The SHAPE v Supreme Litigation: The Interaction of Public and Private International Law Jurisdictional Rules


Written by Dr Rishi Gulati, Barrister, Victorian Bar, Australia; LSE Fellow in Law, London School of Economics

The interaction between public and private international law is becoming more and more manifest. There is no better example of this interaction than the Shape v Supreme litigation ongoing before Dutch courts, with the most recent decision in this dispute rendered in December 2019 in Supreme Headquarters Allied Powers Europe (“SHAPE”) et al v Supreme Site Service GmbH et al (Supreme), COURT OF APPEAL OF ‘s-HERTOGENBOSCH, Case No. 200/216/570/01, Ruling of 10 December 2019 (the ‘CoA Decision’). I first provide a summary of the relevant facts. Second, a brief outline of the current status of the litigation is provided. Third, I make some observations on how public and private international law interact in this dispute. 

1 Background to the litigation

In 2015, the Supreme group of entities (a private actor) brought proceedings (the ‘Main Proceedings’) against two entities belonging to the North Atlantic Treaty Organisation (‘NATO’) (a public international organisation) before a Dutch district court for alleged non-payments under certain contracts entered into between the parties for the supply of fuel (CoA Decision, para 6.1.12). The NATO entities against whom the claims were brought in question were Shape (headquartered in Belgium) and Allied Joint Force Command Headquarters Brunssum (JFCB) (having its registered office in the Netherlands). JFCB was acting on behalf of Shape and concluded certain contracts (called BOAs) with Supreme regarding the supply of fuel to SHAPE for NATO’s mission in Afghanistan carried out for the International Security Assistance Force (ISAF) created pursuant to a Chapter VII Security Council Resolution following the September 11 terrorist attacks in the United States (CoA Decision, para 6.1.8). While the payment for the fuels supplied by Supreme on the basis of the BOAs was made subsequently by the individual states involved in the operations in Afghanistan, ‘JFCB itself also purchased from Supreme. JFCB paid Supreme from a joint NATO budget. The prices of fuel were variable. Monitoring by JFCB took place…’ (CoA Decision, para 6.1.9. The applicable law of the BOAs was Dutch law but no choice of forum clause was included (CoA Decision, para 6.1.9). There was no provision for arbitration made in the BoAs (CoA Decision, para However, pursuant to a later Escrow Agreement concluded between the parties, upon the expiry of the BoAs, Supreme could submit any residual claim it had on the basis of the BOAs to a mechanism known as the Release of Funds Working Group (‘RFWG’). Pursuant to that agreement, an escro account was also created in Belgium. The RFWG comprises of persons affiliated with JFCB and SHAPE, in other words, NATO’s representatives (CoA Decision, para 6.1.10). Supreme invoked the jurisdiction of Dutch courts for alleged non-payment under the BOAs. The NATO entities asserted immunities based on their status as international organisations (‘IOs’) and succeeded  before the CoA meaning that the merits of Supreme’s claims has not been tested before an independent arbiter yet (more on this at 2). 

In a second procedure, presumably to protect its interests, Supreme also levied an interim garnishee order targeting Shape’s escrow account in Belgium (the ‘Attachment Proceedings’) against which Shape appealed (see here for a comment on this issue). The Attachment Proceedings are presently before the Dutch Supreme Court where Shape argued amongst other things, that Dutch courts did not possess the jurisdiction to determine the Attachment Proceedings asserting immunities from execution as an IO (see an automated translation of the Supreme Court’s decision here (of course, no guarantees of accuracy of translation can be made)). The Dutch Supreme Court made a reference for a preliminary ruling to the European Court of Justice (‘CJEU’) (case C-186/19). It is this case where questions of European private international law have become immediately relevant. Amongst other issues referred, the threshold question before the CJEU is:

Must Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (recast) (OJ 2012 L 351, p. 1 [Brussels Recast] be interpreted as meaning that a matter such as that at issue in the present case, in which an international organisation brings an action to (i) lift an interim garnishee order levied in another Member State by the opposing party, and (ii) prohibit the opposing party from levying, on the same grounds, an interim garnishee order in the future and from basing those actions on immunity of execution, must be wholly or partially considered to be a civil or commercial matter as referred to in Article 1(1) of the Brussels I Regulation (recast)? 

Whether the claims pertinent to the Attachment Proceedings constitute civil and commercial matters within the meaning of Article 1 of the Brussels Recast is a question of much importance. If it cannot be characterised as civil and commercial, then the Brussels regime cannot be applied and civil jurisdiction will not exist.  If jurisdiction under the Brussels Recast does not exist, then questions of IO immunities from enforcement become irrelevant at least in an EU member state. The CJEU has not yet ruled on this reference. 

2 The outcome so far 

Thus far, the dispute has focused on questions of jurisdiction and IO immunities. These issues arise in somewhat different senses in both sets of proceedings. 

The Main Proceedings

Shape and JFCB argue that Dutch courts lack the jurisdiction in public international law to determine the claims brought by Supreme as NATO possesses immunities given its status as an IO (CoA Decision, para 6.1.13). The rules and problems with the law on IO immunities have been much discussed, including by this author in this very forum. Two things need noting. First, in theory at least, the immunities of IOs such as NATO are delimited by the concept of ‘functionalism’ – IOs can only possess those immunities that are necessary to protect its functional independence. And second, if an IO does not provide for a ‘reasonable alternative means’ of dispute resolution, then national courts can breach IO immunities to ensure access to justice. According to the district court, as the NATO entities had not provided a reasonable alternative means of dispute resolution to Supreme, the former’s immunities could be breached. The CoA summarised the district court’s decision on this point as follows (CoA Decision, para 6.1.14): 

[T]he lack of a dispute settlement mechanism in the BOAs, while a petition to the International Chamber of Commerce was agreed in a similar BOA agreed with another supplier, makes the claim of an impermissible violation of the right to a fair trial justified. The above applies unless it must be ruled that the alternatives available to Supreme comply with the standard in the Waite and Kennedy judgments: there must be “reasonable means to protest effectively rights”. The District Court concludes that on the basis of the arguments put forward by the parties and on the basis of the documents submitted, it cannot be ruled that a reasonable alternative judicial process is available.

The CoA disagreed with the district court. It said that this was not the type of case where Shape and JFCB’s immunities could be breached even if there was a complete lack of a ‘reasonable alternative means’ available to Supreme (CoA Decision, para 6.7.8 and This aspect of the CoA’s Decision was made possible because of the convoluted jurisprudence of the European Court of Human rights where that court has failed to provide precise guidance as to when exactly IO immunities can be breached for the lack of a ‘reasonable alternative means’, thereby giving national courts considerable leeway. The CoA went on to further find that in any event, Supreme had alternative remedies: it could bring suit against the individual states part of the ISAF action to recover its alleged outstanding payments (CoA Decision, para 6.8.1); and could have recourse to the RFWG (CoA Decision, para 6.8.4). This can hardly be said to  constitute a ‘reasonable alternative means’ for Supreme would have to raise claims before the courts of multiple states in question creating a risk of parallel and inconsistent judgments; the claims against a key defendant (the NATO entities) remain unaddressed; and the RFWG comprises representatives of the defendant completely lacking in objective independence. Perhaps the CoA’s decision was driven by the fact that Supreme is a sophisticated commercial party who had voluntarily entered into the BOAs where the standards of a fair trial in the circumstances can be arguably less exacting (CoA Decision, para 6.8.3).

On the scope of Shape’s and JFCB’s functional immunities, the CoA said that ‘if immunity is claimed by SHAPE and JFCB in respect of (their) official activities, that immunity must be granted to them in absolute terms’ (CoA Decision, para It went on to find:

The purchase of fuels in relation to the ISAF activities, to be supplied in the relevant area of operations in Afghanistan and elsewhere, is directly related to the fulfilment of the task of SHAPE and JFCB within the framework of ISAF, so full functional immunity exists. The fact that Supreme had and has a commercial contract does not change the context of the supplies. The same applies to the position that individual countries could not invoke immunity from jurisdiction in the context of purchasing fuel. What’s more, even if individual countries – as the Court of Appeal understands for the time being before their own national courts – could not invoke immunity, this does not prevent the adoption of immunity from jurisdiction by SHAPE and JFCB as international organisations that, in concrete terms, are carrying out an operation on the basis of a resolution of the United Nations Security Council CoA Decision, para

Acknowledging that determining the scope of an IO’s functional immunity is no easy task, the CoA’s reasoning is somewhat surprising. The dispute at hand is a contractual dispute pertaining to alleged non-payment under the BOAs. One may ask the question as to why a classical commercial transaction should attract functional immunity? Indeed, other IOs (international financial institutions) have included express waiver provisions in their treaty arrangements where no immunities exist in respect of business relationships between an IO and third parties (see comments on the Jam v IFC litigation ongoing in United States courts by this author here). While NATO is not a financial institution, it should nevertheless be closely inquired as to why NATO should possess immunities in respect of purely commercial contracts it enters into. This is especially the case as the CoA found that the NATO entities in question did not possess any treaty based immunities (CoA Decision, para 6.6.7), and upheld its functional immunities based on customary international law only (CoA Decision, para  6.7.1), a highly contested issue (see M Wood, ‘Do International Organizations Enjoy Immunity Under Customary International Law?’ (2013) 10 IOLR 2). It is likely that the CoA Decision would be appealed to the Dutch Supreme Court and any further analysis must await a final outcome.

The Attachment Proceedings

The threshold question in the Attachment Proceedings is whether Dutch courts possess civil jurisdiction under the Brussels Recast to determine the issues in that particular case. If the claim is not considered civil and commercial within the meaning of Article 1 of the Brussels Recast, then no jurisdiction exists under the rules of private international law and the claim comes to an end, with the issue of immunities against enforcement raised by the NATO entities becoming superfluous. This is because if a power to adjudicate does not exist, then the question on the limitations to its exercise due to any immunities obviously becomes irrelevant. Perhaps more crucially, after the CoA Decision, the ongoing relevance of the Attachment Proceedings has been questioned. As has been noted here:

At the public hearing in C-186/19 held in Luxembourg on 12 December, the CJEU could not hide its surprise when told by the parties that the Dutch Appellate Court had granted immunity of jurisdiction to Shape and JCFB. The judges and AG wondered whether a reply to the preliminary reference would still be of any use. One should take into account that the main point at the hearing was whether the “civil or commercial” nature of the proceedings for interim measures should be assessed in the light of the proceedings on the merits (to which interim measures are ancillary, or whether the analysis should solely address the interim relief measures themselves.

Given that a Supreme Court appeal may still be filed in the Main Proceeding potentially reversing the CoA Decision, the CJEU’s preliminary ruling could still be of practical relevance.  In any event, in light of the conceptual importance of the central question regarding the scope of the Brussels Recast being considered in the Attachment Proceeding, any future preliminary ruling by the CJEU is of much significance for European private international law. Summarising the CJEU’s approach to the question at hand, the Dutch Supreme Court said:

The concept of civil and commercial matters is an autonomous concept of European Union law, which must be interpreted in the light of the purpose and system of the Brussels I-bis Regulation and the general principles arising from the national legal systems of the Member States.  In order to determine whether a case is a civil or commercial matter, the nature of the legal relationship between the parties to the dispute or the subject of the dispute must be examined.  Disputes between a public authority and a person governed by private law may also fall under the concept of civil and commercial matters, but this is not the case when the public authority acts in the exercise of public authority.  In order to determine whether the latter is the case, the basis of the claim brought and the rules for enforcing that claim must be examined.  For the above, see, inter alia, ECJ 12 September 2013, Case C-49/12, ECLI: EU: C: 2013: 545 (Sunico), points 33-35, ECJ 23 October 2014, Case C ? 302/13, ECLI: EU: C: 2014: 2319 (flyLal), points 26 and 30, and CJEU 9 March 2017, case C-551/15, ECLI: EU: C: 2017: 193 (Pula Parking), points 33-34 (see the automated translation of the Supreme Court’s decision cited earlier, para 4.2.1). 

There is not the space here to explore the case law mentioned above in any detail. Briefly, if the litigation was taken as a whole with the analysis taking into account the nature of the Main Proceedings as informing the characterisation of the Attachment Proceedings , there would be a close interaction between the scope of functional immunity and the concept of civil and commercial. If an excessively broad view of functional immunity is taken (as the CoA has done), then it becomes more likely that the matter will not be considered civil and commercial for the purposes of the Brussels system as the relevant claim/s can said to arise from the exercise of public authority by the defendants. However, as I said earlier, it is somewhat puzzling as to why the CoA decided to uphold the immunity of the defendants in respect of a purely commercial claim. 

However, it is worth noting that in some earlier cases, while the CJEU seem to take a relatively narrow approach to the scope of the Brussels system (CJEU Case C-29/76, Eurocontrol). More recent case law has taken a broader view.  For example, in Pula Parking, para. 39, the CJEU said ‘Article 1(1) of Regulation No 1215/2012 must be interpreted as meaning that enforcement proceedings brought by a company owned by a local authority…for the purposes of recovering an unpaid debt for parking in a public car park the operation of which has been delegated to that company by that authority, which are not in any way punitive but merely constitute consideration for a service provided, fall within the scope of that regulation’. If the true nature and subject of Supreme’s claims are considered,  it is difficult to see how they can constitute anything but civil and commercial within the meaning of the Brussels system in light of recent case law, with the issue of IO immunities a distraction from the real issues. It will be interesting to see if the CJEU consolidates its recent jurisprudence or prefers to take a narrower approach.

3 The interaction between public and private international law?

In the Main Proceedings, in so far as civil jurisdiction is concerned, already, the applicable law to the BOAs is Dutch law and Dutch national courts are perfectly suited to take jurisdiction over the underlying substantive dispute given the prevailing connecting factors. As the CoA determined that the NATO entities tacitly accepted the jurisdiction of the Dutch courts the existence of civil jurisdiction does not seem to be at issue (CoA Decision, para Clearly, in a private international law sense, Dutch courts are manifestly the suitable forum to determine this claim. 

However, on its face, the norms on IO immunities and access to justice require balancing (being issues relevant to both public and private international law). As the district court found, if an independent mechanism to resolve a purely commercial dispute (such as an arbitration) is not offered to the claimant,    IO immunities can give way to ensure access to justice. Indeed, developments in general international law require the adoption of a reinvigorated notion of jurisdiction where access to justice concerns should militate towards the exercise of jurisdiction where not doing so would result in a denial of justice. Mills has said:

The effect of the development of principles of access to justice in international law also has implications when it comes to prohibitive rules on jurisdiction in the form of the immunities recognised in international law…Traditionally these immunities have been understood as ‘minimal’ standards for when a state may not assert jurisdiction — because the exercise of jurisdiction was understood to be a discretionary matter of state right, there was no reason why a state might not give more immunity than required under the rules of international law. The development of principles of access to justice, however, requires a state to exercise its jurisdictional powers, and perhaps to expand those jurisdictional powers as a matter of domestic law to encompass internationally permitted grounds for jurisdiction, or even to go beyond traditional territorial or nationality-based jurisdiction (A Mills, ‘Rethinking Jurisdiction in International Law’ (2014) British Yearbook of International Law, p. 219).

The Main Proceedings provide an ideal case where civil jurisdiction under private international law should latch on to public international law developments that encourage the exercise of national jurisdiction to ensure access to justice. Not only private international law should be informed by public international law developments, the latter can benefit from private international law as well. I have argued elsewhere that private international law techniques are perfectly capable of slicing regulatory authority with precision so that different values (IO independence v access to justice) can both be protected and maintained at the same time (see here). Similarly, in the Attachment Proceedings, a reinvigorated notion of adjudicative jurisdiction also demands that the private and public properly inform each other. Here, it is of importance that the mere identity of the defendant as an international public authority or the mere invocation of the pursuit of public goals (such as military action) does not detract from properly characterising the nature of a claim as civil and commercial. More specifically, any ancillary proceeding to protect a party’s rights where the underlying dispute is purely of a commercial nature ought to constitute a civil and commercial matter within the meaning of the Brussels system. Once civil jurisdiction in a private international law sense exists, then any immunities from enforcement asserted under public international law ought to give way to ensure that the judicial process cannot be frustrated by lack of enforcement at the end. It remains to be seen what approach the CJEU takes to these significant and difficult questions where the public and private converge.  

To conclude, only a decision on the merits after a full consideration of the evidence can help determine whether Supreme’s (which itself is accused of fraud) claims against Shape et al can be in fact substantiated. In the absence of an alternative remedy offered by the NATO entities, if the Dutch courts do not exercise jurisdiction, we may never know whether its claims are in fact meritorious.