A commercial or private law exception is recognized in principle by all countries which adhere to the restrictive doctrine of State immunity, but its application is so diverse and the criterion by which it is determined so differently formulated as to prevent the articulation of the exception in terms acceptable to all. The purpose underlying the restrictive doctrine is plain: a State engaging in business in competition with private persons or corporations should be answerable in the courts of the country where the business is conducted. The commercial or private law exception is driven by ‘[i]ncreasing concern for private rights and public morality, coupled with the increasing entry of governments into what had previously been regarded as private pursuits’.1 That concern is widespread, but establishing a legal formulation for the exception has proved more difficult; national systems classify public and private acts differently and the existence of such a classification in international law is questionable.2
Chapter 6’s account of the development of the restrictive doctrine of State immunity from its emergence at the beginning of the twentieth century has shown that its application by national courts was varied and differed in pace from State to State; national legislation in the 1970s setting out exceptions based on ‘commercial transaction’ provided indicia as to the scope and matters included in the concept of commerciality; and the ILC’s 1991 draft and the UN’s adoption in 2004 of UNCSI all contributed to its worldwide recognition. However, as mentioned in the Introduction to this book, any assessment of the scope of the exceptions to State immunity and the concept of commerciality has also now to take account of the 2012 Jurisdictional Immunities judgment of the International Court of Justice (ICJ), a judgment based on ‘settled practice’ of customary international law, there being no treaty on State immunity between the parties, Germany and Italy.
Although the ICJ in the Arrest Warrant case and other decisions had given decisions relating to the immunity of a Minister for Foreign Affairs and other officials of a foreign State, it was not until 2012 in the Jurisdictional Immunities case that the ICJ addressed the subject of State immunity in its entirety, declaring that:3
The Court considers that the rule of State immunity occupies an important place in international law and international relations. It derives from the principle of sovereign equality of States, which, as Article 2, paragraph 1, of the Charter of the United Nations makes clear, is one of the fundamental principles of the international legal order.
The ICJ’s actual decision in respect of immunity from jurisdiction was narrowly confined to: ‘under customary international law as it presently stands, a State is not deprived of immunity by reason of the fact that it is accused of serious violations of international human rights law or the international law of armed conflict’.4 Three general aspects, however, of the judgment are of particular relevance to this chapter and provide a fresh and novel slant on the application of the restrictive doctrine: the exclusively procedural nature of the plea of immunity.
Before ruling on the narrow issue of the applicability of State immunity in time of armed conflict, the ICJ indicated three broad requirements necessary to any exception to State immunity: first that the plea was procedural; second that if State immunity was based on sovereign equality it ‘has to be viewed together’ with the principle of territorial sovereignty of the forum State exercising jurisdiction; and third that in determining an exception to immunity the acts of the foreign State were to be characterized as acta jure imperii or jure gestionis. This last relates to the concept of commerciality, the subject of this chapter.
As to the first requirement, the procedural nature of the plea of State immunity and the ICJ’s assessment as foreclosing all regard to other aspects of the claim,5 the court’s approach has already been discussed in Chapter 2 of this book which provides an assessment of ‘this strict (and deeply pragmatic) approach’ of the ICJ basing ‘its reasoning on the straightforward exclusionary proposition that the plea of immunity was a procedural plea independent of the issues raised … relating to State responsibility and their determination’.
Here, it is sufficient to note that the complexities which the application of such a distinction has produced in the determination of the applicable law for the purposes of private international law, as evidenced by conflicting rulings of the Australian and English courts,6 and its abandonment in respect of actions in tort, by reason of its inability to distinguish between damage which relates to substance (the kind of loss) or to procedure (the quantification of loss) provide a sufficient warning against too heavy a reliance on it as a solution in respect of State immunity.7 As Orakhelashvili has written: ‘The essence of the restrictive doctrine is that the “preliminary” or “procedural” nature of State immunity depends on the substantive characterization of the act in question as sovereign or non-sovereign’.8 Whilst for general purposes the procedural nature of the plea may explain the limitation of its application to denial of jurisdiction by the forum court, it should not be treated solely by reason of its position in the proceedings as totally unaffected by other circumstances of the commission of the act, the conduct of the defendant State, that State’s admission of responsibility and the forum State’s compliance with other requirements of international law. Given that claims relating to commercial transactions are generally acte jure gestionis, the relevance of their violation of international law or their lacking alternative means of reparation rarely arises, the underlying complaint to be addressed being their infringement of some commercial or private law rules.
As to the second requirement, the ICJ acknowledged a competing principle of territoriality of the forum State, as already discussed in Chapter 4 on jurisdiction:9
the principle of sovereign equality … has to be viewed together with the principle that each State possesses sovereignty over its own territory and that there flows from that sovereignty the jurisdiction of the State over events and persons within that territory. Exceptions to the immunity of the State represent a departure from the principle of sovereign equality. Immunity may represent a departure from the principle of territorial sovereignty and the jurisdiction which flows from it.
On the basis of this competing principle, whilst acknowledging that national legislation allowed an exception for personal injuries committed in the territory of the forum State, the Court declined to express a view on ‘whether there is in customary international law a “tort exception” to State immunity applicable to acta jure imperii in general’, instead confining itself to declare immune ‘acts committed on the territory of the forum State by the armed forces of a foreign State … in the course of conducting an armed conflict’.10
This second principle of territoriality of the forum State is given due weight as a condition of the exercise by national courts of jurisdiction over foreign States as examined in Chapter 4 on Jurisdiction and also taken account of in national courts’ determination of proceedings relating to the immunity of a foreign State or of its officials. Further, as set out in UNCSI and explained in Chapters 13 to 15 which follow this Chapter, a jurisdictional link to the forum State territory features as a significant requirement in the exceptions to the rule of immunity set out in UNCSI, Articles 5 and 6.
In its third significant general passage the International Court addresses the concept of commerciality indirectly: it first states that
The Court is not called upon to address the question of how international law treats the issue of State immunity in respect of acta jure gestionis. The acts of the German armed forces and other State organs which were the subject of the proceedings in the Italian courts clearly constituted acta jure imperii.
And continues
The Court notes that Italy, in response to a question posed by a member of the Court, recognized that those acts had to be characterized as acta jure imperii, notwithstanding that they were unlawful. The Court considers that the terms ‘jure imperii’ and ‘jure gestionis’ do not imply that the acts in question are lawful but refer rather to whether the acts in question fall to be assessed by reference to the law governing the exercise of sovereign power (jus imperii) or the law concerning non-sovereign activities of a State, especially private and commercial activities (jus gestionis). To the extent that this distinction is significant for determining whether or not a State is entitled to immunity from the jurisdiction of another State’s courts in respect of a particular act, it has to be applied before that jurisdiction can be exercised …11
Apart, however, from declaring that ‘under customary international law as it presently stands, a State is not deprived of immunity by reason of the fact that it is accused of serious violations of international human rights law or the international law of armed conflict’,12 the Court provides little guidance as to the criteria to distinguish acts of a commercial or private law nature from acts performed in exercise of sovereign authority. Though recognizing a category of acts for which immunity may not bar proceedings against a foreign State, it specifically avoids such a task: ‘The Court is not called upon to address the question of how international law treats the issue of State immunity in respect of acta jure gestionis.’13
This cardinal issue, ‘the question of how international law treats the issue of State immunity in respect of acta jure gestionis’, is central to the law on State immunity. Whilst a court may confine itself to its application to a particular set of facts—State acts in a time of armed conflict—a treatise such as this on the subject must address this issue if the plea is to have acceptability as a concept of law rather than as a collective device of the ‘rulers’ to suppress bad news to the ‘masses’.
As noted in Chapter 2, the public/private divide is not confined to a procedural plea relevant to a State, ‘The exercise of public powers, as opposed to the engagement in private relationships, has been a constant theme in political theory and is used as a justification in many branches of the law other than that relating to State immunity.’ The elucidation of the distinction is ‘multi-faceted’ in the words of Cane, one of the number of writers who have analysed the distinction,14 with normative, functional, and even historical explanations derived from Continental analysis and Common law practice. The Continental view is derived from the concept of the State with, in the French form, its separate branch of public or administrative law and its numerous manifestations—jurisdictional, institutional, procedural, substantive, pedagogical, ideological—whereas the English common law view is the outcome of history—the demise of the Star Chamber, the use of prerogative writs alongside ordinary remedies in the High Court and the Dicey view of the ‘supremacy of regular as opposed to arbitrary power’.15 The procedural reforms of the 1970s have, as discussed in Chapter 3, brought about a closer approximation in England to an administrative court with a separate Crown Office List of judges familiar with public law, which serves by a short time limit and an initiatory requirement of obtaining leave of court, to provide an additional deterrent against litigation involving public authorities. This whole area of enquiry, though valuable, provides little assistance in defining the line which State practice currently applies and accepts to distinguish acts on exercise of sovereign authority from acts commercial or of a private law nature.
Whatever the basis of the explanatory scheme which underlies the distinction, this area of the law of State immunity, therefore, continues largely dependent on State practice in the same way as the International Court itself relied upon it in ruling on the scope of State immunity in respect of acta juris imperii; the Court drew extensively, ‘as to its identification and limits on State practice in the form of judicial decisions … accompanied by opinio juris, as demonstrated by the positions taken by States and the jurisprudence of a number of national courts which have made clear that they considered that customary international law required immunity’16 (see Chapter 5). Leaving to others the further analysis of its accepted ‘multi-faceted’ aspects, this chapter will, in reliance on State practice, summarize the acts generally ranked by States, their legislatures, and courts as constituting acta jure imperii, before seeking to examine closer the reasoning which underlies acts performed in the course of ‘commerciality’ and permits the bar of State immunity to be set aside.
No list of such acts in exercise of sovereign activity, however, is set out in legislation, and the courts have generally been slow to provide one. The State’s sovereign interest has been seen to be better served by the formulation of a general rule of immunity in the State’s favour; spelling out the sovereign acts possibly being seen to invite further definition and limitation. Nonetheless, a careful reading of the State Immunity Act 1978 (SIA), for example, reveals some explicit recognition of acts as sovereign; as with agreements between States, SIA, section 3(2), agreements to arbitrate between States, section 9(2), and certain employment contracts with nationals of the foreign State. Any list of activities in exercise of sovereign authority may well include issues which are broadly identified in proceedings between non-State Parties as excluded by a plea of non-justiciability or act of State (see Chapter 3). In addition, some general designation of sovereign acts can be attempted.
Acts in exercise of sovereign authority have been described as ‘political activities’17 and the ‘very core of State authority’. Those acts have been stated as including ‘the activities of the authorities responsible for foreign and military affairs, legislation and exercise of police power and the administration of justice’.18 In the 1991 Third Resolution of the Institut on State immunity the Rapporteur, Ian Brownlie, in focusing on the nature of the claim, adopted a method, formulated very much as the ICJ in the Jurisdictional Immunities judgment has now done, as ‘whether the acts in question fall to be assessed by reference to the law governing the exercise of sovereign power (jus imperii) or the law concerning non-sovereign activities of a State, especially private and commercial activities (jus gestionis)’.19 Professor Brownlie proposed two lists, based, however, not on the distinction into acts jure imperii and jure gestionis, but on criteria indicating competence or incompetence of the legal system of the forum State.2021 His proposals were criticized by other members of the Institut de Droit International as relating to justiciability rather than immunity; and as too widely phrased so as to include prescriptive or legislative jurisdiction as well as adjudicative jurisdiction. Nonetheless, in the Resolution entitled ‘Aspects of Jurisdictional Immunity of States’, adopted by the Institut in 1991, the following indicia against a court taking jurisdiction were set out:
(a) transactions of the State in terms of international law;
(b) internal, administrative, and legislative acts of the State;
(c) issues the resolution of which has been allocated to another remedial context;
(d) the content or implementation of the foreign, defence, and security policies of the State;
(e) intergovernmental agreements creating agencies, institutions, or funds subject to the rule of public international law.22
Let us take each of these headings in turn.
This includes international transactions between States such as the determination of territorial or maritime boundaries,23 State responsibility as a member of an international organization,24 matters arising between States in relation to State succession,25 the terms of an inter-State agreement relating to meeting the cost of training Icelandic pilots in Sweden,26 and lump sum agreements between States which individual claimants seek to reopen.27
As concerning internal administration, the internal organization of the State, this is a matter quintessentially one of sovereign authority because it so closely affects the identity of the State itself. Internal administration is taken account of in the exception for employment contracts in the UNCSI, Article 11, the provisions of which expressly state that they do not apply to employees recruited to perform particular functions in the exercise of governmental authority nor to dismissal or termination which interferes with the organizational restructuring of the staff or security interests of the State (see Chapter 14; as to a possible exception relating to administrative law matters see Chapter 3).
It also relates to the exercise of regulatory control by the State as in registration of title, maintenance of a register of charges,28 expropriation,29 nationalization,30 exchange control,31 regulation by the State of the exploitation of natural resources, and economic development. On the introduction of the restrictive doctrine all these were clearly treated as matters jure imperii; however, in the last decade or so there has been some redrawing of the boundary line. Thus in the US case of Weltover the Supreme Court overruled earlier cases, treating a State’s issuance of bonds to private individuals and subsequent restriction on their payment due to shortage of foreign reserves as acts in exercise of sovereign authority, and held them to relate to commercial activity and hence not to be immune, subject always to a sufficient jurisdictional nexus with the US.32
Claims relating to exploitation of resources where exploratory or relating to licensing have been treated as immune but not commercial drilling. Decisions in the Netherlands,33 the US,34 Germany,35 and France36 agree on this approach.
It covers appointment, terms of service, and termination of the employment of officials of the State, including the implementation of State policies.
Arbitration agreements between States are expressly treated as governmental activities in legislation in common law systems. This category closely resembles the subject-matter for which Lord Wilberforce formulated the general principle of non-justiciability. In one sense it describes the whole rationale of State immunity which permits the defendant State to refer the claim to alternative dispute settlement. In another, it is misleading in that a reference to arbitration contained in an agreement of a private law nature may well be treated as subject to the national court’s jurisdiction.
This covers defence, diplomatic missions abroad, including questions of formation of policy, and the deployment, behaviour, and motivation of personnel of a foreign State operating in another State. As to UNCSI and its exclusion of military activities and diplomatic missions, see Chapter 19. The management of the law enforcement officers of the State are also considered in that chapter. The Swiss Federal Tribunal was not prepared to accept the supervision and enforcement of customs at Pakistani ports by a private commercial company as a non-immune transaction; it categorized the collection of taxes as an attribute of sovereignty with the company engaged to play a leading role in the fight against customs fraud.37 In large part the emergence of independent immunity for central banks has provided a separate head of immune economic activity (see Chapters 16 and 17).
Section 3(3)(b) of the UK SIA broke new ground by boldly declaring non-immune ‘any loan or other transaction for the provision of finance and any guarantee or indemnity in respect of any such transaction or of any other financial obligation’. But that same Act preserved immunity for loans made by international agreement; immunity was preserved where States were parties to the dispute or have otherwise agreed (section 3(2)). The former practice of treating public loans as sovereign acts and hence immune gave way as loans were increasingly raised by States using commercial markets using private law methods and documentation, and are now generally held to be commercial transactions and not immune.38 UNCSI accordingly includes them in its list of commercial transactions.
The treatment of funds transferred under international agreement may give rise to questions not only of immunity but also of non-justiciability. Funds located in a bank pursuant to an agreement for development aid between the Netherlands and Surinam and allocated by Dutch budget decree were held immune from attachment.39
Brownlie’s general approach with the dual listing of the immune and non-immune acts was original and far-sighted, but too advanced for the thinking of the time so that it provoked fierce opposition (see further Chapter 5 under Resolutions: the Institut de Droit International and Chapter 3 where Brownlie’s categories of competence and incompetence of national jurisdiction are discussed in relation to the pleas of act of State and non-justiciability). The two-list approach, or balancing exercise, underlies much State practice but still leaves the decision largely to the court as a matter of discretion. Brownlie’s project was shown in a decision of the New Zealand Court of Appeal, the only court which has given it serious consideration, to be capable of misconstruction as a general doctrine of iniquity enabling the forum court to set aside immunity of a foreign State whenever it considers the latter’s acts to be contrary to the public policy of the forum State.40
It should be warned that the above listing of categories of acte jure imperii takes little account of the general rule of immunity applied in national legislation and State practice, which implicitly treats all acts of the State as acts in the exercise of sovereign authority, leaving the definition of any acts covered by an exception to be expressly formulated (see Chapter 9). In sum, the above list covers the regulatory powers of the State and the officers charged with its enforcement—military, police, other officials of a State are immune from the jurisdiction of national courts of other States. This treatment, as with Brownlie’s approach, rates the two types of acts, jure imperii and jure gestionis, as though on an equal footing (thereby ignoring the State practice of giving implicit precedence to acts in the exercise of sovereign authority).
The search to identify the distinguishing features of the acts which fall either side of the line of immune/non-immune distinction should now be addressed from the opposite direction; that is, the aids and techniques for the identification of the commercial nature of an activity of a State (aids and techniques, if the two-list approach is adopted, which might be applicable in establishing an activity as performed jure imperii to support the retention of immunity in respect of State activities).
Loss of immunity which is attributable to the consent of the defendant State is consistent with the continuance of absolute immunity. Consequently waiver of immunity by the foreign State to the exercise of jurisdiction has been maintained by some States as the true basis for any exception to the principle of State immunity.41
Even express consent as waiver of immunity by the foreign State when reduced to legislative form by appearance in court or the taking of a step in the proceedings involves some extension beyond a statement in words that the national court may proceed. A bolder use of implied waiver was developed so as to result in loss of immunity from the State’s voluntary undertaking of a business of the same kind as carried on by a private person. Here, three legal techniques are combined: consent of the State to the local jurisdiction construed by its engaging in a transaction on that basis;42 conduct of a business whose commerciality distinguishes it from the more usual activity of a State for the public benefit; and engaging in that business with and in the manner of a private person, the private law nature of the transaction engaged in supplying additional evidence that the State voluntarily intended to subject itself to the national court. Thus introduced by way of implied waiver, we find the two tests most frequently employed to determine the non-immunity of a transaction: private law character and commerciality.43
This notion of a commercial transaction imports three elements—activity relating to business and trade; arising from a transaction engaged in voluntarily between two parties; and made expressly or implicitly by reference to the private law of a national jurisdiction. These notions were generalized by the Harvard Research, into the State engaging in business in the territory of the forum State, such business being an industrial, commercial, financial, or other business enterprise in which persons may there engage. A further extension was contained in the Harvard Research’s proposal to remove immunity from proceedings relating to a single act performed in the forum territory, such as a purchase of goods, provided it was connected with the conduct of a business enterprise carried on elsewhere.44 The Harvard Research draft article was here moving away from the more restrictive jurisdictional test in the Institut’s resolution, based on the presence of the foreign State’s enterprise within the forum State, to the business nature of the act wherever it took place. As such it heralded the recognition of ‘commerciality’ as the criterion which the US FSIA adopted in 1976.
The European Convention on State Immunity 1972 adopted no general notion of commerciality, and indeed dispensed with it as a requirement in the exception relating to contracts (Article 4) and to claims for personal injuries (Article 11); but it followed in Article 7 the Harvard Research draft’s removal of immunity for claims relating to commercial activities of a commercial office, agency, or establishment based in the forum State.45 Other articles provided exceptions to immunity for proceedings relating to other business activities such as certain employment contracts, company dealings, patents, and trade marks.46 The removal of immunity by the Brussels Convention on the removal of Immunity for State-owned Ships engaged in trade in 1926 accelerated the evolution of the commercial exception in common law jurisdictions, and led to commerciality as a basis for the removal of immunity being enacted in the US FSIA 1976 and the UK SIA 1978 legislation introducing the restrictive doctrine of State immunity.
National courts have identified a variety of techniques to justify the application of the commercial exception to immunity.
An early method was to proceed by analogy, with the position of the sovereign in person being extended to the artificial entity of the State, it not being difficult in principle to distinguish in respect of an individual holder of office an act performed for the purpose of the State and one performed solely for private purposes. Thus Emperor Maximilian was held immune when sued in respect of a purchase of medals for his army, whereas Queen Isabelle of Spain and King Farouk of Egypt were not in respect of purchases made for jewels or couturier clothes for the latter’s wife.47 Similarly, in relation to Queen Isabella the French court restricted immunity by holding that it had jurisdiction over ‘une action intentée contre un souverain étranger ayant agi à titre de personne privée’; acts performed for the private purposes of the individual holder were distinguished from acts performed for the purposes of the State. The attempt by Italian courts to split the artificial person of the State in two, into an ‘ente politico’ and a ‘corpo morale’, the latter being the legal person exercising civil rights under private law,48 was, however, rapidly rejected. As Lauterpacht declared in his seminal article in 1951, ‘the State always acts as a public person. It cannot act otherwise’.49 The Court of Appeal in Paris in 1912 ruled that no distinction between the public personality and the legal personality should be made, ‘since all acts of a State can have only one goal and one end which are always political, and its unity precludes dualism’.50 But the concept of the State acting in two capacities survived, and with borrowings from French domestic law developed into the distinction between acts of public power, over which there is no domestic jurisdiction, and acts where the French State or its agents act as legal persons under private law, and which are therefore subject to the ordinary courts.
In its original form this criterion contemplated a distinction based on legal acts which a private individual could or could not perform. Thus the purchase of a battleship was not one which a private person could make. But when applied to the purchase of boots or the more modern instance of cigarettes for the army,51 this was rapidly shown to be an over-simplification and to depend more on purpose and motivation. Focus shifted from the capacity of the person performing the act to the relationship of that person with the State and to the form by which it was performed, and hence the criterion changed to an act performed as a private person and in private law form.
In developing a restrictive approach after the First World War, the French courts, as well as using the techniques of dual capacity and implied waiver, elaborated the distinction between acts de puissance and actes de gestion or actes de commerce. In doing so they referred to French internal law’s division of competence between the civil and commercial courts on one hand and the administrative tribunals on the other. From the time of Napoleon, France has had a tradition of strong central government, and the establishment of a separate system of administrative courts recognizes that the needs of the administration differ from those of the private individual and require special powers by which to govern the country.52 The criterion to distinguish the competence of the administrative from the civil courts has remained debatable but the underlying idea recognizes that ‘a public service is any activity of a public authority aimed at satisfying a public need’,53 and to that end the authority must have recourse to methods and prerogatives which would be excluded in relations between private parties; these special powers are ‘des prérogatives exorbitantes du droit commun’, involving unilateral powers of coercion. French courts have broadly applied by analogy the principles on which this distinction was based in internal law to contracts entered into by a foreign State, recognizing that the foreign State, like the French State, must have recourse to the exorbitant prerogatives involving unilateral powers of coercion.54 Clauses in contracts are exorbitant if they are different in their nature from those which would be included in a similar contract under civil law ‘where their object is to confer rights or impose obligations upon the parties quite unlike in their nature to those which anyone would freely agree to in the context of civil or commercial law’.55 The French courts’ application of this distinction is illustrated by proceedings against a foreign State relating to the renting of premises: thus a claim arising out of a lease to the Spanish Tourist Agency for the express purpose of commercial activity and containing the usual covenants relating to repair and payment of rent ‘without resort to an atom of governmental authority—sans recourir à l’exercise d’une parcelle de puissance publique’—was held not to be immune;56 whereas a contract containing special clauses referring to the construction of housing to accommodate staff to implement the Marshall Plan, to its financing by international funds under that Plan, and to exemption from all French taxes, was held immune as an administrative agreement containing exorbitant clauses.57
Both ECSI, Article 4(1) and the UK SIA, section 3(1)(a) provide an exception for State immunity for proceedings relating to an obligation of the State which by virtue of a contract falls to be discharged in the territory of the State of the forum.58 The exclusions in ECSI, and national legislation (SIA, section 3(2)) of contracts between States and contracts made in, and governed by the administrative law of, the defendant State clearly derive from this French civil law retention of immunity for ‘actes de puissance’ and from the civil law division of competence between civil and commercial courts and administrative courts. Like the exception in Article 11 relating to proceedings for personal injury and tangible damage to property, there is no requirement of commerciality in respect of the act out of which the claim arises.
Different aspects of the activity in respect of which proceedings are brought have been relied on in support of the commercial exception to immunity.
The earliest cases on State immunity accepted immunity as a plea where it was shown that the relevant act was performed for a sovereign or governmental purpose. Initially useful to distinguish a public purpose from a private one, the test of the purpose of the act has proved over-inclusive in relation to economic activities of the State. Thus the US Supreme Court applied the purpose test to confer immunity on a ship chartered by a State to carry grain, declaring: ‘We know of no international usage which requires the maintenance and advancement of the economic welfare of a people in time of peace as any less a public purpose than the maintenance and training of a naval force’.59
Further, since by its very nature a State can only act for public purposes, the application of a purpose test invariably favours the State and enlarges its immunity. The German Federal Constitutional Court so noted, stating: ‘the distinction between sovereign and non-sovereign cannot be drawn according to the purpose of the State transaction and whether it stands in a recognizable relation to the sovereign duties of the State. For, ultimately, activities of the State, if not wholly, then to the widest degree, serve sovereign purposes and duties and stand in a still recognizable relationship to them.’60
The use of the nature of the act as a determinant of its character for immunity purposes dates back to 1928 when in a note Switzerland proposed in relation to a restriction of State immunity: ‘The solution … would be to take as a criterion not the ultimate purpose of the act but its inherent nature’.61
By the 1960s Swiss,62 Austrian,63 and German courts had adopted the nature of the act as the method by which to distinguish between acta jure gestionis and acta jure imperii. The German Federal Constitutional Court, in declaring non-immune a contract to repair embassy premises of a foreign State, pronounced the much quoted words:
The distinction between acts jure imperii and acts jures gestionis can only be based on the nature of the act of the State or of the resulting legal relationship, not on the motive or purpose of the State activity … Contrary to the conclusion of the Foreign Minister of Justice, the criterion is not whether the conclusion of the contract was necessary for the orderly conduct of the Embassy’s business and therefore bore a perceptible relation to the sovereign activities of the sending State. Whether a State is entitled to immunity does not depend on the purpose pursued by the foreign State in carrying on a given activity.64
The FSIA, section 1603(d) expressly provides that ‘the commercial character of an activity shall be determined by reference to the nature of the course of conduct or particular transaction or act, rather than by reference to its purpose’ and US courts have resorted to a consideration of purpose ‘only so far as is absolutely necessary to define the nature of the act in question’. In Weltover in the passage cited above, overruling cases which in determining whether an act was immune had regard to a foreign State’s purpose in protecting its economy, the US Supreme Court rejected motive or aim as the relevant question and focused on the type of actions by which a private person engages in trade.65
The UK SIA avoided any express reference to the purpose or nature of the transaction by enumeration of specific transactions, save that the residuary category of ‘any other transaction’ referred to in section 3(3)(c) ‘requires a determination that a State enters or … engages [in it] otherwise than in the exercise of sovereign authority’. Further discussion of the use made of the purpose/nature distinction in UNCSI in the interpretative provision in Article 2(2) will be found in Chapter 13.
In an attempt to overcome these difficulties in an application of the purpose/nature distinction, Lord Wilberforce rejected strict adherence to the nature of the act as a criterion and instead advocated a broader contextual approach:
The conclusion which emerges is that in considering, under the restrictive theory, whether State immunity should be granted or not the court must consider the whole context in which the claim against the State is made, with a view to deciding whether the relevant act(s) on which the claim is based should, in that context, be considered as fairly within an area of activity, trading or commercial or otherwise of a private law character, in which the State has chosen to engage or whether the relevant act(s) should be considered as having been done outside that area and within the sphere of governmental or sovereign activity.66
This contextual approach has been adopted by English courts in deciding subsequent cases both under the SIA67 and under the common law. Claims against acts of visiting armed forces have been determined under common law and the nature of the act, the identity of the parties involved, and the place where it occurred have all been held relevant factors; both the nature and purpose of the activity have been taken into account and the court has adopted Lord Wilberforce’s contextual approach.68
By reason of the uncertainties arising from identification of proceedings as non-immune by reference to the nature of the transaction, an alternative method was to proceed by enumeration of the acts to be treated as immune.
More recently, the criterion of the nature of the act has been applied narrowly so as not to classify the whole transaction but to focus on individual provisions and to remove immunity in respect of those common to commercial transactions: thus in respect of a guarantee given by a State’s central bank on behalf of the Cameroon State to finance the building of a State hospital, the Cour de Cassation denied immunity since it resembled ‘a simple commercial act, performed in the normal course of its banking activities and had nothing to do with the exercise of sovereign authority’ (‘constitue un simple acte de commerce accompli dans l’exercise normal de ses activités bancaires et ne relève en rien de l’exercise de la puissance publique’).69 Similarly, in respect of a claim by a school teacher for maternity and sickness benefits, in denying immunity to the Saudi School, the Cour de Cassation disregarded its lack of separate legal status from the State, public purpose to provide education in accordance with Islamic principles, and the power of dismissal without notice; instead the Court based the exercise of jurisdiction by the national court on the private law nature of the acts complained of, namely refusal to register a school teacher, a non-Saudi national, in the French social system.70 The same approach was echoed by the US Supreme Court in Weltover when it spoke of ‘bonds in almost all respects garden variety debt instruments’.71 The US courts, however, do not pay as much attention to the character of the actor performing the act; many of the French cases in which jurisdiction is accepted relate to activities of State agencies both established for the purpose and expressly authorized by the State to engage in commercial activity.
Such a narrowing of focus may be particularly necessary in a continuing transaction where the acts at their commencement may be of one nature, and in the course of performance be of another. What is the moment in time at which the nature of the transaction is to be determined? Is it at the conclusion of the transaction, its subsequent performance or breach, or the date of the proceedings? As Lord Wilberforce noted, the criterion that the character of the relevant act and not its purpose is decisive may not address this question. It burked or begged:
the essential question, which is: what is the relevant act? It assumes that this is the essential entry into a commercial transaction and that this entry irrevocably confers on later acts a commercial or private law character. Essentially, it amounts to an assertion ‘once a trader always a trader’. But this may be an oversimplification. If a trader is always a trader, a State remains a State and is capable at any time of acts of sovereignty. The question arises, therefore, what the position is where the act on which the claim is founded is quite outside the commercial or private law activity in which the State has engaged, and has the character of an act done jure imperii. The restrictive theory does not and could not deny capability of a State to resort to sovereign, or governmental, action; it merely asserts that acts done within the trading or commercial activity are not immune.72
This problem can arise either where an initial commercial transaction is subject to later governmental acts or where an initial exercise of authority subsequently transforms into commercial activity.
The first situation is illustrated by a claim brought by disappointed customers against an airline, Dominica, a State agency wholly owned by the Dominican Republic in respect of a package vacation tour from Florida which was prematurely terminated when the Dominican immigration officials denied them entry and compelled the airline to their ‘involuntary re-routing’ back to the US.73 The Appeals Court held non-immune as commercial activity the claim for breach of contract relating to the vacation tour by the failure to warn of the risk of non-admission to Dominica, but distinguished as immune the subsequent claim relating to the forcible return of the customers in the involuntary rerouting. The Court said:
The focus of the exception to immunity recognised in section 1605(a)(2) is not on whether the defendant generally engages in a commercial enterprise or activity, as an airline such as Dominica unquestionably does, rather, it is on whether the particular conduct giving rise to the claim in question actually constitutes or is in connection with a commercial activity, regardless of the defendant’s generally commercial or governmental character…. Dominica’s actions in connection with the ‘involuntary re-routing’ were not commercial. Dominica acted merely as an arm or agent of the Dominican government in carrying out this assigned role, and, as such, is entitled to the same immunity from any liability arising from that governmental function as would inure to the government itself.74
Faced with a similar situation in the English case of I Congreso del Partido,75 a majority in the House of Lords held that the commercial transactions of carriage of a cargo of sugar for delivery to Chile remained commercial in character although terminated for political reasons (Cuba’s breaking off relations with Chile after its right-wing coup) because the acts complained of, the discharge and sale of the cargo, were effected under private law. Lord Wilberforce along with Lord Edmund Davies was in the minority in holding in respect of the ship Marble Islands that the Republic of Cuba was never in trading relations with the cargo owners and its actions, being confined to directing transfer of the sugar to North Vietnam and to the enactment of legislation freezing and blocking Chilean assets, must be characterized as being done in the exercise of sovereign authority.
In Kuwait Airways Corpn v Iraqi Airways Co the acts were in the reverse order; the initial acts were clearly acts jure imperii, being the unlawful invasion of Kuwait by Iraq and the seizure of aircraft belonging to the Kuwait civil airlines, but by a majority decision the Lords held that the transfer of title in the stolen aircraft by Iraqi legislative decree to Iraq Airways, a State agency, changed the nature of the acts making the subsequent incorporation of the Kuwaiti aircraft into the Iraqi fleet part of the commercial activity of the operation of a commercial civil airline.76
There seems to be no single method to solve these difficulties of characterization. To some extent they can be avoided by treating the whole issue as one of individuation. Once the events are characterized precisely and as narrowly as is reasonably possible, having regard to the factual and legal issues, as either a commercial transaction or a sovereign act, that characterization is to be treated as final. That approach seems to be borne out in the German cases which apply a presumption that once a State has entered the market a characterization of that act as commercial continues, regardless of the nature of the act constituting its subsequent breach. Such a presumption accords with the current position as to waiver. Just as waiver, unless made expressly revocable, is deemed irrevocable, so the undertaking of a commercial activity might be treated as irrevocable in its nature; a constitutional inability of the executive to bind its future action should not permit the conversion of a commercial act into an exercise of sovereign authority.77
Another approach is to distinguish between the actors, rather than the acts. Thus, where a State agency breached a commercial transaction and a minister sought to intervene to keep it in force, the claim against the agency was held non-immune but that against the State acting through the minister was held to be an exercise of sovereign power (puissance publique).78 Without a plea of immunity a State may be placed in a worse position than its agency, since a plea of force majeure may be available to a State agency, but not the State, who has caused the failure in performance of the contract by a later intervening exercise of sovereign authority.79 This was certainly how Goff J assessed the situation in I Congreso when he allowed the plea of immunity to the Republic of Chile: ‘[I]t is not enough to say that he can, if necessary, plead act of State by way of defence’.80
This particular problem, the characterization of activity which changes over time, is one further illustration of the difficulties presented in borderline cases of the application of the restrictive doctrine.
The restrictive doctrine and the criteria of private law act and commerciality on which it depends have not produced uniformity of practice nor reliable guidance as to when a national court will assume or refuse jurisdiction. The reasons are obvious: first, these criteria lack certainty and do little to reduce the commercial risks of doing business with States. In borderline cases they depend on the particular court’s individuation of some particular element, and reference neither to the nature nor to the purpose of the activity can disguise the arbitrary choices made by the court.81 Secondly, the distinction itself penalizes those governments which seek to conduct their affairs with proper regard to transparency and private interest. In directing particular criticism at the private law test, Henkin writes that it is built on questionable assumptions: ‘it builds on prejudices favouring the liberal State and against the welfare State; it tends to “penalize” governments that do what private persons could do (since in those activities the government loses its immunity) and gives such activities an aura of illegitimacy’.82
Thirdly, the doctrine assumes an unwarranted division of all acts performed by a State. The adoption of either the private law or the commercial nature of the activity as a general criterion produces an unacceptable dichotomy of all activities of the State. But as the hard cases—which present the most difficulty in deciding—illustrate, some activities combine in an inseparable way aspects of both private and public character either in parallel or in sequence, and others lack either aspect as a distinguishing feature. The defendant States in the International Tin case sought to argue that the residuary category was not based on ‘the straightforward dichotomy between acta jure imperii and acta jure gestionis that had become familiar doctrine in public international law’83 and consequently there was a third category of transaction or activity which might be exempt from immunity, being neither ‘sovereign’ nor ‘commercial’ within the definition. Evans J, however, rejected this submission.84
The Australian FSIA, section 11(2)(b), seeks to provide for this third category of activity by declaring expressly that immunity is retained for a proceeding which ‘concerns a payment in respect of a grant, a scholarship, a pension or a payment of like kind’.
It may be that the general tendency to accept a rule of immunity subject to exceptions reflects more accurately the correct relationship between public and private acts which the restrictive doctrine endeavours to formulate. A State charged with the care of a territory and population under international law exercises its jurisdiction for public purposes and is not subject to another State’s jurisdiction save where its acts cause special damage to a private party. On this formulation an exception to the rule of immunity for commercial acts thus becomes just one of the situations recognized as causing such special damage. Crawford makes this point as follows:
Even the notion of ‘trading’ or ‘commercial activity’ as one amongst a number of exceptions is not radically incoherent, since the concept of trade and commerce is a reasonably autonomous one when it is not presented as the other part of a universe of State activities from which the ‘sovereign’ or governmental part is excluded. The point is that the more work concepts such as ‘private’ or ‘commercial’ have to perform in distinguishing non-immune from immune cases, the more difficult and intractable these concepts become.85
In consequence, Crawford would rephrase the commercial exception as ‘contracts or related industrial or commercial activities, not being transactions governed by international law (such as treaties or public international arbitrations) and not being matters recognised as within the domestic jurisdiction of the foreign State’.
In defence of the doctrine must be its continued operation in all the jurisdictions which progressively adopted it during the twentieth century. Schreuer writes that the existence of borderline cases should not be taken as an indication that the distinction is unworkable. ‘A borderline will always remain. But this grey zone can be narrowed if we employ the right criteria and if courts are prepared to look beyond national confines to try to find common international standards.’86
Nor should one forget that the use of a distinction between public and private activity is not unique to the law of State immunity. English law recognizes an analogous distinction in a number of areas of the law, and similar analogous distinctions are to be found in other legal systems, particularly in the implementation of EU and ECHR law.87 In these areas there are difficult borderline cases too but it does not undermine the general usefulness of the distinction as a legal device.
UNCSI confirms that current State practice by adopting a general rule of immunity followed by a list of exceptions based on the distinction between private and public acts. A detailed examination of the scope of UNCSI’s exceptions to State immunity follows in the next chapter.