Argentum Exploration Ltd v Republic of South Africa

On 23 November 1942 SS TILAWA (“the Vessel”) was sunk in the Indian Ocean by enemy action. On board was a cargo of 2364 bars of silver (“the Silver”) being carried from Bombay to Durban. The Silver belonged to the Union of South Africa, now the Republic of South Africa (“the Government”), the appellant. The Silver had been purchased by the Government for the predominant purpose of being made into coin by the South African mint.

Between 29 January and 23 June 2017, the Silver was recovered from the seabed at a depth of some 2 ½ kilometres by the specialist salvage vessel MV SEABED WORKER. The Silver was then carried to the United Kingdom, arriving in Southampton on 2 October 2017 and was subsequently declared to the Receiver of Wreck, pursuant to section 236 of the Merchant Shipping Act 1995. The respondent (“Argentum”) claims to be the salvor of the Silver.

Argentum commenced an in rem claim against the Silver on 1 October 2019, seeking a declaration that it was the owner of the Silver or, in the alternative, salvage. Argentum now accepts that the Government is the owner of the Silver and therefore only the claim for salvage remains.

On 3 March 2020, the Government filed an acknowledgment of service solely for the purpose of challenging the jurisdiction of the court on the basis that it is entitled to immunity in accordance with section 1(1) of the State Immunity Act 1978 (“the SIA”) and/or Article 25 of the International Convention on Salvage 1989 (“the Salvage Convention”) (as given the force of law in the United Kingdom by section 224(1) of the Merchant Shipping Act 1995).

Under section 10(4)(a) of the SIA a state is not immune as respects:

“an action in rem against a cargo belonging to that State if both the cargo and the ship carrying it were, at the time when the cause of action arose, in use or intended for use for commercial purposes”.

It is common ground that in the present case the issue of “use” has to be considered by reference to evidence of the use and intended use of the Vessel and the Silver at the time of the sea carriage in November 1942. It is also common ground that the Vessel was then “in use” for commercial purposes and that there was no subsequent relevant change of use or intended use. The central issue on the appeal is whether in November 1942 the Silver was “in use or intended for use for commercial purposes”.

The judge, Sir Nigel Teare, and the majority of the Court of Appeal (Popplewell LJ and Andrews LJ) held that the Silver was “in use” for commercial purposes, essentially because it was being carried pursuant to a commercial contract of carriage, having been purchased by the Government under a commercial contract of sale. They accordingly concluded that pursuant to section 10(4)(a) of the SIA there is no state immunity. This conclusion was rejected by Elisabeth Laing LJ in her dissenting judgment. She held that as a matter of ordinary language, the Silver, which was sitting in the hold of the Vessel, was not “in use” by the Government for any purpose, whether commercial or otherwise. It was simply being carried. It was, however, “intended for use” for a non-commercial purpose, namely, to be minted into coinage. In those circumstances section 10(4)(a) does not apply and the Government is entitled to claim immunity. The Government contends that she was right so to conclude. Argentum supports the reasoning and conclusion of the majority.

On 17 November 1942 the Silver was despatched from the Bombay Mint and shipped on board the Vessel bound for Durban. The Silver was sold by the Government of India to the Government on free on board (“fob”) terms. Although sold on fob terms, it was the Government of India as seller which arranged the contract of carriage with the owners of the Vessel. It is common ground that it did so on behalf of the Government as purchaser and that the Government was a party to the contract of carriage.

The Vessel was a privately owned passenger/cargo liner engaged in commercial carriage.

The Silver had been purchased by the Government in order for it to be made into coin by the South African Mint. The judge found that the Silver was procured for the production of coin for both the Union of South Africa (a sovereign purpose) and Egypt (a commercial purpose) and that it was likely that the greater part of the consignment on board the Vessel would be used for Union coinage. The intended use of the Silver was therefore for a predominantly sovereign purpose.

The Vessel was sunk by two torpedoes fired from a Japanese submarine in the Indian Ocean on 23 November 1942.

The Silver was salved from the seabed between 29 January and 23 June 2017. It was transhipped from the salvage vessel, the MV SEABED WORKER, onto another vessel, the MV PACIFIC ASKARI, in the contiguous zone off the coast of South Africa on 3 September 2017. The Silver was then carried to the United Kingdom, arriving in Southampton on 2 October 2017 and subsequently declared to the Receiver of Wreck. It was brought to the United Kingdom because Argentum understood that the Silver belonged to the UK Government.

The Government had first become aware of the possibility of recovering the Silver from other salvors, namely a company called Odyssey Marine Exploration Inc (“Odyssey”), who had approached the then-Deputy President of the Government in September 2016 with a view to securing a salvage contract. The Government signed a contract with Odyssey on 14 February 2018. The judge found that the Government had not formed any intention to enter into a salvage contract with Odyssey until 13 October 2017 at the earliest, by which time the Silver had already been safely landed in the United Kingdom. Accordingly, as at the latest date by which the cause of action for salvage could have accrued, the Government had no intention as to the use of the Silver if and when salved.

Argentum commenced a claim in rem against the Silver on 1 October 2019. On 20 November 2020 it issued a claim in personam against the Government claiming salvage. The in personam claim has now been served out of the jurisdiction on the Government. It has indicated that the proceedings will be challenged on jurisdictional and time bar grounds. Under Article 23.1 of the Salvage Convention any action for salvage is time barred if proceedings have not been instituted within two years of the day on which the salvage operations are terminated.

In Jurisdictional Immunities of the State (Germany v Italy) [2012] ICJ Rep 99, para 56 the International Court of Justice (“the ICJ”) referred to the conclusion of the International Law Commission (“the ILC”) in 1980 that the rule of state immunity had been “adopted as a general rule of customary international law solidly rooted in the current practice of States” (Yearbook of the International Law Commission, 1980, Vol II (2), p 147, para 26). The ICJ considered that that practice showed that “whether in claiming immunity for themselves or according it to others, states generally proceed on the basis that there is a right to immunity under international law, together with a corresponding obligation on the part of other states to respect and give effect to that immunity”. It continued (at para 57):

“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. This principle 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.”

While state immunity in international law is primarily a matter of customary international law, it will be necessary to refer to the following multilateral treaties.

Until the 1970s the common law within the United Kingdom granted to foreign states a near absolute immunity from actions in personam and an absolute immunity in Admiralty actions in rem. Immunity was absolute in the sense that it failed to distinguish between the sovereign and non-sovereign activities of a state. Although there was support in some early cases for denying immunity in the case of state-owned trading vessels (The Charkieh (1873) LR 4 A & E 59, per Sir Robert Phillimore at pp 99–100; The Parlement Belge (1879) 4 PD 129 per Sir Robert Phillimore at pp 148–149), the judgment of the Court of Appeal in The Parlement Belge (1880) 5 PD 197 was erroneously understood (for example in The Porto Alexandre [1920] P 30) as authority for the absolute immunity of state property and as requiring the absolute immunity of state-owned ships from actions in rem. (See the discussion in The Philippine Admiral [1977] AC 373, per Lord Cross of Chelsea at pp 391–394, and in Benkharbouche v Embassy of the Republic of Sudan [2017] UKSC 62; [2019] AC 777 per Lord Sumption at paras 43–44.) In Cia Naviera Vascongada v Steamship Cristina (The Cristina) [1938] AC 485, 490 Lord Atkin, with whom Lord Wright agreed, expressed the principle of immunity in absolute terms:

“The foundation for the application to set aside the writ and arrest of the ship is to be found in two propositions of international law engrafted into our domestic law which seem to me to be well established and to be beyond dispute. The first is that the courts of a country will not implead a foreign sovereign, that is, they will not by their process make him against his will a party to legal proceedings whether the proceedings involve process against his person or seek to recover from him specific property or damages.

The second is that they will not by their process, whether the sovereign is a party to the proceedings or not, seize or detain property which is his or of which he is in possession or control.”

Lord Atkin and Lord Wright considered that the second principle extended to property only used for the commercial purposes of the sovereign and to personal private property. It is, however, significant that the other three members of the House of Lords expressed doubts about whether there was a sufficient international consensus to support such an absolute rule (Lord Thankerton at pp 494–496; Lord Macmillan at p 498; Lord Maugham at pp 518–520).

In the years following the Second World War there emerged a trend in the decisions of courts of a number of nations in favour of the restrictive theory of state immunity. In the Tate Letter of 1952, the US State Department favoured restrictive immunity and that line was then taken up by the US Federal Courts. (See, for example, Alfred Dunhill of London Inc v Republic of Cuba (1976) 425 US 682, 701–703.) In the Federal Republic of Germany, in 1963 the Bundesverfassungsgericht adopted a theory of restrictive immunity founded on the juridical character of the conduct in question (Claim against the Empire of Iran (1963) 45 ILR 57, 79–82). Thereafter, the general trend in the decisions of national courts and the writings of international jurists was away from absolute immunity and towards a more restrictive theory.

In 1975, in The Philippine Admiral [1977] AC 373, an appeal from the Supreme Court of Hong Kong, the Judicial Committee of the Privy Council recognised this trend. It explained (at p 397G–H):

“This restrictive theory seeks to draw a distinction between acts of a state which are done jure imperii and acts done by it jure gestionis and accords the foreign state no immunity either in actions in personam or in actions in rem in respect of transactions falling under the second head.”

The Privy Council declined to follow The Porto Alexandre and applied the restrictive theory of state immunity to actions in rem against a state-owned trading vessel. As the vessel was being operated as an ordinary trading vessel and as it was not even asserted that she would not continue to be used in this way while owned by the state, there was no entitlement to immunity. In coming to this conclusion, the Privy Council accepted that to apply the restrictive theory to actions in rem while leaving actions in personam to be governed by the absolute theory would produce a very illogical result. While it was no doubt open to the House of Lords to decide otherwise (ie to allow an action in personam to be brought against a foreign sovereign state on a commercial contract) the Privy Council considered it unlikely that it would do so. Nevertheless, the Privy Council rejected a submission that the matter should be left to the executive to ratify the Brussels Convention and the European Convention on State Immunity and to secure implementing legislation.

“But their Lordships—while recognising that there is force in that argument—are not prepared to accept it. Thinking as they do that the restrictive theory is more consonant with justice they do not think that they should be deterred from applying it so far as they can by the thought that the resulting position may be somewhat anomalous.” (at p 403B–C)

Any resulting anomaly was short-lived. In 1977 the Court of Appeal held in Trendtex Trading Corpn v Central Bank of Nigeria [1977] QB 529 that the restrictive theory of immunity should be applied generally. The claimant claimed against the Central Bank of Nigeria payments due in respect of the Bank’s breaches and repudiation of a letter of credit which it had issued. A majority of the Court of Appeal (Lord Denning MR and Shaw LJ) held that even if the Bank were part of the government of Nigeria, effect should be given to international law which recognised no immunity from suit for a government department in respect of ordinary commercial transactions.

The decision of the Court of Appeal in Trendtex was not appealed to the House of Lords. However, the issue as to the scope of state immunity at common law arose for decision in I Congreso del Partido [1983] AC 244. Pursuant to a contract for the sale of sugar by a Cuban state enterprise to the claimants, a Chilean company, two cargoes of sugar were dispatched to Chile on board the Playa Larga and the Marble Islands. The vessels were under voyage charters to Cubazucar from Mambisa, another Cuban state enterprise. Following a revolution in Chile in September 1973 the Marble Islands cargo and the undelivered balance of the Playa Larga cargo were diverted on the orders of the Cuban government and not delivered. The claimants brought three actions in rem against the owners of the I Congreso, a sister ship of the Playa Larga and the Marble Islands, a vessel constructed in Sunderland to be used for normal trading purposes, of which Mambisa, on behalf of the Republic of Cuba, had just taken delivery. In each action it was alleged that Mambisa or the Republic of Cuba would be liable to the claimants in an action in personam. The Republic of Cuba applied to set aside the writs as impleading a foreign sovereign. Although there was disagreement among their Lordships as to its application to the Marble Islands claim, the House of Lords gave its approval to the restrictive theory of state immunity in respect of both actions in rem and in personam. Lord Wilberforce, in a passage of particular relevance to the present appeal at p 261D–G, stated:

“Sitting in this House I would unhesitatingly affirm as part of English law the advance made by The Philippine Admiral … with the reservation that the decision was perhaps unnecessarily restrictive in, apparently, confining the departure made to actions in rem. In truth an action in rem as regards a ship, if it proceeds beyond the initial stages, is itself in addition an action in personam – viz the owner of the ship (see The Cristina [1938] AC 485, 492 per Lord Atkin, p 504 per Lord Wright), the description in rem denoting the procedural advantages available as regards service, arrest and enforcement. It should be borne in mind that no distinction between actions in rem and actions in personam is generally recognised elsewhere so that it would in any event be desirable to liberate English law from an anomaly if that existed. In fact there is no anomaly and no distinction. The effect of The Philippine Admiral … if accepted, as I would accept it, is that as regards state-owned trading vessels, actions, whether commenced in rem or not, are to be decided according to the ‘restrictive’ theory.”

Lord Wilberforce explained that when a claim is brought against a state and state immunity is claimed, it is necessary to consider what is the relevant act which forms the basis of the claim: whether it is an act of a private law character such as a private citizen might have entered into or a sovereign or public act. He stated his conclusion on this point as follows (at p 267B-D):

“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) upon 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.”

Before leaving this account of the development of a restrictive immunity at common law, it is appropriate to record that in his judgment in Benkharbouche, with which the other members of the Supreme Court agreed, Lord Sumption explained (at para 52) that the adherence in the United Kingdom to a theory of immunity which failed to distinguish between sovereign and non-sovereign activities of states was largely founded on an erroneous view of international law which never warranted extending immunity beyond what sovereigns did in their capacities as such.

By the date of the decision of the House of Lords in I Congreso in July 1981, the SIA had come into force on 22 November 1978. (As the statute did not have retroactive effect, I Congreso was decided on common law principles.) It had become apparent that there was an urgent need for legislation in this field. While it would, no doubt, have been possible for the judges to complete the reform of the common law of state immunity in this jurisdiction, this would have required elaboration in many cases over many years. In the meantime, the law would have been left in a state of uncertainty. As Lord Wilberforce explained in I Congreso (at p 260 C–D), while it had become clear that international law in a general way gave support to a restrictive theory of state immunity, the precise limits of the doctrine were still in the course of development and were in many respects uncertain. Furthermore, in the event, the SIA did not adopt a straightforward dichotomy between sovereign and non-sovereign acts in its approach to immunity from adjudicative and enforcement jurisdiction. (See Alcom Ltd v Republic of Colombia [1984] AC 580 per Lord Diplock at p 600C–D.)

What was needed was not the incremental development of the common law through judicial decisions but a new statutory scheme providing detailed and comprehensive rules governing both adjudicative and enforcement jurisdiction in cases involving foreign and Commonwealth states. There was also a commercial need to bring domestic law in the United Kingdom into line with the new international reality of restrictive immunity. In this regard, it was highly significant that the United States had given effect to restrictive immunity in the Foreign Sovereign Immunities Act of 1976.

In addition, the United Kingdom wished to become a party to the Brussels Convention and the ECSI, two international conventions which gave effect to the restrictive theory in different fields. These would require implementation into domestic law by legislation. In particular, the ECSI included provision for the reciprocal recognition of judgments against contracting states. As a result, in defining exceptions to immunity from adjudicative jurisdiction, it did not seek simply to identify non-sovereign activities but limited those exceptions to those cases where there was also a sufficient jurisdictional link between the contracting state and the subject matter of the proceedings for the purposes of recognition. The implementation of this system could only be achieved by legislation in the United Kingdom. The United Kingdom ratified both conventions in 1979.

The long title of the SIA is “An Act to make new provision with respect to proceedings in the United Kingdom by or against other States; to provide for the effect of judgments given against the United Kingdom in the courts of States parties to the European Convention on State Immunity; to make new provision with respect to the immunities and privileges of heads of State; and for connected purposes”.

Part I concerns proceedings in the United Kingdom by or against other states. Section 1 establishes a general immunity from jurisdiction.

“(1)

A State is immune from the jurisdiction of the courts of the United Kingdom except as provided in the following provisions of this Part of this Act.

(2)

A court shall give effect to the immunity conferred by this section even though the State does not appear in the proceedings in question.”

The following sections then set out exceptions to immunity in cases of submission to the jurisdiction (section 2), commercial transactions and contracts to be performed in the United Kingdom (section 3), contracts of employment (section 4), personal injuries and damage to property (section 5), ownership, possession and use of property (section 6), patents, trade-marks etc (section 7), membership of bodies corporate etc (section 8), arbitrations (section 9), ships used for commercial purposes (section 10) and value added tax, customs duties etc (section 11).

The various jurisdictional connecting factors provided for in the ECSI are generally implemented in all of the exceptions to immunity, with the exception of commercial transactions in section 3 and Admiralty proceedings in section 10. Thus, for example, under section 5 a state is not immune as respects proceedings in respect of death or personal injury or damage to or loss of tangible property only where it is caused by an act or omission in the United Kingdom.

As Fox and Webb point out (The Law of State Immunity (revised and updated 3^rd Ed, 2015), at p 175), the SIA makes considerable departures from any application of a distinction between acta jure imperii and acta jure gestionis in removing immunity for all contracts performable wholly or in part in the United Kingdom and for specified transactions of sale of goods, provision of services or loans (section 3(1)(b) and 3(3)) and for claims relating to personal injuries caused by an act or omission of the state in the United Kingdom (section 5).

Section 3 which establishes an exception to immunity from the adjudicative jurisdiction of UK courts in the case of commercial transactions and contracts to be performed in the United Kingdom provides:

“(1)

A State is not immune as respects proceedings relating to—

(a)

a commercial transaction entered into by the State; or

(b)

an obligation of the State which by virtue of a contract (whether a commercial transaction or not) falls to be performed wholly or partly in the United Kingdom.

(2)

This section does not apply if the parties to the dispute are States or have otherwise agreed in writing; and subsection (1)(b) above does not apply if the contract (not being a commercial transaction) was made in the territory of the State concerned and the obligation in question is governed by its administrative law.

(3)

In this section ‘commercial transaction’ means—

(a)

any contract for the supply of goods or services;

(b)

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; and

(c)

any other transaction or activity (whether of a commercial, industrial, financial, professional or other similar character) into which a State enters or in which it engages otherwise than in the exercise of sovereign authority;

but neither paragraph of subsection (1) above applies to a contract of employment between a State and an individual.”

It has been noted above that the general inclusion of jurisdictional connecting factors in respect of the exceptions to immunity from adjudicative jurisdiction in the SIA is departed from in the case of commercial transactions in section 3(1)(a). However, it is adhered to in the case of the second limb of the general commercial exception relating to contractual obligations of the state in section 3(1)(b) where the jurisdictional link of performance wholly or partly in the United Kingdom is retained.

Section 6(4) concerns cases of indirect impleader where a state is not directly impleaded but where the proceedings relate to property which is in the possession or control of a state or in which a state claims an interest.

“(4)

A court may entertain proceedings against a person other than a State notwithstanding that the proceedings relate to property—

(a)

which is in the possession or control of a State; or

(b)

in which a State claims an interest,

if the State would not have been immune had the proceedings been brought against it or, in a case within paragraph (b) above, if the claim is neither admitted nor supported by prima facie evidence.”

These general provisions should be contrasted with the more specific provisions which apply to Admiralty proceedings within section 10.

Section 10 makes provision for Admiralty proceedings and proceedings on any claim which could be made the subject of Admiralty proceedings.

“(1)

This section applies to—

(a)

Admiralty proceedings; and

(b)

proceedings on any claim which could be made the subject of Admiralty proceedings.

(2)

A State is not immune as respects—

(a)

an action in rem against a ship belonging to that State; or

(b)

an action in personam for enforcing a claim in connection with such a ship,

if, at the time when the cause of action arose, the ship was in use or intended for use for commercial purposes.

(3)

Where an action in rem is brought against a ship belonging to a State for enforcing a claim in connection with another ship belonging to that State, subsection (2)(a) above does not apply as respects the first-mentioned ship unless, at the time when the cause of action relating to the other ship arose, both ships were in use or intended for use for commercial purposes.

(4)

A State is not immune as respects—

(a)

an action in rem against a cargo belonging to that State if both the cargo and the ship carrying it were, at the time when the cause of action arose, in use or intended for use for commercial purposes; or

(b)

an action in personam for enforcing a claim in connection with such a cargo if the ship carrying it was then in use or intended for use as aforesaid.

(5)

In the foregoing provisions references to a ship or cargo belonging to a State include references to a ship or cargo in its possession or control or in which it claims an interest; and, subject to subsection (4) above, subsection (2) above applies to property other than a ship as it applies to a ship.

(6)

Sections 3 to 5 above do not apply to proceedings of the kind described in subsection (1) above if the State in question is a party to the Brussels Convention and the Claim relates to the operation of a ship owned or operated by that State, the carriage of cargo or passengers on any such ship or the carriage of cargo owned by that State on any other ship.”

Section 10 was enacted in order to enable the United Kingdom to ratify the Brussels Convention. However, it does not precisely reflect the provisions of that Convention the interpretation of which had in the past given rise to certain difficulties. It should also be noted that this subject matter falls outside the scope of the ECSI, article 30 of which provides that it is not to apply to proceedings in respect of claims relating to the operation of seagoing vessels owned or operated by a contracting state or to the carriage of cargoes and of passengers by such vessels or to the carriage of cargoes owned by a contracting state and carried on board merchant vessels. The purpose of article 30 was to exclude matters covered by the Brussels Convention which was in force between a number of the member states of the Council of Europe. (See The Philippine Admiral at p 401G–H.)

Whereas sections 2 to 11 of the SIA are in general concerned with immunity from the adjudicative jurisdiction of the UK courts, section 10 is different in that it is a hybrid provision making specific provision concerning immunity from both adjudicative and enforcement jurisdiction in Admiralty proceedings or proceedings on any claim which could be made the subject of Admiralty proceedings. Section 13(2)(b) and 13(4) also apply to enforcement jurisdiction in such proceedings. (See Alcom per Lord Diplock at p 600F-G.)

Enforcement jurisdiction in general is dealt with in the procedural provisions in sections 12 to 14, in particular section 13(2) to 13(6), 14(3) and 14(4). Section 13(2), 13(3) and 13(4) are the most relevant to this appeal.

“(2)

Subject to subsections (3) and (4) below—

(a)

relief shall not be given against a State by way of injunction or order for specific performance or for the recovery of land or other property; and

(b)

the property of a State shall not be subject to any process for the enforcement of a judgment or arbitration award or, in an action in rem, for its arrest, detention or sale…

(3)

Subsections (2) and (2A) above do not prevent the giving of any relief or the issue of any process with the written consent of the State concerned; and any such consent (which may be contained in a prior agreement) may be expressed so as to apply to a limited extent or generally; but a provision merely submitting to the jurisdiction of the courts is not to be regarded as a consent for the purposes of this subsection.

(4)

Subsection (2)(b) above does not prevent the issue of any process in respect of property which is for the time being in use or intended for use for commercial purposes; but, in a case not falling within section 10 above, this subsection applies to property of a State party to the European Convention on State Immunity only if—

(a)

the process is for enforcing a judgment which is final within the meaning of section 18(1)(b) below and the State has made a declaration under Article 24 of the Convention; or

(b)

the process is for enforcing an arbitration award.”

Section 17(1) defines “commercial purposes” in Part I of the SIA as meaning “purposes of such transactions or activities as are mentioned in section 3(3) above”.

As we have seen, one purpose of the enactment of the SIA was to enable the United Kingdom to ratify the Brussels Convention.

Fox and Webb at pp 117-8 explain that the Brussels Convention was an early attempt to abolish immunity in respect of a particular area of trade: the operation of trading ships owned or controlled by States. They further state:

“The Brussels Convention was intended to give effect to the distinction between public and private acts but the complexity of its drafting means that its application depends on the ship coming within the precise conditions listed in the provisions. On this account the UK State Immunity Act, instead of merely giving effect in English law to the Brussels Convention, contains a specific section dealing with Admiralty proceedings in rem and in personam relating to claims against ships, their sister ships, and their cargoes (section 10) … The application of the Brussels Convention to ships conforming with its listed conditions may on occasion afford a wider immunity than does the UK SIA to government ships. But the complexity of the Convention has discouraged its wider ratification by States, particularly as national courts have shown themselves prepared, in reliance on general principles of international law, to accept jurisdiction as a non-immune proceeding relating to a private law activity over claims arising out of the carriage of goods or passengers on merchant ships owned or operated by States.”

In I Congreso at p 260E–H Lord Wilberforce rejected a submission that the Brussels Convention was a statement of generally accepted international law. He described the Brussels Convention as “a limited agreement between a limited number of states” and continued: “At the very most it may, together with its progressive, though not numerous, ratifications and accessions, be evidence of the gradual seepage into international law of a doctrine of restrictive immunity.”

So far as material the Brussels Convention provides as follows:

“Article 1

Sea-going ships owned or operated by States, cargoes owned by them, and cargoes and passengers carried on State-owned ships, as well as the States which own or operate such ships and own such cargoes shall be subject, as regards claims in respect of the operation of such ships or in respect of the carriage of such cargoes, to the same rules of liability and the same obligations as those applicable in the case of privately-owned ships, cargoes and equipment.

Article 2

As regards such liabilities and obligations, the rules relating to the jurisdiction of the Courts, rights of action and procedure shall be the same as for merchant ships belonging to private owners and for private cargoes and their owners.

Article 3

(1)

The provisions of the two preceding Articles shall not apply to ships of war, State-owned yachts, patrol vessels, hospital ships, fleet auxiliaries, supply ships and other vessels owned or operated by a State and employed exclusively at the time when the cause of action arises on Government and non-commercial service, and such ships shall not be subject to seizure, arrest or detention by any legal process, nor to any proceedings in rem.

Nevertheless, claimants shall have the right to proceed before the appropriate Courts of the State which owns or operates the ship in the following cases: —

(i)

Claims in respect of collision or other accidents of navigation;

(ii)

Claims in respect of salvage or in the nature of salvage and in respect of general average;

(iii)

Claims in respect of repairs, supplies or other contracts relating to the ship;

and the State shall not be entitled to rely upon any immunity as a defence.

(2)

The same rules shall apply to State-owned cargoes carried on board any of the above-mentioned ships.

(3)

State-owned cargoes carried on board merchant ships for Government and non-commercial purposes shall not be subject to seizure, arrest or detention by any legal process nor to any proceedings in rem.

Nevertheless, claims in respect of collisions and nautical accidents, claims in respect of salvage or in the nature of salvage and in respect of general average, as well as claims in respect of contracts relating to such cargoes, may be brought before the Court which has jurisdiction in virtue of Article 2.”

The structure of the Brussels Convention is therefore to make States subject to the same rules of liability in respect of vessels owned or operated by them and cargoes owned by them (Article 1) and the same rules of jurisdiction and procedure (Article 2) as are applicable to privately owned vessels and cargoes. Article 3 then disapplies or modifies the application of Articles 1 and 2 in relation to (a) a list of ships which may loosely be called “State Vessels”; (b) state-owned cargoes carried on board State Vessels and (c) certain state-owned cargoes carried on board merchant ships.

Most claims for salvage arise under a salvage contract. In the present case there was no salvage contract and the claim is one for voluntary salvage under Article 12(1) of the Salvage Convention which entered into force on 14 July 1996. Article 12(1) provides that: “Salvage operations which have had a useful result give right to a reward.” Article 13(1) sets out the criteria by which the reward shall be fixed. Article 13(2) requires that payment “shall be made by all of the vessel and other property interests in proportion to their respective salved values.” Article 25 of the Salvage Convention provides:

"Article 25 – State-owned cargoes

Unless the State owner consents, no provision of this Convention shall be used as a basis for the seizure, arrest or detention by any legal process of, nor for any proceedings in rem against, non-commercial cargoes owned by a State and entitled, at the time of the salvage operations, to sovereign immunity under generally recognized principles of international law.”

The Salvage Convention was given the force of law in the United Kingdom by section 224(1) of the Merchant Shipping Act 1995. At the time of the enactment of the SIA the right to voluntary salvage was governed by maritime law. The right to salvage under maritime law is an exceptional example of a liability being imposed on the owner of property salved irrespective of whether they requested or consented to the salvage. The liability does not arise as a result of any personal undertaking or breach of duty: it is imposed, as a matter of public policy for the advantage of trade, on the owner of the property which is salved. As explained by Bowen LJ in Falcke v Scottish Imperial Insurance Company (1886) 34 Ch D 234 at pp 248–9:

“The general principle is, beyond all question, that work and labour done or money expended by one man to preserve or benefit the property of another do not according to English law create any lien upon the property saved or benefited, nor, even if standing alone, create any obligation to repay the expenditure. Liabilities are not to be forced upon people behind their backs any more than you can confer a benefit upon a man against his will.

There is an exception to this proposition in the maritime law. I mention it because the word ‘salvage’ has been used from time to time throughout the argument, and some analogy is sought to be established between salvage and the right claimed by the Respondents. With regard to salvage, general average, and contribution, the maritime law differs from the common law. That has been so from the time of the Roman law downwards. The maritime law, for the purposes of public policy and for the advantage of trade, imposes in these cases a liability upon the thing saved, a liability which is a special consequence arising out of the character of mercantile enterprises, the nature of sea perils, and the fact that the thing saved was saved under great stress and exceptional circumstances. No similar doctrine applies to things lost upon land, nor to anything except ships or goods in peril at sea.”

The necessary elements of a claim for voluntary salvage at maritime law are: (i) the services have been rendered by a volunteer; (ii) the property saved is a recognised subject of salvage; (iii) the property was in danger at the time that the services were performed and (iv) the services have resulted in the preservation of property of value. See The Goring [1988] AC 831, 845F–G per Lord Brandon of Oakbrook, adopting the definition which was common ground between the parties (see the argument of Anthony Clarke QC recorded at p 833G–H).

Recognised subjects of salvage are: “a ship, her apparel, and her cargo; of freight in danger, and saved by reason of the saving of the ship or cargo; and of flotsam, jetsam, or lagan, being each of them part of the cargo of a ship” – per Lord Esher MR in The Gas Float Whitton (No 2) [1896] P 42, 49. Kennedy and Rose on the Law of Salvage, 10th ed (2021), at paras 3-022 to 3-023 states that a further de facto requirement of a claim for salvage is that the activity being undertaken at the time of the casualty be a maritime adventure – ie “an activity of a common nature involving passage through maritime waters”.

Salvage gives rise to a maritime lien. A maritime lien arises by operation of law, at the time of the event creating it and, once created, it is enforceable against purchasers of the property, whether or not they have notice of it. It takes priority over all other claims (whether those claims arose before or after the creation of the maritime lien) save for possessory liens (Jackson: Enforcement of Maritime Claims 4th ed (2005), at para 18.2).

A maritime lien has been described as “a privileged claim … over a thing belonging to another” “to be carried into effect by legal process” and “a subtraction from the absolute property of the owner in the thing” (The Ripon City [1897] P 226, 242 per Gorell Barnes J). In The Tervaete [1922] P 259, 264 Bankes LJ cited these observations with approval and observed at p 266 that the value of property to which a maritime lien attaches will necessarily be affected: “A vessel to which a maritime lien extends for any substantial amount must necessarily be worth less in the market than if she was free from any lien”.

The legal process by which a maritime lien is “carried into effect” is an action in rem. In The Bold Buccleugh (1851) 7 Moo PC 267, 284 Sir John Jervis described a maritime lien as being a claim or privilege to be carried into effect by legal process and proceedings in rem as “a process to make perfect a right inchoate from the moment the lien attaches”.

More recently in the Federal Court of Australia, Allsop CJ and Edelman J in Reiter Petroleum Inc v The Ship “Sam Hawk” [2016] FCAFC 26, [2016] 2 Lloyd’s Rep 639, para 49 stated the “essential characteristics” of the maritime lien to be as follows:

“….a security or privilege attaching to the ship itself, travelling with the ship (irrespective of any change in ownership, or in encumbrance, of the ship) and arising from circumstances concerned with the ship…a charge on the ship…attaching automatically by operation of law, upon the occurrence of the relevant events…adhering to the ship, notwithstanding a change of ownership; enforced only by maritime proceedings against the ship, and in that sense inchoate, until perfection by the bringing of maritime proceedings; its importance being both as to access to the property for payment irrespective of ownership, and as to priority over other claimants to the ship represented by the fund to which it attaches, even secured claims such as mortgages.”

A ‘true’ or ‘core’ in rem claim is a claim against the property itself irrespective of its ownership – see generally Meeson and Kimbell on Admiralty Jurisdiction and Practice, 5th ed (2018), paras 3.1 to 3.33. As they state at para 3.2:

“There is a category of in rem claim which could be described as truly in rem because it is brought against a ship irrespective of her present ownership and irrespective of any link with liability in personam on the part of the owner of the ship at the time the claim is brought. This category comprises claims to enforce maritime liens and mortgages, claims for forfeiture, droits of Admiralty, and claims relating to possession or ownership. These are claims where in substance there is a claim to the ship in whole or in part. That is to say claims which are true in rem claims are, as the name suggests, directed against the ship as res and not against any person who has an interest in the ship such as an owner.” (emphasis in original)

The nature of a core in rem claim and the fact that it does not involve personal liability was explained by Fletcher Moulton LJ in The Burns [1907] P 137, 149 as follows:

“…the action in rem is an action against the ship itself. It is an action in which the owners may take part, if they think proper, in defence of their property, but whether or not they will do so is a matter for them to decide, and if they do not decide to make themselves parties to the suit in order to defend their property, no personal liability can be established against them in that action.”

As set out in section 20 of the Senior Courts Act 1981, which deals with the Admiralty jurisdiction of the High Court, core in rem claims comprise any claim to the possession or ownership of a ship or to the ownership of any share therein (section 20(2)(a)); any question arising between the co-owners of a ship as to possession, employment or earnings of that ship (section 20(2)(b)); any claim in respect of a mortgage of or charge on a ship or any share therein (section 20(2)(c)); any claim for the forfeiture or condemnation of a ship or of goods which are being or have been carried, or have been attempted to be carried, in a ship, or for the restoration of a ship or any such goods after seizure, or for droits of Admiralty (section 20(2)(s)); and any case in which there is a maritime lien for the amount claimed (section 21(3)). The recognised categories of claim to which a maritime lien attaches are (i) damage done by a ship; (ii) salvage; (iii) seamen’s wages; (iv) bottomry and respondentia and (v) masters’ wages and disbursements (Meeson and Kimbell at para 1.47, Jackson at para 18.1).

As a salvage claim gives rise to a maritime lien it is a core in rem claim. The only other Admiralty jurisdiction claims which may be made against cargo are claims for forfeiture, condemnation, restoration or droits of Admiralty arising under section 20(2)(s) which are also core in rem claims.

There are also statutory in rem claims. These are the claims set out in section 20(2) which are not core in rem claims. Although in form a statutory in rem claim is against the ship, unlike a core in rem claim it is directed against the person who is the owner of the ship. The defendant is sued but through service on the ship.

The most important practical advantages of proceedings in rem are that they confer a right of arrest, thereby enabling security for a claim to be obtained, and they establish jurisdiction through service on the res within the jurisdiction. As Teare J explained in LD Commodities Rice Merchandising LLC v The Owners and/or Demise Charterers of the Vessel “Styliani Z” (The Styliani Z) [2015] EWHC 3060 (Admlty), [2016] 1 Lloyd’s Rep 395, para 20:

“…an Admiralty action in rem nevertheless has a characteristic which distinguishes it from an Admiralty action in personam. That characteristic is that the in rem claim form may be served within the jurisdiction on the vessel named in the claim form as the vessel against which the action is brought (by fixing a copy of the claim form on the outside of the vessel in a position which may be reasonably expected to be seen). The action may then proceed to trial even though the owners of the vessel are out of the jurisdiction and have not been served personally with the proceedings or acknowledged service of proceedings. The vessel may also be arrested by the Admiralty Marshal and when judgment is given (or pendente lite) the vessel can be sold by the Admiralty Court and the judgment satisfied from the proceeds of sale. In practice, in the great majority of cases, the owners’ P&I Club will provide security for the claim and instruct solicitors to accept service so that the action will proceed in a manner indistinguishable from an action in personam. But that practice should not obscure the distinguishing characteristics of an action in rem.”

After there has been acknowledgment of service of an action in rem, the action becomes in personam. It does not, however, lose its character of being an action in rem but continues as a hybrid action – see The Maciej Rataj [1992] 2 Lloyd’s Rep 552, 559.

The principal reasons given by the judge and the majority of the Court of Appeal for concluding that the Silver was “in use” for commercial purposes were as follows:

In her dissenting judgment, Elisabeth Laing LJ stated that whether something is a cargo is primarily a factual, not a legal, question. A thing is a cargo if it is being carried on a ship (para 138). Whether a cargo has an intended use is a question of fact, to be answered on any evidence which shows what use was intended for the cargo by the state. It is to be answered by reference to the evidence about that intended use at the time when the casualty occurred (para 139(v)). As a matter of ordinary language, the Silver which was sitting in the hold of the ship was not being used by the Government for any purpose, commercial or otherwise. It was being carried, and that is all. It was the subject of commercial arrangements for its carriage, but that is not the relevant inquiry (para 142). The approach of the judge, of Popplewell LJ and of Andrews LJ means that section 10(4)(a) could never apply to a state-owned cargo carried on a commercial or merchant vessel, because such a cargo will always be carried pursuant to commercial arrangements of the kind which they describe in their judgments, and will therefore never have immunity. That is a counter-intuitive result for state-owned cargoes such as armaments (para 140). Section 10(4)(a) should be construed, so far as possible, so as to be consistent with article 3(3) of the Brussels Convention. The Silver was being ‘carried’ for the relevant purposes under article 3(3), as it was being carried to South Africa in order to be minted into coinage, and substantially for a Governmental and non-commercial purpose, as the judge found (paras 148, 156 and 157).

The issues on this appeal may be stated as follows:

Before the judge and the Court of Appeal the Government contended that since the cause of action in salvage arose in 2017 when the Silver was salved, it was its use or intended use in 2017 which was determinative and it had no commercial use or intended use at that time. This argument was rejected by the judge, not least because it would be likely to mean that a state would always be immune from liability in salvage in wreck cases. He considered that, applying section 10(4)(a) intelligently rather than mechanically, regard must be had to the use of the vessel and cargo when the vessel was carrying the cargo. He concluded that “it is appropriate to have regard to the status of the vessel and cargo in 1942 when deciding whether the vessel and cargo were, at the time the cause of action for salvage arose in 2017, in use or intended for use for commercial purposes” (para 123).

The Court of Appeal reached a similar conclusion but on the basis that the phrase “when the cause of action arose” is referring to “the point of time at which the relevant aspect of the cause of action for salvage in maritime law arises… The ingredient of the maritime law cause of action in salvage to which use of ship or cargo is relevant comprises the maritime circumstances which make the property a recognised subject matter of salvage” (para 70).

The Government now accepts that the judge’s approach is correct and that it is appropriate to have regard to the use and intended use of the vessel and cargo at the time when the vessel was carrying the cargo in 1942. There is therefore now no material dispute between the parties. The only difference between them is whether the Court of Appeal’s analysis of when the cause of action arose is correct.

We agree with the Government that section 10(4)(a) is referring to when the cause of action arose, not the events which later led to that cause of action arising. That can only have been in 2017. Before then there was no act of salvage. In 1942 and indeed for many years thereafter there was not only no act of salvage but no possibility of any such acts given the depth of the wreck. Moreover, for most of that period the claimant, Argentum, did not exist.

We agree with the judge’s approach and conclusion on this issue. It is appropriate to have regard to the use and intended use of the vessel and cargo at the time when the vessel was carrying the cargo. In cases where the cause of action arises later in time, it is also appropriate to have regard to whether there has been any change in use or intended use in the intervening period. If, as in the present case, there has been no such change then it is the status of the vessel and cargo at the time of carriage which will be determinative.

Section 10(4)(a) of the SIA provides that a state is not immune as respects an action in rem against a cargo belonging to that state if both the cargo and the ship carrying it were, at the time when the cause of action arose, in use or intended for use for commercial purposes. It is common ground between the parties that the Vessel was in use for commercial purposes. It is also now common ground between the parties that the Silver on board the Vessel was not intended for use for commercial purposes. The judge found that the intended use of the silver in 1942 was predominantly the sovereign purpose of minting coins for the Union of South Africa. As a result, the central issue between the parties on this appeal is whether the Silver was “in use … for commercial purposes” within section 10(4)(a) of the SIA when being carried on board the Vessel in 1942.

The Government submits that the conclusion of the judge and the majority in the Court of Appeal that the Silver was in use by the Government when it was being carried on board the Vessel is inconsistent with the natural meaning of the words “in use” and ignores the words “or intended for use” in section 10(4)(a). It submits that the effect of that reading is to remove the distinction drawn in section 10(4) between the threshold for immunity in respect of in rem claims (section 10(4)(a)) and in personam claims (section 10(4)(b)) against cargo. It further submits that that reading is inconsistent with the true meaning of article 3(3) of the Brussels Convention which is intended to be implemented by section 10(4).

On behalf of Argentum it is submitted that when section 10 is read in conjunction with section 17(1) and section 3(3) of the SIA the words “commercial purposes” in section 10 mean for the “purposes of such transactions or activities as are mentioned in section 3(3)”. The transactions in section 3(3), so far as relevant include “any contract for the supply of goods or services” and “any other transaction … into which a State enters or in which it engages otherwise than in the exercise of sovereign authority”. The activities mentioned in section 3(3) are any activity “into which a State enters or in which it engages otherwise than in the exercise of sovereign authority”. As a result, it is submitted, section 10(4)(a) expressly contemplates that a cargo in transit can be “in use” for the purposes of a contract for the supply of goods or services, or a transaction or activity into which the state enters or in which it engages otherwise than in the exercise of sovereign authority. Argentum submits that it is difficult to conceive what could otherwise have been intended. It submits that the transactions or activities which determine the use to which the Government was putting the Silver in 1942 are the international commercial contract of sale on fob terms, the international commercial contract of carriage on board a merchant ship and the bills of lading issued thereunder.

At the outset Argentum’s submission encounters the difficulty that to say that the Silver was “in use” by the Government while it was being carried on the Vessel simply does not accord with the ordinary and natural meaning of those words. As Elisabeth Laing LJ explained in her dissenting judgment in the Court of Appeal (at para 142), as a matter of ordinary language a cargo which was sitting in the hold of a ship was not being used for any purpose, commercial or otherwise. While it was undoubtedly the subject of commercial arrangements for its carriage, it would be a distortion of language to say that it was being used for the purposes of those arrangements. As Elisabeth Laing LJ succinctly put it: “It was being carried, and that is all.” In this regard, we note the observation in 1991 of the International Law Commission in its commentary to article 16(5) of the Draft Articles on Jurisdictional Immunities of States and their Property that “the word [sic] ‘intended for use’ has been retained because the cargo is not normally used while it is on board the ship and it is therefore its planned use which will determine whether the State concerned is or is not entitled to invoke immunity.” (The draft article became article 16(4) of the UNCSI. The effect of article 16(4) is that the exception to immunity created by article 16(3) does not apply to “any cargo owned by a state and used or intended for use exclusively for government non-commercial purposes”.)

On behalf of Argentum it is asked what meaning is to be given to the expression “in use” in this context if a cargo which is being carried is not being used for the purposes of the sale and carriage arrangements relating to it. This question prompted some discussion at the oral hearing of this appeal. It was suggested that a cargo of gas or coal might be in use while being carried if it were used, with permission, to fuel the vessel on which it was carried. Whether or not it should be regarded in such circumstances as cargo, let alone cargo in use by the owner, this example seems a little far-fetched. Perhaps a more realistic instance might be the use of a cargo by its owner as stock in trade while it was in transit, for example by selling goods afloat. No such use was made of the Silver in this case. It does seem that examples of a cargo being used for the purposes of its owner while it is being carried will be rare. However, it is not necessary to identify a situation in which a cargo may be considered to be “in use” by its owner. The term “in use or intended for use for commercial purposes” is a convenient composite expression applied in section 10(4)(a) to both the vessel and its cargo. In reality, it is likely that “use” will generally apply to the vessel and “intended for use” will generally apply to the cargo. As Elisabeth Laing LJ explained (at para 139(iv)), a court considering the application of section 10(4)(a) is not required to conclude that a cargo must have an actual use or an intended use. It is simply required to ask and to answer those questions in that order.

The words “in use or intended for use for commercial purposes” also appear in section 13(4) of the SIA. Section 13(2)(b) provides that, subject to certain exceptions, the property of a State shall not be subject to any process for the enforcement of a judgment or arbitration award or, in an action in rem, for its arrest, detention or sale. An exception is established by section 13(4): the rule in section 13(2)(b) does not prevent the issue of any process in respect of property which is for the time being in use or intended for use for commercial purposes. There are important differences between section 10(4)(a) and section 13(4). First, as explained above (at paras 36, 47–56), section 10(4) is a hybrid provision (in that, because of the nature of Admiralty proceedings in rem, it is concerned with both adjudicative and enforcement jurisdiction), while section 13(4) is concerned only with enforcement jurisdiction. Secondly, section 10(4)(a) focuses on use or intended use “at the time when the cause of action arose” whereas section 13(4) focuses on use or intended use “for the time being”. However, both section 10(1)(a) and sections 13(2)(b) and (4) expressly apply to actions in rem. In these circumstances it appears that the words “in use or intended for use for commercial purposes” were intended to bear the same meaning in both sections and that decided cases on section 13(4) may cast light on the meaning of the same words in section 10(4)(a). (See, generally, Bennion, Bailey and Norbury on Statutory Interpretation, 8th ed (2020), para 21.3; R (Good Law Project) v Electoral Commission [2018] EWHC 2414; [2019] 1 All ER 36, para 33 per Leggatt LJ.)

In Alcom v Republic of Colombia (“Alcom”) [1984] AC 508 the issue of law was whether the English courts had jurisdiction in garnishee proceedings (now third party debt proceedings) to order the attachment of the whole or part of the balance standing to the credit of a foreign state in a current account maintained at a London branch of a commercial bank by the Colombian diplomatic mission, upon which it drew for the purpose of meeting the expenditure incurred in the day-to-day running of the mission. Lord Diplock, with whose speech the other members of the Appellate Committee agreed, considered (at p 602E-F) that for the purposes of execution in garnishee proceedings the customer’s right to withdraw his credit balance was a single not a composite chose in action. The crucial question of construction was whether a debt with these legal characteristics was “property which is for the time being in use or intended for use for commercial purposes” within section 13(4) of the SIA. He continued (at pp 602F – 603A):

“To speak of a debt as ‘being used or intended for use’ for any purposes by the creditor to whom the debt is owed involves employing ordinary English words in what is not their natural sense, even if the phrase ‘commercial purposes’ is given the ordinary meaning of jure gestionis in contrast to jure imperii that is generally attributed to it in the context of rights to sovereign immunity in public international law; though it might be permissible to apply the phrase intelligibly to the credit balance in a bank account that was earmarked by the state for exclusive use for transactions into which it entered jure gestionis. What is clear beyond all question is that if the expression ‘commercial purposes’ in section 13(4) bore what would be its ordinary and natural meaning in the context in which it there appears, a debt representing the balance standing to the credit of a diplomatic mission in a current bank account used for meeting the day-to-day expenses of running the mission would fall outside the subsection.”

He then went on to consider the extended meaning given to “commercial purposes” by section 17(1) of the SIA which imports the comprehensive definition of “commercial transaction” in section 3(3). He concluded (at p 604 D-E):

“Unless it can be shown by the judgment creditor who is seeking to attach the credit balance by garnishee proceedings that the bank account was earmarked by the foreign state solely (save for de minimis exceptions) for being drawn upon to settle liabilities incurred in commercial transactions, as for example by issuing documentary credits in payment of the price of goods sold to the state, it cannot, in my view, be sensibly brought within the crucial words of the exception for which section 13(4) provides.”

In the light of the reasoning of the majority of the Court of Appeal in the present case, it is significant that what was relevant in Alcom was the purpose for which the property was used or intended to be used and not the nature of the relationship or activity which gave rise to it.

In SerVaas Inc v Rafidain Bank [2012] UKSC 40; [2013] 1 AC 595 (“SerVaas”) the claimant, which had obtained a judgment against the Iraqi Ministry of Industry for the price for supply of equipment, machinery and related services for a state-owned factory in Iraq, sought a third party debt order against a bank, which conducted business as a commercial bank, in respect of a debt payable by the bank to the Republic of Iraq by way of dividend under a scheme of arrangement. The debt consisted of commercial debts previously acquired by Iraq by way of assignment from creditors of the bank (“the admitted claims”). The claimant accepted that Iraq intended to use the admitted claims for sovereign purposes, namely payment to the UN Development Fund for Iraq, but contended that, since the nature of the transaction which gave rise to the bank’s liability was entirely commercial, the admitted claims were “in use … for commercial purposes” within section 13(4) of the SIA and were therefore not immune from execution under section 13(2)(b). Lord Clarke of Stone-cum-Ebony, with whom the other members of the Supreme Court agreed, held that the nature of the origin of the debts was not relevant to the question whether the property was in use for commercial purposes (para 15). He observed (para 16):

“As to the language of section 13(4), I would accept Mr Howard QC’s submission on behalf of Iraq that the expression “in use for commercial purposes” should be given its ordinary and natural meaning having regard to its context. I would further accept his submission that it would not be an ordinary use of language to say that a debt arising from a transaction is “in use” for that transaction. Parliament did not intend a retrospective analysis of all the circumstances which gave rise to property, but an assessment of the use to which the state had chosen to put the property.”

He then contrasted the language of section 13(4) with that of section 3(1) (proceedings “relating to” a commercial transaction) and section 10 (claims “in connection with” a ship) and continued (para 17):

“In enacting section 13(4), Parliament could have referred to property that “related to” a commercial transaction, or arose “in connection with” a commercial transaction as being susceptible to enforcement. It chose not to do so, which suggests that it intended a difference in meaning. Property will only be subject to enforcement where it can be established that it is currently “in use or intended for use” for a commercial transaction. It is not sufficient that the property “relates to” or is “connected with” a commercial transaction. I would accept Mr Howard’s submission that this is consistent with the different treatment of the two categories of immunity in the Act.”

Referring to Alcom, Lord Clarke observed (para 19):

“…[T]he critical point for present purposes is the proposition that the judgment creditor must show that the bank account was earmarked by the state solely for being drawn down upon to settle liabilities incurred in commercial transactions. The essential distinction is between the origin of the funds on the one hand and the use of them on the other. As Stanley Burnton LJ said in the instant case at para 33, it was not suggested by Lord Diplock in the Alcom case … that if the moneys in the bank account resulted from commercial transactions, that might be relevant to the question whether the account was used or intended for use for commercial purposes.”

Lord Clarke in SerVaas (paras 20 and 21) also found support for this approach in three first instance decisions (AIC Ltd v Federal Government of Nigeria [2003] EWHC 1357 (QB) (Stanley Burnton J); AIG Capital Partners Inc v Republic of Kazakhstan [2005] EWHC 2239 (Comm); [2006] 1 WLR 1420 (Aikens J); Orascom Telecom Holding SAE v Republic of Chad [2008] EWHC 1841 (Comm); [2009] 1 All ER (Comm) 315 (Burton J). See also LR Avionics Technologies Ltd v Federal Republic of Nigeria [2016] EWHC 1761 (Comm); [2016] 4 WLR 120 (Males J) at para 40.) At paras 23–28, Lord Clarke also noted that US cases on the immunity of states from execution under the Foreign Sovereign Immunities Act 1976 (28 USC, Chapter 97, sections 1602-1611) draw the same distinction between the source of the property and its use when applying the words “property … used for a commercial activity” in section 1610(a) (Connecticut Bank of Commerce v Republic of Congo (2002) 309 F 3d 240 (US Court of Appeals, 5^th Circuit); Af-Cap Inc v Republic of Congo (2007) 475 F 3d 1080, 1087 (US Court of Appeals, 9^th Circuit); Ministry of Defense and Support for the Armed Forces of the Islamic Republic of Iran v Cubic Defense Systems Inc (2007) 495 F 3d 1024 (US Court of Appeals, 9^th Circuit); EM Ltd v Republic of Argentina (2007) 473 F 3d 463, 468 (US Court of Appeals, 2^nd Circuit). See also the decision of the majority in the Court of Appeal in Hong Kong in FG Hemisphere Associates LLC v Democratic Republic of Congo [2010] HKCA 19.) In Connecticut Bank of Commerce v Republic of Congo, Judge Garza, delivering the majority opinion in the 5^th Circuit Court of Appeals observed (at p 251):

“What matters under the statute is not how the Congo made its money, but how it spends it. The amenability of these royalties and taxes to garnishment depends on what they are ‘used for’, not on how they were raised.”

Contrary to the submission of Argentum, the observations of Lord Clarke in SerVaas apply with equal force to the meaning of “in use or intended for use for commercial purposes” in section 10(4)(a). While it is correct that immunity from both adjudicative and enforcement jurisdiction are in play in section 10(4)(a), it is inconceivable that the words “in use or intended for use for commercial purposes” should bear a different meaning in the context of adjudicative jurisdiction. In section 10(4)(a) it is the use or intended use to which the state has decided to put the property concerned and not the transactions or activities from which the property originated which determine whether there is immunity.

In the present proceedings, the majority in the Court of Appeal were led into error by their concentration on the cause of action in salvage which they permitted to dominate their interpretation of section 10(4)(a). This led Popplewell LJ to take an unduly technical approach to the meaning of “cargo” in section 10(4)(a), focussing on the contractual means by which the Silver became cargo and then asking whether those transactions were commercial. However, as Elisabeth Laing LJ observed (at para 138), whether something is a cargo is primarily a question of fact. The history of how it was acquired or became cargo is not relevant to its use. It also led Popplewell LJ to misidentify the context under section 10(4). As the Government points out, the context is not whether property qualifies as cargo but whether something which is obviously cargo because it is being carried on a ship should be immune from an action in rem. Section 10(4)(a) is not directed at the circumstances which render the Silver cargo, but at whether the cargo is in use or intended for use for commercial purposes.

The reading of “in use” proposed by Argentum is open to the further objection that it would distort the statutory scheme. Section 10(4)(a) and section 10(4)(b) address different situations. Paragraph (a) is concerned with an action in rem against a State-owned cargo; paragraph (b) is concerned with an action in personam for enforcing a claim in connection with a such a cargo. The circumstances in which immunity is denied are different under these respective heads. If a party wishes to establish that a state is not immune from an in personam claim under section 10(4)(b), it will be enough to establish that the ship carrying the cargo was in use or intended for use for commercial purposes. If a party wishes to establish that a state is not immune from an in rem claim under section 10(4)(a), it will need to establish that both the ship and the cargo were either in use or intended for use for commercial purposes. If it were right to say that when a cargo is carried on a commercial vessel it is in use for commercial purposes, the additional test in paragraph (a) would be satisfied by every such cargo. As a result, the test under (a) would be the same as under (b) which was clearly not the legislative intention. The distinction between the two limbs of section 10(4) would be negated and the additional threshold criteria in paragraph (a) made redundant. In the result, section 10(4)(a) could never apply to a state-owned cargo carried pursuant to a commercial contract and Parliament’s intention to make separate provision for actions in rem and actions in personam in respect of cargo would be defeated.

There are compelling reasons why more stringent criteria should be satisfied before immunity is denied in the case of actions in rem.

First, the mere issue of a claim in rem gives the claimant the status of a secured creditor and encumbers the property with that claim. The Court of Appeal so decided in In re Aro Co Ltd [1980] Ch 196. That case involved a statutory in rem claim for damages for short delivery of cargo. The claimant had issued a claim in rem against the vessel ARO but had not served the writ or arrested the vessel. The vessel had previously been arrested by another creditor and the claimant entered a caveat against that arrest, which meant that notice to it had to be given if the arrest was to be lifted. Subsequently a petition for the winding up of the ship-owning company was presented and the critical question was whether the issue of the in rem claim meant that the claimant was to be regarded as a secured creditor at the time of the commencement of the winding up.

In giving the judgment of the court Brightman LJ (sitting with Stephenson and Brandon LJJ) stated at p 206A-B:

“Following on the issue of the writ, the plaintiffs had an immediate right, without having served the writ, to have the ship arrested by the Admiralty Marshal. Nothing would have been required of the plaintiffs, as no caveat against arrest had been entered, except an affidavit verifying the basic facts essential to the existence of their statutory right of action in rem. The justification for this procedure is that ships are owned and trade internationally, and unless a claimant can gain immediate security for a claim he may never have the opportunity effectively to pursue it.”

He identified the critical question as being whether, immediately before the presentation of the winding up petition, the claimant could properly assert as against all the world that the vessel ARO was security for its claim (p209C). He concluded that it could, reasoning as follows (p 209C-D):

“If it is correct to say, as was not challenged in the court below and is not challenged in this court, that after the issue of the writ in rem the plaintiffs could serve the writ on the Aro, and arrest the Aro, in the hands of a transferee from the liquidator and all subsequent transferees, it seems to us difficult to argue that the Aro was not effectively encumbered with the plaintiffs' claim. In our judgment the plaintiffs ought to be considered as secured creditors…”

Secondly, in rem claims against cargo are core in rem claims and the property is thereby made the subject matter of the claim. Claims for salvage give rise to a maritime lien over the cargo and proceedings in rem are the means by which that maritime lien is carried into effect. It is the process which perfects that inchoate right and gives effect to the consequent “subtraction from the absolute property of the owner in the thing” (see paras 47–49 above).

Thirdly, jurisdiction is established by the presence of the property within the jurisdiction. There is no need to meet the requirements for service out of the jurisdiction such as establishing that the cause of action falls within a jurisdictional gateway and that the domestic court is clearly the appropriate forum. The state which owns the property may have no connection or dealings with the jurisdiction and the presence of the property may be happenstance, as, for example, if cargo is within a jurisdiction because of an intermediate port of call during the course of a vessel’s voyage.

Fourthly, the establishment of jurisdiction by issue and service of the in rem claim would otherwise put the state in the difficult position of facing a choice between appearing and defending the claim or staying away and losing its property. As explained by Lord Wright in The Cristina at p 505:

“A judgment in rem is a judgment against all the world, and if given in favour of the plaintiffs would conclusively oust the defendants from the possession which on the facts I have stated they beyond question de facto enjoy. The writ by its express terms commands the defendants to appear or let judgment go by default. They are given the clear alternative of either submitting to the jurisdiction or losing possession. In the words of Brett LJ the independent sovereign is thus called upon to sacrifice either its property or its independence. It is, I think, clear that no such writ can be upheld against the sovereign State unless it consents. It is therefore given the right, if it desires neither to appear nor to submit to judgment, to appear under protest and apply to set aside the writ or take other appropriate procedure with the same object. It may be said that it is indirectly impleaded, but I incline to think that it is more correct to say that it is directly impleaded.”

Fifthly, issuing a claim in rem gives rise to a right to arrest the property. The threat of arrest will often pressurise a defendant into giving security for the claim made.

In all these ways proceedings in rem are far more intrusive into the rights of a state over its property than proceedings in personam. This is quite apart from whether the Admiralty jurisdiction enforcement powers of arrest, appraisal and sale are engaged.

In practice the issue of proceedings in rem will often be for the purpose of arresting the vessel and arrest will follow issue. The adjudicative and the enforcement aspects of the jurisdiction are intertwined.

Argentum stressed that section 10 of the SIA is concerned with immunity from adjudicative jurisdiction, whereas section 13 is concerned with immunity from enforcement jurisdiction and the two stages need to be kept distinct. However, the nature of a claim in rem makes it difficult to disengage its adjudicative and enforcement aspects. As Lord Diplock observed in Alcom at p 600F–G, the Admiralty jurisdiction in rem may be regarded for the purposes of the SIA as “hybrid”.

The correct reading of section 10(4)(a) becomes even clearer when we consider articles 1 to 3 of the Brussels Convention. As explained above, the SIA was enacted, among other purposes, in order to enable the United Kingdom to become a party to the Brussels Convention. The Act broadly implements the provisions of the Convention into domestic law within the United Kingdom. Reference has been made above (at para 39) to the complexity of the drafting of the Brussels Convention which resulted in Parliament enacting section 10 of the SIA, instead of merely giving effect to the Convention in domestic law. Fox and Webb suggest (at p 118) that the application of the Brussels Convention to ships conforming with its listed conditions may on occasion afford a wider immunity than does the SIA to government ships. This is, however, of no relevance to the present issue which concerns cargo. In these circumstances, the relevant provisions of the State Immunity Act are to be construed, so far as is possible, so as to conform to the Convention. (See Kabab-Ji SAL v Kout Food Group [2021] UKSC 48; [2021] Bus LR 1717 per Lord Hamblen and Lord Leggatt at para 31 and Crawford, Brownlie’s Principles of Public International Law, 9th ed (2019), at p 62).

The relevant provisions of the Brussels Convention have been considered at paras 41–42 above. Article 3(3), which is in point here, is concerned with state-owned cargoes carried on board merchant ships for Government and non-commercial purposes. It is convenient to set out article 3(3) once again.

“State-owned cargoes carried on board merchant ships for Government and non-commercial purposes shall not be subject to seizure, arrest or detention by any legal process nor to any proceedings in rem.

Nevertheless, claims in respect of collisions and nautical accidents, claims in respect of salvage or in the nature of salvage and in respect of general average, as well as claims in respect of contracts relating to such cargoes, may be brought before the Court which has jurisdiction in virtue of Article 2.”

On behalf of the Government, it is submitted that the effect of the first paragraph of article 3(3) is that if a cargo is being carried for governmental and non-commercial purposes it is immune from seizure, arrest or detention and from actions in rem. The effect of the second paragraph of article 3(3), the Government submits, is that claimants still have the right to proceed in personam in respect of broad categories of claims (collision and accident, salvage and general average and claims in respect of contracts relating to such cargoes) before a court which has jurisdiction under article 2. In other words, the second paragraph of article 3(3) left in place the immunity from in rem proceedings established by the first paragraph but provided that there should be no immunity for in personam claims. Before the Court of Appeal, the submission on behalf of Argentum was that the effect of the second paragraph of article 3(3) was a complete withdrawal of immunity in respect of claims of this kind. In our view, the reading supported by the Government is clearly to be preferred and the second paragraph of article 3(3) must be confined to in personam claims. This follows from the immediately preceding prohibition of proceedings in rem in the first paragraph of article 3(3) and from the very wide category of claims expressly permitted by the second paragraph to be brought before courts having jurisdiction under article 2. On the reading previously suggested by Argentum, the grant of immunity in respect of in rem claims by the first paragraph of article 3(3) would be deprived of any content.

Our conclusion as to the meaning of article 3(3) is supported by two commentators, Van Slooten and Gidel, both of whom were closely involved in the drafting of the Brussels Convention. They show that the original draft of article 3(3) was much narrower in scope and granted immunity only to state-owned cargoes that were carried on state-owned vessels. Van Slooten describes the development as follows:

“One cannot conceive, in fact, … how a State would accept that one could seize a cargo which it intended for a governmental purpose that was transported on a merchant ship. Such a situation is not purely hypothetical, the dangers it poses are not fanciful. Many states use this procedure for the transport of arms, supplies and munitions intended for the use of their armed forces, in their colonies and overseas possessions. The Conference ended up by adopting this point of view …” (G van Slooten, The Brussels Convention on the juridical status of State ships, Revue de droit international, Vol. VII (1926), pp 476–477) (unofficial translation).

Similarly Gidel writes:

“If government-owned cargo is transported on board a merchant ship, it only benefits from immunity from seizure or other legal proceedings in rem if it has a governmental and non-commercial purpose (article 3(3)).

The Goetenburg and Genoa drafts were much too severe with respect to state-owned cargoes: they did not exempt such cargoes from seizure under any circumstances when transported on board private ships. Immunity was only accorded to them if they were transported for a governmental and non-commercial aim on board ships owned or operated by States. The Brussels Convention has very opportunely reduced the rigour of these provisions.” (G Gidel, Le Droit International Public de la Mer, Le Temps De Paix, Tome II, Les Eaux Intérieures, Chapter 5, p 367) (unofficial translation).

The intention behind article 3(3) was clearly that a state should be immune from interference with its cargo which was transported on a merchant ship and intended for use for a governmental purpose.

At the oral hearing before the Supreme Court, Argentum proposed for the first time an alternative reading of article 3(3). While accepting that the first paragraph of article 3(3) applies to both the adjudicative and enforcement aspects of an action in rem, it submitted that the exception to the first paragraph created by the second paragraph applies, in respect of salvage, to permit a claim in rem or in personam but only in respect of its adjudicative aspect as opposed to its enforcement aspect. In our view, this reading is untenable. First, there is no warrant in article 3 for such a distinction between the adjudicative and enforcement aspects of an action in rem in respect of salvage. Secondly, as already explained, it is not possible to separate the adjudicative and enforcement aspects of an action in rem in the manner suggested. Thirdly, it is clear that article 3(3) intended to distinguish between seizure, arrest, detention and actions in rem, on the one hand (which are all treated in the same way), and actions in personam on the other. The intention behind the second paragraph of article 3(3) was to acknowledge that, notwithstanding the prohibition of the seizure, arrest or detention by any legal process or any proceedings in rem, in respect of state-owned cargoes carried on board merchant ships for Government and non-commercial purposes, the specified actions in personam were permitted. Fourthly, the word “nevertheless” which introduces the second paragraph of article 3(3) indicates not an exception to the first paragraph but an acknowledgement that other forms of action are permitted. We note that the word is used in the same sense in the second paragraph of article 3(1).

We conclude therefore that the second paragraph of article 3(3) leaves in place the immunity from in rem proceedings established by the first paragraph, but provides that there should be no immunity for in personam claims. This accords with the understanding of the drafter of section 10(4) of the SIA which draws a clear distinction between actions in rem in section 10(4)(a) and actions in personam in section 10(4)(b). We consider that section 10(4) accurately gives effect to article 3(3) of the Convention. It follows that Argentum’s proposed reading of section 10(4) is inconsistent with article 3(3) of the Brussels Convention.

For these reasons, the Government is immune as respects an action in rem in respect of the Silver carried on board the Vessel, because the cargo was, at the time when the cause of action arose, intended for use for non-commercial purposes.

The majority of the Court of Appeal rejected a submission on behalf of the Government that, because the Silver was not in use at all when carried on board a ship, the focus of the Court’s inquiry should be on the Government’s intended use for the Silver. Popplewell LJ observed (at para 95):

“Under customary international law, to which section 10(4) is intended to give effect, the sovereign/commercial test applies to the nature of the activity being undertaken, not its purpose, as the citations from the speech of Lord Wilberforce in I Congreso … make clear. This is mirrored in the language of the 1978 Act which identifies what is sovereign and what commercial in section 3 by reference to “transactions” and “activities”. Use of cargo, rather than its intended use, must therefore be the essential focus of the provision for state-owned cargoes, because it addresses the nature of the activity, not its purpose.”

Returning to the nature of a claim for salvage, he concluded (at para 97):

“Moreover, the specific context of section 10(4)(a) is salvage, including salvage of wreck. The intended use of the cargo on completion of the voyage is legally and logically irrelevant to such a claim, which arises before completion of the voyage. The maritime cause of action in salvage depends upon the property being cargo when on board the vessel, which demands an inquiry as to use at that time, not intended use thereafter.”

Similarly, Andrews LJ considered (at para 123) that “the intended use of the cargo by the state after the voyage is complete is an irrelevant consideration, even if that intention was formed before the cargo was loaded”. She could see “no principled justification for the question of immunity turning on the state’s intentions… rather than on its actions”.

This approach is erroneous in a number of respects. First, for the reasons given in relation to Issue 2, above, it is not appropriate to allow the nature of the cause of action in salvage to influence the reading of the SIA in this way. Actions in rem against cargo within section 10(4)(a) are not limited to actions in salvage. Section 10(4) was intended to permit ratification of the Brussels Convention which simply refers in article 3(1) and 3(3) to “proceedings in rem”. In domestic law an action in rem for condemnation of the goods or for droits of Admiralty may also be brought against cargo. (See sections 20(2)(s) and 21(2) of the Senior Courts Act 1981).

Secondly, while the SIA was, no doubt, intended to be consistent with rules of customary international law on state immunity in that it granted immunity where required by customary international law, the SIA did not seek to replicate precisely the distinction between acta jure imperii and acta jure gestionis which was at the time of its enactment emerging in customary international law and coming to be accepted at common law in England and Wales. Indeed, in 1978 the limits of the restrictive theory of state immunity were uncertain and the difficulty of encapsulating that distinction with precision in the developing common law was one of the reasons for the enactment of the SIA. Furthermore, as we have seen, the inclusion of jurisdictional connecting factors in some of the statutory heads of non-immunity from adjudicative jurisdiction led to the result that immunity was accorded under the SIA in a number of areas in respect of non-sovereign activities where immunity was not required by customary international law. (See the discussion at paras 26 and 29–32 above.) As a result, caution is required when seeking to apply to the interpretation of the SIA the reasoning of Lord Wilberforce in his speech in I Congreso, which was decided on the basis of the common law. Moreover, in so far as Admiralty actions in rem are concerned, section 10(4) was not intended to give effect to customary international law but was intended to enable the United Kingdom to ratify the Brussels Convention.

Thirdly, the view expressed by the majority of the Court of Appeal as to the operation of the sovereign/non-sovereign distinction in customary international law is an over-generalisation. It is correct that where proceedings relate to the activities of a state, the state is in general entitled to immunity in customary international law only in respect of acts done in the exercise of sovereign authority (Benkharbouche per Lord Sumption at paras 8, 37). Thus, in the case of immunity from adjudicative jurisdiction the focus is generally on the juridical character of the activity and not the purpose for which it was carried out. This is reflected in the cases in this jurisdiction concerning our domestic common law immediately prior to the coming into force of the SIA. Accordingly, in Trendtex Trading Corpn v Bank of Nigeria Lord Denning MR explained (at p 558D–F) that if a state enters into a commercial agreement for the purchase of goods there is no immunity from suit, regardless of the purpose to which the state may intend to put the goods. Immunity depends on the nature of the transaction not the purpose behind it. In I Congreso Lord Wilberforce (at p 262A) affirmed the reasoning of Lord Denning in Trendtex “that if the act in question is of a commercial nature, the fact that it was done for governmental or political reasons does not attract sovereign immunity”. Having addressed some of the difficulties of distinguishing the sovereign or non-sovereign character of relevant acts, in particular whether a breach of a commercial agreement may be a sovereign act, Lord Wilberforce concluded (at p 267B-D, cited at para 22 above) that the court must consider the whole context of the claim in order to decide whether the relevant acts on which the claim is based should be considered as within an area of activity of a private law character or within the sphere of governmental activity.

To the extent that the SIA employs the distinction between sovereign and non-sovereign activities as a basis for delimiting the scope of immunity from adjudicative jurisdiction, it is, once again, the juridical nature of the activity undertaken by the state and not the purpose for which it was undertaken that is determinative.

This is not, however, a rule of universal application. In the case of proceedings concerning the property of a state different considerations apply. As Lord Sumption explained in Benkharbouche (at para 41) (while considering the decision of the US Supreme Court in The Schooner Exchange v McFaddon (1812) 11 US 116), in the context of the immunity of property of a foreign state the immunities recognised by international law have generally been wider than those available in actions for breach of duty. In particular, in determining the immunity of state property both the use and the intended use of the property by the state are highly material considerations. The exercise of jurisdiction by the courts of one state over the property of a foreign state which is in use or intended for use by that foreign state for sovereign purposes may well constitute an impermissible interference with sovereign functions of the foreign state.

In The Philippine Admiral two actions in rem were brought against a state-owned vessel, one in respect of goods and services supplied to the vessel and the other claiming damages for breach of charterparty. The Judicial Committee of the Privy Council, deciding the appeal from the Supreme Court of Hong Kong on common law principles, held that the restrictive theory of immunity applied. Its formulation of the test is highly significant for present purposes.

“The question then arises whether the Philippine Admiral can properly be regarded as a mere trading vessel or was at the relevant time for one reason or another a ship publicis usibus destinata. In order to answer that question one must consider both the past history of the vessel in question since she became the property of the foreign state and also the use to which she is likely to be put by that state in the future.” (at p 403, emphasis added)

As we have seen, the SIA similarly employs a test of use or intended use of property in order to determine whether a state’s property is immune from enforcement jurisdiction. (See Alcom v Republic of Colombia and SerVaas (considered above at paras 72 and 73–76 respectively.) Section 13(2)(b) provides that the property of a state is not subject to any process for the enforcement of a judgment or arbitration award or, in any action in rem, for its arrest, detention or sale. However, section 13(4) provides that this does not prevent the issue of any process in respect of property which is for the time being in use or intended for use for commercial purposes. Section 10(4) of the SIA, which is a hybrid provision in the sense that it relates to both adjudicative and enforcement jurisdiction, employs the same test.

In the present case, the claim is not founded on any commercial activity or alleged breach of duty on the part of the Government. The claim to salvage is not based on any contract of salvage but on the fact that the Government was the owner of the cargo which was salved. In these circumstances, as a matter of customary international law it is both permissible and necessary to have regard to the use and intended use of the cargo. Such an approach is also consistent with the Brussels Convention.

In these circumstances it is not necessary to rule on a further submission by Mr Wordsworth KC on behalf of the Government that, in customary international law, purpose may be a relevant consideration in delimiting sovereign and non-sovereign activities when it throws light on the juridical character of the activities. (See I Congreso per Lord Wilberforce at pp 271–272 (considering the Marble Islands); Holland v Lampen-Wolfe [2000] 1 WLR 1573 per Lord Millett at pp 1586H-1587D; Benkharbouche per Lord Sumption at para 8.)

Fourthly and in any event, it is necessary to apply the provisions of the SIA which are entirely unambiguous in this regard. Section 10(4)(a) is clear in requiring account to be taken of both the use and the intended use of the cargo when deciding whether there is an entitlement to immunity. In the present case the Silver was not in use for commercial purposes when it was simply being carried as cargo. However, it was not in dispute that the intended use of the Silver was the sovereign purpose of minting currency. As a result, section 10(4)(a) did not remove the general immunity conferred by section 1(1) of the SIA.

Argentum submits that if, applying ordinary rules of construction to section 10(4)(a) of the SIA, the Government would be entitled to immunity, such immunity would exceed that required in customary international law and would therefore be a disproportionate interference with the right of access to a court under article 6 ECHR. It submits that section 10(4)(a) must therefore be read down under section 3 of the Human Rights Act 1998, in accordance with the principles set out in Ghaidan v Godin-Mendoza [2004] UKHL 30; [2004] 2 AC 557 and Sheldrake v Director of Public Prosecutions [2004] UKHL 43; [2005] 1 AC 264, so as to deny immunity to the Government.

In a line of authority beginning with Al-Adsani v United Kingdom (2001) 34 EHRR 11 the European Court of Human Rights has taken the view that a grant of state immunity requires to be justified under article 6 because it is a restriction on access to a court. In Al-Adsani it stated (at para 56):

“… [M]easures taken by a High Contracting Party which reflect generally recognised rules of public international law on State immunity cannot in principle be regarded as imposing a disproportionate restriction on the right of access to court as embodied in Article 6(1). Just as the right of access to court is an inherent part of the fair trial guarantee in that Article, so some restrictions on access must likewise be regarded as inherent, an example being those limitations generally accepted by the community of nations as part of the doctrine of State immunity.”

(See, to similar effect, McElhinney v Ireland (2001) 34 EHRR 13, para 37; Fogarty v United Kingdom (2001) 34 EHRR 12, para 36; Cudak v Lithuania (2010) 51 EHRR 15 at para 57; Sabeh El Leil v France (2011) 54 EHRR 14, para 49; Jones v United Kingdom (2014) 59 EHRR 1, para 189.)

To the extent that immunity is required by customary international law, it might be thought that article 6 is not engaged because that provision is concerned with access to the jurisdiction enjoyed by states in accordance with international law. (See Holland v Lampen-Wolfe per Lord Hope of Craighead at pp 1577G-1578D, per Lord Millett at p 1588B–D; Matthews v Ministry of Defence [2003] UKHL 4; [2003] 1 AC 1163 per Lord Millett at paras 101-106; Jones v Ministry of the Interior of the Kingdom of Saudi Arabia [2006] UKHL 26; [2007] 1 AC 270 per Lord Bingham of Cornhill at para 14.) We see great force in the view expressed by Lord Bingham in Jones:

“Based on the old principle par in parem non habet imperium, the rule of international law is not that a state should not exercise over another state a jurisdiction which it has but that (save in cases recognised by international law) a state has no jurisdiction over another state.”

However, that is not the view taken by the Strasbourg court. In Benkharbouche the Supreme Court (at paras 30, 75) left the controversy unresolved as there was in that case no binding rule of international law denying jurisdiction. It is not necessary to address it in this appeal for the following reasons.

First, approaching the matter purely in terms of the Strasbourg jurisprudence, the right of access to a court secured by article 6 is not absolute but is subject to limitations permitted by implication and Contracting States enjoy a margin of appreciation. In enacting section 10(4)(a) of the SIA the United Kingdom Parliament had the legitimate aim of giving effect to the restrictive theory of immunity in international law so as to enable it to become a party to the Brussels Convention. The measure is also proportionate. The Strasbourg court explained in Cudak v Lithuania (at para 55) that the Court “must be satisfied that the limitations applied do not restrict or reduce the access left to the individual in such a way or to such an extent that the very essence of the right is impaired”. In the present case Argentum has two possible routes by which to pursue a claim for salvage: an action in rem or an action in personam. The immunity claimed by the Government under section 10(4)(a) applies only to proceedings in rem and it is common ground that it does not preclude proceedings in personam. Furthermore, it is relevant here that an action in rem of this sort is unusual and is not generally found outside common law jurisdictions.

Secondly, and more generally, in the circumstances of this case the grant of immunity from proceedings in rem on the basis of the intended use of the property for sovereign purposes conforms with and is required by general principles of international law. The hybrid nature of proceedings in rem and the intrusion they would represent into a foreign state’s rights over its property (see paras 79–87 above) make it both appropriate and necessary that where the property is intended for use for sovereign purposes the state should be entitled to invoke immunity. This is apparent from cases such as The Philippine Admiral which concerned actions in rem and which was decided on the basis of principles of customary international law given effect in the common law. It is also apparent from the close analogy with the law on immunity from enforcement jurisdiction both in customary international law and under the SIA. (See Alcom and Servaas considered above at paras 72-76.) Similarly it is reflected in the provisions of the Brussels Convention (article 3) and the UNCSI (article 16(3), (4)). Section 10(4)(a) of the SIA gives effect to the restrictive theory of state immunity. It is a lex specialis required in order to prevent unjustifiable interference with a foreign state’s public property.

Accordingly, whether one considers that article 6 of the ECHR is not engaged because immunity is required by international law or that compliance with international law is a justifiable interference with article 6 rights, the answer is the same.

For the reasons set out above, the Government is entitled under section 1 of the SIA to immunity from this in rem claim against the Silver. The exception to immunity under section 10(4)(a) does not apply because the Silver was not, at the time when the cause of action arose, in use or intended for use for commercial purposes. Furthermore, this result accords with article 25 of the Salvage Convention which is implemented into domestic law. The Silver was a non-commercial cargo owned by a state and entitled, at the time of salvage operations, to sovereign immunity under generally recognised principles of international law.

For these reasons we would allow the appeal.

Finally, we should record that on 3 May 2024 we were informed by the parties that a settlement had been arrived at in this matter on 26 April 2024. The parties have agreed that we should nevertheless hand down the judgment and we are satisfied that it is appropriate to do so.