1.1. Conflicting assumptions about ‘judicial lawmaking’

Whether courts are supposed to make or develop the law, as opposed to merely applying it, is one of (international) law's perennial questions.Footnote ⁵ Like many perennial questions, it eschews a simple answer. Debates typically proceed on the basis of two commonly shared assumptions.

The first assumption is that international courts have no express legislative mandate.Footnote ⁶ This is frequently acknowledged,Footnote ⁷ and it should not come as a surprise. Wary of activist judges, states have often attempted to circumscribe the judicial role rather narrowly by including safeguard clauses in the constitutional documents of international courts and tribunals.Footnote ⁸ The Statute of the ICJ is no exception. Nothing in it empowers the Court to make or develop law.Footnote ⁹ Its decisions are binding only between the parties and in respect of the particular dispute.Footnote ¹⁰ They are not formal sources of law but, at best, ‘material’ ones – meaning that whatever ‘law’ can be found in them must be anchored in a formal source to be binding and applicable in the relationships between states other than the parties to the dispute.Footnote ¹¹

The second common assumption is that even without a legislative mandate, the Court, through its decisions, can influence international law. In fact, it seems largely agreed that, in practice, its contribution to the formation and development of international law is immense.Footnote ¹² Again, this is not surprising. The International Court of Justice is the principal judicial organ of the United Nations, and its decisions are bound to have significant repercussions beyond the strict confines of the question before it. The Court refers (almost exclusively) to itself and will only break from its line of jurisprudence in rather exceptional circumstances;Footnote ¹³ states and their lawyers rely on ICJ case law in formulating claims, treating it as authoritative on the existing law;Footnote ¹⁴ scholars seek to ground their arguments in ICJ decisions;Footnote ¹⁵ the International Law Commission (ILC) draws on the Court's jurisprudence when codifying international law.Footnote ¹⁶ The UN General Assembly stated early on that the Court should be ‘utilized to the greatest practicable extent in the progressive development of international law’.Footnote ¹⁷ Even the Court itself (always formally denying that it can act as a legislatorFootnote ¹⁸) has acknowledged that its decisions have implications for the relations between states other than the parties to a dispute before it,Footnote ¹⁹ quite apart from a number of its judges having clearly taken a position in favour of the Court developing or making international law, whether in separate and dissenting opinions,Footnote ²⁰ or writing in an extrajudicial capacity.Footnote ²¹ As a result of these factors, there are areas of substantive international law which can hardly be understood without a knowledge of the Court's case law: any student of international law seeking to assess the state of the law on questions such as the use of force, maritime delimitation, or the legal personality of international organizations will immediately be referred to the landmark ICJ decision on the matter, which is seen as an accurate expression of ‘the law’.

On the face of it, the two assumptions seem difficult to reconcile. One way of addressing the tension between them is to distinguish between the theory and reality of international lawmaking – hence some commentators suggest that while it should not do so, the Court in reality does ‘make’ or ‘develop’ the law.Footnote ²² Yet that in itself is simplistic. A more appropriate way of explaining the role of the Court may be to view its pronouncements as contributions to the process of legal development and norm creation. As the above-mentioned examples illustrate, in a system relying on treaty-making and ‘amorphous processes of state practice and opinio juris’,Footnote ²³ the potential contribution of judicial decisions is considerable:Footnote ²⁴ through its jurisprudence, the ICJ can clarify the content of unwritten law, whether custom or general principle; it can advance a particular interpretation of a treaty; it can fill gaps in the law by relying on analogous reasoning; it can refine existing principles through their clarification or modification, or the carving out of exceptions; and so forth.Footnote ²⁵ Yet even where it does so, the Court does not make or develop the law single-handedly; it operates within the broader context of legal development and in many respects is constrained by it. Two obvious ‘obstacles to the rapid judicial development of the law’Footnote ²⁶ (which talk about ‘judge-made law’ risks overlooking) are of particular importance. First, as is the case with all courts, the ICJ's role is reactive; it depends on cases instituted before it. Unlike domestic courts, however, international jurisdiction is consensual rather than compulsory. Since 1946, the Court has been seized of a timid average of 2.3 ICJ cases per year.Footnote ²⁷ And while there are patterns of repeated involvement (notably maritime delimitation and, perhaps, the use of force), this has prevented the Court from developing the law in any systematic or comprehensive way.

Second, and more importantly, the Court's influence on the process of legal development is interstitial. It no doubt has a chance to influence the law through its decisions, but its influence is limited in time.Footnote ²⁸ Once it has rendered its decisions, the case – and with it the legal issues that it had raised – is out of the Court's hands. Nothing prevents other actors from ignoring, overruling, or limiting the impact of the Court's contribution.Footnote ²⁹ ICJ decisions are not per se relevant contributions to the process of legal development, but only to the extent that they are acceptable to the international legal community. Outside the bounds of Article 59 of the Statute, their authority is persuasive only. And so they must ‘persuade’.

1.2. ‘Rules of thumb’ on the ICJ's contribution to legal development

In the light of these considerations, it may be preferable to avoid terms such as ‘judicial lawmaking’ and instead speak of the Court's role as an ‘agent’Footnote ³⁰ or actor participating in the process of legal development. When seeking to assess the agent's influence on the process, legal arguments about the normative value of ICJ decisions provide only the starting point for the enquiry. The real question, instead, is to assess to what extent a particular ICJ decision has in fact shaped the law in a given area. As far as specific holdings are concerned, this can of course be done retrospectively – by evaluating the impact of a given ICJ pronouncement on the subsequent development of the law. By contrast, it is much more difficult to assess in the abstract under which conditions ICJ pronouncements are likely to be influential. Despite the wealth of debate about the conceptual problems of ‘judge-made law’, this question is hardly addressed in any detail. If anything, commentators have typically advanced a number of rather simple ‘rules of thumb’.Footnote ³¹

Two of these ‘rules of thumb’ seem to have gained particular currency.Footnote ³² One refers to the reasoning supporting a particular pronouncement. Schwarzenberger, commenting on the likely impact of judicial decisions on the formation of international law, noted that much depended on ‘the fullness and cogency of the reasoning’ and that it ‘was not accidental that the least convincing statements on international law made by the International Court of Justice excel by a remarkable economy of argument’.Footnote ³³ Others rely on the common-law distinction between ratio and obiter,Footnote ³⁴ with Amerasinghe suggesting that ‘[m]ore authority naturally attaches to the former than to the latter.’Footnote ³⁵ Lastly, it is argued that the concerns raised by activist lawmaking had affected the attitude of the Court and its members. A recent study qualifies them as ‘reluctant lawmakers’, fully aware that they ought not to be perceived to make law.Footnote ³⁶ In a similar vein, Judge Shahabuddeen (writing extrajudicially) concludes his detailed analysis by observing that the ICJ typically navigated ‘from case to case, like the ancient Mediterranean mariners, hugging the coast from point to point and avoiding the dangers of the open sea of system and science’.Footnote ³⁷

The picture emerging from these brief considerations is that while questions of judicial lawmaking may have been rather overworked conceptually, we are still a far cry from assessing with any certainty the circumstances under which a particular judicial pronouncement is likely to shape the law. The subsequent discussion will not solve this problem. Yet it will approach it inductively, by assessing the impact of the two key holdings of the Barcelona Traction case and by drawing a number of tentative lessons from that decision's history.

2.1. Diplomatic claims on behalf of corporations

1. As for diplomatic protection, it is submitted that the Court's approach to nationality continues to govern the law of diplomatic claims. To recall, the Court rejected Belgium's claim to exercise diplomatic protection on behalf of a company that was registered in Canada but was effectively controlled by Belgian shareholders. Its reasoning on this point was based on three key considerations.

(i) The Court emphasized the distinction between shareholders and company, which were ‘separated . . . by numerous barriers’, including the ‘separation of property rights’.Footnote ³⁸ This approach was indeed accepted in many domestic legal systems, but rather than saying that domestic law was merely a fact (as the Permanent Court of International Justice (PCIJ) had doneFootnote ³⁹), the Court, accepting a renvoi, held that international law would be ‘called upon to recognize’ the domestic distinction between corporation and shareholder, as ‘international law had not established its own rules’.Footnote ⁴⁰

(ii) Controlled by municipal law, international law had to respect the sharp distinction between shareholders and company. The obvious way to implement this was to make nationality dependent on the formal criterion of incorporation, provided at least that there was some real connection going beyond mere registrationFootnote ⁴¹ – one might call this the ‘incorporation plus X’ approach.

(iii) In contrast, the Court was adamant that shareholder rights could not be equated with rights of the company. While some states had previously acted on behalf of a company controlled by shareholders bearing their nationality, the Court (without really assessing it) considered this practice to be insufficient.Footnote ⁴² In its view, the shareholders’ state could therefore exercise diplomatic protection for the company only exceptionally:Footnote ⁴³ (a) if special treaty rules so provided; (b) in special cases concerning the treatment of enemy property and nationalizations; (c) if the company had ceased to exist; and (d) perhaps if claims were raised against the state of incorporation (‘protection by substitution’). But no special treaty applied, nor did any of the exceptions. What is more, the Court saw no need to assess in any detail whether Spain had violated direct rights of the shareholders (which Belgium could, of course, have espoused by way of diplomatic protection).Footnote ⁴⁴

2. The Court's holding on nationality disappointed many commentators.Footnote ⁴⁵ With respect to the Court's reasoning, criticism indeed seemed justified. In particular, the Court's ‘domestic analogy’ rested on rather shaky foundations,Footnote ⁴⁶ and one could have expected a more detailed analysis of actual instances of diplomatic protection on behalf of shareholders.Footnote ⁴⁷ As for the outcome, there was widespread concern that by stressing the separation between company and shareholders, the Court had adopted a formal or ‘rigid’ approach, ignored the realities of international business, and divorced law from real life.Footnote ⁴⁸

Looked at from a distance, this last criticism seems exaggerated. As the brief summary shows, the Court's approach was by no means as ‘formal’ as is sometimes suggested. The Barcelona Traction company was not an ‘empty shell’, and the Court emphasized its ‘manifold links’ with Canada.Footnote ⁴⁹ One might even say that while the Court refused to make the nationality of corporations dependent on a genuine link requirement (as developed in its Nottebohm judgmentFootnote ⁵⁰), its ‘incorporation plus X’ test came rather close to it: in both instances, international legal rules on nationality rely on domestic legal acts but complement this formal approach by adding a substantive criterion to prevent abuse. Perhaps more importantly, the Court's allegedly formalistic approach to nationality had obvious advantages. It relied on a rather simple test and – unlike competing approaches emphasizing more tangible ties such as control of business or the ‘social seat’ of a company – produced clear and predictable results. It thus avoided problems of multiple claims brought by different states, which, as the Court stated, would have created ‘an atmosphere of confusion and insecurity in international economic relations’.Footnote ⁵¹ And it was flexible in that the Court admitted the possibility of special rules deviating from the general approach.

3. These factors may help to explain the continuing relevance of the Court's holding on nationality. To be sure, in many respects international law has moved on. Since 1970, states have agreed on a wide range of treaty rules laying down special requirements for claims relating to the treatment of foreign corporations. Very often, these do not follow the Court's Barcelona Traction approach of requiring ‘incorporation plus X’. Some define nationality on the basis of some form of control;Footnote ⁵² others are more formal than the ICJ had been, in that they consider incorporation as such to be sufficient.Footnote ⁵³ Perhaps more importantly, a huge number of bilateral investment treaties (BITs) establish special mechanisms for investment protection and blur the line between shareholders and company which the Court had emphasized.Footnote ⁵⁴ More often than not, these BITs expressly state that shareholdings should be treated as investments for the purposes of mixed arbitration – and, surprisingly, these jurisdictional provisions have been used to assimilate substantive shareholder rights with rights of the company.Footnote ⁵⁵ Finally, investment treaties now frequently permit claims in defence of indirect investments, which allows claimants to circumvent the strictures of nationality rules. In short, it can hardly be questioned that many disputes that would have given rise to diplomatic claims on behalf of corporations at the time of Barcelona Traction are today addressed within special legal frameworks, notably by way of investment arbitration. In its recent Diallo judgment, the ICJ noted that ‘in contemporary international law, the protection of the rights of companies and the rights of shareholders is essentially governed by bilateral or multilateral agreements for the protection of foreign investments’.Footnote ⁵⁶ Yet, as that case shows, diplomatic claims on behalf of corporations remain a possibility.

Where such claims are brought by means of diplomatic protection, the more convincing approach is, indeed, that the ‘incorporation plus X’ test laid down in Barcelona Traction continues to govern. There have, of course, been attempts to discard it altogether. According to Orrego Vicuña, for example, ‘the aggregate of the practice [as summarized in the last section] demonstrates forcefully that the criteria of the Barcelona Traction case no longer prevail and that shareholders should be entitled to protection’.Footnote ⁵⁷ Yet that view is attractive only at first sight. It ignores the distinction between the general rule and special provisions: if the latter were concluded to disapply the former, then they implicitly affirm the general rule, even if simply as the default rule. The whole point of the maxim lex specialis derogat legi generali is to maintain the relevance of the general rule, giving occasional priority to any potential special rule providing differently.Footnote ⁵⁸ With respect to using investment treaties as evidence there is a further problem, not always acknowledged by those criticizing Barcelona Traction. As observed by Kate Parlett, investment arbitration is not ‘a morphed form of delegated diplomatic protection’;Footnote ⁵⁹ rather, Article 27 of the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (ICSID Convention) suggests that it is an alternative to diplomatic protection rather than a redefinition of its conditions.

Given these conceptual problems, reports about the death, or ‘bypassing’,Footnote ⁶⁰ of Barcelona Traction seem greatly exaggerated. The allegedly technical holding remains, rather, in good shape. In fact, whenever the international community, since 1970, has had to address general rules on diplomatic protection of corporations, it has relied on the much-criticized Barcelona Traction holding. In the ELSI case, a chamber of the Court affirmed the central message of Barcelona Traction, even though it accepted, exceptionally, a claim of ‘protection by substitution’ on the basis of a special bilateral treaty regime.Footnote ⁶¹ In his report on diplomatic protection of corporations, the ILC's Special Rapporteur John Dugard observed that Barcelona Traction ‘dominate[d] all discussion of this topic’.Footnote ⁶² On the basis of his work the ILC discussed at length whether to replace the ‘incorporation plus X’ test by any of the competing theories (siège social, centre of business, and so forth), but decided against it. In essence, draft articles 9–12 affirm the central features of the Court's Barcelona Traction ruling – hence the ILC echoes the warning against multiple actions,Footnote ⁶³ and affirms that ‘[t]he most fundamental principle of the diplomatic protection of corporations is that a corporation is to be protected by the state of nationality of the corporation and not by the state or states of nationality of the shareholders’.Footnote ⁶⁴ While the 2006 Draft Articles on Diplomatic Protection attempt to specify the required tangible ties between a company and its state of incorporation,Footnote ⁶⁵ the ILC's work on balance probably is best seen as an attempt – perhaps akin to that of a glossator explaining the meaning of a provision of Justinian's codex – to concretize the message of Barcelona Traction, not to move away from it. And to the extent that the ILC might have been perceived as moving away, the ICJ's recent judgment in the Diallo case clearly reaffirms Barcelona Traction's key holdings:Footnote ⁶⁶ while recognizing the increasing importance of special legal frameworks, the Court continues to emphasize the distinction between a company and its shareholders, stresses the importance of incorporation to determine the nationality of the former, and seems to take a more cautious approach than the ILC (or indeed the Court itself, in ELSI) to the problem of ‘protection by substitution’.Footnote ⁶⁷

To sum up, since 1970, special rules on nationality and special legal frameworks for the vindication of investor rights have multiplied. Yet while the exceptions have become more numerous, they remain what they were in 1970: exceptions to a general rule based on the ‘incorporation plus X’ test. Despite the amount of criticism, that general rule remains largely unchanged. And so the law of diplomatic protection remains premised on the Barcelona Traction approach – developments since 1970 may be more than a footnote, but little more than a coda, to it.

2.2. Obligations erga omnes

1. The Court's second important holding – on obligations erga omnes – has prompted debates of a different character. On the face of it, it has been less controversial. Only few commentators have openly criticized it, while many hail it as an inspiring dictum. The question is not whether the Court was right to ‘invent’ the notion of obligations erga omnes as an enforcement concept. Instead, debate centres on two other issues: what did the Court mean by it? and does it matter?

The Court itself is responsible for much of the confusion surrounding the erga omnes concept, as it introduced it in a rather mysterious way. The relevant passage appears, without much advance warning, in paragraphs 33–34 of the Barcelona Traction case, which state that

This pronouncement continues to fascinate and puzzle commentators (including at times the Court and its members).Footnote ⁶⁹ It is phrased in a rather complicated way, not least because it relies on a curious Latin concept (obligations erga omnes) that had been used previously to describe third-party effects of treaties or judgments,Footnote ⁷⁰ but was now applied to the field of law enforcement. Yet, on consideration, matters are not that complicated. The essence of the erga omnes concept can be described in four simple steps.

(i) International law draws a distinction between the general rules governing the treatment of aliens, and a special set of rules protecting fundamental values.

(ii) To this special set of rules protecting fundamental values applies a special regime of standing. The right to raise claims in response to violations is not restricted to the state of nationality (as it is under diplomatic protection).

(iii) Instead, certain fundamental values, being the concern of the international community as a whole, can be protected by each and every state.

(iv) Finally, these rights of protection do not have to be conferred expressly by treaty, but can (also) exist without a special written ‘empowerment’ – and would then flow from general international law.

Looked at in this rather sober way, one might say that obligations erga omnes are not that mysterious after all. The idea behind the concept is certainly known to many domestic legal systems, which accept that ‘technical rules of locus standi’ may need to be modified where important interests are at stake, so as to permit the effective protection of community interests.Footnote ⁷¹ That said, to have embraced the concept of obligations erga omnes certainly was a giant leap for the ICJ. After all, only four years earlier, in 1966, the same Court had relied on ‘technical rules of locus standi’Footnote ⁷² to dismiss an extremely high-profile ‘public interest claim’ brought by Ethiopia and Liberia against South Africa.Footnote ⁷³ As is well known, South West Africa was a disaster for the Court – from a legal but also ‘from a public relations point of view’Footnote ⁷⁴ – and it required the Court to mitigate damage, which it did in two ways: by recognizing the UN's termination of the South African mandate,Footnote ⁷⁵ and by launching the erga omnes concept. In so doing, it accepted that for a narrowly defined circle of community obligations, international law should be prepared to accept law enforcement by many states, even if this might create ‘an atmosphere of confusion and insecurity in international [economic] relations’.Footnote ⁷⁶

2. What, then, have been the effects of this ‘other’ Barcelona Traction dictum? In practice, one popular answer is (or at least was) that they are next to nothing. Hugh Thirlway, in his otherwise excellent review of ICJ jurisprudence in the British Year Book, suggests that obligations erga omnes are a ‘purely theoretical category’ and the passage ‘little more than an empty gesture’.Footnote ⁷⁷ Putting it rather more bluntly, Alfred Rubin labelled obligations erga omnes the product of ‘the wishful thinking of some publicists who have no money to spend, no troops to send, no children likely to die in a military action’.Footnote ⁷⁸ These may seem extreme statements, yet they draw support from the fact that, forty years after Barcelona Traction, the Court still has to hear its first full-blown ‘erga omnes case’. There have, of course, been instances of public interest litigation before the Court, including the Nuclear Tests cases brought by Australia and New Zealand,Footnote ⁷⁹ Portugal's East Timor case,Footnote ⁸⁰ or the pending proceedings between Belgium and Senegal;Footnote ⁸¹ yet, typically, applicant states in these cases have sought to emphasize their special interest in the subject matter before the Court.Footnote ⁸²

Still, the cautious reading of Thirlway, Rubin, and others seems to lose ground. On consideration, the much more convincing view is that – despite the absence of proper ICJ cases – the erga omnes concept has been a remarkable success. There are two main arguments to support this claim. First, the cautious reading may be based on an unrealistic view of the ICJ. While recent decades have witnessed an increase in the number of cases, states very rarely institute ICJ proceedings; thus one should not make too much of the absence of erga omnes cases.

Second, and more importantly, even without proper ICJ cases, the erga omnes concept has left its mark on international law. It has ‘developed apace’Footnote ⁸³ and ‘spilled over’ into other areas of law, notably the law of state responsibility. The ILC's 2001 Articles (not binding in law, but formulated in close co-operation with governments) in particular take up the idea of ‘law enforcement in the public interest’. Drawing on Barcelona Traction, Article 48 of the ILC's text recognizes the right of each state to invoke another state's responsibility if ‘the obligation breached is owed to the international community as a whole’ (i.e. an obligation erga omnes).Footnote ⁸⁴ While Article 48 merely spells out the meaning of the Barcelona Traction dictum, that dictum has also been applied to justify other forms of law enforcement. Much of the debate has centred on the countermeasures – that is, coercive measures taken in response to serious and well-attested violations of obligations erga omnes.Footnote ⁸⁵ Whether such a right exists remains a matter for debate. The ILC seemed unable to expressly recognize it, and in its Article 54 left the matter open.Footnote ⁸⁶ However, practice suggests a more liberal approach. On frequent occasions, states have asserted a right to suspend treaties, freeze foreign assets, or impose embargoes in response to erga omnes breaches, against other states such as Zimbabwe, Belarus, Yugoslavia, South Africa, and so forth.Footnote ⁸⁷ Given this rather widespread practice, much suggests that the ‘erga omnes rationale’ has modified the rules governing countermeasures.Footnote ⁸⁸

There may also be other spillover effects, outside the field of responsibility.Footnote ⁸⁹ Some argue that the erga omnes concept should govern questions of jurisdiction;Footnote ⁹⁰ others proclaim erga omnes effects on concepts such as waiver or estoppel;Footnote ⁹¹ and the ICJ in the Wall opinion seemed to imply that states were under a duty not to recognize effects of erga omnes breaches.Footnote ⁹² Finally, treaties with express law enforcement clauses are now called ‘erga omnes partes’ treaties, as if only the ‘erga omnes’ label could justify a broad approach to standing.

Some of these ‘other’ alleged erga omnes effects may admittedly be questionable. To take but one example, the frequent references to obligations erga omnes partes seem to ignore the fact that the erga omnes concept is intended to close an enforcement gap, and thus is hardly necessary where a treaty expressly provides for standing in the public interest. Yet the brief survey shows that the erga omnes concept, far from being a purely theoretical category, clearly has a place in contemporary international practice and jurisprudence. In fact, the real problem today seems to be one of over-use: there is a tendency, among ‘publicists . . . without money to spend’,Footnote ⁹³ but also among members of the International Court, to use the erga omnes concept as a legal vade mecum that can conveniently be used to explain all sorts of legal effects. In the long run, this inflationary reliance may be the real challenge for the erga omnes concept.Footnote ⁹⁴ Yet it clearly shows that the Court's ‘PR exercise’ has successfully placed a concept on the legal agenda, and that this concept has developed (if the term may be permitted in this context) a considerable amount of ‘traction’.

3.1. Courts can be both reluctant and enthusiastic lawmakers

The first lesson relates to the attitude of the Court when engaging in legal development. Barcelona Traction provides evidence of two different attitudes: in line with what is perceived to be its general approach, the Court was a ‘reluctant’Footnote ⁹⁵ agent of legal development, but it also – contrary to the common perception – interpreted its role much more enthusiastically.

With respect to the question of diplomatic claims, the Court, in Barcelona Traction, could hardly avoid shaping the law. No generally agreed test governing the nationality of corporations had been accepted, nor had the matter been addressed by any major treaty;Footnote ⁹⁶ so the Court's pronouncement was very likely to be applied outside the specific case before it. That the Court was reluctant and circumscribed in its approach to the question it had to answer is evident, hence its choice of a simple and straightforward general rule not requiring much in the way of proof (incorporation) over other possibilities (siège social, the strongest link) which would have been less predictable.

By contrast, in its pronouncement on erga omnes obligations the Court was not reluctant or circumscribed at all. Taking up Lord Devlin's above-quoted remark, one might say that, rather than ‘hugging the coast point by point’, the Court boldly set sail for ‘the open sea’.Footnote ⁹⁷ What is more, it did so of its own accord, as no gap was waiting to be filled, no outstanding issue had to be decided lest there be a non liquet. The Court could perfectly well have spared the world paragraphs 33 and 34, and no one would have realized – because no one expected them to be there in the first place. In fact, it may well be that, precisely because no specific outcome was at stake, the Court considered itself free to engage in its exercise of ‘enthusiastic’ legal development. The experience of Barcelona Traction certainly suggests that where the Court – exceptionally, no doubt – decides to ‘leave the coast’,Footnote ⁹⁸ it may also leave behind it its usual concerns for straightforward rules ensuring legal certainty. Yet, more fundamentally, Barcelona Traction clearly shows that common assertions about the cautious nature of the Court need to be taken with a grain of salt. It is by no means always a ‘reluctant lawmaker’,Footnote ⁹⁹ but, at least occasionally, goes out of its way enthusiastically to ‘strik[e] out a path towards new developments in the law’.Footnote ¹⁰⁰

3.2. Popular rules of thumb are of limited usefulness

The second observation relates to the ‘rules of thumb’ put forward by commentators to assess the likely impact of judicial pronouncements on the development of international law. As noted above, two such rules of thumb are popular among writers: the preference for ratio over obiter, and the preference for well-reasoned judicial pronouncements.Footnote ¹⁰¹ Curiously, neither of them is borne out by the Barcelona Traction case.Footnote ¹⁰²

3.3. Residual rules are more likely to make a lasting impact

In addition to questioning popular rules of thumb, Barcelona Traction offers some insights into when a judgment is likely to make an impact. In this vein, the third lesson to be drawn from a rereading of the case is that the Court's pronouncements are more likely to influence the development of the law if they posit a general, residual rule that admits of exceptions. This is brought out very clearly by the remarkable ‘success’ of the Court's holding on nationality.

In the circumstances of the case, any attempt by the Court to formulate a general test governing the nationality of corporations – whether relying on incorporation, seat, strongest link, or any other potential criterion – would have acquired general relevance. The case squarely raised an issue of major practical and theoretical relevance and required the Court to address it on the basis of general international law. While the Court thus could hardly avoid making a pronouncement that would acquire general relevance, it ‘secured’ its approach by admitting the possibility of exceptions. It effectively articulated a general, default rule that allowed for further development through the carving out of exceptions, the refinement of the scope of application of the general rule, and so forth. In this respect, the Barcelona Traction case affirms Lauterpacht's observation (preceding the judgment) that

Indeed, since 1970 both ‘ordinary legislative processes’ and ‘future judicial decisions’ have built on the Court's general rule of nationality in Barcelona Traction. What is more, states have progressively elaborated on the principle, even if that has been primarily through the making of special rules, with the concomitant reduction in the practical significance of the general rule. Still, as further demonstrated by the Court in its recent Diallo judgment, the Barcelona Traction general rule of nationality of corporations has retained its status as the fallback position.Footnote ¹¹⁴

In fact, experience since 1970 suggests that while the broad and residual rule on nationality enunciated in Barcelona Traction may be easy to disapply in particular circumstances, it is almost impossible to reverse. The obvious way to reverse it would be through the conclusion of a general multilateral treaty – as had happened to the Lotus holding on jurisdiction.Footnote ¹¹⁵ Yet the prospects of such an eventuality are rather slim when the issue is as politically sensitive and divisive as the diplomatic protection of corporations. By the same token, it is difficult to imagine that the Court's dictum should be reversed by a body of international practice consistent enough to give rise to the emergence of new rule of custom – especially if diverging approaches can be explained as leges speciales. In short, Barcelona Traction suggests that residual, default rules that admit of exceptions are rather likely to make a lasting impact on the law.

3.4. Judicial pronouncements will shape the law if they take up a societal demand

The preceding considerations highlight one particular feature of the Court's statement on nationality, but, in and of themselves, cannot explain the tremendous influence that both Barcelona Traction pronouncements have exercised. So it may be asked whether, despite their diversity, these two pronouncements share a common trait that can explain their impact. It is submitted that, ultimately, both pronouncements were able to shape the law because they responded to a clear societal demand.

With respect to the nationality of corporations, the international society, in 1970, seemed to be in need of a general rule, which the Court provided. As noted above, no general test governing the nationality of corporations had been accepted, hence there was a gap in the law. Of course, not every gap in the law requires to be filled. Yet where the gap concerns an issue as important and politically sensitive as the diplomatic protection of corporations, there arguably is a societal demand for legal certainty. In Barcelona Traction the Court responded to that need by laying down a straightforward default rule that was relatively easy to apply, but admitted of exceptions. Its pronouncement clarified the state of the law in an important area and thereby enhanced legal certainty. Conversely, it prompted states that considered the default rule to be insufficient or unacceptable to ‘contract out’ of it by way of treaty. The Barcelona Traction case thus enunciated a general rule and indirectly encouraged states to formulate special rules for special circumstances. As a result of both factors, it succeeded in bringing at least a measure of legal certainty to a hitherto rather underregulated area of international law.

The Court's ‘other’ Barcelona Traction dictum, the pronouncement on obligations erga omnes, fulfilled a very different function, but it, too, responded to a societal demand. The Court's dictum launched a concept that accommodated a generally felt interest in some form of enforcement action in defence of community interests. After the 1966 South West Africa judgment, international law was in need of such a concept – one that sent a political signal, that reopened the door to the notion of community interest, and that was broad enough (and maybe mysterious enough) to be applied outside its initial field of application. The Court, having felt the repercussions of South West Africa, was keenly aware of this societal demand. By ‘inventing’ the erga omnes concept, it was able to translate it into a general legal concept that would provide a framework for debates about law enforcement in the public interest. It thereby not only responded to a societal demand, but also ‘[gave] general and articulate formulation to developments implicit, though as yet not clearly accepted, in actual international custom or agreement of States’.Footnote ¹¹⁶

The experience with both pronouncements thus suggests that a judgment is most likely to shape the law if it responds to a societal demand or concern, and translates this demand into legal form. This may seem trite, but it arguably constitutes the most important ‘lesson’ that can be drawn from the Barcelona Traction case. Just as the law more generally, so a specific pronouncement by the Court, cannot ‘be divorced from the general framework of normative argument in the society within which it operates’.Footnote ¹¹⁷ As both of the Court's pronouncements in this case were in line with ‘general framework[s] of normative argument[s]’Footnote ¹¹⁸ within the international society, they were unlikely to be reversed through subsequent international lawmaking (whether treaty or custom). Instead, they shaped the process of legal development because states and other actors could be expected to ‘build practice around them’Footnote ¹¹⁹ – by applying the principles articulated or even by contracting out of, and thus indirectly affirming, them.