(a) The ILC’s ‘master plan’.

While the PCIJ and ICJ were influential in shaping general principles of responsibility and specific rules implementing them (the second and third levels of normative decisions mentioned in the preceding section), it is important to note that the conceptual decision to think of responsibility as an overarching category comprising the general conditions for, and consequences flowing from, wrongful conduct was taken by the ILC. Unlike other changes of direction in the law, the crucial decision can indeed be traced with relative precision: in 1963, having failed to agree on a regime of State responsibility for injuries to aliens, the Commission decided to change tack – in a rare attempt to bring a project ‘back to life’²² by moving from the specific to the general/abstract. After some debate, a subcommittee was set up to study ways of rescuing the Commission’s work on the topic proposed to ‘give priority to the codification of general rules governing the international responsibility of States’.²³ In retrospect, it seems clear that on the journey towards the current law of responsibility, this was the decisive fork in the road, and the Commission’s subsequent endorsement of the subcommittee’s recommendation – and its appointment of Roberto Ago, the key figure in the subcommittee’s deliberations, as Special Rapporteur – was to change the legal landscape. For it was in the Commission’s engagement with Ago’s reports that, for better or worse, the contemporary notion of responsibility took shape, was ‘encoded’.²⁴ And while many of the specific rules (level 3) and some of the organising principles (level 2) would be revisited at a later stage, the strategic decision to understand responsibility as the crucial concept ‘between illegality and liability’²⁵ would stand. In that respect, developments since the early 1960s have followed the ILC’s ‘master plan’.

(b) Foundational decisions by the PCIJ

But of course, the Commission did not decide out of the blue to embark on its most ambitious codification project. In its subcommittee, the view prevailed that responsibility could be approached from a general angle as questions of attribution, or consequences, were governed by common principles. Trying to persuade the ILC subcommittee to move away from the study of responsibility in particular areas (such as the treatment of aliens), Mustafa Kamil Yasseen suggested that ‘the first step must be to define the general theory of responsibility. That theory exists.’²⁶ Not everyone agreed at the time; hence the continued attraction of ‘going (or staying) sectoral’.²⁷ But Yasseen’s view – that ‘[a general theory of responsibility] exists’ and awaits codification, which the subcommittee adopted – certainly seemed plausible. In fact, with the benefit of hindsight, one wonders why it took so long to emerge: the trees were there; it was time to think of the forest.

In distilling general principles of responsibility (from which a ‘general theory’ could be deduced), the Permanent Court was highly influential. Alongside scholarship²⁸ and diplomatic and arbitral practice, its jurisprudence had established a number of fundamental propositions, on which the codification effort (embarked upon in the 1960s) would draw. Three of them stand out:

First, a string of PCIJ decisions had affirmed the autonomy of international responsibility from domestic laws. In fact, it may well be the principle most frequently affirmed by the Permanent Court. This is true for its two ‘variations’: (i) violations of constitutional law do not render conduct internationally wrongful;²⁹ and (ii) compliance with domestic law cannot justify violations of international law.³⁰ As regards the latter, more important, variation, the judgment in the Polish Nationals case contains the quintessential formulation; in it, the PCIJ affirmed that ‘a State cannot adduce as against another State its own Constitution with a view to evading obligations incumbent upon it under international law or treaties in force’.³¹ Seventy years later, that principle would be affirmed, with due reference to the PCIJ’s formative jurisprudence, in Article 3 of the ILC’s text.³²

Secondly, the PCIJ’s jurisprudence could be read to foreshadow the emergence of responsibility as a separate notion ‘between illegality and liability’ – a notion the relevance of which failed to convince Allott.³³ In Phosphates in Morocco, the Court referred to attribution and illegality as the two key conditions³⁴ and noted that where these conditions were met, ‘international responsibility would be established immediately as between the two States’.³⁵ While this paved the way for appreciating responsibility as a notion combining conditions for, and consequences of, wrongfulness, the reference to ‘two States’ betrayed a bilateralist mindset³⁶ that reduced responsibility to reciprocal relations involving rights of claimants and corresponding duties of respondents – a restriction which haunts debates to this day.³⁷ Moreover, the Court’s State-centred interpretation of diplomatic protection claims – by which a State was ‘in reality asserting its own rights’³⁸ – would add a further restriction of lasting impact.³⁹

Thirdly, in a much-cited passage, the PCIJ would formulate, in a general way, the most important automatic consequence ‘immediately arising’⁴⁰ from responsibility. In Factory, it noted that a breach of international law ‘involves an obligation to make reparation’; this was said to be ‘a principle of international law, and even a general conception of law’.⁴¹ In a later passage of the same case, the PCIJ then explored the content of the obligation:

This was followed by a statement on the relationship between two potential forms of reparation:

The impact of these statements has been no less than remarkable. Hardly supported by argument, they have become cornerstones of the regime of consequences of responsibility: relied upon to support the existence of a general duty to make reparation⁴⁴ and the primacy of restitution over compensation.

Taken together, the three instances show the remarkable role of the PCIJ in preparing the ground for the emergence, and gradual formulation, of the modern law of State responsibility between the 1960s and 2001. The PCIJ was not the architect of the modern notion – as a judicial body deciding specific cases it had limited powers to design broad normative frameworks. However, the PCIJ’s jurisprudence (partly drawing on earlier arbitral practice, partly relying on ‘conceptualist reasoning’)⁴⁵ recognised a number of important general principles that shape or influence the law of responsibility to this day and that, in the terminology introduced above, form part of the ‘second level’ of normative decisions.

(c) The ICJ’s continuing relevance

The ICJ’s work has been influential, too, but its impact has typically been at a different level. At least for the last four decades,⁴⁶ the World Court has decided responsibility cases against the backdrop of the ILC’s work. And quite clearly, this has affected its impact on legal development.

(a) Some comparative remarks

With respect to State responsibility, three aspects of this substantial influence stand out.

First, State responsibility has been a permanent feature of the PCIJ and ICJ caseload, and their jurisprudence has been a constant source of influence. The PCIJ’s first contentious case (Wimbledon) involved questions of responsibility, as did Corfu Channel, the case of the ICJ; since then, questions of responsibility have been a regular feature of ICJ proceedings. Of the various branches and sections of public international law, the law of treaties (and perhaps the law of claims, unless it is viewed as part of responsibility) may be the only other with a similarly long-standing record of jurisprudence. Other areas have either come before the Court sporadically⁶⁷ or they have come and gone.⁶⁸ Responsibility has stayed and little suggests that this should change. This does not ensure influence on the development of the law but it is an enabling factor.

Secondly, whereas in many other areas of international law the PCIJ and ICJ have only pronounced on specific issues – such as maritime delimitation within the law of the sea,⁶⁹ or the relationship between human rights and general international law⁷⁰ – the Courts’ jurisprudence on questions of responsibility has left footprints all over the field: from general principles formulated by the PCIJ to specific rules clarified by the ICJ, few parts of the contemporary law of responsibility have completely escaped judicial scrutiny. Whereas typically the PCIJ and ICJ address specific and selected issues of a given area of law, in the field of responsibility their jurisprudence has been a pervasive factor.

Thirdly, PCIJ and ICJ pronouncements on questions of responsibility have throughout been treated as authoritative. Binding only ‘between the parties and in respect of that particular case’,⁷¹ PCIJ and ICJ decisions have to persuade to be relevant. And, while generally, the PCIJ and ICJ have been rather successful, persuasion often is a matter of degree: there are (rare) instances in which judgments have been overruled (such as Lotus) or deliberately bypassed,⁷² and a number of decisions have remained highly controversial.⁷³ By contrast, the brief survey given above suggests that in the field of responsibility, jurisprudence is indeed accorded ‘a truly astonishing deference’.⁷⁴ PCIJ statements continue to be seen as ‘the law’, and on more than one occasion, the ICJ has been recognised as the supreme arbiter deciding the fate of controversial ILC provisions. Despite the wealth of jurisprudence, and the sensitive character of some of the PCIJ and ICJ cases on questions of responsibility, it is hard to think of any equivalent to Lotus or Fisheries.⁷⁵ As a general matter, PCIJ and ICJ decisions have been ‘bright beacons’: rather than ‘flicker[ing] and [dying] near-instant deaths’,⁷⁶ they have been remarkably persistent.

(b) Three lessons

The development of the modern law of State responsibility in many respects displays unique features. However, it yields a number of general lessons about the role of international courts as ‘law-formative agencies’.⁷⁷ In concluding, three such lessons merit being briefly spelled out.

First, the preceding survey suggests that the impact of PCIJ and ICJ pronouncements on the development of international law in a given area is a natural by-product of their dispute settlement activity. This sounds trite, but it may be worth stressing since much of the scholarship seeks to explain factors accounting for the precedential impact of judicial pronouncements, which is said to depend on the attitude of courts,⁷⁸ on the strength of their reasoning⁷⁹ or on the (essential or obiter) character of a particular pronouncement.⁸⁰ Experience in the field of State responsibility does not bear out these distinctions: it includes ratio and dicta,⁸¹ well-reasoned statements, and mere assertions.⁸² And who could say whether the PCIJ and ICJ, since the 1920s, have been ‘activist’ or ‘restrained’? If the PCIJ and ICJ have been influential players in the development of State responsibility (unlike in other areas of international law), then this would primarily seem to reflect the fact that they have decided responsibility cases for nine decades. The first lesson can be formulated in refreshingly simple terms: ‘The impact of international courts and tribunals on the evolution of international law largely depends upon how many cases are brought before them.’⁸³ This would explain the relatively high influence on the development of State responsibility.

Secondly, the substantial impact of PCIJ and ICJ pronouncements on State responsibility may also reflect the fact that the law of responsibility is particularly receptive to judicial development. It belongs to the core of general international law, on which the world’s ‘generalist’ court is particularly trusted. It is based on organisational principles that require to be spelled out and applied – which are much better suited to be concretised in dispute settlement processes than, for example, specialised fields of law comprising vast numbers of specific provisions (such as the law of the sea, international humanitarian, or international environmental law). As importantly, as a field of law, State responsibility lacks specialised institutions to administer the application of the law – in the way human rights committees or Meetings of the Parties (MoPs) and Conferences of the Parties (CoPs) ‘manage’ their respective treaties, moulding them in a process of regular engagement and adaptation. What is more, as has been shown, the ILC, as the key institution overseeing the codification process, seemed to co-operate harmoniously with the Court. Experience in the field of State responsibility thus suggests that, in addition to numbers of cases, the impact of international courts on legal development may depend on the ‘make-up’ of the field, which can be receptive (such as State responsibility, but also diplomatic protection, or the law of treaties) or not.

Thirdly, the survey highlights how the role of courts as agencies of legal development can change over time. Where the PCIJ could lay down general principles, the ICJ would operate within the parameters of responsibility established by the ILC. Conversely, the ICJ retained an important role, as through its decisions it could engage with normative propositions and confirm or modify the legal status of specific draft articles. This markedly differs from the much more limited role of international courts that only become involved after the completion of a codification exercise – as happened, for instance, with respect to the law of the sea or international humanitarian law. A third lesson to be drawn from this brief survey is that international courts, as agencies of legal development, depend on the right circumstances or ‘setting’: they are influential during the formative stages of the law and during long-term and on-going codification attempts.

4.1 State responsibility for non-State actors: the ordinary principles

The International Maritime Organisation (IMO) has, since 2009, issued ‘interim’ guidance to flag States and shipowners on the use of PCASP to protect vessels from pirate attack in the High Risk Area.⁴⁷ The embarkation of PCASP is increasingly common with up to 15–25 per cent of all vessels transits through the High Risk Area now carrying them.⁴⁸ This State practice clearly indicates that the use of PCASP does not per se violate international law.⁴⁹ PCASP have also proved extraordinarily effective; so far, no vessel with embarked PCASP has been taken by pirates.⁵⁰ However, as the Enrica Lexie incident shows, having armed persons defend a vessel can have tragic consequences. When, then, might flag States incur responsibility if PCASP are embarked upon their vessels?

First, the use of PCASP may indirectly implicate the responsibility of a flag State. States are generally not directly responsible for the acts of individuals unless they are State agents or an exceptional rule applies.⁵¹ However, States may come under a duty to mitigate the risk that activities within their jurisdiction pose to other States (and their nationals). A State must take all ‘necessary steps immediately’ to prevent or mitigate injury to other States arising from dangers within its jurisdiction.⁵² This suggests that when States know their flag vessels are using PSCAP they are subject to an international duty to take necessary steps to mitigate any risk to other States (and their nationals) arising from that use. Such a duty would not involve strict liability; at most it would be a standard of ‘due diligence’, the content of which depends upon relevant treaty obligations owed other States.⁵³ In exercising rights of freedom of navigation on the high seas States must, under UNCLOS, ‘have due regard for the interests of other States in their exercise of the freedom of the high seas’.⁵⁴ This obligation alone could sustain a finding that a due diligence obligation exists regarding activities conducted aboard a vessel under a flag State’s jurisdiction. Further, under the Convention every State must also ‘effectively exercise its jurisdiction and control in administrative, technical and social matters over ships flying its flag’.⁵⁵ The term ‘administrative, technical and social matters’ is not defined elsewhere in the Convention.⁵⁶ It obviously extends, however, at least to such matters as are expressly provided for in the Convention concerning construction and seaworthiness, the crewing of vessels, and matters regarding communication and avoidance of collisions.⁵⁷ It should also be interpreted to include criminal jurisdiction generally, as a flag State must be able to effectively discipline seafarers in criminal cases involving ‘incidents of navigation’ (discussed above).⁵⁸

Taken together, these basic duties under UNCLOS to exercise ‘due regard’ and to ‘effectively exercise…jurisdiction’ suggest that if PCASP embarked on a vessel wrongfully injure foreign nationals or a foreign vessel, the flag State may risk incurring international responsibility. As above, this would not be vicarious liability for the acts of the PCASP themselves. Nonetheless, liability could arise from a failure of ‘due diligence’, such as a failure to take the minimum necessary steps to have an adequate regulatory regime in place governing PCASP and their acts.

What would such a minimum regime look like? Percy notes that private security companies in general are ‘poorly regulated’ because ‘they are inherently difficult to regulate’.⁵⁹ At the international level, regulatory efforts tend to find difficulty in securing agreement on effective measures; and ‘domestic regulation alone could never regulate an industry that operates almost entirely abroad’.⁶⁰ Further, the fast pace of industry change has meant that regulation tends to be backward-looking rather than focused on present challenges.⁶¹ One might think such concerns were lessened where the activities in question, while still extraterritorial, are not extra-jurisdictional. Flag States should be in a relatively good position to regulate maritime PCASP through their direct jurisdictional control over their vessels on the high seas. Whether this theoretical advantage is borne out in practice depends, of course, on the nature of regulation and the capacity of the flag State. A principal forum for regulatory efforts regarding maritime PCASP has been the IMO.

In May 2012 the IMO issued Revised Interim Recommendations for Flag States regarding the use of Privately Contracted Armed Security Personnel in the High Risk Area.⁶² These recommendations run only to two pages. Essentially, the IMO Recommendations note that the permissibility of PCASP aboard vessels ‘is a matter for flag States’, which may individually ‘determine if and under which conditions [their use] will be authorized’.⁶³ Flag States are encouraged to consider whether PCASP ‘would be an appropriate measure’ in the High Risk Area,⁶⁴ taking into account ‘the possible escalation of violence which could result from the use of firearms’.⁶⁵ Notably, flag States should also ‘require the parties concerned to comply with all relevant requirements of flag, port and coastal States’ as a vessel carrying PCASP may need to take those personnel (and their weapons) through a variety of jurisdictions.⁶⁶ If a flag State then decides to permit PCASP, the IMO recommends that any policy governing their use include consideration of:⁶⁷

1. 1. ‘the minimum criteria or minimum requirements with which PCASP should comply’, taking into account the relevant IMO guidance to shipowners on PCASP⁶⁸

2. 2. ‘a process for authorizing the use of [those] PCASP’ which meet with the flag State criteria

3. 3. ‘a process by which shipowners, ship operators or shipping companies may be authorized to use PCASP’

4. 4. ‘the terms and conditions under which the authorization is granted and the accountability for compliance associated with that authorization’

5. 5. ‘references to any directly applicable national legislation pertaining to the carriage and use of firearms by PCASP…and the relationship of PCASP with the Master while on board’

6. 6. ‘reporting and record-keeping requirements’.

The reference to the IMO Guidance to Shipowners does incorporate some further detail (as noted below).⁶⁹ Nonetheless, this IMO list of relevant considerations for flag States does more to highlight than to dispel the potential difficulties. First, while PCASP will have to comply with any flag State firearms regulation, the relevant port State laws will also apply. Where port States are hostile to weapons being carried aboard foreign ships, this may require that certain ports be avoided, or that weapons are ‘bonded’ aboard the vessel, or deposited at a government-run ‘floating armoury’,⁷⁰ or simply thrown overboard.⁷¹ Secondly, there is the question of the ‘relationship of PCASP with the Master’. The IMO Interim Guidance to Shipowners further recommend that a ‘documented [PCASP] command and control structure should provide…a clear statement recognizing that at all times the Master remains in command and retains the overriding authority on board’.⁷² Under international law, ultimate authority for decisions regarding safety aboard a merchant vessel rests with the master. For example, Regulation 8(a) of Chapter XI-2 of the SOLAS Convention provides: ‘The master shall not be constrained by the Company, the charterer or any other person from taking or executing any decision which, in the professional judgement of the master, is necessary to maintain the safety and security of the ship.’ The final, and perhaps most significant, difficulty in regulating the use of force by PCASP is in ‘reporting and record-keeping requirements’ and ensuring ‘accountability for compliance’ with relevant standards. The most important issue here is, obviously, standards governing the use of force and post-shooting incident inquiries. There may be real problems in conducting an adequate review of shooting incidents occurring far from the flag State’s territory, let alone mounting a criminal investigation.

Nonetheless, a flag State which fails to take adequate measures to prevent such abuses before the fact (through establishing an adequate regulatory framework) or to investigate them afterwards may incur responsibility towards the State whose nationals are injured. As regards the establishment of an adequate regulatory framework, Percy notes that the complexity of regulating the private security industry has tempted States generally to rely on industry self-regulation, such as the voluntary International Code of Conduct for Private Security Service Providers (ICoC) initiative convened by the Swiss government and the Geneva Centre for the Democratic Control of Armed Forces.⁷³ An International Standards Organisation (ISO) certification for PCASP is also being developed in consultation with the industry.⁷⁴ A completely ‘hands-off’ approach by States, simply relying on such third-party certification, would probably not satisfy due diligence obligations. Nonetheless, instruments such as the ICoC may provide evidence of the internationally accepted minimum standards for the conduct of PCASP on issues such as the use of force. Incorporating such third-party standards into national regulations would, however, likely be desirable.

States may already have national laws applicable to PCASP activities, even if they were not designed for such a purpose. For example, UK law does not require direct government approval of a decision to use PCASP, simply the submission of a ‘counter-piracy plan’.⁷⁵ However, aboard a UK-flagged vessel a person in possession of a firearm will likely require a certificate under the Firearms Act 1968;⁷⁶ and any UK national ‘supplying’ or ‘delivering’ weapons across an international border will require a relevant trade licence (unless covered by an employer licence).⁷⁷ While this latter law was designed to regulate arms traders, it (unintentionally) also applies, for example, to persons carrying corporate-owned weapons who at the end of a voyage will return them to their employer.⁷⁸ Whether such laws are appropriately adapted for regulating PCASP or PMSCs is another question.

Of course, a State must not only enact PCASP regulations but also take action regarding ‘accountability for compliance’. A key question will be accountability for the use of force. The question then arises as to the international standard for the use of force in self-defence by private individuals.⁷⁹ As regards the appropriate standard for the use of force by PCASP the IMO Interim Guidance to Shipowners states that:

This language closely tracks the ICoC, which borrows much of its drafting from the UN Basic Principles on the Use of Force and Firearms by Law Enforcement Officials.⁸¹ The UN Basic Principles are a soft-law instrument adopted by consensus by 127 States at the Eighth UN Congress on the Prevention of Crime and Treatment of Offenders in 1990.⁸² The document is thus not binding but represents persuasive evidence of widespread State consensus (opinio juris) as to the applicable law. While the UN Basic Principles apply only to State officials, the replication of these standards in other instruments applicable to private individuals, such as the ICoC and IMO Recommendations, may also suggest that they are increasingly accepted as having broader application. Failure to make adequate efforts to ensure compliance with such basic standards through, inter alia, appropriate post-incident investigation might thus incur State responsibility.

However, this is not to suggest that flag State implementation of international minimum standards suffices. The International Court of Justice and the International Tribunal for the Law of the Sea have defined ‘an obligation to act with due diligence’ as involving:

This makes the obvious point that due diligence involves not only identifying the risks that may be posed by ‘private operators’ and taking appropriate steps (including implementing any relevant international minimum standards) but also monitoring those private operators with an appropriate degree of vigilance. In the present case the IMO Recommendations simply point the way towards States devising effective regulatory systems of their own. They represent little more than a bare minimum supplemented by industry self-regulation. As Percy observes: ‘[w]hether or not the bare minimum is sufficient to regulate an industry that has significant lethal potential is questionable’; and given that this is a sector dealing ‘in lethal force’ one may also question the appropriateness of relying on self-regulation.⁸⁴ Neither formal compliance with the IMO Recommendations nor reliance on self-regulation will likely be enough to discharge any duty of due diligence. The key will be effective monitoring and enforcement. Weaknesses of flag State supervision is an acknowledged reality in matters such as fisheries management, and, if oversight of PCASP is no better, this could be deeply worrying.

4.2 Responsibility for State ‘complicity’ with corporate human rights breaches

Even so, this is a relatively narrow field within which States might incur responsibility. A question arises as to whether a State might also incur responsibility under a broader standard of ‘complicity’ in corporate misconduct. Such an approach could be based on an extension of Article 16 of the ILC Articles on State Responsibility dealing with situations where a State aids or assists another State’s internationally wrongful act:

One may suggest that the same principle should cover situations where a State assists certain types of corporate misconduct breaching human rights or international criminal law. McCorquodale and Simons have argued for just such an approach:

McCorquodale and Simons argue that it follows that a State may incur international responsibility for complicity in corporate acts which ‘if committed by that home state would constitute internationally wrongful acts…at least where the [state’s] aid or assistance “contributed significantly to that act”’.⁸⁶ The argument remains, however, speculative. Certainly, the ILC Articles do not preclude the individual responsibility of natural persons for international crimes. The idea that corporations can incur ‘international responsibility’ for international crimes is less well established. Direct criminal responsibility for corporate actors was famously excluded from the jurisdiction of the International Criminal Court. Further, the possible extension of Article 16 to cover complicity in corporate conduct is, at present, at best a suggestion de lege ferenda unsupported by State practice.

To the extent the argument is limited to violations of human rights law one may question whether we really need an expanded concept of State complicity at all, given that a different standard of attribution may already apply. Human rights bodies have long set a lower requirement for State responsibility for the actions of actors who are not State organs than the stringent test articulated by the ICJ.⁸⁷ That is, for the ICJ to find a State responsible for the acts of a non-State entity it requires either:⁸⁸

1. (a) the exercise of actual or complete control by a State over a non-State entity (that control being based on a position of complete dependence of the entity on the State) or

2. (b) ‘effective control’ over the particular conduct alleged to give rise to State responsibility. Essentially, there is State responsibility ‘if the person or group of persons is in fact acting on the instructions of, or under the direction or control of, that State’ in a particular case.⁸⁹

The European Court of Human Rights, by contrast, will find a State responsible for the actions of State-like entities where they are either under the ‘decisive influence’ of a State (as in the Ilaşcu case)⁹⁰ or which are acting within territory under the ‘effective overall control’ of a State (as in the Loizidou case where the governmental acts of the Turkish Republic of Northern Cyprus were held attributable to Turkey).⁹¹ Neither is a particularly stringent standard, and the latter may seem particularly lax. Milanović cogently suggests that cases such as Loizidou are not, strictly speaking, about attribution at all.⁹² Rather, they are about the circumstances in which a State has positive duties to secure or ensure human rights in territory under its control, for example the right to life. Thus, during an occupation, a State may incur international responsibility not only for wrongs committed by its own forces but also:

The suggestion in such cases is clearly that the State has a ‘due diligence’ obligation to take measures to prevent such violations. If this line of analysis is correct, human rights jurisprudence adds nothing unique to the present discussion. It simply provides a separate foundation for the same conclusion: States have an obligation to exercise due diligence to secure respect for human rights within spaces under their jurisdiction or control. This could readily extend to taking effective measures to minimise the risks of unlawful violence by PCASP or PMSCs operating from flag vessels.

Overview

One salient aspect that should be borne in mind when examining the relationship between the two systems of responsibility for collateral civilian casualties is the ramifications of the concept of military necessity. The following appraisals aim to address some of the confusion that may arise when the concept of military necessity is invoked as an attempt to deny responsibility for collateral civilian casualties on two levels (exculpation of criminal responsibility of a defendant; and/or exoneration of State responsibility).

The analysis starts with the thesis that the operational scope of this concept is confined to those specific IHL provisions that expressly or implicitly set it forth, so that any endeavour to plead for a further ambit of derogability to escape responsibility under the general context of IHL is futile. Next, the analysis turns to defending the argument that with the concept of military necessity already ingrained in the normative parameters of all IHL rules, reliance on it is excluded at the level of the secondary rules on State responsibility. The underlying assumption is that the principle of proportionality, which necessitates ascertaining the elements of military objective and military advantage, as well as the relative degree of ‘excessiveness’, is a specific form of the notion of military necessity.³³

No room for the concept of military necessity save under the specific rules that incorporate it

The concept of military necessity in the present-day context of IHL is completely differentiated from its historical counterpart that was affiliated to the Prussian doctrine of Kriegsraison geht vor Kriegsmanier,³⁴ which was premised on the very wide interpretation of the right of self-preservation.³⁵ This concept does not have any place outside specific exceptional clauses that provide for it.³⁶ Conversely, the treaty provisions that do not contain such saving clauses must never allow any plea of military necessity.³⁷ Otherwise, the very express clauses or provisions that integrate this concept would be deemed superfluous.³⁸ Once there is a finding that specific IHL rules have been flouted, there is no room for the concept of military necessity to operate again to exonerate the recalcitrant State’s responsibility,³⁹ as if this were an implied and overarching principle of IHL. Indeed, this chapter argues that in view of the close parallel development of customary law and conventional rules, there is no residual customary law that would allow the broader scope of application ratione materiae of the concept of military necessity beyond the clauses that incorporate it.⁴⁰

This understanding has been fully established in the case law.⁴¹ In the Krupp case, the United States Military Tribunal III-A at Nuremberg held that if the laws of war did not contain any escape valve of military necessity, then derogation was not permissible from those rules.⁴²Admittedly, despite the Krupp ruling, the Trial Chamber of the International Criminal Tribunal for the former Yugoslavia (ICTY) in some instances erred in reverting to the argument put forward by the defendants in the Krupp case, holding that ‘[t]argeting civilians is an offence when not justified by military necessity’.⁴³ This interpretation overlooks the non-derogable nature of the outlawing of a deliberate attack on civilians.⁴⁴ Such a construction has been duly rectified by the ICTY Appeals Chamber, which has reaffirmed that ‘there is an absolute ban on targeting civilians in customary international law’.⁴⁵ Allowing States to plead a general and implicit concept of military necessity outside the expressly recognised rules to justify deviations from otherwise unqualified IHL rules ‘would risk making the law…subservient to the exigencies of war’.⁴⁶

No room for military necessity under Article 25(2)(a) Articles on Responsibility of States for Internationally Wrongful Acts (ARSIWA)

If we reach a conclusion under IHL (namely, on the level of the primary law) that the notion of military necessity does not justify a certain conduct or omission, on the level of the secondary law of State responsibility, this notion can no longer serve as a ground for precluding an internationally wrongful act within the meaning of Article 25(2)(a) ARSIWA. At first glance, the contention that the state of necessity under the law of peace cannot be invoked in the context of IHL that deals with much greater exigencies than other fields of international law seems paradoxical. Yet, when we grasp that IHL already takes into account such exigencies, and above all that ‘[i]l n’y a pas une seule norme du droit des conflits armés qui ne réponde à une mise en balance entre les intérêts humanitaires et les intérêts issus des nécessités de la situation de belligérance’,⁴⁷ such a paradox should soon dissolve. Put succinctly, ‘a balance between military necessity and humanitarian concerns has been made in advance’⁴⁸ on the level of primary rules. As a consequence, once violations of the IHL rules on proportionality are found, the State concerned is precluded from invoking the concept of military necessity under Article 25(2)(a) of the International Law Commission’s (ILC) ARSIWA.⁴⁹

Article 25(2)(a) ARSIWA assumes this line of thought. It stipulates the non-availability of the plea of necessity in case ‘the international obligation [in the primary rule] in question excludes the possibility of invoking necessity’. The ILC commentary explains that:

The third sentence of the above passage contemplates the specific IHL rules that contain the concept of military necessity expressly, such as Articles 23(1)(g) the Hague Regulations; Articles 33(2), 34 and 54 First Geneva Convention 1949 (GCI); Article 126(2) GCIII; and Articles 49(2) and 53 GCIV.⁵¹ It also suggests that IHL rules deal essentially with emergency situations, and that such exigencies are already taken into account within the normative substance of the rules.⁵²

Most crucially, the ILC commentary highlights what should be obvious: the concept of military necessity is relevant only in respect of the obligations under IHL, and this concept is not given another part to play in the sphere of the law on State responsibility.⁵³ In other words, the concept of military necessity works exclusively in relation to the primary rules (that is, the rules of IHL), which is separate from the ‘general (secondary) rule of necessity’⁵⁴ under the law on State responsibility. The ILC commentary observes that:

The ARSIWA furnish another clear indication that the notion of ‘military necessity’ under IHL is a normative species that should be distinguished from the concept of ‘necessity’ under Article 25. Subparagraph (1)(a) of that provision calls for the absence of any other alternative to invoke the concept of necessity, a stringent requirement that is unfamiliar to the intrinsically elusive notion of military necessity under IHL.⁵⁶

Confusion as to the primary and secondary rules of necessity

The foregoing discussions notwithstanding, even the ICJ was once trapped in a confusion that suggests the ‘double counting’ of necessity on two levels: first, military necessity operating on the level of primary rules (IHL) and, secondly, the general or secondary concept of necessity deployed under the law of State responsibility.⁵⁷ The ICJ in its Wall Advisory Opinion ruled that:

This reasoning would ‘immunise’ an act not justified by military necessity under a state of necessity codified under Article 25 ARSIWA. The Court’s reasoning is incongruous as it itself acknowledges that the considerations of military exigencies are already inherent in the primary rules of IHL.⁵⁹ Venturini criticises that ‘[t]his method, which firstly applies international humanitarian law as lex specialis and subsequently goes back to lex generalis for further evaluation, tends to blur the traditional separation between jus in bello and jus ad bellum’.⁶⁰ Needless to say, such an approach would dilute the IHL’s foundational idea that this body of law is ‘agnostic’ or ‘eclectic’ with respect to the moral or deontological rationales of the casus belli.⁶¹

The ICJ Wall opinion entails the effect of enabling Article 25 ARSIWA to ‘give states “two bites at the apple” of necessity’.⁶² The Court’s dictum is at odds with the idea that IHL was crafted as a ‘closed system’, immune to all justifications, including any ground of necessity under the law on State responsibility.⁶³ The grounds of necessity codified under Article 25 ARSIWA are prevented from playing a ‘role de lege ferenda’ in allowing the introduction of ‘new rules of exception over the normative threshold’ into the primary rules of IHL.⁶⁴

1 Practice may be somewhat obscure, but third-party countermeasures are nevertheless part of the reality of international law

The saving clause in Article 54 ASR was based in part on the conclusion that State practice on third-party countermeasures was limited, embryonic, selective and confined to only a small number of mainly Western States.³⁹ But the ILC underestimated the true extent, consistency and geographical spread of State practice. There is, in fact, a considerable amount of practice evidencing the gradual recognition of an entitlement of States to adopt third-party countermeasures in response to gross and systematic breaches of communitarian norms.⁴⁰ No doubt a similar conclusion motivated the Institut de Droit International to endorse a decentralised regime of third-party countermeasures at its Krakow session in 2005.⁴¹

It is not the purpose of this chapter to revisit this (still very contentious) matter in any detail. It suffices here to point to the widespread support for the third-party countermeasures adopted by a large and remarkably diverse group of States against Syria in 2011. The adoption of these third-party countermeasures, which have included the freezing of internationally protected assets (such as those belonging to President Al-Assad and the Central Bank of Syria) and a prima facie unlawful membership suspension from the League of Arab States, underline their continuing relevance in the limited toolbox of communitarian law enforcement.⁴²

In February 2012, on the proposal of then French President Sarkozy, the so-called ‘Group of Friends of the Syrian People’ was created.⁴³ This is a large and diverse diplomatic coalition of States and international organisations, which was created as a direct response to the Security Council’s inability to take resolute action on Syria.⁴⁴ In April 2012, the ‘Friends of the Syrian People International Working Group on Sanctions’ (The Working Group) was formed:

At a meeting in September 2012, sixty Member States of the Working Group ‘welcomed the increasing pressure placed on the [Syrian] regime by the wide range of sanctions [e.g. the freezing of State and Head of State assets] adopted by different states and organisations’ and observed that they had ‘seriously affected the Syrian regime and have reduced its ability to crack down on the Syrian people’.⁴⁶ They added that ‘the members of the Group have adopted effective, proportional and coordinated sanctions…[and] urge[d] other countries to follow suit’.⁴⁷ More specifically:

The Working Group has reaffirmed the same point in several subsequent meetings.⁴⁹ As a minimum, these repeated statements, at least insofar as they relate to the freezing of assets belonging to President Al-Assad and the Central Bank of Syria, are indicative of a willingness of a very large number of States to adopt prima facie unlawful unilateral coercive measures for which the justification can seemingly only be explained in legal terms by the concept of third-party countermeasures.

A few comments about practice of a more general nature are also warranted. The evaluation of State practice on third-party countermeasures admittedly raises some difficult questions. One such difficulty concerns the relative obscurity of practice; the primary evidence in the form of statements from States is rarely conclusive. If a State often provides some form of explanation (however brief and perfunctory) for its adoption of a unilateral coercive measure it rarely provides a clear statement to explain which of the sometimes many possible unilateral coercive measures it actually relies upon in a given case to justify its action in legal terms. Then ILC member Opertti Badan alluded to this difficulty when he observed that the concept of third-party countermeasures belongs to ‘an area in which the borderline between international law per se and foreign relations [is] fairly indistinct’.⁵⁰ It is true that relevant examples from practice were ‘not always officially designated as [third-party] countermeasures’.⁵¹ Indeed, as Special Rapporteur Crawford has observed, States appear to have an ‘implied preference for other concepts’.⁵² The fact that foreign policy considerations – and the concomitant sensitivities surrounding the use of non-forcible coercion by individual States – appear to play a prominent role in this area helps to explain the relative obscurity of practice on third-party countermeasures and the considerable complexities involved in assessing it. These complexities are evident in the apparent unwillingness of States to rely expressly on the concept of third-party countermeasures in practice; this is a factor that has significantly clouded the legal issues involved and sometimes led to confusion in the analysis of opinio juris.

These problems are compounded in those situations where States have strongly supported third-party countermeasures in their practice, yet opposed them in the Sixth Committee of the General Assembly during the ILC’s work on State responsibility. As an illustration, the United States and the United Kingdom opposed the concept of third-party countermeasures in the Sixth Committee as an ‘inappropriate vehicle’ and as a ‘highly destabilizing’ means of enforcing fundamental norms, yet they have both on numerous occasions resorted to third-party countermeasures in practice.⁵³ For its part, Tanzania opposed third-party countermeasures by asserting before the Sixth Committee that ‘it could hardly be refuted that [third-party] countermeasures were a threat to small and weak States’.⁵⁴ And yet Tanzania has itself adopted third-party countermeasures against South Africa, Nigeria, Burundi and Zimbabwe.⁵⁵ Likewise, Botswana, another ardent critic in the Sixth Committee,⁵⁶ expressed support in the Security Council for the third-party countermeasures adopted against Burundi by certain African States which in that instance at least ‘deserve[d] the commendation of the international community’.⁵⁷

While some hesitation may be understandable in accepting the relevance of the mentioned practice to the development of the law on third-party countermeasures, this appears to be an inevitable consequence of the complex interplay between law and politics in this area. Special Rapporteur Riphagen may have been right to observe that ‘the more serious the breach of an international obligation, the less likely it is to find an objective legal appraisal of the allowable responses to such a breach’.⁵⁸ Still, in the final analysis, as States have relied on the rationale of the concept and no other justifications have been available, the concept of third-party countermeasures is needed to explain the practice of States in legal terms.⁵⁹ In short, third-party countermeasures cannot be described as a ‘neologism’ – a category ‘completely invented’ by ex post facto rationalisations of practice.⁶⁰ If the matter was still in doubt a decade ago, third-party countermeasures are today part of the reality of international law.

2 Third-party countermeasures and UN Charter Chapter VII enforcement measures are distinct and may operate in parallel

The influential view in the ILC and the Sixth Committee, already noted above, that a decentralised regime of third-party countermeasures would be contrary to the UN Charter as it would encroach on the powers of the Security Council under Chapter VII, does not withstand scrutiny. At least in part, it is a view initially informed by Special Rapporteur Ago’s conclusion from the 1970s that, as a matter of lex lata, obligations erga omnes were not enforceable by a decentralised regime of third-party countermeasures; but by the organised international community, in the form of the competent organs of the UN, notably the Security Council.⁶¹ From historical experience, in which countermeasures ‘were frequently used as an instrument of intervention or aggression’,⁶² it was not possible to underestimate the risks of abuse involved in pressing recognition of the concept of unilateral third-party countermeasures and the introduction of another circumstance precluding wrongfulness, which ‘sooner or later might extend to the use of force’.⁶³ During the debate on the ILC Articles in the Sixth Committee, several States insisted in categorical terms that third-party countermeasures could only be adopted within the UN framework.⁶⁴ Cameroon’s position aptly sums up these concerns. It stated:

It was against the background of these concerns that much of the debate in the ILC and Sixth Committee on third-party countermeasures came to focus on the potential role of the Security Council as an institutional safeguard against abuse. But these concerns are not reflected in State practice; and in addition, they conflate the analytical distinction between enforcement measures under Chapter VII of the UN Charter and third-party countermeasures.

A striking feature of the very substantial amount of State practice in the UN period is that third-party countermeasures have largely been adopted without any intervention of the Security Council whatsoever. In fact, in several cases, such as those involving Argentina, Iraq, Burundi, Yugoslavia and Sudan, third-party countermeasures have even been adopted while the Security Council has been actively seized of these matters.⁶⁶ Likewise, the third-party countermeasures adopted against Syria in 2011 continue (without much controversy)⁶⁷ to operate in parallel with the action taken against it by the Security Council in September 2013.⁶⁸ The practice accumulated over several decades provides a strong indication that third-party countermeasures do not contradict the UN Charter, nor unduly encroach on the powers of the Security Council.

The preferred policy option of the Special Rapporteurs on first reading had nevertheless been to link the law of collective security to the law of State responsibility. As it happened, this approach, in its various forms,⁶⁹ was overwhelmingly rejected as incompatible with the existing powers of the Security Council under the UN Charter.⁷⁰ In any event, the ILC deemed such an institutional safeguard unnecessary given the likely involvement of the Security Council in addressing the most serious illegalities.⁷¹ Ultimately, the complex relationship between third-party countermeasures and the Security Council is safeguarded by Article 59 ASR, which is based on what could perhaps be described as the principles of co-existence and co-ordination.

The principle of co-existence is based on the premise of two distinct spheres of application: the Security Council deals with the political aspects of maintaining or restoring international peace and security, whereas the law of State responsibility (including third-party countermeasures) deals with the legal aspects of serious breaches.⁷² As Jordan explained before the Sixth Committee:

The principle of co-existence, that is to say, the seemingly clear-cut distinction between the law of State responsibility and the law of collective security,⁷⁴ appears to have been the source of some confusion among a number of States in the ILC’s work on State responsibility; almost by definition, the Security Council cannot resort to third-party countermeasures. As Spain emphasised before the Sixth Committee:

In the ILC debate, Simma observed that ‘it seemed not entirely clear to some States’ that third-party countermeasures and enforcement measures under Chapter VII of the UN Charter could not be assimilated.⁷⁶ Put simply, there is no a priori role for the Security Council in a regime of third-party countermeasures.

A reason for the apparent confusion could be found, as Special Rapporteur Crawford suggested, in what France termed the ILC’s ‘ambiguous’⁷⁷ definition of countermeasures on first reading.⁷⁸ It gave the unfortunate impression of including within its definition both institutional (that is, Chapter VII of the UN Charter) and decentralised forms of coercive measures, as opposed to clearly spelling out the critical distinction of a previous wrongful act that separates Chapter VII enforcement measures from countermeasures. As Crawford explained, ‘collective responses of that sort were not countermeasures; they were measures authorized by a competent international organization and did not belong in the framework of article 30 [1996] [defining countermeasures on first reading]’.⁷⁹ France (followed by Crawford and Simma) indeed articulated the need for this issue to be clarified on second reading, but the matter was not taken further.⁸⁰ The ambiguity, then, essentially remains the same in Article 22 ASR.⁸¹ Still, the basic point of distinction remains intact. Countermeasures are a separate genus from Security Council action; they perform different functions and can therefore, in principle, co-exist autonomously in response to the same wrongful conduct. Third-party countermeasures cannot be said to encroach on the (distinct) powers of the Security Council, let alone contravene the UN Charter in any a priori sense. That said, even where conceptual boundaries are clearly demarcated in theory, functional ones may not be in practice. This brings us to the more complex principle of co-ordination.

In reality, it is not uncommon for the Security Council – in the maintenance or restoration of international peace and security under Chapter VII of the UN Charter – to adopt enforcement measures aimed at ensuring the cessation of wrongful conduct and/or reparation for injury. In these situations, there is inevitably a point of convergence with the law of State responsibility and, in particular, with the law of countermeasures. The ILC recognised this fact by pointing out that at least ‘[cessation] is frequently demanded not only by States but also by…the Security Council in the face of serious breaches of international law’.⁸² It follows that enforcement measures under Chapter VII of the UN Charter, such as various forms of asset freezes, embargoes or other suspensions of treaty rights – to the extent that they overlap with the law of State responsibility – may act either as a complement to the possible concurrent use of third-party countermeasures or operate to constrain such use in a given case.

The extent of the actual interplay or co-ordination between the law of State responsibility (including countermeasures) and the law of collective security raises questions of considerable complexity with no obvious answers from practice. As a general matter, it can be observed that the UN Charter is not a self-contained regime. It can therefore be argued that third-party countermeasures will remain available (subject to overall compliance with its safeguards regime, notably proportionality)⁸³ to the extent that the Security Council is ineffective; or their use has not been proscribed by the specific language of a given Security Council Resolution; or would otherwise contradict Charter obligations in the circumstances.

3 There is no requirement of a ‘widely acknowledged breach’, although in practice joint statements on alleged serious illegalities limit the risk of abuse

The major and potentially most effective criticism of countermeasures undoubtedly concerns the inherent risk of abuse associated with the auto-interpretation of allegedly wrongful conduct. It is a risk exacerbated by the factual inequalities between States. As the arbitral tribunal in the Naulilaa case was careful to emphasise, ‘the first requirement – sine qua non – of the right to take reprisals is a motive furnished by an earlier act contrary to the law of nations’.⁸⁴ Likewise, the ILC stressed that the existence of a prior breach of international law entitling a State to invoke the responsibility of the wrongdoing State is a ‘fundamental prerequisite’ of any lawful countermeasure; its establishment ‘presupposes an objective standard’.⁸⁵ A State that resorts to countermeasures in the erroneous belief that a breach has occurred does so at its own peril – caveat actor.⁸⁶ Still, in an essentially decentralised system, ‘each State establishes for itself its legal situation vis-à-vis other States’.⁸⁷ This raises the spectre that any number of States could individually resort to third-party countermeasures contrary to the obligations incumbent upon them on the basis of the mere assertion or claim of wrongful conduct.

These concerns are understandable and help explain the ‘extreme sensitivity’⁸⁸ of the topic. A decentralised regime of third-party countermeasures is an institution that affects the very foundations of international law; and, at least in theory, poses a threat of some seriousness to sovereignty and the freedom of action of States within the law (the domaine réservé). The fear that a future incarnation of Lord Palmerston might abusively exclaim ‘Civis mundi sum!’ as mere pretext for ‘power politics’⁸⁹ underlies the deep-rooted concern among some States (especially – but by no means limited to – developing States) that a regime of decentralised and self-assessed third-party countermeasures would be ‘used as an instrument of intervention and aggression’.⁹⁰ In short, on this view, the concept of third-party countermeasures merely ‘provided a superficial legitimacy for the bullying of small States on the claim that human rights must be respected’.⁹¹

The proposed solution favoured by the ILC Special Rapporteurs on first reading was therefore to involve the United Nations in the ‘objective’ prior determination of breach. For his part, Special Rapporteur Riphagen’s proposal to involve the Security Council in this process was based on his belief that:

In a similar vein, as already alluded to above, Special Rapporteur Arangio-Ruiz proposed a far more elaborate and ambitious institutional safeguards regime for third-party countermeasures informed by the ‘indispensable role of international institutions’.⁹³ In essence, his scheme was based on a two-phase procedure in which the General Assembly or the Security Council would first make a political determination (under Chapter VI of the UN Charter), and the International Court of Justice (ICJ) would later make a decisive legal determination concerning the possible existence of a serious breach.⁹⁴ However, this regime was resoundingly rejected in both the ILC and Sixth Committee as inconsistent with the UN Charter; it was simply unrealistic – far removed from States’ conception of international law as a decentralised system of law⁹⁵ – and, in any event, woefully ineffective. By the time the ICJ would finally authorise the use of individual third-party countermeasures any serious breach (such as genocide) would likely already have been consummated.⁹⁶

On second reading, Special Rapporteur Crawford identified the risk of abuse posed by auto-interpretation as an important ‘due process’⁹⁷ issue for the target State, but his proposed solution was more attuned to what remained a fundamentally decentralised system of international law enforcement. In the absence of a prior judicial determination of the existence of a previous wrongful act, it is often difficult to assess with confidence whether there has indeed been a violation of international law, let alone a serious violation. As Crawford explained:

He concluded that for decentralised third-party countermeasures to be permissible, ‘some formula such as a “gross and reliably attested breach” was called for’, alternatively formulated as ‘gross and well-attested breach’ or ‘serious and manifest breach’.⁹⁹ Ultimately, the ILC left the formula for the prior ascertainment of breach open by the agnostic arrangement embodied in Article 54 ASR. Still, the ILC did make clear that an institutional procedure for the prior ascertainment of breach would contradict the UN Charter and, in any event, be unnecessary given the likely involvement of the main UN political organs in addressing serious breaches.¹⁰⁰

Finally, in its Krakow Resolution of 2005, the Institut de Droit International proposed its solution to the problem of auto-interpretation. On Rapporteur Gaja’s proposal, it concluded that:

The Institut explained that ‘the reference to the wide acknowledgment of the existence of a breach is designed to limit the risk of unilateral assessment…diminish the risk of abuses and ensure that States genuinely seek to protect an interest of the international community’.¹⁰² Importantly, this position found support in practice on third-party countermeasures which has ‘generally related to infringements of obligations erga omnes that were indeed widely acknowledged’.¹⁰³

Indeed, resort to third-party countermeasures is often preceded by joint statements adopted by States as part of a process of multilateral diplomacy in international fora.¹⁰⁴ As part of this (often lengthy) diplomatic process, the responsible State will normally have been publicly requested in a joint statement to comply with its secondary obligations and will have been notified of the possible imminent adoption of third-party countermeasures against it in the event that it does not promptly return to international legality. Still, practice does not seem to indicate that States have co-operated in this way out of a sense of legal obligation. Instead, these joint statements appear predominantly motivated by an overriding practical imperative: co-ordinating an effective response to the most serious breaches of international law. Two separate considerations appear to bear out this conclusion.

First, as the ILC itself has recognised, it is ‘open to question’¹⁰⁵ whether general international law at present entails a positive duty of co-operation in response to serious breaches. The prevalence of the phenomenon of co-operation in response to serious breaches of community norms, especially in the context of international organisations, does not diminish the force of this conclusion. Quite simply, such co-operation is ‘often the only way of providing for an effective remedy’.¹⁰⁶ Secondly, a duty of co-operation would seemingly be inconsistent with the instrumental function of countermeasures. As a minimum, States have an individual or autonomous entitlement to claim cessation of breaches of erga omnes obligations within the meaning of Article 48 ASR. A requirement of a joint statement, by which a breach must first be ‘widely acknowledged’, would be inconsistent with the autonomous entitlement of States to at least claim cessation for serious breaches under general international law, a claim par excellence enforced by third-party countermeasures. In the absence of clear opinio juris to the contrary, it can be presumed that practical considerations of effectiveness have predominantly influenced practice in this field. In principle, then, a State may lawfully resort to third-party countermeasures (assuming a relevant breach has actually occurred) even in those rare circumstances where the breach is not ‘widely acknowledged’. Perhaps most importantly, practice also indicates that third-party countermeasures – whatever the mode of establishment of the individual breach – have overall not been adopted in an abusive manner but on the well-founded belief of a serious infringement of a community norm.

Introduction

I do not consider this report to be a ‘bad decision’ as some have called it,⁶⁰ though it could have been clearer. The Appellate Body’s use of the ILC Articles was hesitant but not necessarily incorrect.⁶¹

Whilst the Appellate Body’s interpretation of ‘public body’ corresponds with the head of attribution set out in Article 5 of the ILC Articles, it neglected to clarify how secondary rules (of State responsibility) are relevant to interpreting a provision which sets out the scope of primary rules, the breach of which results in a finding of State responsibility (but does not itself contain an obligation the breach of which might result in State responsibility). Nor did it convincingly establish the basis in the Vienna Convention for interpreting the term ‘public body’ against the background of Article 5 of the ILC Articles.

Article 31(3)(c) of the Vienna Convention sets forth three elements (‘rules of international law’, ‘relevant’ and ‘applicable in the relations between the parties’). Yet the third element becomes obsolete if the rule is accepted as reflecting customary international law or a general principle of (international) law. In those circumstances, the rule evidently applies in the relations between the parties because of its general application (and irrespective of whether ‘the parties’ is held to mean ‘the parties to the dispute’ or ‘the parties to the treaty being interpreted’).⁶²

Based on that consideration, the Appellate Body rightly focused on two elements: the status of the ILC Articles and their relevance to the question being considered. Logically, the Appellate Body first examined the second element because the issue of the status of the ILC Articles is inconsequential if those articles are found not to be relevant to the meaning of the term ‘public body’.

Relevance of the ILC Articles

The Appellate Body defined relevance in function of the subject matter of both Article 1.1(a)(1) of the SCM Agreement and Articles 4, 5 and 8 of the ILC Articles: all set out rules relating to the attribution of conduct to a State.⁶³ Leaving aside conceptual distinctions between the character of each set of rules, the status of the ILC Articles and the extent to which each set was defined by reference to conduct and/or entity, it appears clear that the Appellate Body accepted the relevance of, in particular, Article 5 to interpreting the term ‘public body’ in Article 1.1(a)(1). Yet, despite the ILC Articles’ relevance, the Appellate Body decided to forego determining the status of Article 5 as a matter of public international law because the outcome of its interpretive exercise did ‘not turn on’ it. That reasoning enabled the Appellate Body to avoid basing its analysis more firmly on Article 31(3)(c) of the Vienna Convention. In so doing, the Appellate Body injected a high standard of relevance into Article 31(3)(c) without explaining either the basis for that standard, or how to distinguish it from the normative weight that is to be given to a rule satisfying the conditions set out in Article 31(3)(c). Nor did it reflect upon how that standard affects the relationship between the general rule of interpretation in Article 31 of the Vienna Convention and that in Article 32. Those matters are discussed in the following section of this chapter.

Under the SCM Agreement, WTO Members can be held responsible for violating that agreement if they made a financial contribution, thereby conferring a benefit, which is contrary to the substantive obligations set out in that agreement. Article 1 of the SCM Agreement defines a ‘subsidy’ by reference to (i) certain types of conduct, (ii) the entity performing the conduct and (iii) the conferral of a benefit as a result of that conduct. Under Article 1.1(a)(1), the types of conduct described in items (i)–(iv) all involve ‘government’ conduct for the purposes of the SCM Agreement. The use of the word ‘government’ in each of the items makes that plain. The meaning of that word is further explained in Article 1.1(a)(1) as meaning ‘a government or any public body within the territory of a Member’. Evidently, Article 1 of the SCM Agreement forms part of the description of the scope of application of the obligations under the SCM Agreement the breach (or rather nullification or impairment) of which may entail the responsibility of the WTO Member awarding the subsidy. Those obligations apply to ‘government’ conduct as defined in items (i)–(iv) in Article 1.1(a)(1).

Thus, it appears undisputed that contributions that cannot be linked to a WTO Member cannot constitute a subsidy for which that Member can be held responsible because it violates the substantive obligations in the remainder of the SCM Agreement.⁶⁴ That required link is expressed in three different forms in Article 1.1 of the SCM Agreement: (i) the government itself, (ii) any public body within the territory of a Member and (iii) a private body entrusted or directed to carry out one or more of the functions illustrated in (i)–(iii) of Article 1.1(a)(1) of the SCM Agreement which would normally be vested in the government.⁶⁵ The first and second forms are collectively named ‘government’ in the SCM Agreement.

The Appellate Body appeared to take the same starting point: the measure defined under Article 1.1(a)(1) of the SCM Agreement is a subsidy only if it is attributable to a State. Thus, the element of attribution was treated as an intrinsic element of the definition. According to the Appellate Body, the connecting factor in the ILC Articles was conduct, whereas under Article 1.1(a)(1) of the SCM Agreement it was both the particular conduct and the type of entity.⁶⁶ It did not appear to recognise that, for example, the distinction made between Articles 4 and 5 of the ILC Articles is based also on the character of the entity performing the conduct, in particular whether or not that entity qualifies as an organ of State.⁶⁷ Nor did the Appellate Body take account of the separate functions of both types of rule – one being secondary norms and the other being primary rules (though of a type affecting the scope of application of an agreement). Whilst that distinction might be, as James has pointed out, somewhat artificial,⁶⁸ the respective functions of each type arguably cannot be ignored in determining whether one set is relevant to interpreting the other.

However, the Appellate Body was correct in characterising, albeit in an indirect manner, the matter under appeal as one regarding the attribution of conduct. Conduct that is not that of a WTO Member and, in particular, conduct that is not government conduct as defined in Article 1.1 cannot lead to consequences under the SCM Agreement. It is in this context that questions of attribution arise and that the ILC Articles’ provisions on attribution became materially relevant. That context is not confined to the issue of the scope of the SCM Agreement. In principle, attribution underlies most findings of inconsistency with WTO obligations because the latter are mostly obligations owed by WTO Members and in principle a nexus between a measure and a Member is required in order for those measures to be subject to dispute settlement proceedings.⁶⁹ However, Article 1.1(a)(1) of the SCM Agreement is distinct, though not unique, when compared to many other WTO provisions in that it expressly internalises attribution in defining the material scope of application of that agreement.⁷⁰ On that basis, the argument according to which the ILC Articles, and thus also the rules on attribution, become relevant only if something wrongful has happened lacks merit.⁷¹ Indeed, the characterisation of a measure as a subsidy is separate from the determination of whether that measure is inconsistent with the substantive obligations set out in the SCM Agreement (that is, whether the subsidy is wrongful as a matter of WTO law). In the context of the SCM Agreement, the issue of attribution pertains to the former inquiry but not to the latter.

The Appellate Body has been criticised for using in this decision the ILC Articles that ‘have nothing to do with international trade’.⁷² That argument fails to acknowledge that the Appellate Body’s reasoning reflects the logic that if WTO Members are to be held responsible for breaches of the treaties to which they consented to be bound, rules on attribution are indispensable for determining whose or what entity’s conduct entails that responsibility. In that regard, WTO law is not distinct from public international law in general. In most cases decided before panels and the Appellate Body, that question of attribution is taken for granted because the measure at issue is obviously one that can be attributed to the State (usually because, on its face, it falls under the general rule of attribution set out in Article 4). Yet, with respect to certain provisions or agreements, such as the SCM Agreement, the issue of attribution is internalised in the terms defining the scope of (the obligations assumed under) the treaty.⁷³ In those circumstances, it becomes difficult to distinguish, in conceptual terms, the material scope of the primary rules the breach of which results possibly in State responsibility and the secondary rules offering the background against which to assess whether the conduct is that which can be attributed to the State.

Here, the Appellate Body thus accepted that attribution formed part of the material definition of a subsidy. However, whilst the SCM Agreement sets out the types of entity whose conduct, if corresponding with the forms of financial contribution listed in items (i)–(iii) and resulting in conferral of a benefit, can entail the responsibility of a WTO Member under that agreement when that conduct is inconsistent with the substantive obligations in the SCM Agreement, it did not define the terms ‘government’ and ‘public body’.

The text of Article 1.1(a)(1) of the SCM Agreement, when read in isolation, can be interpreted in two ways. First, financial contributions made by a government, defined as encompassing both ‘a government’ and ‘any public body within the territory of a Member’, result in the responsibility of the WTO Member concerned for violating the SCM Agreement if found to be inconsistent with substantive obligations in that agreement independently from whether that ‘government’ was entrusted with the exercise of governmental authority. By contrast, financial contributions made by a ‘private body’ lead to the same result only if it was entrusted or directed to carry out a governmental function. However, that interpretation does not to distinguish ‘government’ from ‘public body’. Nor does it explain why collectively they are termed ‘government’ for the purpose of the SCM Agreement. Secondly, ‘government’ and ‘any public body within the territory of a Member’ can be read as having separate meanings (despite their collective denomination as ‘government’), which are also distinct from those given to the term ‘private’ bodies described in Article 1.1(a)(iv). Whilst the text does not prescribe that public bodies must be directed or entrusted to perform certain functions in the same manner as private bodies, nor is their conduct or status as such sufficient to characterise them as belonging to the government. However, that interpretation does not resolve how to interpret the term ‘public body’ as meaning something different from the term ‘government’ as it first appears in Article 1.1(a)(1).

The Appellate Body opted for the second reading. It appears to have accepted that the first type of ‘government’ reflected the notion of the government in the strict sense or of those entities whose conduct is described in Article 4.1 of the ILC Articles – that is to say, ‘conduct of any State organ shall be considered an act of that State under international law, whether the organ exercises legislative, executive, judicial or any other functions’. That provision ‘defines the core cases of attribution’.⁷⁴ If the body is a State organ under internal law,⁷⁵ then its conduct is attributable to the State under Article 4.1. Under the most basic rule of attribution – which is accepted as reflecting customary international law⁷⁶ – that conduct must be exercised in official capacity.⁷⁷ The second type was read by the Appellate Body as broadly corresponding to the conduct described in Article 5, which applies only if the entity or person is not an organ of State under Article 4 and apparently has a legal personality separate from that of the State.⁷⁸

Article 5 of the ILC Articles applies to a person or entity that is ‘empowered by the law of [a] State to exercise elements of the governmental authority…provided the person or entity is acting in that capacity in the particular instance’. The provision is intended to cover ‘para-statal entities, which exercise elements of governmental authority in place of State organs…[and] situations where former State corporations have been privatized but retain certain public or regulatory functions’.⁷⁹ According to its commentary, such entities may include ‘public corporations, semi-public entities, public agencies of various kinds and even, in special cases, private companies’.⁸⁰ In all circumstances, Article 5 applies solely if those entities are ‘empowered by the law of the State to exercise functions of a public character…and the conduct of the entity relates to the exercise of the governmental authority concerned’.⁸¹ It can cover conduct of private and public entities.⁸² What constitutes ‘governmental authority’, according to the commentary, depends on ‘the particular society, its history and traditions’ and ‘the content of the powers, but [also] the way they are conferred on an entity, the purposes for which they are to be exercised and the extent to which the entity is accountable to government to their exercise’.⁸³

Leaving aside how the Appellate Body formally used the ILC Articles in order to interpret ‘public body’ in Article 1.1(a)(1) of the SCM Agreement and its focus on Article 5 of those articles, it thus appears to have read the terms ‘government’ (the first form) and ‘public body’ (the second form) as reflecting the distinction between the conduct described in Article 4.1 of the ILC Articles (conduct of State organs) and Article 5 of the same articles (conduct of persons and entities that are not State organs within the meaning of Article 4.1).

That leaves the question of how Article 1.1(a)(1)(iv) of the SCM Agreement (conduct of a private body entrusted or directed to carry out one or more of the type of functions illustrated in (i)–(iii) which would normally be vested in the government) fits in that structure. Article 5 of the ILC Articles makes no distinction between private and public entities. In principle, a private body can also be empowered by the law of a State to exercise elements of the governmental authority. In its report in US – Countervailing Duty Investigation on DRAMS, the Appellate Body accepted that position.⁸⁴ There, it held that ‘“entrustment” occurs where a government gives responsibility to a private body, and “direction” refers to situations where the government exercises its authority over a private body’, and whether either form of instruction occurs ‘will hinge on the particular facts of the case’.⁸⁵ At first glance, it thus would seem that Article 1.1(a)(1)(iv) of the SCM Agreement also describes heads of attribution which can be characterised as corresponding with that set out in Article 5 of the ILC Articles – at least on the understanding that transfer of authority resembles entrustment or direction. If that is indeed the consequence of reading together the Appellate Body’s decisions in both cases, do the qualitative tests for attributing conduct of a ‘public body’ and a ‘private body’ differ?

That remains unclear. Whilst the Appellate Body in US – Anti-dumping and Countervailing Duties (China) considered what it called the juxtaposition of the collective term ‘government’ and ‘private body’, it did so to draw conclusions on the apparent ‘nexus’ between the term ‘government’ in the strict sense and the term ‘public body’.⁸⁶ In that context, it did identify one distinction between a ‘public body’ and a ‘private body’ within the meaning of Article 1.1(a) of the SCM Agreement: a public body, just like ‘government’ in the strict sense, has ‘the requisite attributions to be able to entrust or direct a private body, namely, authority in the case of direction and responsibility in the case of entrustment’⁸⁷ and this cannot be an attribute of a private body. The use of the term ‘requisite’ suggests that the governmental functions, which the public body must be authorised to exercise, must include this attribute. That might also indicate that (without having been expressly articulated by the Appellate Body in that manner) – in terms of heads of attribution – three different forms of government authority are reflected in the text of Article 1.1(a)(1) of the SCM Agreement and that the form relevant to the attribution of conduct by a public body is considerably stricter than those forms expressed in Article 1.1(a)(iv) and possibly stricter than that articulated in Article 5 of the ILC Articles. The first form, with respect to ‘public body’, appears to be wider than that described in Article 1.1(a)(iv) and includes also the attribute described above. Against that background, future controversies will thus concern evidentiary standards for establishing each form and the applicable standard of review. Only then will it become clear what the proper standards are for determining whether an entity is a ‘public body’ or a ‘private body’ within the meaning of Article 1.1(a)(1) of the SCM Agreement.

Status of the ILC Articles

The Appellate Body’s decision to forgo determining the status of in particular Article 5 of the ILC Articles is unhelpful but not surprising.⁸⁸ Had the outcome of the Appellate Body’s interpretation of the term ‘public body’ turned on Article 5, it would have been necessary for it ‘to resolve definitively’ whether and to what extent Article 5 of the ILC Articles reflected customary international law.⁸⁹ That was the Appellate Body’s explanation for not examining the legal status of Article 5: the provision, whatever its status, provided ‘further support’ for its analysis.⁹⁰

This explanation is difficult to reconcile with the Appellate Body’s so-called holistic approach to interpreting the covered agreements based on the ordinary meaning of their wording, context, object and other elements of interpretation included in the general rule in Article 31 of the Vienna Convention. It is also an unhelpful contribution to the wider debate about how to interpret those agreements against the background of international law and the incomplete case law addressing that question. By refusing to determine the status of Article 5, whilst accepting that that status of being a rule of international law is an essential element for applying Article 31(3)(c) of the Vienna Convention,⁹¹ the Appellate Body thus accepted, albeit in an implicit manner, that it did not interpret on the basis of Article 31(3)(c) the term ‘public body’ against the background of Article 5 which, as is also apparent from its analysis, it nonetheless took to be relevant in that regard.⁹²

What was then the basis for the Appellate Body’s use of Article 5 of the ILC Articles? Possibly Article 32 of the Vienna Convention which provides for recourse to supplementary means of interpretation in order to, inter alia, ‘confirm the meaning resulting from the application of article 31’. If that is, indeed, how the Appellate Body used Article 5 of the ILC Articles, then the interpretation of the covered agreements against the background of public international law:

1. (a) is subject to the conditions of Articles 31(3)(c) of the Vienna Convention only if the interpretation turns on, in the sense of depending on, those rules of public international law; but

2. (b) is subject to no conditions under Article 32 of the Vienna Convention if the interpretation is merely supported or confirmed by those rules of public international law.

The distinction between approaches (a) and (b) would then be based on the degree to which a rule of international law is accepted to be ‘relevant’. The Appellate Body appears to read that term in Article 31(3)(c) of the Vienna Convention as implying a high standard so as to mean: pertinent to or bearing upon, or possibly decisively important to the meaning of, a term or phrase in the covered agreements. Yet, that reading of Article 31(3)(c) of the Vienna Convention is at odds with how the Appellate Body uses the different elements of interpretation in the general rule in Article 31. In general, its position has been that, for example, the ordinary meaning, the intent of the parties, the factual context and the circumstances surrounding the conclusion of the treaty are not ‘rigid components’, thus confirming the holistic approach towards treaty interpretation.⁹³ In principle, none of the elements is controlling even if the weight to be given to each element, with respect to a particular text, might not be the same. In that regard, the Appellate Body did not explain how its position on the meaning of ‘relevant’ in Article 31(3)(c) of the Vienna Convention can be reconciled with the distinction it made in EC and Certain Member States – Large Civil Aircraft between the three elements for applying Article 31(3)(c) of the Vienna Convention and the normative weight to be ascribed to a rule satisfying those elements: the latter was controlled by the introductory phrase to Article 31(3) ‘[t]here shall be taken into account, together with the context’.⁹⁴

But perhaps the Appellate Body did not intend its statement on the status and relevance of Article 5 of the ILC Articles to have such a wider implication on the meaning of Article 31(3)(c) of the Vienna Convention and the relationship between that provision and Article 32 of the Vienna Convention. Rather, the statement might need to be taken as an expression of the Appellate Body’s reluctance to examine what constitutes customary international law. Where the status of a norm under international law is not definitely settled elsewhere, and in particular in the form of a pronouncement of the International Court of Justice, the Appellate Body is hesitant, if not defiant, to inquire itself into the evidence of that status.⁹⁵ So far, the Appellate Body has not exposed the theoretical underpinning for that position. An inquiry into evidence of the customary international law status of a rule requires an in-depth analysis of the material sources showing State practice and that the rule is accepted as international law (opinio iuris), involving possibly a considerable body of factual elements requiring sorting for determining their legal relevance.⁹⁶ That material evidence must be put before the Appellate Body and possibly be subject to a debate between parties that, so far, is usually lacking from submissions to interpret the WTO agreements against the background of customary international law by virtue of Article 31(3)(c) of the Vienna Convention. Jurisdictional limitations to the Appellate Body’s inquiry in that regard do not appear to exist. If the Appellate Body is tasked to interpret the WTO agreements by using customary principles of treaty interpretation, including Articles 31–3 of the Vienna Convention, it must properly apply those principles, including the conditions set forth in Article 31(3)(c) of the Vienna Convention. Examining and taking a position on the evidence of the existence of a norm of customary international law are part of the judicial function upon which the Appellate Body acts in that context.

Putting aside those jurisdictional issues, the Appellate Body’s avoidance of that type of inquiry could also be taken to signal a degree of judicial comity:⁹⁷ it prefers to defer to the judgment of other international courts and tribunals, in particular the International Court of Justice, on what is customary international law. That position might become difficult to sustain in the light of the growing body of dispute-settlement activity in the context of which material sources are examined and positions are taken on what constitutes customary international law. It is no longer possible or necessary to await, in relation to a particular rule, a pronouncement of the ICJ on that question.⁹⁸ Leaving aside the reasons of the Appellate Body for avoiding that inquiry, it should not ignore the question of whether a provision is a norm of customary international law or decide on the uncertainty regarding that status without at least appreciating the landscape of the available case law in that regard. Indeed, one Panel has suggested that the text of draft Article 7.2 ‘might be considered as reflecting customary international law’.⁹⁹ In particular with regard to the ILC Articles, there is a significant body of awards of arbitral tribunals in which the status of those provisions is considered. For example, the Iran–United States Claims Tribunal appears to consider that Part One of the ILC Articles, including Article 5, is an authoritative statement of current international law on State responsibility.¹⁰⁰ Other arbitral tribunals under the ICSID Convention have accepted that Article 5 reflects a ‘generally recognized rule’ or established customary international law.¹⁰¹ Without taking a position on its status, other international tribunals, courts or other bodies have also referred to Article 5.¹⁰²

Whilst the Appellate Body’s avoidance of inquiring itself into whether a rule or principle reflects customary international law (or a general principle of international law) might imply that a considerable body of normative activity and developments in international law are excluded from the scope of Article 31(3)(c) of the Vienna Convention (when used in the context of WTO dispute settlement), its interpretation of ‘public body’ in US – Anti-dumping and Countervailing Duties (China) shows that that formal impediment need not be an obstacle to using, for example, Article 5 provided that the latter is sufficiently relevant. Even if the Appellate Body’s use of Article 5 in this case does not contribute to a better understanding of its position on the meaning of Article 31(3)(c) of the Vienna Convention, it nonetheless shows that, in the practice of WTO dispute settlement, the technical and often theoretical debate about the meaning of Article 31(3)(c) of the Vienna Convention increasingly is of little use. Instead, whether to use public international law and the weight to attribute to it (and possibly normatively relevant instruments that cannot (yet) be characterised as a source of public international law within the meaning of Article 38(1) of the ICJ Statute) depend, just like with all other elements of interpretation, on its relevance and the context in which it might be used.

(a) Lex Specialis exclusion

Article 55 of the ILC Articles, entitled Lex Specialis, states:

Article 55 thus specifies that the Articles do not apply where special rules of international law govern the conditions for the existence of a wrongful act, or the content or implementation of international responsibility. The commentary to Article 55 refers to the examples of the World Trade Organization Dispute Settlement Understanding and the European Convention on Human Rights as regimes that displace the rules contained in the ILC Articles.¹² However, Article 55 also makes clear that the Articles have a residual character, and that they apply to the extent that they are not excluded by a special rule.

To the extent that investment treaty arbitration could be said to form part of a lex specialis regime of international responsibility, containing separate rules governing State responsibility, according to Article 55, the general rules set out in the ILC Articles do not apply.

Investment treaty arbitration has been described as constituting a lex specialis regime in the context of the distinction between investment treaty arbitration from diplomatic protection. The ILC, in its Draft Articles on Diplomatic Protection, concluded that international investment arbitration constitutes a lex specialis regime and the rules applicable in the investment treaty context do not govern diplomatic protection claims or contribute to the development of customary rules generally applicable to diplomatic protection claims.¹³ The same point was made by the ICJ in Diallo in 2007, distinguishing between direct claims by investors under ICSID and diplomatic protection claims.¹⁴ This has also been acknowledged by investment tribunals and commentators.¹⁵

In general however, insofar as international responsibility is concerned, investment treaties do not prescribe their own rules for invocation of responsibility. Certainly, they do prescribe procedural rules governing the invocation of responsibility. For example, an investment treaty may allow an investor to commence ICSID proceedings, in which the ICSID arbitration rules govern the procedure of the arbitration. However, these procedural issues are not matters which are governed by the ILC Articles on State Responsibility. In contrast, investment treaties do not ordinarily specify the basis on which conduct is to be attributable to the respondent State. In the absence of specific rules in the treaty, Article 55 of the ILC Articles would not preclude the application of the rules on attribution set out in the Articles themselves.

If an investment treaty excludes or modifies the general rules of State responsibility, Article 55 preserves the applicability of the special rule set out in the investment treaty as the governing rule. An example may be found in the expropriation provision in a multitude of investment treaties, which provides in practice for the payment of compensation for lawful expropriation. That specific rule would, according to Article 55, apply to the exclusion of the general rule under international law, according to which restitution is a primary remedy.¹⁶

The relevant question is whether and to what extent the applicable bilateral investment treaty (BIT) sets out rules governing responsibility that differ from the generally applicable rules. This will require an examination of the applicable BIT. In the absence of specific rules governing the existence of an internationally wrongful act, or the content or implementation of international responsibility, investment claims cannot be said to form part of a lex specialis regime which excludes, pursuant to Article 55, the applicability of the general rules.

(b) Exclusion of application of legal consequences according to Article 33(2)

Part Two of the ILC Articles on State Responsibility addresses the legal consequences for an internationally wrongful act. These include the obligation to make reparation (restitution, compensation, satisfaction) and specific consequences arising from breaches of peremptory norms, such as the obligation not to recognise such a situation as lawful, nor to render aid or assistance in maintaining such situation. Article 33, entitled ‘Scope of international obligations set out in this Part’ provides:

1. 1. The obligations of the responsible State set out in this Part may be owed to another State, to several States, or to the international community as a whole, depending in particular on the character and content of the international obligation and on the circumstances of the breach.

2. 2. This Part is without prejudice to any right, arising from the international responsibility of a State, which may accrue directly to any person or entity other than a State.¹⁷

Thus Article 33(1) purports to limit the scope of application of the rules governing consequences of an internationally wrongful act to situations in which the obligations of the responsible State are owed to other States or to the international community as a whole; arguably the consequences may be different when the responsible State breaches an obligation owed to other actors.¹⁸ At the same time, Article 33(2) expressly recognises and preserves the possibility of a State’s responsibility towards a non-State actor under international law.

Article 33(2) appears clear in its terms: the provisions of Part Two, which include the obligation to make reparation, do not necessarily apply in respect of rights granted to persons or entities other than States. As noted by the commentary, ‘[t]he articles do not deal with the possibility of the invocation of responsibility by persons and entities other than States, and paragraph 2 makes this clear’.¹⁹ However, this exclusion is not a general one: it applies only to rules governing invocation of responsibility, dealt with in Part Two and in parts of Part Three of the ILC Articles.²⁰

(a) Countermeasures as a circumstance precluding wrongfulness

International law permits an injured State to take non-forcible countermeasures in response to an internationally wrongful act by another State, provided that certain conditions are met. Countermeasures have been described as ‘a feature of a decentralised system by which injured States may seek to vindicate their rights and to restore the legal relationship with the responsible State which has been ruptured by the internationally wrongful act’.²¹ The origins of countermeasures may be traced to the practice of ‘reprisals’, traditionally used to denote otherwise unlawful measures of self-help.²²

In codifying the rules on non-forcible countermeasures, the ILC Articles on State Responsibility provide that wrongfulness of the measure taken by the injured State is precluded, if and to the extent that it constitutes a countermeasure. This general rule is codified in Article 22:

Chapter II of Part Three (Articles 49–54) of the ILC Articles sets out conditions and limitations on the taking of countermeasures by an injured State and addresses the conditions of the implementation of countermeasures. The objects and limits of countermeasures are elaborated in Article 49, which provides:

1. 1. An injured State may only take countermeasures against a State which is responsible for an internationally wrongful act in order to induce that State to comply with its obligations under Part Two.

2. 2. Countermeasures are limited to the non-performance for the time being of international obligations of the State taking the measures towards the responsible State.

3. 3. Countermeasures shall, as far as possible, be taken in such a way as to permit the resumption of performance of the obligations in question.

(b) Countermeasures and third parties

Article 49 makes clear that countermeasures must be directed against the State committing the prior wrongful act – the responsible State. The wrongfulness of a measure taken as against a third State is not precluded. The ILC commentary to Article 49 states:

While this paragraph of the commentary refers only to the rights of third States, the following paragraph elaborates with respect to the effects of countermeasures as against third States and refers in that context to other parties, contemplating that countermeasures may not preclude the wrongfulness of acts taken in violation of rights of third parties other than States:

The commentary therefore implies that the effect of Article 49(1) is that the wrongfulness of a countermeasure is not precluded as against third parties, where those third parties have individual rights which are affected by the measure. That is to say, a countermeasure may affect the position or interests of third parties. However, it may not affect the rights of third parties.

(c) Countermeasures in practice: a trio of NAFTA awards

In three NAFTA cases, Mexico invoked countermeasures as a circumstance precluding the wrongfulness of any breach of its obligations under NAFTA vis-à-vis the investor. In all three cases, Mexico’s invocation of countermeasures was rejected.

The facts relevant to the pleas of countermeasures were materially identical in all three cases, as they all concerned the same measure imposed by Mexico. The proceedings were initiated against Mexico by American agricultural companies, relating to the imposition of a 20 per cent tax by Mexico on soft drink bottlers using the sweetener High Fructose Corn Syrup (HFCS). In response to its alleged violation of the national treatment standard in Article 1102 of NAFTA, Mexico argued that it had imposed the tax as a countermeasure against two violations of NAFTA by the United States. The alleged breaches of NAFTA by the US related to access of Mexico’s surplus sugar produce to the US market.²⁵ All three NAFTA tribunals have issued redacted awards.²⁶

The exclusive competence of the Commission to conclude international agreements

The Lisbon Treaty, for the first time, lists the exclusive competences of the Union. Article 3(2) TFEU indicates that the EU has an exclusive competence for the conclusion of international agreements. Article 216 TFEU reiterates the exclusive competence of the Union to ‘conclude agreements with one or more third countries or international organisations where the Treaties so provide or where the conclusion of an agreement is necessary in order to achieve, within the framework of the Union’s policies, one of the objectives referred to in the Treaties, or is provided for in a legally binding Union act or is likely to affect common rules or alter their scope’. Such agreements are binding upon the institutions of the Union and on its Member States.

When a legislative act of the Union expressly provides for the exclusive competence of the Union – such as for the Common Commercial policy² – this competence exists ‘even if the specific matters covered by those agreements are not yet, or are only very partially, the subject of rules at [the EU] level’.³

In addition to these express competences, Article 216 also foresees a competence to act arising from its internal competence. Such a possibility flows from the AETR judgment, in which the CJEU held that, even if the EU Treaties did not expressly confer a competence upon the EU to conclude an international agreement in a particular field, such a competence could also flow from other provisions of the Treaties and from measures adopted by the EU institutions. The Court, considering that the attainment of the objective pursued by those rules and the objectives of the Treaties themselves would be compromised if Member States were free to adopt international agreements affecting EU rules,⁴ held that ‘each time the [EU], with a view to implementing a common policy envisaged by the Treaty, adopts provisions laying down common rules, whatever form these may take, the Member States no longer have the right, acting individually or even collectively, to undertake obligations with third countries which affect those rules’.⁵ This judgment establishes a parallelism between the internal and external competences of the EU insofar as the EU institutions have exercised their competence at the internal level by adopting common rules to implement a common policy. This condition has been given a broad interpretation as it also includes the adoption of rules outside the scope of a specific common policy. In Opinion 2/91, the Court stressed that the authority of the AETR decision ‘cannot be restricted to instances where the [Union] has adopted [EU] rules within the framework of a common policy. In all areas corresponding to the objectives of the Treaty, Article [4(3) TEU] requires Member States to facilitate the achievement of the [Union]’s tasks and to abstain from any measure which could jeopardise the attainment of the objectives of the Treaty.’⁶

However in Opinion 1/94,⁷ the Court adopted a more restrictive view requiring that the policy area in question be fully harmonised at EU level for it to become an exclusive competence. But subsequently the Court held that the exclusive competence of the EU institutions could flow from the fact that an international agreement fell into an area largely covered by EU rules, in particular in ‘areas where there are harmonising measures’.⁸

Even where the EU has adopted such common rules, it will only benefit from an exclusive competence to conclude international agreements if a Member State’s action ‘might affect those rules or alter their scope’.⁹ In Opinion 2/91, the Court adopted an extensive interpretation of this condition, considering it is fulfilled when the commitments arising from an international agreement were merely liable to affect EU rules even though there was no contradiction between the international agreement and the Directives at stake.¹⁰ The Court went even further when it ruled that a mere proposal submitted by Greece to the International Maritime Organisation (IMO), which initiated a procedure potentially leading to the adoption by the IMO of new rules, had an effect on EU rules. According to the Court, Greece ‘took an initiative likely to affect the provisions of the Regulation’.¹¹

The ability of the Member States to enter into international agreements

Article 2(1) TFEU provides that ‘[w]hen the Treaties confer on the Union exclusive competence in a specific area, only the Union may legislate and adopt legally binding acts, the Member States being able to do so themselves only if so empowered by the Union or for the implementation of Union acts’. Such exclusivity means that the competence has been entirely transferred by the Member States to the Union. Any action by a Member State in the same area is a priori in conflict with EU law. Yet, EU Member States have continued to conclude international agreements in their own name with or without the Union. Although they may appear extraneous to EU law, such agreements are still relevant for the Union.¹²

Article 4(2) provides a non-exhaustive list of areas in which the Union does not hold an exclusive competence, but shares its competence with the Member States, including, inter alia, the internal market, environmental protection, transport and energy. Regarding the external relations of the EU, when a specific field falls within a shared competence, it is likely to lead to the conclusion of a mixed agreement signed by the Union and the Member States on the one hand and third parties on the other. The category of shared external competence is broad and encompasses different situations. This may result from an explicit provision in the treaty or from the fact that the agreement covers different matters falling both within EU and Member States competences. In these areas, Member States may only exercise their power to the extent that the Union has not exercised – or ceased to exercise – its competence. A Protocol on the exercise of shared competences annexed to the Lisbon Treaty specified that ‘when the Union has taken action in a certain area, the scope of this exercise of competence only covers those elements governed by the Union act in question and therefore does not cover the whole area’.¹³ This leaves a certain margin for Member States’ intervention.

Agreements concluded by the Member States without the participation of the Union cover a broad range of situations, such as when agreements were concluded by the Member States before they adhered to the Union and when agreements concern a matter falling outside the competence of the EU. Such agreements are generally not binding on the EU and the CJEU has no jurisdiction to give rulings on their interpretation. However, the Union is bound by them whenever they are integrated into the EU legal order through an express ‘renvoi’ or when the EU has taken over the powers formally vested in the Member States.¹⁴ Moreover, in a situation where the Union itself is not bound by an agreement but all the Member States are, the Court indicated that the latter fact is ‘liable to have consequences for the interpretation of…the provisions of secondary law which fall within the field of application of [the international agreement]’.¹⁵ This finding is based on ‘the customary principle of good faith, which forms part of general international law, and of Article [4(3) TEU]’ considering that it was ‘incumbent upon the Court to interpret those provisions taking account of [the international agreement concerned]’.¹⁶

Even where an agreement concerns a matter falling within the exclusive competence of the EU, Member States may still act, provided that they obtain an authorisation from the EU.¹⁷ Such an authorisation will, for instance, be granted when the EU is not formally represented within an international organisation, or cannot itself negotiate and conclude an agreement because the international body reserves the right to adhere to States, excluding regional integration organisations such as the EU. The Court has acknowledged the fact that the International Labour Organisation was not open to EU adherence and that, therefore, Union competence ‘may, if necessary, be exercised through the medium of the Member States acting jointly in the [Union]’s interest’.¹⁸ In this situation, the Council adopts a decision expressly authorising the Member States to conclude an agreement or a general framework of regulations establishing the procedure to be followed by the MemberStates.¹⁹

Position of the EU in international organisations

The possibility for the EU to negotiate and conclude international agreements has been recognised by the treaties but the question of its participation in international organisations or bodies has not been entirely dealt with.²⁰ Yet, in practice, it is sometimes difficult for the EU to express its opinion and to maintain a strong and credible position at the international level.

No provision directly relates to the participation of the EU in international organisations. Article 211 TFEU provides that ‘the Union and the Member States shall cooperate with third countries and with the competent international organisations’. Article 218(9) lays down the rules applicable in establishing ‘the positions to be adopted on the Union’s behalf in a body set up by an agreement, when that body is called upon to adopt acts having legal effects’. This Article arguably does not apply to the discussions and negotiations within these organisations.

As a general matter, the founding treaties of international organisations do not foresee the adherence of regional organisations such as the EU. Therefore, the status of the Union depends not only on the allocation of competences within the EU but also, and most importantly, on the rules of the organisations in question. The EU has full membership in a limited number of organisations including, for instance, the Food and Agricultural Organisation (FAO). Within the World Trade Organization (WTO) the Union holds a special status similar to full membership except for the voting rights.²¹ The Member States are often also members of other organisations so that many competences are shared between them and the EU. The EU may also be granted an observer status which implies that it ‘can attend meetings of a body or an organisation, but without voting rights’.²² This is, for instance, the case within agencies of the United Nations (UN). Such a status entails important limitations, such as the fact that the EU’s presence may be limited to formal meetings or an intervention coming after all of the other interventions.

The representation of the EU

Whether it can directly participate in international organisations or in the negotiation of international agreements or not, the question of the representation of the Union and its Member States matters. Article 218 TFEU lays down the rules generally applicable to the negotiation and conclusion of agreements between the Union and third countries or international organisations. The Treaties confirm the essential role played by the Commission. Article 17(1) TEU provides that ‘[w]ith the exception of the common foreign and security policy, and other cases provided for in the Treaties, [the Commission] shall ensure the Union’s external representation’.

Despite this clear statement, the issue of the representation of the Union is not entirely solved. The situation is complicated when – and this is most often the case – the matter discussed is of concern for both the EU and the Member States, and the Union has been granted a mere observer status. In such a case, the EU is generally represented by both the Commission and the Member State holding the presidency of the Union.²³ This system is however not optimal as it creates confusion for third countries and requires careful discussions and statements by each party taking the floor.

One of the most interesting situations is when the EU has an exclusive competence in a given area but is not a member of the international organisation concerned, or cannot therefore conclude an agreement itself. According to Article 4(3) TFEU, ‘pursuant to the principle of sincere cooperation, the Union and the Member States shall, in full mutual respect, assist each other in carrying out tasks which flow from the Treaties’ and further facilitate the achievement of the Union’s tasks, refraining from ‘any measure which could jeopardise the attainment of the Union’s objectives’. It follows that when the Union has competences that are at least exclusive and cannot be a member of an international organisation, the Member States must defend the EU’s interests.²⁴

The need for co-ordination

When Member States conclude international agreements, they must co-operate with each other as well as with the Union. For instance, the Court stated that ‘where it is apparent that the subject-matter of an agreement or convention falls in part within the competence of the [EU] and in part within that of the Member States, it is essential to ensure close cooperation between the Member States and the [EU] institutions, both in the process of negotiation and conclusion and in the fulfilment of the commitments entered into. That obligation to cooperate flows from the requirement of unity in the international representation of the [EU].’²⁵

When an international agreement is to be negotiated solely by the Member States, a common position must be reached. In AETR, the Court indicated that the Council and the Commission should ‘reach agreement…on the appropriate methods of cooperation with a view to enduring most effectively the defence of the interests of the Union’.²⁶ In Opinion 2/91, the Court emphasised that co-operation was ‘all the more necessary’ since the Union could not conclude the agreement by itself. It added that ‘[i]t is therefore for the [Union] institutions and the Member States to take all the measures necessary so as best to ensure such cooperation both in the procedure of submission to the competent authority and ratification of Convention No 170 and in the implementation of commitments resulting from that Convention’.²⁷

Compliance with EU law by the Member States

When entering into international agreements, Member States are not totally free to exercise their powers as they wish. The duty of sincere co-operation imposes on them to ‘exercise their international powers without detracting from Union law or from its effectiveness’²⁸ and requires them to facilitate the achievement of the Union’s tasks, as well as to abstain from any measure which could jeopardise the attainment of the objectives of the Treaty. For instance, the Court held that in relation to a bilateral agreement regarding a matter falling within the competence of the Member States, ‘the latter may not disregard [EU] rules but must exercise their powers in a manner consistent with [EU] law’. It added that ‘when giving effect to commitments assumed under international agreements, be it an agreement between Member States or an agreement between a Member State and one or more non-member countries, Member States are required…to comply with the obligations that [EU] law imposes on them’.²⁹

In Commission v. Greece, the Court further clarified that ‘[t]he mere fact that the [Union] is not a member of an international organisation in no way authorises a Member State, acting individually in the context of its participation in an international organisation, to assume obligations likely to affect [Union] rules promulgated for the attainment of the objectives of the Treaty’.³⁰ The duty of co-operation is therefore applicable to both the Commission and the Member States. The Court went further in deciding that even if the Commission had not entirely fulfilled its duty of co-operation – the Commission could arguably have done more to reach a common position – it does not entitle a Member State to ‘unilaterally adopt, on its own authority, corrective or protective measures designed to obviate any breach by an institution of rules of [EU] law’.³¹

The Court confirmed that this duty is of general application and does not depend either on whether the Union competence was exclusive, or on any right of the Member States to enter into obligations towards third countries.³² Where an EU Member State fails to comply with EU law, the Commission has powers to bring the infringement to an end and, where necessary, may refer the case to the ECJ under Article 258 TFEU for failure of the Member State to fulfil its obligations under EU law. Not only are Member States not allowed to conclude an agreement that would expressly violate EU rules, but they must also implement their international obligations in such a way as to take account of their EU obligations.³³

1 Postal services: the Universal Postal Convention³⁴

The Universal Postal Union (UPU) is an inter-governmental organisation acting as the primary forum for co-operation between postal sector players. In 2012, the UPU Congress, bringing together all the member countries, met to define the future world postal strategy and to review the Universal Postal Convention, including the system of terminal dues. Terminal dues correspond to the remuneration for the costs of handling and delivering cross-border mail in the country of destination. They are an important source of revenue for postal operators, in particular for those in transition economies and developing nations. The proposal made in view of the 2012 UPU Congress was trying to bring terminal dues closer to costs, which, in practice, would still be far from the actual costs of handling international mail.

The review of the terminal dues system raised two issues of considerable importance for the EU and its Member States. First, in application of the EU Treaties and the case law on the allocation of competences between the EU and the Member States, it appears that terminal dues is a subject matter that falls within the exclusive competence of the EU for two reasons. First, the regulation of terminal dues fallsa priori within the scope of the Common Commercial Policy. Secondly, the AETR doctrine seems to be applicable as the EU has developed a policy aiming to complete the internal market for postal services and has adopted common rules regulating the provision of postal services, including terminal dues – such as the Postal Directive. Besides, any action of the Member States might affect those rules or alter their scope. Therefore, the question can be raised as to whether, by participating in the 2012 Universal Postal Convention and voting on the proposed system of terminal dues, the Member States would infringe EU rules on the repartition of competence.

As the EU is not a member of the UPU, Member States were thus to co-ordinate their position in order to negotiate and conclude the agreement on its behalf. In the past, efforts have been made towards co-ordination between Member States and the EU. In 2004, the Commission adopted a Communication reaffirming the importance of ensuring ‘that the Commission participates to the fullest extent possible in work in the UN system which concerns issues for which it is responsible within the EU’.³⁵ The Commission called for a common EU position during the negotiations and co-ordination among the Member States and with the Commission.³⁶ Despite the Council Resolution that followed, the Member States showed ‘little apparent coordination’ in the 2004 and 2008 Congresses, submitting individual proposals, sometimes inconsistent with each other.³⁷ Yet, the Commission clarified in the above-mentioned Communication that, whether the EU has exclusive competence or not, all Member States are obliged to somehow co-ordinate their positions.³⁸ Although this goal was reiterated by the Council in 2008, no common position was found.³⁹

The second issue raised by these negotiations was linked to the compatibility of the proposed system of terminal dues with EU law. The UPU Target System of Terminal Dues seems to diverge not only from the Postal Directive⁴⁰ but also from the EU Treaties, and more specifically from its competition rules. However, the case law has clearly stated that, when entering into an international agreement, Member States cannot infringe EU rules. This principle, based notably on the duty of sincere co-operation, may be considered as being even stronger when the Member States are acting on behalf and in the interests of the Union. In view of the 2012 UPU Congress, the Commission stated that it was essential to ensure compatibility between the UPU system and the EU framework and that it was ‘necessary to ensure that the rules and the positions taken by the Member States in the coming UPU Congress are compatible, complementary and coherent with [EU] legislation in particular with that included in [the Postal Directive]’.⁴¹

To conclude, the Member States started the negotiations with divergent views – most probably due to divergent economic incentives – despite the necessity for them to defend the EU position. The situation was further complicated by the impossibility for the Union to directly participate in the negotiation and in the conclusion of the agreement. Generally speaking, the decision-making process at the UPU is problematic.

2 Telecommunications: the International Telecommunications Regulations

Telecommunications is subject to heavy regulation at the national as well as the European level. It was therefore logical for the EU and the Member States to react actively to the decision to hold a World Conference on International Telecommunications (WCIT or the Conference) in Dubai in December 2012, with the aim to revise the International Telecommunications Regulations (ITRs). The International Telecommunication Union (ITU) is a specialised agency of the UN which is responsible for information and communication technologies. It currently has a membership of 193 countries. The ITRs define the general principles for the provision and operation of international telecommunications. The Conference had been tasked with reviewing and updating ITRs that had remained unchanged since 1988. The new regulations would reflect the changes that the internet and new forms of communications have had on telecoms.

The main concern of the EU – and the US – was the position defended by countries such as Russia or China to increase regulation and strengthen control of the information spread on the internet. As only ITU members are full participants with the right to vote – other ITU members, including telecommunications operators and regional organisations, having an observer status – the EU and its Member States had to co-ordinate their position. Contrary to the negotiations on terminal dues in the postal sector, the Member States managed here to agree on a common position, consisting of maintaining the current status and avoiding heavier regulation. Reaching a common position was essential for the EU and the Member States in order to defend their position with credibility.

Discussions took place between the EU institutions, the Member States, telecommunications operators and other bodies. In August 2012, the Commission submitted a proposal for a common EU position to submit for discussion in the Council.⁴² It was proposed that, in relation to matters falling within its competence, Member States should ensure that any changes to the ITRs would be compatible with EU law and would not restrict future developments of the EU acquis. As regards matters falling outside of the EU competence, Member States were required to adopt common positions. This was underlined by the Cypriot president, who warned that any changes that might restrain future developments in the area of telecommunications and the internet were precluded.⁴³

Beyond the substantial issues raised by these negotiations, the role of the Commission in drafting the common position was questioned. An issue was raised as to whether the Commission – which was not formally represented at the ITU – had the legal right to draft a position that was binding on EU governments. Normally, the Commission is mandated to negotiate on behalf of the EU in areas where the Union holds an exclusive competence. A motion for a Resolution submitted to the European Parliament proposed for the Council to negotiate on behalf of the EU. The Parliament amended it and proposed for the Commission to carry out such negotiations.⁴⁴ In October 2012, the legal service of the Council of the EU indicated that the Commission has the ability to speak on behalf of the Member States on matters where the latter have transferred their lawmaking power to the EU. But the fact that the Commission is not formally represented remained problematic. If it is considered that the EU holds an exclusive competence in the field of telecommunications, a difficulty arises from the fact that the Commission – and more generally the Union – does not have a formal seat at the ITU but holds an ‘observer member’ status. The Union requested the admission of the Commission as an ‘observer in an advisory capacity’ which would not allow it to negotiate on behalf of the EU but would allow officials from the EU executive to ask for the floor and provide advice to Member State delegations on points relating to the EU position. This approach was finally approved.⁴⁵

The proposal for a Common Position put forward by the Commission was adopted by the Council in November 2012. Despite these minor disagreements regarding the role of the Commission, the EU was able – for the first time in the ITU framework – to agree on a position before going to the Conference. This most probably sent an important signal to third countries. When preparing these negotiations, the long-term interests of the EU were carefully balanced in order to allow the Member States and the Union to defend their positions in future negotiations.

Finally, a proposal made in the final discussions introduced elements that were unacceptable to EU Member States including a possible extension of the ITRs to cover internet issues. EU Member States, alongside the United States, did not agree on these proposals. In the opinion of EU participants, the final text risked threatening the future of the open internet and internet freedoms, as well as having the potential to undermine future economic growth. The EU was concerned about this possible harm not only within the EU, but globally, including in developing countries.

A Facts and background

In 1985, the French and United Kingdom governments issued an invitation calling for tenders to develop, finance, construct and operate a ‘Fixed Link’ across the Channel between France and the UK, to be financed entirely by private investment.⁵ France-Manche S.A. and The Channel Tunnel Group Ltd submitted a joint response⁶ and on 20 January 1986, the French president and the British prime minister announced that they had been selected as the Concessionaires of the Fixed Link.⁷ The Treaty concerning the Construction and Operation by Private Concessionaires of a Channel Fixed Link between the UK and France (Treaty of Canterbury) sets out the international legal framework to permit the construction and operation of this project.⁸ Pursuant to its Article 1(2), the Fixed Link includes the tunnels themselves and associated terminal areas and freight facilities.

Article 10(1) establishes an Intergovernmental Commission (IGC) ‘to supervise, in the name and on behalf of the two Governments, all matters concerning the construction and operation of the Fixed Link’. Subsequently, a Concession Agreement was signed on 14 March 1986 by the ministre de l’équipement, du transport et de l’habitat for France and the Secretary for Transport for the UK, on the one hand, and France-Manche S.A. and The Channel Tunnel Group Ltd, on the other, which granted both companies a concession to develop, finance, construct and operate the Fixed Link for a term of fifty-five years.⁹ This period was extended to ninety-nine years by the Concession Extension Agreement of 13 February 1998.

The Fixed Link consists of ‘a fixed twin-bored tunnel rail link under the English Channel, a service tunnel and terminal areas at Coquelles in the French Département du Pas de Calais and Cheriton in the County of Kent’.¹⁰ Since its opening in 1994, it has accommodated the transport of road vehicles through trains and shuttles, at the time said to be the largest privately financed infrastructure project in history. The Coquelles terminal has a kilometres-long perimeter fence, supplemented by additional fencing, and is situated in a rural area, approximately 3 km from the town of Calais. The Partial Award includes a map of the region (see Figure 23.1).

Figure 23.1 Map of the Calais Region

B Legal claims: Sangatte and SeaFrance

By a Notice of a Request for Arbitration dated 17 December 2003, the Channel Tunnel Group Ltd and France-Manche S.A. initiated arbitral proceedings against the UK and France under the auspices of the Permanent Court of Arbitration (PCA). Their two main claims related to (1) the allegedly inadequate protection of the Fixed Link against disruptions by clandestine migrants at Sangatte, linked to a complaint about the UK’s civil penalty regime, and (2) the allegedly improper financial support to the SeaFrance sea link. Although both claims were brought against both respondents, the tribunal later held that there was no dispute between the Concessionaires and the UK as concerned the SeaFrance claim so, to the extent that this claim was directed against the UK, the case was dismissed entirely for want of jurisdiction.¹¹ Analysis of the SeaFrance claim hence provides no further relevance to the present discussion concerning shared responsibility.

With regard to the Sangatte claim, the Concessionaires asserted that the Fixed Link was inadequately protected against interference by clandestine migrants based at the Sangatte Hostel, especially from 1999 to 2003. Thousands of migrants from Kosovo, Afghanistan, Iraq and Somalia broke into the perimeter of the terminal site at Coquelles, hoping to smuggle themselves into shuttle trains destined for the UK. The claimants alleged that ‘this caused significant disruption to the operation of the Fixed Link and cost a great deal in additional protective measures’.¹²

Linked was a complaint concerning the UK’s civil penalty regime, introduced in April 2000 pursuant to Part II of the Immigration and Asylum Act (the 1999 Act),¹³ under which persons or companies responsible for transporting clandestine migrants into the UK were subjected to a system of penalties. At first this regime applied only to road transport vehicles, but in 2001 it was extended to the claimants. Close to £400,000 in penalty charges were imposed on the Concessionaires, which were all subsequently remitted, up to the point where the UK entirely exempted claimants from the system in 2002. Later that same year, the UK amended the 1999 Act so as to re-render operators of freight shuttle trains liable for carrying clandestine migrants to the UK, but at the time of the Partial Award, no penalties had been imposed upon the Concessionaires under the amended 1999 Act.

The UK also imposed liability on the claimants under the Immigration Act 1971, as modified by the Channel Tunnel (International Arrangement) Order (the 1993 Order), which incorporated the Sangatte Protocol into English law,¹⁴ for the costs of detention and removal of clandestine migrants arriving in the UK via the Fixed Link. The Partial Award refers to an amount of over £100,000 paid by the claimants on this account.

A Applicable law and jurisdiction

The law to be applied by the tribunal consisted of the relevant provisions of the Treaty and the Concession Agreement, supplemented by English and French law insofar as necessary for the implementation of particular obligations. Furthermore, recourse could be had to ‘the relevant principles of international law and, if the parties in dispute agree, to the principles of equity’.¹⁵

In order to establish the scope of its jurisdiction,¹⁶ the tribunal distinguished three questions:

The tribunal identified the source of its competence to determine the parties’ respective rights and obligations as the Treaty (‘but only insofar as it is given effect to by the Concession Agreement’) and the Concession Agreement (‘whether or not it goes beyond merely giving effect to the Treaty’).¹⁸

B Findings on joint and several responsibility

Chapter V of the Eurotunnel Partial Award specifically deals with the claimants’ thesis of ‘Joint and Several Responsibility’. The Treaty of Canterbury, the Concession Agreement and their implementation were, at least in part, acts of the French and UK governments, so one contentious issue concerned ‘the basis on which the Respondents may be held responsible’.¹⁹ The question was whether the claimants needed to show precisely the degree to which any breach of obligations owed to them was specifically due to one or other government, or whether they could invoke some principle of solidary or collective responsibility.

C Findings on the Sangatte claim

The Sangatte claim consisted of two allegations which were brought jointly against both respondents: (1) failure to protect the Coquelles site against clandestine migrant incursions and (2) favouring the Société Nationale des Chemins de Fer Français (French Railways National Society or SNCF) Terminal and Port of Calais to the detriment of the Fixed Link.

A Reciprocity of joint and several responsibility?

Arguably, Clause 20, which establishes a system of joint and several liability of the Concessionaires vis-à-vis the Principals but not vice versa, may seem prima facie ‘unfair’ as it contradicts the reciprocity of contractual obligations, creating a potential imbalance in the relationship between the contracting parties. However, although the fact of private corporations being joint and severally liable in the absence of a similar liability as between the States may seem to lack symmetry, there is a clear reason why enterprises which run commercial ventures are often assumed to be joint and severally liable while governments are not.⁵⁹ Under private law, the system of joint and several liability aims at strengthening the position of the creditor, extending the assets available for recovery in case the debtor goes bankrupt. The traditional position is that States and public bodies, even when engaging in commercial activities, cannot be equated with private corporations because, amongst other things, they cannot go bankrupt. In case of contractual breach, the creditor will still be able to obtain a (financial) remedy,⁶⁰ which could explain the position taken when the Eurotunnel Concession Agreement was signed. Although the current financial crisis may weaken this argument, it remains indisputable that States cannot simply be equated with private companies in the anticipation of bankruptcy.

B The Eurotunnel case as a precedent for determining shared responsibility

C Determining damages

Regarding the precise division of the damages flowing from the establishment of shared responsibility in the confidential settlement of the Eurotunnel case, one can only resort to speculative guesswork. Inasmuch as the IGC’s conduct is concerned, it is most likely that the responsibility of the States involved has been divided on a fifty-fifty basis. Such estimation is based on the formulation of the Concession Agreement, and more precisely the stipulations under Clause 27 which give both Principals an equal role in the Agreement’s implementation and the functioning of the IGC and the Safety Authority. Resulting from such equal power to drive the IGC’s actions, and in the absence of evidence to the contrary, it would appear that each State should bear an equal measure of responsibility if and when the IGC fails to act to an extent that such omission is found to be in breach of the Agreement.

However, the same reasoning will not necessarily have been applied concerning the determination of the compensation for breaches due to acts or omissions of the respondents directly (as opposed to through the IGC). The tribunal held that ‘it was incumbent on the Principals, acting through the IGC and otherwise, to maintain conditions of normal security and public order in and around the Coquelles terminal’.⁶⁸ The liability for the Principals’ actions covered by the term ‘and otherwise’ is where the distinction between the consequences of primary versus secondary liability could come into play, because the settlement may have, as expressed in Lord Millett’s words, treated ‘the United Kingdom liable to the Claimants…in a very small amount’.⁶⁹ It is impossible to establish whether UK protests against French conduct would have had any effect. Hence, in the absence of any information on actual UK actions that have effectively contributed to the damage, it could even be possible that the UK share of reparation for this part of the Sangatte shared responsibility claim (as opposed to the IGC-related part) remained limited to some form of satisfaction, rather than actual compensation.