ‘discontinuous with pre-existing law’ (Spiro, at 24) or establishing a ‘harmful rule of international law’ (Kochenov, at 2), the decision adhered to a number of international rules established in the realm of nationality. First, the ICJ reiterated the Hague Convention formula: ‘[i]t is for each State to determine under its own law who are its nationals’ (Art 1 Convention on Certain Questions relating to the Conflict of Nationality Laws, 1930). Originally a German citizen, Nottebohm immigrated to Guatemala in 1905. After the outbreak of the Second World War, he became a Liechtenstein national by naturalisation in 1939, thereby losing his German nationality. The Court respected Liechtenstein’s sovereignty, holding that ‘[i]t is for Liechtenstein to confer its nationality by naturalization’ in accordance with its legislation (at 20).

Second, the Court recognised that State sovereignty on nationality law is not unlimited. The Court relied, once again, on the Hague Convention and rightly noted that national law on the attribution of nationality ‘shall be recognized by other States insofar as it is consistent with international conventions, international custom and the principles of law generally recognized with regard to nationality’ (at 23). In the words of the International Law Commission (ILC), on which critics heavily rely to sustain their views, ‘[a]lthough a State has the right to decide who are its nationals, this right is not absolute’ (ILC Draft Articles on Diplomatic Protection, 2006, at 32). This limitation derives from the fact that the attribution of nationality may carry consequences on the international plane, that is, with respect to other States. Accordingly, it is international law that determines whether a given nationality should be recognised by, or opposable to, other States. On this basis, the ICJ correctly held that the ‘international effect’ of the lawfully acquired Liechtenstein nationality, i.e. its opposability to Guatemala, was a matter of international law (at 21).

Third, the Court went on to determine the only question it was bound to answer, namely ‘whether the nationality conferred on Nottebohm can be relied upon as against Guatemala’ for diplomatic protection purposes (at 17). The Court noted that ‘[i]nternational arbitrators’ have answered the very same question in ‘numerous cases of dual nationality’ by resorting to the ‘real and effective nationality’ test (at 22). This test, the Court rightly observed, requires showing ‘stronger factual ties’, also (fortuitously) referred to by the Court as ‘genuine connections’, between the person concerned and the State whose nationality is invoked (at 22 and 23). This test, more commonly known as the rule of dominant and effective nationality, has been widely recognised as a rule of customary international law (see ILC Draft Articles, 2006, Art 7). In applying the rule, the Court found that, since Nottebohm’s ‘actual connections with Liechtenstein were extremely tenuous’ when compared to the existing ‘long-standing and close connection between him and Guatemala’, Liechtenstein nationality should not be attributed international effect (at 25 and 26).

It is difficult to consider this approach as ‘bad law. The Court fully followed ‘pre-existing’ accepted rules of international law to determine the question that fell within the scope of its mandate. As such, the Court did not, nor did it intend to, create a ‘genuine link requirement’ as a condition for the international recognition of nationality, let alone as a condition for the attribution of nationality. If anything, one could argue, as critics do, that the ICJ erred in applying the rule of real and effective nationality to a factual scenario that was distinct from that of the cases where the rule normally applies. Indeed, Nottebohm was not a dual national (that is, a national of both States party to the dispute). He only held the nationality of the claimant State, a fact that the Court also acknowledged. This does not, however, make the decision wrong as a matter of law.

2. Nottebohm is not ultimately about ‘genuine links’

At any rate, a closer reading of the majoritys reasoning may tell critics that Nottebohm is in fact not ultimately about genuine links. In this respect, Robert Sloane offers a more convincing and, often overlooked, reading of the ICJ’s holding. He persuasively argues that Nottebohm should ‘be read as a narrow decision in which the ICJ tacitly invoked a general principle of law, viz., abuse of rights, to prevent what it saw as a manipulative ascription of nationality’ (Sloane, at 1). In his view, therefore, the language expounding the genuine link requirement was dicta and had little to do with the Court’s concerns.

A detailed description in the judgment of the way in which Nottebohm acquired and invoked his Liechtenstein nationality attests to this reading. The Court noted that Liechtenstein nationality ‘was conferred in exceptional circumstances of speed and accommodation’ (at 26). In the Court’s view, the ‘sole aim’ of Nottebohm to acquire this nationality was ‘to substitute for his status as a national of a belligerent State that of a national of a neutral state’ (at 26). This objective was twofold; first, to circumvent the application of the international law of war relative to enemy aliens; and second, as that first effort failed, to enable Liechtenstein to bring a diplomatic protection claim on his behalf. The newly acquired nationality was, in other words, one of convenience.

As Sloane notes, ‘the majority saw this as a clear abuse of rights’ and thus decided that ‘Guatemala is under no obligation to recognize a nationality granted in such circumstances’ (at 20). Space precludes an examination of the abuse of rights principle. Suffice it to say that, conceived as an important particularisation of the principle of good faith, abuse of rights is overwhelmingly accepted as a general principle of law or as part of customary international law. The principle requires, inter alia, that ‘every right […] be exercised honestly and loyally. Any fictitious exercise of a right for the purpose of evading either a rule of law or a contractual obligation will not be tolerated’ (Bin Cheng, at 123). This is what Nottebohm seemingly did by exercising his right to acquire Liechtenstein’s nationality to bypass a rule of law (that of war) and as a way to expand Liechtenstein’s right to bring a diplomatic protection claim.

Whichever interpretation of the decision is the most compelling, Nottebohm keeps surfacing in international legal practice, making the ruling a relevant and effective tool for the modern international regulation of nationality.

3. Nottebohm is present in contemporary international legal practice

Opponents of Nottebohm also consider that, as ‘individuals become more highly mobile and are enabled to maintain multiple citizenships’, ‘the prospect of sorting supposedly authentic citizenship from instrumental citizenship is a fool’s errand’ (Kochenov, at 70). Contrary to this view, however, recent developments show that the criteria used by the ICJ to establish ‘authentic citizenship’ are influencing some areas of international law, including investment treaty arbitration and cases involving claims by dual nationals.

As I discussed in a previous blog post, claims by dual nationals have continued to increase and tribunals have already rendered a number of decisions. Certain tribunals have considered that, whether as custom or treaty law, the rule of dominant and effective nationality should apply to ascertain the legal standing of dual nationals. In these cases, tribunals considered Nottebohm as relevant authority to determine the dominant and effective nationality of the investor.

To mention the most recent example, in Carrizosa v Colombia the Tribunal followed Nottebohm by observing that it had the discretion to ascertain ‘whether full international effect was to be attributed to the nationality invoked in the context of the exercise of protection’ under the applicable investment treaty (¶ 183). The Tribunal then held that ‘the approach set out by the ICJ in the Nottebohm case’ should apply to determine ‘which of the two nationalities asserted by the claimants is the predominant’ (Carrizosa v Colombia, Award, 7 May 2021, ¶ 184). The Tribunal found that ‘[a]s the ICJ observed in Nottebohm, different factors fall to be taken into consideration’ (¶ 193), including ‘habitual residence […] family ties, his participation in public life, attachment shown by him for a given country and inculcated in his children, etc’ (¶ 183, citing Nottebohm at 22). After examining the factual ties between the claimants and their States of nationality, the Tribunal ruled that their dominant nationality was that of the respondent State, Colombia, and thus it declined jurisdiction to hear the dispute. 

The proposed rationale for Nottebohm in terms of abuse of rights is also present in investment jurisprudence that addresses the manipulation of corporate nationality. The seminal case here is Philips Morris v Australia. The claimant, Philip Morris (PM) Asia Ltd, acquired all shares of PM Australia to become a protected investor under the Australia–Hong Kong bilateral investment treaty (BIT). This BIT determines corporate nationality merely by reference to the State of incorporation. The strategic restructuring, which made PM Asia a prima facie qualified Hong Kong national, took place when the dispute with Australia was clearly foreseeable. The Tribunal declined jurisdiction on grounds of abuse or right, holding inter alia that the ‘main and determinative, if not sole, reason for the restructuring was the intention to bring a claim under the BIT’ (Philips Morris v Australia, Award, 17 December 2015, ¶ 584). Similarly, Liechtenstein nationality was acquired to avail Nottebohm of international protection through Liechtenstein’s espousal of the claim. 

On a final note, the policy considerations that led to the tacit application of the abuse of rights principle in Nottebohm also resemble contemporary arguments for regulating ascriptions of nationality through CBI schemes. Whilst individuals may use these schemes for a number of legitimate reasons (e.g. greater mobility thanks to visa-free travel), they can also create very specific risks. For instance, the Commission notes that CBI programmes enable tax evaders to ‘circumvent the reporting obligations laid down by the EU legislation on automatic exchange of financial account information’ (Commission’s Report, at 17). If materialised, this risk can lead to the type of non-bona fide acquisition of Liechtenstein nationality in Nottebohm, which was also acquired to evade a rule of the law of war. By implication, an argument can be made that if an individual acquires a given nationality through CBI schemes to, for instance, avoid compliance with EU legislation, such an ascription may be covered under the abuse of rights doctrine.


Scrutiny of Nottebohm suggests that current resentments towards the decision are misplaced. The ICJ did not fabricate a novel doctrine of international law in the field of nationality. Rather, the Court rightfully resorted to an existing rule of customary law to determine whether the legitimate exercise of national law by a State should engraft obligations under international law on another State. The rationale behind the ruling was to prevent abusive nationality practices. 

Irrespective of whether one joins the critical chorus or not, it is hard to deny that Nottebohm continues to have salience in areas of international law like investment treaty arbitration. An increasing number of decisions in this field shows, as did Nottebohm in 1955, that the conferral of nationality serves a transitional purpose, cementing the protection of rights beyond domestic borders. This case law reinforces the principle that, whereas the conferral of nationality is a matter of internal law, its international legal functions are a matter of international law.

So perhaps the moral left by the ICJ is that, considering its cross-border dimension, nationality should no longer be confined to the “reserved domain” of national legal competence. Sooner or later, a discussion about the instrumentalization of nationality at the international plane must grapple with efforts to codify the international legal regulation of nationality. In the meantime, I hope that the reflections shared here invite the opponents of Nottebohm to reconsider its intrinsic value and inspire conversation about reconciliation.

Print Friendly, PDF & Email

Leave a Comment

Comments for this post are closed


José-Miguel Bello y Villarino says

December 10, 2021

Thank you for the post. Glad to see another voice in this ring (and one reading Nottebohm properly!)

For those interested in the topic and an analysis on how Nottebohm plays out in the EU context, you can find my take in "If Mr Nottebohm had a golden passport: a study of the obligations for third countries under international law regarding citizenships-for-sale", published last year on the Cambridge International Law Journal (including the insightful analogy with Philips Morris v Australia!).

Tamás Hoffmann says

December 11, 2021

Dear Dr. Olmedo,

With respect, I'm afraid that you commit exactly the same mistake as the ICJ when you analyze the Nottebohm case as if it was a dual nationality situation.
In reality, Nottebohm had already given up his citizenship more than 15 years before the judgment and 4 years before his arrest in Guatemala so he had no other state to request international protection from than Liechtenstein. In effect, the ICJ decision meant that no single country in the world could provide him diplomatic protection.

All the caselaw cited by the Court in defence of its position pertained to dual nationality cases, just like the recent Carrizosa v Colombia case invoked by you.

Given the fact that Nottebohm's case was not actually a dual nationality situation, do you still think that the Court was justified in treating it as such?

P.S. The Nottebohm jurisprudence seeped into international criminal law as well and led to the Yugoslavia Tribunal declaring that in an inter-ethnic conflict, people of different ethnicity can be treated as foreign nationals for the sake of application of Geneva Convention IV. I think it is a horrible and potentially dangerous approach but it still seems to be relatively unchallenged.

Javier Garcia Olmedo says

December 12, 2021

Thank you for your comment, Tamás. I am aware that Nottebohm was not a dual national and that he was left stateless by the ICJ for international law purposes. I mentioned that in the post, a fact that was acknowledged by the court. My analysis however focuses on the court's reasoning. I am not judging the side effects of the decision, in other words. My argument is that the ICJ, as any other international court or tribunal, enjoys discretion to determine the international recognition of nationality. The Court answered the question whether Nottebohm's sole nationality should be opposable to Guatemala.

Dapo Akande says

December 13, 2021

Dear Javier,

Thanks for a very interesting post. I am curious about why you say that "The Court fully followed ‘pre-existing’ accepted rules of international law to determine the question that fell within the scope of its mandate." You accept that those earlier rules relate to dual nationality cases, whereas Nottenbohm was not such a case so why was this a case of fully following the earlier rules?

Many thanks!


Tamás Hoffmann says

December 13, 2021

Dear Javier,

I think you have misunderstood my comment but Dapo had already clarified it. So could you please tell me why you think that it is irrelevant that Nottebohm's was not a dual nationality case when the ICJ only quoted dual nationality cases to justify the opposability of citizenship and you yourself did the same thing in your post? I haven't read the Court's judgment again recently but my recollection is that the ICJ treated Nottebohm as if he was dual national - even though he was very obviously not.



Javier says

December 14, 2021

Many thanks, Dapo and Tamás

The question before the Court was not whether the claim should be espoused on behalf of a dual national, but rather whether the nationality conferred upon Nottebohm created an obligation on the part of Guatemala to recognise the effect of that nationality for diplomatic protection purposes. That is, whether Liechtenstein nationality should have international effects.
The Court first recognised two rules that apply in the field nationality, including dual nationality. The first is that ‘[i]t is for each State to determine under its own law who are its nationals’ (Hague Convention), and the second is that the attribution of nationality ‘shall be recognized by other States insofar as it is consistent with international conventions, international custom and the principles of law generally recognized with regard to nationality’ (Hague Convention). It is by application of these two rules that the Court considered that it should not:

‘go beyond the limited scope of the question which it has to decide, namely whether the nationality conferred on Nottebohm can be relied upon as against Guatemala in justification of the proceedings instituted before the Court. It must decide this question on the basis of international law; to do so is consistent with the nature of the question and with the nature of the Court's own function’ (at 17).

The Court then noted that the question of ‘whether full international effect was to be attributed to the nationality invoked’ (at 22) has been determined in cases where the person concerned happened to hold two nationalities. It is important to emphasise that, as in Nottebohm, those cases of dual nationality also involved the invocation of one nationality by the applicant State and arbitrators had to ascertain whether that nationality was one which could be relied upon as against the respondent state. The rule applicable in those cases, on which the Court relied, is known today as the rule of dominant and effective nationality, and requires the search for factual connections between the person concerned and the States whose nationality is involved.

In short, the Court followed different existing rules of international law in the field of nationality:
a) The Court respected Liechtenstein’s decision to naturalise Nottebohm in the exercise of its sovereignty.
b) The Court acknowledged, however, that the opposability of a lawfully acquired nationality to another state is a matter for international law to determine.
c) The Court resorted to an existing rule of customary law to find that, based on the particular factual situation of Liechtenstein nationality, the latter should not have international effect for the purposes of founding the claim against Guatemala.

As I mentioned in the post, one can criticise the ICJ in that the factual premise of the case was different when compared to cases where the rule of dominant and effective nationality applies: Nottebohm only had one nationality. However, in my view, this was not relevant since the primary question before the Court and arbitrators hearing disputes involving dual nationals was whether the nationality invoked, that of the claimant state, should be recognised by the respondent state.

I hope this clarifies your queries.
Best wishes,


Dapo Akande says

December 14, 2021

Dear Javier,

Many thanks for your response and for setting out your position so clearly. And thanks for a stimulating debate too!

I suppose the key question is whether, as you say, the applicable rule on which the Court relied is one which requires "the search for factual connections between the person concerned and the States whose nationality is involved", or one which, in cases in which the person on behalf of whom the claim is brought is a dual/multiple national, requires (or, at the time of Nottembohm, required) such a search. At least today, such a search for factual connections is thought by many, including the ILC in its Draft Articles on Diplomatic Protection (Art. 7) to be required only in cases where a claim is brought on behalf of a person of dual/multiple nationality by one state of nationality against another.

Javier says

December 14, 2021

Many thanks, Dapo

Yes, the rule of dominant and effective nationality is so far, as practice has shown, only applicable to dual nationality cases involving the nationality of both states party to the dispute. However, given the growing instrumentalisation of nationality, I wonder whether the rule, or more accurately the search for factual connections, should be extended to other strategic nationality scenarios.

Think for instance of the case Aven v Costa v Rica. The investor was a dual Italian-US national bringing a claim against Costa Rica under DR-CAFTA through his US home state nationality. Costa Rica invoked the rule of dominant and effective nationality to argue that the dominant nationality of the investor was that of Italy and thus the investor did not qualify for protection under the treaty (Italy is not a party to DR-CAFTA).

Interestingly, Costa Rica asserted that Mr. Aven’s invocation of DR-CAFTA constituted illegitimate ‘treaty shopping’, for he ‘used his Italian nationality in the establishment and operation of its alleged investment’ and ‘yet when it [became] a convenient moment’ he took advantage of his US nationality with ‘the purpose of acquiring the benefits of investment treaty protection’.

The tribunal rejected the argument, holding that the rule of dominant and effective nationality only applies in cases involving dual home-host state nationals.

Tamas Hoffmann says

December 16, 2021

Dear Javier,

Thank you for further clarifying your argument. However, how could you demonstrate that Nottebohm was not "bad law".

As you correctly quote the decision, the Court should have demonstrated that to "go beyond the limited scope of the question which it has to decide, namely whether the nationality conferred on Nottebohm can be relied upon as against Guatemala in justification of the proceedings instituted before the Court. It must decide this question on the basis of international law".

To put it simply, the ICJ should have demonstrated that there is a rule of international law - presumably a customary norm - that allows the opposability of a lawfully acquired nationality to another state in case of a SINGLE CITIZENSHIP.

You claim that the "Court resorted to an existing rule of customary law to find that, based on the particular factual situation of Liechtenstein nationality, the latter should not have international effect for the purposes of founding the claim against Guatemala." However, if we actually read the judgment itself, it becomes exceedingly clear that the ICJ only proved that there is a customary rule that allows opposability in cases of multiple nationality. There was not a single example in the judgment that does not fall into this category and the Court jumps without explanation to concluding that "According to the practice of States, to arbitral and judicial decisions and to the opinions of writers, nationality is 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." (p. 23)

Now, I completely agree with Sloane that the real reason behind this "creative" approach was the apparent bad faith acquisition of the citizenship of Liechtenstein, that was explicitly raised by Guatemala during the proceedings. However, in that case the ICJ should have admitted that even if that would have cast a bad light on Liechtenstein.

To conclude, I really cannot see why you claim that the Court proved the existence of a customary rule when it actually could not adduce a single example. What is even more interesting, that even in your post when you submit that "Nottebohm is present in contemporary international legal practice" you quote a dual nationality case that hardly proves the Nottebohm doctrine. I agree with you that the concept of abuse of rights should have a more pertinent role in legal doctrine but we hardly need to defend an old and arguably defective judgment for that...