Dr Saeed Shehabi & Anor v The Kingdom Of Bahrain

The issue on this appeal is whether a foreign state whose agents, located abroad, cause spyware to be installed remotely on the computers of individuals located in the United Kingdom, causing those individuals psychiatric injury when they discover that the state has been spying on them in this way, is entitled to immunity from civil proceedings. That depends on whether the proceedings from which the state claims to be immune are ‘in respect of … personal injury … caused by an act or omission in the United Kingdom’ within the meaning of section 5 of the State Immunity Act 1978.

Three issues have been argued on this appeal:

The judge, Mr Justice Julian Knowles, decided these issues in favour of the claimants. The defendant state, the Kingdom of Bahrain, appeals.

Another case, decided by the same judge on materially the same facts six months earlier, was Al-Masarir v Kingdom of Saudi Arabia [2022] EWHC 2199 (QB), [2023] QB 475. The arguments in that case overlapped with, but were not the same as, the arguments in the present case. In Al-Masarir Saudi Arabia argued that section 5 did not apply to acts done by a state in the United Kingdom in the exercise of sovereign or governmental authority, but only to acts of a private law nature. Mr Justice Julian Knowles rejected that argument. Saudi Arabia appealed, but the appeal was dismissed before it could be heard because Saudi Arabia failed to comply with an order for security for costs. Bahrain does not advance this argument in the present case. In the present case Bahrain contends that psychiatric injury does not amount to ‘personal injury’ within the meaning of section 5. That was not an argument advanced by Saudi Arabia in Al-Masarir.

The legal framework within which the issues arise was common ground. Section 1(1) of the State Immunity Act 1978 Act is headed ‘General immunity from jurisdiction’ and provides:

‘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.’

The effect of this provision is that, in order for a state to be subject to the jurisdiction of the courts of the United Kingdom, the proceedings must be of a kind specified in sections 2 to 11 of the Act. If none of those provisions apply, the court lacks jurisdiction.

Relevant for present purposes is section 5, headed ‘Personal injuries and damage to property’, which provides:

‘5. A State is not immune as respects proceedings in respect of–

(a)

death or personal injury; or

(b)

damage to or loss of tangible property,

caused by an act or omission in the United Kingdom.’

The burden of proving that the claim falls within section 5 as one of the exceptions to the general immunity provided by section 1 lies on the claimants. This must be established on the balance of probabilities as a preliminary issue: JH Rayner (Mincing Lane) Ltd v Department of Trade and Industry [1989] Ch 72, 193-194 (Lord Justice Kerr) and 252 (Lord Justice Ralph Gibson). It was common ground between the parties that the act or omission referred to in section 5 must be an act or omission of the state or of those for whom the state is responsible.

The claimants, who describe themselves as pro-democracy activists, are prominent members of the Bahraini opposition movement. The first claimant, Dr Saeed Shehabi, is a journalist and activist who is also the founder of a Bahraini political party called Al Wefaq. He has lived in the United Kingdom since 1973. He was granted asylum in 1985 and British citizenship on 14^th June 2002.

The second claimant, Mr Moosa Mohammed, is a Bahraini citizen, but has lived in the United Kingdom since 2006. He was granted refugee status on 7^th August 2007 and has been granted indefinite leave to remain here.

The claimants’ case is that from around September 2011 Bahrain’s servants or agents, likely operating remotely from outside the United Kingdom, hacked – or infected – their computers with a spyware program called ‘FinSpy’ while they and their computers were in the United Kingdom; that this amounted to harassment under the Protection from Harassment Act 1997; that they suffered psychiatric injury as a consequence when they learned that their computers had been hacked in this way; that this amounts to personal injury within section 5 of the State Immunity Act 1978; and therefore that Bahrain is not immune.

Spyware is a type of computer program which allows a remote operator to take control of a target’s device (e.g. their computer or mobile phone) and then to use that device to carry out surreptitious remote eavesdropping and surveillance of the target by the collection and transmission to a remotely located server of video, audio and data. The spyware program is usually deposited on the target’s device by the target unwittingly opening an infected email or attachment that has been sent by the remote operator. FinSpy is one such program.

The claimants’ case is that the operation of that spyware resulted in the covert and unauthorised accessing by Bahrain of information stored on, or communicated or accessible via, the claimants’ laptops. This has enabled Bahrain to collect much, if not all, of the data processed on the laptops, including messages, emails, calendar records, instant messaging, contacts lists, browsing history, photos, databases, documents and videos. It has also permitted Bahrain to track the location of the claimants via their laptops, to intercept calls made on them, and to eavesdrop on the claimants by covert use of the laptops’ microphones and cameras.

Bahrain denies the claimants’ allegations, but the judge found, on the basis of expert evidence, that the claimants had discharged the burden upon them of proving on the balance of probabilities that their computers were infected by spyware by Bahrain’s servants or agents. It will be open to Bahrain to challenge that conclusion at trial, but it has not been challenged on this appeal.

It appears that the claimants learned of the hacking of their computers in or around August 2014 when WikiLeaks published on its website documents concerning Bahrain’s use of FinSpy and an organisation called Bahrain Watch identified the claimants as targets of such hacking.

The claimants frame the case in the tort of harassment. The ingredients of that tort were described by Mr Justice Nicklin in Hayden v Dickenson [2020] EWHC 3291 (QB) as follows:

‘40. s.1 Protection from Harassment Act 1997 (“PfHA”) provides, so far as material:

“(1)

A person must not pursue a course of conduct - (a) which amounts to harassment of another, and (b) which he knows or ought to know amounts to harassment of the other.

(1A) [omitted]

(2)

For the purposes of this section …, the person whose course of conduct is in question ought to know that it amounts to … harassment of another if a reasonable person in possession of the same information would think the course of conduct amounted to harassment of the other.

(3)

Subsection (1) does not apply to a course of conduct if the person who pursued it shows –

(a)

that it was pursued for the purpose of preventing or detecting crime,

(b)

that it was pursued under any enactment or rule of law or to comply with any condition or requirement imposed by any person under any enactment, or

(c)

that in the particular circumstances the pursuit of the course of conduct was reasonable.”

41.

s.3 provides that any actual or apprehended breach of s.1(1) may be the subject of a civil claim by anyone who is or may be the victim of the course of conduct. Remedies in a civil claim include interim and final injunctions and damages for “any anxiety caused by the harassment and any financial loss resulting from the harassment”: s.3(2).

42.

s.7(2) provides: “References to harassing a person include alarming the person or causing the person distress”; and in subsection (3) (b): “A ‘course of conduct’ must involve, in the case of conduct in relation to a single person (see section 1(1)), conduct on at least two occasions in relation to that person.” Conduct can include speech (s.7(4)).

43.

A defendant has a defence if s/he shows: (i) that the course of conduct was pursued for the purpose of preventing or detecting crime; and/or (ii) that in the particular circumstances the pursuit of the course of conduct was reasonable (s.1(3)).

44.

The principal cases on what amounts to harassment are: Thomas v News Group Newspapers [2002] EMLR 4; Majrowski v Guy’s and St Thomas’s NHS Trust [2007] 1 AC 224; Ferguson v British Gas Trading Ltd [2009] EWCA Civ 46; Dowson v Chief Constable of Northumbria Police [2010] EWHC 2612 (QB); Trimingham v Associated Newspapers Ltd [2012] EWHC 1296 (QB); [2012] 4 All ER 717; Hayes v Willoughby [2013] 1 WLR 935; R v Smith [2013] 1 WLR 1399; Law Society v Kordowski [2014] EMLR 2; Merlin Entertainments LPC v Cave [2015] EMLR 3; Levi v Bates [2016] QB 91; Hourani v Thomson [2017] EWHC 432 (QB); Khan v Khan [2018] EWHC 241 (QB); Hilson v Crown Prosecution Service [2019] EWHC 1110 (Admin); and Sube v News Group Newspapers Ltd [2020] EMLR 25. From these cases, I extract the following principles

i)

Harassment is an ordinary English word with a well understood meaning: it is a persistent and deliberate course of unacceptable and oppressive conduct, targeted at another person, which is calculated to and does cause that person alarm, fear or distress; “a persistent and deliberate course of targeted oppression”: Hayes v Willoughby [1], [12] per Lord Sumption.

ii)

The behaviour said to amount to harassment must reach a level of seriousness passing beyond irritations, annoyances, even a measure of upset, that arise occasionally in everybody’s day-to-day dealings with other people. The conduct must cross the boundary between that which is unattractive, even unreasonable, and conduct which is oppressive and unacceptable. To cross the border from the regrettable to the objectionable, the gravity of the misconduct must be of an order which would sustain criminal liability under s.2: Majrowski [30] per Lord Nicholls; Dowson [142] per Simon J; Hourani [139]-[140] per Warby J; see also Conn -v- Sunderland City Council [2007] EWCA Civ 1492 [12] per Gage LJ. A course of conduct must be grave before the offence or tort of harassment is proved: Ferguson v British Gas Trading Ltd [17] per Jacob LJ.

iii)

The provision, in s.7(2) PfHA, that “references to harassing a person include alarming the person or causing the person distress” is not a definition of the tort and it is not exhaustive. It is merely guidance as to one element of it: Hourani [138] per Warby J. It does not follow that any course of conduct which causes alarm or distress therefore amounts to harassment; that would be illogical and produce perverse results: R v Smith [24] per Toulson LJ.

iv)

s.1(2) provides that the person whose course of conduct is in question ought to know that it involves harassment of another if a reasonable person in possession of the same information would think the course of conduct involved harassment. The test is wholly objective: Dowson [142]; Trimingham [267] per Tugendhat J; Sube [65(3)], [85], [87(3)]. “The Court’s assessment of the harmful tendency of the statements complained of must always be objective, and not swayed by the subjective feelings of the claimant”: Sube [68(2)].

v)

Those who are “targeted” by the alleged harassment can include others “who are foreseeably, and directly, harmed by the course of targeted conduct of which complaint is made, to the extent that they can properly be described as victims of it”: Levi v Bates [34] per Briggs LJ.

vi)

Where the complaint is of harassment by publication, the claim will usually engage Article 10 of the Convention and, as a result, the Court’s duties under ss.2, 3, 6 and 12 of the Human Rights Act 1998. The PfHA must be interpreted and applied compatibly with the right to freedom of expression. It would be a serious interference with this right if those wishing to express their own views could be silenced by, or threatened with, proceedings for harassment based on subjective claims by individuals that they felt offended or insulted: Trimingham [267]; Hourani [141]. …’

As appears from this summary, the facts of the present case are some way removed from a typical case of harassment. In the present case, presumably, if the hacking took place as alleged, Bahrain’s intention was that the claimants would never find out that their computers had been hacked, in which case they would never suffer alarm, fear or distress as a result, and Bahrain would continue to spy on their activities and communications while the claimants remained in ignorance. However, we are not concerned with such matters on this appeal. We must proceed on the basis that, if established, and subject to the issue of state immunity, the claimants’ allegations will entitle them to a remedy in the tort of harassment.

We are concerned with the scope of section 5 of the State Immunity Act 1978. The circumstances in which the common law, following the development of international law, moved from a near absolute principle of state immunity to a restrictive theory, distinguishing between ‘acta jure imperii’ and ‘acta jure gestionis’, are well known. The story is traced by the Supreme Court in Argentum Exploration Ltd v Republic of South Africa [2024] UKSC 16, [2024] 2 WLR 1259 at [17] to [22]. This was the background to the 1978 Act.

However, the Act did not attempt simply to enact the restrictive theory of state immunity as it had so far developed, but provided what the Supreme Court in Argentum at [25] described as ‘a new statutory scheme providing detailed and comprehensive rules governing both adjudicative and enforcement jurisdiction in cases involving foreign and Commonwealth states’. Its long title 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.’

That statutory scheme must be interpreted in accordance with the usual principles of statutory interpretation. These have been authoritatively explained in the judgment of Lord Hodge in R (O) v SSHD [2022] UKSC 3, [2023] AC 255:

‘28. Having regard to the way in which both parties presented their cases, it is opportune to say something about the process of statutory interpretation.

29.

The courts in conducting statutory interpretation are “seeking the meaning of the words which Parliament used”: Black-Clawson International Ltd v Papierwerke Waldhof-Aschaffenburg AG [1975] AC 591, 613 per Lord Reid. More recently, Lord Nicholls of Birkenhead stated: “Statutory interpretation is an exercise which requires the court to identify the meaning borne by the words in question in the particular context.” (R v Secretary of State for the Environment, Transport and the Regions, Ex p Spath Holme Ltd [2001] AC 349, 396). Words and passages in a statute derive their meaning from their context. A phrase or passage must be read in the context of the section as a whole and in the wider context of a relevant group of sections. Other provisions in a statute and the statute as a whole may provide the relevant context. They are the words which Parliament has chosen to enact as an expression of the purpose of the legislation and are therefore the primary source by which meaning is ascertained. There is an important constitutional reason for having regard primarily to the statutory context as Lord Nicholls explained in Spath Holme, 397: “Citizens, with the assistance of their advisers, are intended to be able to understand parliamentary enactments, so that they can regulate their conduct accordingly. They should be able to rely upon what they read in an Act of Parliament.”

30.

External aids to interpretation therefore must play a secondary role. Explanatory notes, prepared under the authority of Parliament, may cast light on the meaning of particular statutory provisions. Other sources, such as Law Commission reports, reports of Royal Commissions and advisory committees, and Government White Papers may disclose the background to a statute and assist the court to identify not only the mischief which it addresses but also the purpose of the legislation, thereby assisting a purposive interpretation of a particular statutory provision. The context disclosed by such materials is relevant to assist the court to ascertain the meaning of the statute, whether or not there is ambiguity and uncertainty, and indeed may reveal ambiguity or uncertainty: Bennion, Bailey and Norbury on Statutory Interpretation, 8th ed (2020), para 11.2. But none of these external aids displace the meanings conveyed by the words of a statute that, after consideration of that context, are clear and unambiguous and which do not produce absurdity. In this appeal the parties did not refer the court to external aids, other than explanatory statements in statutory instruments, and statements in Parliament which I discuss below. Sir James Eadie QC for the Secretary of State submitted that the statutory scheme contained in the 1981 Act and the 2014 Act should be read as a whole.

31.

Statutory interpretation involves an objective assessment of the meaning which a reasonable legislature as a body would be seeking to convey in using the statutory words which are being considered. Lord Nicholls, again in Spath Holme, 396, in an important passage stated:

“The task of the court is often said to be to ascertain the intention of Parliament expressed in the language under consideration. This is correct and may be helpful, so long as it is remembered that the ‘intention of Parliament’ is an objective concept, not subjective. The phrase is a shorthand reference to the intention which the court reasonably imputes to Parliament in respect of the language used. It is not the subjective intention of the minister or other persons who promoted the legislation. Nor is it the subjective intention of the draftsman, or of individual members or even of a majority of individual members of either House. … Thus, when courts say that such-and-such a meaning ‘cannot be what Parliament intended’, they are saying only that the words under consideration cannot reasonably be taken as used by Parliament with that meaning”.’

I would add that section 5 of the 1978 Act is expressed in plain and straightforward language. That language is the primary source by which its meaning must be ascertained.

Further, as explained by the Supreme Court in General Dynamics United Kingdom Ltd v State of Libya [2021] UKSC 22, [2022] AC 318, at [59], the 1978 Act must be understood in the context of the twin (and equally important) principles of international law on which the law of state immunity is based, summarised in these terms by the International Court of Justice in Jurisdictional Immunities of the State (Germany v Italy) [2012] ICJ Rep 99:

‘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 and the jurisdiction which flows from it. Immunity may represent a departure from the principle of territorial sovereignty and the jurisdiction which flows from it.’

Beyond this, however, it would be regrettable if the true meaning of section 5 could only be understood by reference to the substantial volume of external material, extending over more than 2,600 pages, cited in the course of this appeal. Indeed, it is notable that in Al-Adsani v Government of Kuwait (1996) ILR 536, Lord Justice Ward said of section 5 of the 1978 Act that ‘the Act is as plain as plain can be’, an observation endorsed by Lord Bingham and Lord Hoffmann in Jones v Ministry of the Interior of the Kingdom of Saudi Arabia [2006] UKHL 26, [2007] 1 AC 270 at [13] and [38].

Finally as to the general approach to the interpretation of section 5, section 1 is sometimes described as containing the general rule, to which the provisions of sections 2 to 11 are exceptions. But that does not mean that they should be interpreted restrictively, in the way that (for example) a contractual exceptions clause would be interpreted. As Lord Sumption explained in Benkarbouche v Embassy of the Republic of Sudan [2017] UKSC 62, [2019] AC 777:

‘39. I do not regard these considerations as decisive of the present issue. No one doubts that as a matter of domestic law, Part I of the State Immunity Act is a complete code. If the case does not fall within one of the exceptions to section 1, the state is immune. But the present question is whether the immunity thus conferred is wider than customary international law requires, and that raises different considerations. In the first place, it is necessary to read the grant of the immunity in article 5 of the United Nations Convention together with the exceptions which follow, as an organic whole. The exceptions are so fundamental in their character, so consistent in their objective and so broad in their effect as to amount in reality to a qualification of the principle of immunity itself rather than a mere collection of special exceptions. …’

It appears that Lord Sumption was speaking mainly about the United Nations Convention on Jurisdictional Immunities of States and their Property 2004 (‘the UN Convention’), but this court made clear in The Prestige (Nos. 3 & 4) [2021] EWCA Civ 1589 [2022] 1 WLR 3434 that the same approach applies to interpretation of the 1978 Act:

‘38. … we were not attracted by Mr Young’s argument that a restrictive interpretation should be put on the scope of the definition within subsection (3)(c) because the structure of the Act was one which provided immunity in section 1 and section 3(1)(a) was an exception, thereby giving rise to a strict interpretation of the exceptions if it was to be removed. We do not accept that the structure of the Act provides any basis for such a restrictive approach to construction of the exception sections, for the reasons articulated by Lord Sumption JSC in Benkarbouche at para 39.’

Bahrain’s first ground of appeal is that the judge was wrong to hold that the alleged acts of computer surveillance took place in the United Kingdom rather than abroad.

The claimants’ pleaded case is that Bahrain’s acts in the United Kingdom included the following:

‘(1) Transmitting executable files for installing FinSpy on the SS Computer and the MM Computer ("the Devices"), which were at all material times located in England.

(2)

Installing FinSpy on the Devices, including by overwriting the hard disk and/or Master Boot Record with malicious code.

(3)

Running the spyware on the Devices.

(4)

Executing FinSpy to the Devices' Central Processing Units, and reading data to, and writing it from, the Devices' Random Access Memory.

(5)

Storing information gathered by the spyware on the Devices' hard disks.

(6)

Using the Devices' computer network interface controller to send and receive data via a wired or wireless network and telecommunications equipment within the UK.

(7)

Using the Devices' battery power to transmit and receive data and commands, and to use other hardware components in the Devices.

(8)

Exfiltrating or causing to be exfiltrated information held on, available from and/or transmitted via the Devices.

(9)

Activating or causing to be activated the Devices' microphones and/or cameras, and recording information with the same.

(10)

Recording and transmitting keystrokes and mouse movements made on the Devices.’

The judge held that remote manipulation from abroad of a computer located in the United Kingdom is an act within the United Kingdom. His decision did not depend on the precise technical means by which this manipulation occurred, as detailed in the claimants’ pleading, but was more broadly based. He relied on cases on extradition (R v Governor of Brixton Prison, ex parte Levin [1997] QB 65) and jurisdiction (Ashton Investments Ltd v OJSC Russian Aluminium (Rusal) [2006] EWHC 2545 (Comm), [2007] 1 All ER (Comm) 857), concluding that:

‘143. I do not consider there is any meaningful distinction between the facts of Levin, where there was nefarious real time manipulation from Russia of a computer in the US, and this case of a spyware attack implanting software from abroad by trickery onto a device in the UK, which then, at regular intervals under the control of the spyware program, sends data back to a C&C server abroad. The technicalities may be different, but the principle is the same. In both cases a foreign entity has taken control of a computer located in the UK in order to obtain data.

144.

I therefore conclude that infecting a computer located in the UK with spyware from abroad is an act done in the UK for the purposes of s 5.’

Professor Dan Sarooshi KC for Bahrain submitted that when a person located abroad uses a computer to infect with spyware a computer in the United Kingdom, the act in question is to be regarded as having taken place abroad and not in the United Kingdom for the purposes of section 5 of the 1978 Act. The act of hacking takes place where the actor instigates it or conducts the manipulation, not where the target device is located. In the present case, the claimants’ evidence is that the infection of their computers had been carried out remotely by agents of the Bahraini government, probably located in Bahrain. It is not suggested that the acts in question were carried out by anyone present within the territory of the United Kingdom. The judge’s decision confused the act (what the actor does to the computer located abroad) with the effects of that act (what happens as a result to the computer located in the United Kingdom).

Professor Sarooshi relied on the recent judgment of the European Court of Human Rights in Wieder v United Kingdom (2024) 78 EHRR 8 (Applications nos. 64371/16 and 64407/16), a decision not available to the judge which, he submitted, held that an act of surveillance comparable to the alleged hacking in the present case takes place where the communications are intercepted and examined, and not where the victim’s device is located.

He relied also on the Computer Misuse Act 1990 which, he submitted, recognises that when a person abroad secures unauthorised access to a computer in the United Kingdom, the offending acts are committed abroad and not in the United Kingdom. In such a case, Parliament has provided in section 4(2) that the United Kingdom courts will only have jurisdiction if there is ‘at least one significant link with [the] domestic jurisdiction’, a provision which would be unnecessary if the act in question were regarded as having been committed here.

Mr Ben Silverstone for the claimants submitted that it was unreal to suggest that no acts occurred in the United Kingdom. The claimants had pleaded a series of acts taking place here and it made no difference that other acts had been performed by Bahrain’s agents abroad. It was wrong to conflate the act and the actor, there being no requirement in section 5 that the actor had to be present within the United Kingdom.

Mr Silverstone submitted that the claimants’ interpretation is supported by the European Convention on State Immunity 1972 (‘ECSI’) and the UN Convention, both of which refer expressly to the location of the author of the acts, a point deliberately omitted by section 5 of the 1978 Act. He relied also on cases concerned with jurisdiction under paragraph 3.1(9) of Practice Direction 6B and on the cases cited by the judge.

In my judgment, as a straightforward use of language, the remote manipulation from abroad of a computer located in the United Kingdom is an act within the United Kingdom. The true position in such a case is that the agents of the foreign state commit acts both in this country and abroad. To distinguish between what happens abroad and what happens here, characterising the former as an act and the latter as merely the effect of the act, is artificial and unprincipled. The reality is that a foreign state which acts in this way is interfering here with the territorial sovereignty of the United Kingdom.

I agree with the judge that what was said in R v Governor of Brixton Prison, ex parte Levin and Ashton Investments Ltd v OJSC Russian Aluminium (Rusal) is helpful. That is not because those cases set out any applicable principle of law. It would be dangerous to incorporate principles of extradition law, or the procedural law governing service out of the jurisdiction, into the very different context of state immunity, and for that reason I did not find some of the other jurisdiction cases relied on by the judge of assistance. Rather, they are helpful because they demonstrate on comparable facts that to describe the act of hacking as taking place here is a natural and appropriate use of language.

In Levin the applicant had used his own computer in St Petersburg to gain access to a United States bank’s computer at Parsipenny in the state of New Jersey, in order to transfer funds into various bank accounts controlled by him. The United States sought his extradition from the United Kingdom. The applicant contended that the appropriation of funds had taken place in Russia, where his computer keyboard was situated, and that the English court therefore had no jurisdiction. The Divisional Court (Lord Justice Beldam and Mr Justice Morison) rejected this contention:

‘For the reasons we have already indicated, the operation of the keyboard by a computer operator produces a virtually instantaneous result on the magnetic disk of the computer even though it may be 10,000 miles away. It seems to us artificial to regard the act as having been done in one rather than the other place. But, in the position of having to choose on the facts of this case whether, after entering the computer in Parsipenny, the act of appropriation by inserting instructions on the disk occurred there or in St Petersburg, we would opt for Parsipenny. The fact that the applicant was physically in St Petersburg is of far less significance than the fact that he was looking at and operating on magnetic disks located in Parsipenny. The essence of what he was doing was done there. …’

Similarly, the essence of what Bahrain’s agents are alleged to have done in the present case was done in the United Kingdom where the claimants’ computers were hacked.

Ashton Investments involved a claim for breach of confidence, unlawful interference with business and conspiracy. The allegation was that the defendants in Russia had hacked into the claimant’s computer system in London in order to obtain confidential and privileged information. The claimant sought permission to serve a claim form on the defendants in Russia pursuant to (what was then) CPR 6.20(8). In order to rely on this jurisdictional gateway, the claimant had to show either that damage was sustained within the jurisdiction or that the damage sustained resulted from an act committed within the jurisdiction. Sitting as a Deputy High Court Judge, Mr Jonathan Hirst QC held that they succeeded on both grounds:

‘62. Ashton’s computer server was in London. That is where the confidential and privileged information was stored. The attack emanated from Russia but it was directed at the server in London and that is where the hacking occurred. In my view, significant damage occurred in England where the server was improperly accessed and the confidential and privileged information downloaded. The fact that it was transmitted almost instantly to Russia does not mean that the damage occurred only in Russia. If a thief steals a confidential letter in London but does not read it until he is abroad, damage surely occurs in London. It should not make a difference that, in a digital age of almost instantaneous communication, the documents are stored in digital form rather than hard copy and information is transmitted electronically abroad where it is read. The removal took place in London. …

63.

I also consider that substantial and efficacious acts occurred in London, as well as Russia. That is where the hacking occurred and access to the server was achieved. This may have been as a result of actions taken in Russia but they were designed to make things happen in London, and they did so. Effectively the safe was open from afar so that its contents could be removed. It would be artificial to say that the acts occurred only in Russia. On the contrary, substantial and effective acts occurred in London.’

As can be seen from these citations, in both cases it was considered artificial to say that the act of hacking occurred only in the foreign state. I respectfully agree.

In my judgment this conclusion is in accordance, not only with the language of section 5 of the 1978 Act, but also with the principles underpinning state immunity in international law. That is because the hacking by a foreign state of a computer located in this jurisdiction is an interference with the territorial sovereignty of the United Kingdom, as already noted. For this purpose it makes no difference where the agents of the foreign state are located.

Wieder v United Kingdom, on which Professor Sarooshi relied, does not detract from this conclusion. If anything, it supports it. The case concerned the bulk interception of communications by the United Kingdom intelligence agencies. The applicant’s communications, sent from outside the United Kingdom, had been intercepted when they passed through the United Kingdom. The United Kingdom argued that this was outside the territorial scope of the European Convention on Human Rights, Article 1 of which provides that a state’s obligation to secure the rights and freedoms defined in the Convention is territorial. The argument was that the interception of communications did not fall within a state’s jurisdictional competence when the sender or recipient was outside the territory of the state in question. Perhaps not surprisingly, the European Court of Human Rights rejected this argument. I agree with Mr Silverstone that this decision provides no support to the defendant’s case.

However, it is of some interest that the Court was prepared to look broadly at the substance of the interference in determining where the act of interference occurred:

‘93. … Although there are important differences between electronic communications, for the purposes of Article 8 of the Convention, and possessions, for the purposes of Article of Protocol No. 1, it is nevertheless the case that an interference with an individual’s possessions occurs where the possession is interfered with, rather than where the owner is located (see, for example, Anheuser-Busch Inc v Portugal [GC], no. 73049/01, ECHR 2007-I). Similarly, in the specific context of Article 8, it could not seriously be suggested that the search of a person’s home within a Contracting State would fall outside that State’s territorial jurisdiction if the person was abroad when the search took place. …’

In modern terms, the hacking of a person’s computer is equivalent to burglars breaking in and stealing the contents of their safe. Just as the latter is an act within the United Kingdom, so too is the former.

Although Professor Sarooshi also relied on the Computer Misuse Act 1990, in my judgment that Act is of no assistance in interpreting section 5 of the State Immunity Act 1978. The contexts and purposes of the two Acts are completely different and in any event the Computer Misuse Act was passed some 12 years after the 1978 Act. The way in which Parliament chose to legislate for computer misuse in 1990 can shed no light on whether the hacking from abroad of a computer located in the United Kingdom is an act within the United Kingdom for the purposes of section 5 of the 1978 Act.

I agree with Mr Silverstone that the claimants derive some limited support from the contrast between Article 11 of the ECSI and section 5 of the 1978 Act. Article 11 provides:

‘A Contracting State cannot claim immunity from the jurisdiction of a court of another Contracting State in proceedings which relate to redress for injury to the person or damage to tangible property, if the facts which occasion the injury or damage occurred in the territory of the State of the forum, and if the author of the injury or damage was present in that territory at the time when those facts occurred.’

Although I shall have to consider Article 11 more fully in connection with ground 2 below, it is immediately apparent that there are several differences between the terms of Article 11 and those of section 5 of the 1978 Act. In particular, the requirement in Article 11 for ‘the author of the injury or damage’ to be present in the territory of the forum state at the time when ‘the facts which occasion the injury or damage occurred’ is not included in section 5, and plainly this omission was deliberate.

For all these reasons, therefore, I would reject ground 1.

The second ground of appeal is that the exception to immunity in section 5 of the 1978 Act requires that each and every act or omission causing injury that forms the basis for liability must occur in the United Kingdom. Thus the exception does not apply where the acts in question are committed partly in the United Kingdom and partly abroad.

The judge held that the plain grammatical meaning of section 5 was that immunity would not apply if an act or omission causing personal injury took place in the United Kingdom, even if other causative acts took place abroad:

‘97. In my judgment, the grammatical meaning of s 5, and in particular the use of the indefinite article (death or personal injury caused by 'an act or omission') (emphasis added) means what it says. There has to be an act or omission in the UK which is causative of the requisite damage on a more than de minimis basis. Parliament did not say ‘the act or omission’, still less, ‘acts or omissions occurring entirely within the UK’, both of which would have been more supportive of the Defendant's interpretation of s 5. This suggests the Claimants’ contention is the correct one.’

In reaching this conclusion, he followed his earlier judgment in Al-Masarir.

Professor Sarooshi criticised the judge’s approach, submitting that he had placed too much reliance on the grammatical meaning of section 5. Professor Sarooshi identified five points which, he submitted, demonstrated that the exception to immunity applied only if all of the acts causing personal injury occurred in the United Kingdom. These were: (1) the fact that it was the purpose of section 5 to implement in domestic law the provisions of Article 11 of the ECSI; (2) the decision of the Court of Appeal in Al-Adsani; (3) principles of international comity, including the act of state doctrine; (4) the personal injury exception to state immunity contained in the UN Convention drafted by the International Law Commission; and (5) the decisions of United States courts dealing with the United States Foreign Sovereign Immunities Act 1976, which was to be regarded as the precursor to our 1978 Act.

To be clear, Professor Sarooshi’s submission was that all of the acts causing the personal injury had to take place in United Kingdom, not that the injury itself must occur here. He disclaimed reliance on any case as to where ‘in substance’ a tort is committed (cf. in the jurisdictional context, Metall & Rostoff AG v Donaldson Lufkin & Jenrette Inc [1990] 1 QB 391; and in the context of state immunity, contrast Heiser v Islamic Republic of Iran [2019] EWHC 2074 (QB) at 148, where Mr Justice Stewart pointed out that section 5 ‘is not concerned with where the substance of the tort is committed’) and did not espouse the ‘entire tort’ theory such as seen in the United States cases considered below.

Mr Silverstone submitted that the judge was right to regard the meaning of section 5 as clear (‘as plain as plain can be’, in Lord Justice Ward’s words in Al-Adsani) and that the points relied on by Professor Sarooshi could not overcome this. But in any event, he submitted that those points did not lead to the conclusion for which Professor Sarooshi contended.

I agree with the judge that the language of section 5 is clear and unambiguous in this respect. A foreign state does not have immunity for personal injury caused by an act in the United Kingdom, even if other causative acts take place abroad. Since the language of the section is clear and unambiguous, there is no scope to arrive at a different interpretation based on the external aids on which Professor Sarooshi relied, although I also agree with Mr Silverstone that those aids do not support Bahrain’s interpretation.

I consider that the claimants’ interpretation of section 5 is in accordance with the fundamental principles of international law on which state immunity is based. Once again, a foreign state which hacks a computer located in the United Kingdom interferes with the territorial sovereignty of the United Kingdom even if some of the acts in question take place abroad. Legislation which is broadly similar to the State Immunity Act 1978 has been enacted in numerous jurisdictions and there are international conventions to similar effect, even if such legislation does not (or does not yet) represent customary international law. Accordingly, if State A interferes with the territorial sovereignty of State B by doing an act in State B which is liable to cause death or personal injury to persons in State B, it takes the risk that it will be subject to civil proceedings in State B. Such proceedings are in accordance with principles of international comity.

I consider in turn the matters on which Professor Sarooshi relied.

Professor Sarooshi submitted that one of the important reasons for enactment of the 1978 Act was to permit the United Kingdom to become a party to the ECSI, and specifically that the purpose of section 5 of the Act was to implement Article 11 of the ECSI in domestic law. For this purpose he relied on the statement by the European Court of Human Rights in Al-Adsani v United Kingdom (2002) 34 EHRR 11 at [22] that section 5 ‘was enacted to implement the 1972 European Convention on State Immunity’. He relied also on what are described as ‘Notes on Clauses’ provided to Ministers during the Report Stage of the passage of the State Immunity Bill in the House of Lords. These Notes stated that a Bill was required to amend the law of state immunity to enable the United Kingdom to ratify ECSI, which it had signed but not yet ratified. They stated in relation to what became section 5 that:

‘This Clause implements Article 11 of the Convention. As paragraph 47 of the Explanatory Report points out the necessary jurisdictional links in this context were based on Article 10(4) of the Convention of the 1^st February 1971 on the Recognition and Enforcement of Foreign judgments in Civil and Commercial Matters. The Report on that Convention indicates the intention that a court should have jurisdiction for the purpose of the Convention in respect of a death or personal injury or damage to tangible property which occurs outside the territory of the court, provided that it was caused by an act or omission in the territory of the forum and the person who was responsible for the act or omission was present in the territory when the act or omission occurred. Thus if a motor vehicle was negligently repaired in United Kingdom by the servant of a foreign State and as a result of that negligence and accident occurred in the territory of another State which caused death, personal injury, loss of or damage to tangible property it would be possible to take proceedings against the State in the appropriate court of the United Kingdom and the State would not be immune from the jurisdiction. Article 11 of the Convention requires the author of the injury or damage to be present in the territory at the time when the facts which occasion the injury or damage occurred. These words have not been repeated in the Bill because it is difficult to imagine circumstances when Clause 5 would apply when the person responsible for the act or omission was outside the United Kingdom.’

I would observe that although this Note states that the purpose of section 5 was to implement Article 11 of the ECSI, it also makes clear that the decision to depart from the terms of Article 11, and in particular to omit the requirement for the author of the injury or damage to be present in the forum state, was deliberate. Although the Note said that it was difficult to imagine circumstances when the person responsible for the act or omission would be outside the United Kingdom, it did not say that in such circumstances (if they occurred) a foreign state would have immunity. In any event, many things which it would have been difficult to imagine in 1978 are now common aspects of our daily life.

More fundamentally, however, these Notes on Clauses are not an admissible aid to interpretation of the Act at all. They were provided to Ministers, essentially as a briefing note, but in 1978 were not even made available to Members of Parliament (although it appears that this practice had changed by the early 1990s). As explained by Mr Justice Sales in R (Public & Commercial Services Union) v Minister for the Civil Service [2010] EWHC 1027 (Admin), [2010] ICR 1198:

‘55. In my judgment, notes on clauses (as distinct from published explanatory notes) are not a proper aid to the interpretation of an Act of Parliament, whether they are circulated to MPs (as happened in relation to the 1990 Act) or not (as in relation to the 1972 Act). Although in the former case, unlike the latter case, it might be argued that there are some grounds for saying that the notes on clauses form part of the contextual background against which the Bill was passed by Parliament as a collective body, so that they should be taken to have an interpretive role and status analogous to that of statements in a White Paper proposing legislation, or in clear statements by a promoter of a Bill in Parliament or in modern form Explanatory Notes, I think that there is an important difference from all these cases. Notes on clauses when not cited in debate are private documents not available to the public at large, unlike White Papers, statements reported in Hansard and published Explanatory Notes. An Act of Parliament creates law applicable to all citizens. In my judgment, it is fundamental that all materials which are relevant to the proper interpretation of such an instrument should be available to any person who wishes to inform himself about the meaning of that law. That is not the position in relation to notes on clauses and for that reason I do not consider they are a legitimate aid to construction of an Act of Parliament. …’

The Supreme Court dealt with a similar argument concerning Notes on Clauses in In re Scottish Independence Referendum Bill [2022] UKSC 31, [2022] 1 WLR 5435 at [26], although in this case (unlike in 1977-78) the Notes were published both online and in hard copy, saying that:

‘26. We do not, however, attach particular weight to the Notes on Clauses. The document was drafted by government officials and has no endorsement by the United Kingdom Parliament. It is much less significant than the language carefully chosen by the parliamentary drafter and enacted by Parliament.’

I would add that Bahrain sought in the court below to rely on statements made in Parliament during the passage of the Bill. The judge held that these statements did not satisfy the requirements set out in Pepper v Hart [1993] AC 593, and that they were therefore inadmissible. That decision has not been challenged on appeal. Notes on Clauses which were private to Ministers are an a fortiori case.

In fact it is too simple, and therefore inaccurate, to say that the purpose (or even a purpose) of the 1978 Act was to implement the ECSI as a matter of domestic law. The true position is that the Act gave broad effect to the ECSI, but departed from it in a number of respects. This is made clear by a number of statements of high authority. For example, in La Générale des Carrières & des Mines v FG Hemisphere Associates LLC [2012] UKPC 27, [2013] 1 All ER 409, Lord Mance said that:

‘10. The Act was aimed at giving broad effect to (though not following precisely the wording of) the European Convention on State Immunity (Basle, 16 May 1972; Misc 31 (1972); Cmnd 5081), which was agreed under the aegis of the Council of Europe at Basle on 16 May 1972 and which entered into force on 11 June 1976.’

To similar effect, in General Dynamics Lord Lloyd-Jones pointed out at [48] and [76(4)] that the 1978 Act ‘deliberately diverges from the ECSI’ in relation to the enforcement of arbitration awards against a state. He concluded, therefore, that its provisions cast little light on the correct reading of section 12 of the 1978 Act, which is concerned with service of court process on states. Other examples where the Act diverges to a greater or lesser extent from the Convention include contractual obligations (cf. section 3(1)(b) of the Act and Article 4(1) of the ECSI), employment contracts (cf. section 4(1) and Article 5(1)), and companies (cf. section 8(1) and Article 6(1)). It should therefore come as no surprise to find that section 5 is another provision which diverges from the equivalent provision in the ECSI.

As already explained, it is apparent from a comparison of section 5 of the 1978 Act with Article 11 of the ECSI that Parliament deliberately departed from the terms of Article 11. Section 5 is not concerned with ‘the facts’ which occasion the injury or damage, but with ‘an act’ of the foreign state, while the requirement for ‘the author of the injury or damage’ to be present in the forum state has been deliberately omitted. In these circumstances the terms of Article 11 are of no real help in interpreting section 5. The European Court of Human Rights was therefore mistaken, save in a very broad and general sense, to say in Al-Adsani that section 5 ‘was enacted to implement [Article 11 of] the 1972 European Convention on State Immunity’, a statement which was in any event unnecessary for its decision.

The claimant in Al-Adsani alleged that he had been detained and tortured in Kuwait by officials of the government of Kuwait, and that after leaving Kuwait and coming to live in the United Kingdom, he received death threats from agents of the government of Kuwait, including threats emanating from the Embassy of Kuwait in London. He claimed as a result to have suffered both physical and psychological injuries, and sought to bring proceedings in the English courts against the government of Kuwait, contending that a state’s immunity did not extend to acts of torture. This court held, so far as the alleged torture in Kuwait was concerned, that the exception to immunity in section 5 did not apply, so that Kuwait was immune pursuant to section 1. This was the context for Lord Justice Ward’s comment that ‘the Act is as plain as plain can be’ to which I have already referred. Lord Justice Stuart-Smith said that it was clear that ‘the Act is a comprehensive code and is not subject to overriding considerations’. On this issue the decision was affirmed in Jones v Ministry of the Interior of the Kingdom of Saudi Arabia.

However, the claim in relation to the threats allegedly made in the United Kingdom failed for a different reason, on the facts. The judge was not satisfied that the threats emanated from a person or persons acting at the behest of the government of Kuwait or for whom the government was responsible, and the claimant therefore failed to discharge the burden to satisfy the court of these threats on the balance of probabilities.

Professor Sarooshi relied on this case as demonstrating that a state is immune as regards acts outside the United Kingdom. I agree, although that is already clear from the language and scheme of the 1978 Act. However, the case does not address the issue with which we are concerned, which is whether immunity applies when there is an act causing personal injury in the United Kingdom as well as other causative acts which occur outside the United Kingdom. If anything, because Al-Adsani implies that the threats allegedly made in the United Kingdom would have been justiciable if they had been sufficiently proved, it suggests that the state will not be immune as regards an act within the United Kingdom even where that act is related to and follows on from further acts of the state occurring outside the United Kingdom. But as the point does not appear to have been argued, that may be reading too much into the decision.

Professor Sarooshi submitted that to deprive Bahrain of immunity would infringe principles of international comity because, contrary to the act of state doctrine, it would involve the English court sitting in judgment on acts performed within the territory of a foreign state. That would arise because the court would be adjudicating on the acts of the agents of the government of Bahrain who, located in Bahrain, caused the spyware to be installed on the claimants’ computers. He relied on the principle most recently stated by Lord Lloyd-Jones in ‘Maduro Board’ of the Central Bank of Venezuela v ‘Guaidó’ Board of the Central Bank of Venezuela (reported as Deutsche Bank AG v Receivers Appointed by the Court) [2021] UKSC 57, [2023] AC 156:

‘135. It appears therefore that a substantial body of authority, not all of which is obiter, lends powerful support for the existence of a rule that courts in this jurisdiction will not adjudicate or sit in judgment on the lawfulness or validity under its own law of an executive act of a foreign state, performed within the territory of that state. The rule also has a sound basis in principle. It is founded on the respect due to the sovereignty and independence of foreign states and is intended to promote comity in inter-state relations. While the same rationale underpins state immunity, the rule is distinct from state immunity and is not required by international law. It is not founded on the personal immunity of a party directly or indirectly impleaded but upon the subject matter of the proceedings. The rule does not turn on a conventional application of choice of law rules in private international law nor does it depend on the lawfulness of the conduct under the law of the state in question. On the contrary it is an exclusionary rule, limiting the power of courts to decide certain issues as to the legality or validity of the conduct of foreign states within their proper jurisdiction. It operates not by reference to law but by reference to the sovereign character of the conduct which forms the subject matter of the proceedings. In the words of Lord Cottenham [in Duke of Brunswick v King of Hanover (1848) 2 HL Cas 1], it applies “whether it be according to law or not according to law”. I can, therefore, see no good reason to distinguish in this regard between legislative acts, in respect of which such a rule is clearly established (see paras 171-179 below), and executive acts. The fact that executive acts may lack any legal basis does not prevent the application of the rule. In my view, we should now acknowledge the existence of such a rule.’ (My emphasis)

It is, perhaps, sufficient to say that, as Lord Lloyd-Jones explained, the act of state doctrine is distinct from state immunity. It cannot, therefore, cast much light on the true meaning of section 5 of the 1978 Act. Moreover, it is not at all clear in my judgment that allowing these proceedings to continue on the merits would infringe the act of state principle. As appears from the words which I have emphasised, that principle applies to the act of a foreign state ‘performed within the territory of that state … within [its] proper jurisdiction’. However, the English proceedings will be concerned with the act of infecting the claimants’ computers within the United Kingdom. The fact that the agents who caused this to happen were located in Bahrain at the time, if that proves to be the case, is immaterial. It is equally immaterial whether whatever was done in Bahrain was lawful under the law of Bahrain. The act of state principle does not protect a state which chooses, by an act committed within the United Kingdom, to infringe the territorial sovereignty of this country.

In this connection I find persuasive the analysis of Justice LeBel in the Canadian Supreme Court case of Kazemi v Iranian Republic of Iran [2014] 3 SCR 176, commenting on section 6 of the Canadian State Immunity Act 1985 which is broadly equivalent to our section 5:

‘72. By contrast an interpretation of s. 6(a) that requires the tort causing the personal injury or death to have occurred in Canada upholds the purposes of sovereign equality without leading to absurd results. It accords with the theory of sovereign equality to allow foreign states to be sued in Canada for torts allegedly committed by them within Canadian boundaries. As explored above, sovereignty is intimately tied to independence. State independence relates to the “exclusive competence of the State in regard to its own territory” (Island of Palmas Case (or Miangas), United States of America v Netherlands, Award (1928), II R.I.A.A. 829 at p. 838; Fox & Webb, at p. 74). If a foreign state is committing torts within Canadian controlled boundaries, Canada has the competence (derived from its independence) to bring the foreign state within Canada’s adjudicative jurisdiction. There would thus be a sufficient connection with the forum state to justify bringing the foreign state’s actions under Canadian scrutiny. In this way, the territorial tort exception to state immunity maintains an appropriate balance between “the principles of territorial jurisdiction and state independence” (Laroque, Civil Actions for Uncivilised Acts, at p.258. It enables a forum state to exercise jurisdiction over foreign states within its borders without allowing the forum state to “sit in judgment of extraterritorial state conduct” (ibid.). …’

Article 12 of the UN Convention, headed ‘Personal injuries and damage to property’, provides as follows:

‘Unless otherwise agreed between the States concerned, a State cannot invoke immunity from jurisdiction before a court of another State which is otherwise competent in a proceeding which relates to pecuniary compensation for death or injury to the person, or damage to or loss of tangible property, caused by an act or omission which is alleged to be attributable to the State, if the act or omission occurred in whole or in part in the territory of that other State and if the author of the act or omission was present in that territory at the time of the act or omission.’

The Convention is not in force. It has been signed, but not yet ratified, by the United Kingdom.

Although Article 12 has some similarities to Article 11 of the ECSI, there are also differences. For example, like the ECSI, Article 12 of the UN Convention provides that immunity for personal injury is only lost if the author of the act or omission was present in the territory of the forum state at the time of the act or omission. On the other hand, unlike the ECSI, Article 12 provides that immunity is lost if the act or omission occurred ‘in whole or in part’ in the territory of the forum state.

Professor Sarooshi submitted that, in practice, immunity would only be lost under Article 12 if all the acts causing personal injury occurred in the forum state. While that may be so, and while it has been said (as a general comment) that the Convention ‘powerfully demonstrates international thinking’ on state immunity (Jones v Ministry of the Interior of the Kingdom of Saudi Arabia per Lord Bingham at [8]), I cannot see how a convention concluded in 2004, which is not in force and which does not purport to represent existing customary international law, can have any bearing on the interpretation of the 1978 Act. It is not even a consistently safe guide to the content of customary international law, for the reasons explained by Lord Lloyd-Jones in General Dynamics at [52] to [54] and, as Lord Lloyd-Jones also explained at [50], it is necessary to approach general statements about the Convention with caution.

In the United States the applicable legislation is the Foreign Sovereign Immunities Act 1976. This was described by Lord Clarke in SerVaas Inc v Rafidain Bank [2012] UKSC 40, [2013] 1 AC 595 at [23] as ‘a leading precursor’ of the 1978 Act, where the issue concerned the meaning of the expression ‘in use … for commercial purposes’ in section 13(4) of the 1978 Act. At [28] Lord Clarke treated United States cases on the equivalent section of the 1976 Act as ‘strong persuasive authority’.

The United States equivalent to section 5 of the 1978 Act is known as ‘the non-commercial tort exception’. It provides that there is no immunity from an action involving ‘personal injury or death, or damage to or loss of property, occurring in the United States and caused by the tortious act or omission of [a] foreign state or of any official or employee of that foreign state while acting within the scope of his office or employment’. However, the United States courts have developed what is known as ‘the entire tort’ doctrine, pursuant to which immunity will only be lost if ‘the entire tort – including not only the injury but also the act precipitating that injury’ occurs in the United States: e.g. Jerez v Republic of Cuba 775 F.3d 419 (2014), quoting Asociacion de Reclamantes v United Mexican States 735 F.2d 1517 (1984). (It should be noted, however, that the ‘entire tort’ doctrine does not apply to the ‘terrorist exception’ contained in section 1605A of the 1976 Act which was introduced by amendment in 1996). Professor Sarooshi does not suggest that this entire tort doctrine forms any part of United Kingdom law under the State Immunity Act 1978.

It was by reference to this United States doctrine that the United States Court of Appeals for the District of Columbia Circuit dismissed a claim on similar facts to the present case. In Doe, aka Kidane v Federal Democratic Republic of Ethiopia 851 F.3d 7 (2017) the claimant alleged that he was tricked in the United States into downloading a computer program which enabled Ethiopia to spy on him from abroad. The state tort on which he relied was ‘intrusion upon seclusion’ which, under Maryland law, required the formation of an intent by the defendant as well as the interference with the claimant’s computer. The court said that the tortious intent aimed at the claimant was formed outside the United States, as was the tortious act of computer programming, albeit that Ethiopia’s placement of the FinSpy virus on the claimant’s computer was completed in the United States. It held, therefore, that the entire tort did not occur in the United States and Ethiopia was entitled to immunity. (I note in passing that the court appears to have regarded the placement of the FinSpy virus on the claimant’s computer as an act occurring in the United States, which may have some relevance to ground 1 in the present case).

Although Professor Sarooshi relied on this decision, in my judgment it does not assist him. As he accepted, the legal principle on which it is based does not form part of United Kingdom law. Moreover, as with the UN Convention, it is necessary to be cautious about general statements regarding United States authorities, applying different legislation by reference to (in the case of the non-commercial tort exception) different legal concepts.

There is a particular need for caution in a context such as the present where the United States courts place some emphasis on the legislative history of the provisions in question. For example, in the Asociacion de Reclamantes case the court recognised that ‘the statutory provision is susceptible of the interpretation that only the effect of the tortious action need occur here’, but nevertheless applied a different interpretation because, quoting the House Report:

‘The legislative history makes clear that for the exception of 1605(a)(5) to apply “the tortious act or omission must occur within the jurisdiction of the United States”.’

No doubt, in some cases, of which SerVaas was one, decisions of United States courts on equivalent provisions of the Foreign Sovereign Immunities Act 1976 will constitute strong persuasive authority as to the meaning of provisions of the 1978 Act. But that will generally depend on the closeness of the language of the two provisions and on the absence of particular domestic principles of United States law.

For these reasons, I would also reject ground 2 of this appeal.

The claimants’ case, supported by expert evidence, is that the discovery of the hacking of their computers caused them psychiatric injury. The first claimant is alleged to have developed an adjustment disorder, while the second claimant is alleged to have undergone a significant exacerbation of the adjustment disorder from which he already suffered. An adjustment disorder is a recognised condition, described by the judge as involving otherwise normal emotional and behavioural reactions that manifest more intensely than usual (considering contextual and cultural factors), causing marked distress, preoccupation with the stress or its consequences, and functional impairment.

The third and final ground of appeal is that standalone claims of psychiatric injury do not constitute ‘personal injury’ within the meaning of section 5 of the 1978 Act.

The judge held that psychiatric injury does constitute ‘personal injury’ within the meaning of section 5 of the 1978 Act. He relied on two decisions of the Employment Appeal Tribunal to that effect, Military Affairs Office of the Embassy of the State of Kuwait v Caramba-Coker [2003] UKEAT/1054/02 and Federal Republic of Nigeria v Ogbonna [2011] UKEAT/585/10, [2012] 1 WLR 139, citing extensively from the decision of Mr Justice Underhill in the latter case and pointing out that it had been referred to with approval by this court in Zu Sayn-Wittgenstein-Sayn v His Majesty Juan Carlos de Borbón y Borbón [2022] EWCA Civ 1595, [2023] 1 WLR 1162.

For the purpose of the state immunity issue the judge was satisfied, on the basis of expert evidence, that the claimants had sufficiently proved their case on the facts. That point has not been challenged on appeal (but see [14] above).

Professor Sarooshi accepted that there would be no immunity for an act in the United Kingdom causing psychiatric injury if this was associated with physical injury, but submitted that ‘standalone’ psychiatric injury was not within the meaning of ‘personal injury’ in section 5 of the 1978 Act. He recognised that English courts now regard the term ‘personal injury’ as including both physical and psychiatric injury, but submitted that this approach should not be applied to the 1978 Act on the ground that it was not clearly established in 1978 even as a matter of English law, let alone in international law. He relied again on Article 11 of the ECSI, pointing out that the equally authentic French text refers to ‘un préjudice corporel’; on Article 12 of the UN Convention, which in its French text refers to ‘l’intégrité physique d’une personne’; and on Canadian case law, which holds that ‘Only when psychological distress manifests itself after a physical injury will the exception to state immunity be triggered’ (Schreiber v Federal Republic of Germany [2002] 3 RCS 269 and Kazemi v Islamic Republic of Iran at [75]). Accordingly Ogbonna was wrongly decided.

Professor Sarooshi relied also on the Notes on Clauses supplied to Ministers during the passage of the 1978 Bill through Parliament, which stated that:

‘Where there has been injury to the person or damage to property, the rule of non-immunity applies equally to any accompanying claim for non-material damage resulting from the same acts, e.g. a claim for damages for pain and suffering or for mental shock.’

However, as I have already held that the Notes on Clauses are not admissible, I need say no more about them.

This ground of appeal was addressed by Professor Philippa Webb on behalf of the claimants. She submitted, in summary, that as a matter of statutory interpretation, section 5 of the 1978 Act should not be tied to concepts of ‘personal injury’ as they were understood in 1978 (the ‘always speaking’ principle); that the decision of Mr Justice Underhill in Ogbonna is correct; that even in 1978, ‘personal injury’ was understood as including standalone psychiatric injury in English law; that there was no settled international law concept of ‘personal injury’ which excluded stand-alone psychiatric injury’; and that the Canadian case law is irrelevant.

Broadly speaking, I accept Professor Webb’s submissions.

It is a general principle of statutory interpretation that a statute is not frozen in time at the date of its enactment, but should be interpreted taking into account changes that have occurred since its enactment. The principle was explained by the Supreme Court in News Corp UK & Ireland Ltd v Revenue & Customs Commissioners [2023] UK 7, [2024] AC 89:

‘27. It is clear that the modern approach to statutory interpretation in English (and UK) law requires the courts to ascertain the meaning of the words used in a statute in the light of their context and the purpose of the statutory provision: see, eg, Quintavalle, para 8 (per Lord Bingham); Uber BV v Aslam [2021] UKSC 5; [2021] ICR 657, para 70; Rittson-Thomas v Oxfordshire County Council [2021] UKSC 13; [2022] AC 129, para 33; R(O) v Secretary of State for the Home Department [2022] UKSC 3; [2022] 2 WLR 343, paras 28-29. 28. Within that modern approach, it is also a well-established principle of statutory interpretation that, in general, a provision is always speaking: see, eg, Royal College of Nursing of the United Kingdom v Department of Health and Social Security (“Royal College of Nursing”) [1981] AC 800; R v Ireland [1998] AC 147, 158-159 ; Quintavalle; Owens v Owens [2018] UKSC 41, [2018] AC 899 (approving [2017] EWCA Civ 182, [2017] 4 WLR 74); Test Claimants in the Franked Investment Income Group Litigation v Revenue and Customs Commissioners [2020] UKSC 47, [2022] AC 1. See also Craies on Legislation (ed Daniel Greenberg), 12^th ed, (2022) ch 21; and Bennion, Bailey and Norbury on Statutory Interpretation 8^th ed, (2020) ch 14.

29.

What is meant by the always speaking principle is that, as a general rule, a statute should be interpreted taking into account changes that have occurred since the statute was enacted. Those changes may include, for example, technological developments, changes in scientific understanding, changes in social attitudes and changes in the law. Very importantly it does not matter that those changes could not have been reasonably contemplated or foreseen at the time that the provision was enacted. Exceptionally, the always speaking principle will not be applied where it is clear, from the words used in the light of their context and purpose, that the provision is tied to an historic or frozen interpretation. A possible example (referred to by Lord Steyn in R v Ireland at p 158) is The Longford (1889) 14 PD 34 where the word “action” in a statute was held not to be apt to cover an Admiralty action in rem: at the time the statute was passed, the Admiralty Court “was not one of His Majesty’s Courts of Law” (p 38).

30.

The great merit of the always speaking principle is that it operates to prevent statutes becoming outdated. It would be unrealistic for Parliament to try to keep most statutes up to date by continually passing amendments to cope with subsequent change.’

I see no reason to suppose that section 5 of the 1978 Act was intended to be ‘tied to an historic or frozen interpretation’. On the contrary, it was recognised that the law of state immunity was undergoing a process of development. The position might be different if there had been a settled understanding in international law that personal injury did not include standalone psychiatric injury, but (as discussed below) there was no such understanding.

An interesting and relevant example of the ‘always speaking’ principle, discussed in News Corp at [33], is R v Ireland [1997] UKHL 34, [1998] AC 147. In that case the House of Lords decided that ‘grievous bodily harm’ and ‘actual bodily harm’ in sections 18, 20 and 47 of the Offences against the Person Act 1861 included standalone psychiatric injuries. After tracing developments in the understanding of mental injury and pointing out that ‘The civil law has for a long time taken account of the fact that there is no rigid distinction between body and mind’, Lord Steyn said:

‘ … the distinction between physical and mental injury is by no means clear cut. … But I would go further and point out that although out of considerations of piety we frequently refer to the actual intention of the draftsman, the correct approach is simply to consider whether the words of the Act of 1861 considered in the light of contemporary knowledge cover a recognisable psychiatric injury. It is undoubtedly true that there are statutes where the correct approach is to construe the legislation "as if one were interpreting it the day after it was passed:" The Longford (1889) 14 PD 34. Thus in The Longford the word "action" in a statute was held not to be apt to cover an Admiralty action in rem since when it was passed the Admiralty Court "was not one of His Majesty's Courts of Law:" (see pp. 37, 38.) Bearing in mind that statutes are usually intended to operate for many years it would be most inconvenient if courts could never rely in difficult cases on the current meaning of statutes. Recognising the problem Lord Thring, the great Victorian draftsman of the second half of the last century, exhorted draftsmen to draft so that "An Act of Parliament should be deemed to be always speaking": Practical Legislation (1902), p.83; see also Cross, Statutory Interpretation, 3^rd ed. (1995), p.51; Pearce & Geddes, Statutory Interpretation in Australia, 4th ed. (1996), pp.90-93. In cases where the problem arises it is a matter of interpretation whether a court must search for the historical or original meaning of a statute or whether it is free to apply the current meaning of the statute to present day conditions. Statutes dealing with a particular grievance or problem may sometimes require to be historically interpreted. But the drafting technique of Lord Thring and his successors have brought about the situation that statutes will generally be found to be of the "always speaking" variety: see Royal College of Nursing of the United Kingdom v. Department of Health and Social Security [1981] AC 800 for an example of an "always speaking" construction in the House of Lords.

The proposition that the Victorian legislator when enacting sections 18, 20 and 47 of the Act 1861, would not have had in mind psychiatric illness is no doubt correct. Psychiatry was in its infancy in 1861. But the subjective intention of the draftsman is immaterial. The only relevant enquiry is as to the sense of the words in the context in which they are used. Moreover the Act of 1861 is a statute of the "always speaking" type: the statute must be interpreted in the light of the best current scientific appreciation of the link between the body and psychiatric injury.’

The real distinction, as Lord Steyn explained it, was between a mental condition which amounted to a recognised psychiatric injury and one (such as alarm, fear or distress) which did not. That is an important distinction in the context of the tort of harassment, discussed above, which is committed when a course of unacceptable and oppressive conduct, targeted at another person, causes that person alarm, fear or distress. That is not the same as psychiatric injury.

As it is common ground that, whatever the position in 1978, English law now regards psychiatric injury as falling within the term ‘personal injury’, section 5 should be interpreted in this way unless there are compelling reasons to the contrary.

The first EAT case on which the judge relied was Caramba-Coker. In that case Mr Justice Keith held that a claim to have suffered severe depression (as distinct from injury to feelings) as a result of discrimination was a claim for personal injury within the meaning of section 5 of the 1978 Act. However, the reasoning was fairly brief. In the second EAT case, Ogbonna, the point was fully considered. The defendant state’s argument (see at [14]) was that ‘whatever might be its meaning in a purely domestic context … the phrase “personal injury” should be interpreted as it would be understood as a matter of international law; and that as [a] matter of international law a claim for compensation for harm to a claimant’s mental health would be regarded as a claim for personal injuries if, but only if, it was consequent on a physical injury in the sense of some damage to the body as opposed to the mind’.

Mr Justice Underhill accepted at [15] that in general the 1978 Act should be interpreted so far as possible to conform to any recognised international norm. He was not persuaded, however, that the international law materials on which the defendant state relied demonstrated any ‘recognised meaning in international law to the phrase “personal injury” which is more limited than the natural meaning of those words in domestic law’. Those materials included Article 11 of the ECSI, Article 12 of the UN Convention, and the explanatory reports and commentaries on those Conventions. Rather, the distinction drawn was not between physical and mental injury, but between injury to the person on the one hand and such other forms of injury as damage to economic interests or to reputation on the other.

Mr Justice Underhill considered that this distinction explained the decision of the Canadian Supreme Court in Schreiber v Germany. The personal injuries claimed in that case consisted of ‘mental distress, denial of liberty and damage to reputation’, but mental distress fell short of any recognised psychiatric injury, and the decision in Schreiber that a claim for personal injury required a physical injury should be read in that light. But even if Schreiber was to be read as excluding standalone psychiatric injury from the concept of personal injury, that did not justify the conclusion that the phrase ‘personal injury’ had a recognised meaning in international law which would exclude psychiatric injury.

Mr Justice Underhill said that he was glad to reach that conclusion:

‘28. … Not only is the distinction urged on me by Mr Pipi [counsel for Nigeria] one which would mean that the concept of personal injury in section 5 of the Act was different from its meaning elsewhere in English law but it would give rise to what would frequently be difficult, and frankly artificial, debates about the extent to which a particular injury in respect of which a claim was made was physical or mental. The whole trend of recent authority has been to recognise that these kinds of distinction are difficult both conceptually and evidentially.’

Subject to what I say below about Schreiber, I respectfully agree with this reasoning.

In Zu Sayn-Wittgenstein-Sayn the claimant brought a claim against the former king of Spain, alleging that he had embarked upon a course of conduct against her which amounted to harassment. The first issue in the case was whether the ex-king was immune in respect of acts alleged to constitute part of that course of conduct which had occurred before his abdication on the basis that the acts in question were within the sphere of governmental or sovereign activity. This court held that he was immune as the acts were of a sovereign nature.

However, an issue also arose whether the injury allegedly suffered by the claimant amounted to personal injury within the meaning of section 5 of the 1978 Act. She claimed to have suffered ‘great mental pain, alarm, anxiety, distress, loss of well-being, humiliation and moral stigma’ and to suffer from sleep deprivation and nightmares. There was no allegation of physical injury. The judge, Mr Justice Nicklin, held that this was not a case of personal injury within section 5 because there was no claim for any recognised psychiatric injury, citing Ogbonna. In a judgment with which Lady Justice King and Lord Justice Popplewell agreed, Lady Justice Simler approved his decision and reasoning:

‘71. The judge addressed the argument advanced on the respondent’s behalf by reference to the personal injury exception in section 5 SIA as follows:

“76.

Although, based on my decision, the point does not arise, I should deal, finally, with the submission that, had an immunity subsisted, the claimant's claim could nevertheless continue on the basis of section 5 of the SIA. I would have rejected that argument. The claimant's claim is for pure harassment. The loss she claims does not include a claim for any recognised psychiatric injury (see [10] above). As such, I do not accept that the claimant's claim is, or includes, a claim for personal injury. A claim for distress and anxiety arising from an alleged course of conduct amounting to harassment is not, without more, a personal injury claim. Neither of the authorities relied upon by Mr Lewis QC assists the claimant. The claimant in Jones v Ruth [2011] EWCA Civ 804, [2012] 1 WLR 1495 was pursuing a claim for psychiatric injury (i.e. a claim for personal injury). Nigeria v Ogbonna [2012] 1 WLR 139 is authority only for the proposition that "personal injury", as used in section 5 of the SIA, should be given its normal meaning in domestic law; i.e. to include a claim for a recognised psychiatric injury (see [27] per Underhill J). The short point is that, in her Particulars of Claim, the claimant makes no claim that she has been caused a recognised psychiatric injury by the alleged harassment. Her claim is therefore not a claim for personal injury within the terms of section 5 of the SIA; it is a claim for distress caused by the alleged harassment.”

72.

Mr Lewis [counsel for the claimant] accepted that the original pleading did not specifically use the phrase “personal injury” or adduce a medical expert report as to any asserted psychiatric injury suffered by the respondent, as is required for a personal injury claim by CPR PD 16, para 4. However, the Particulars of Claim pleaded a claim at paragraph 7.1 for damages caused by anxiety and damage to the respondent’s health caused by harassment. Moreover, he relied on the clearly pleaded claim at paragraphs 56.1 and 56.3, for damages for anxiety, distress and depression. Although in writing he submitted this sufficiently pleaded a recognised psychiatric injury, he accepted in the course of the hearing, that it did not, and that personal injury was not in fact pleaded in the original Particulars of Claim.

73.

However, he maintained that these passages made clear that the respondent intended to claim damages for injury to her health, and it was open to her to provide further particulars documenting the extent of her injuries (which she has now done in the draft Re-Amended Particulars of Claim, including by reference to an expert medical report). Certainly, by the time of the hearing before the judge and having raised reliance on section 5 of the SIA, it was clear that she regarded her claim as a claim for personal injury, and the amended pleading demonstrates that this is the case she intends to run. The amendment would cure any defect and she should have been given the opportunity to cure any defect in her pleading, if there is one.

74.

I do not accept these submissions and can see no error in the judge’s conclusion in respect of section 5 of the SIA. The claim was plainly not pleaded as a personal injury claim nor were damages for personal injury claimed in the prayer. As the judge correctly held, a claim for distress and anxiety arising from an alleged course of conduct amounting to harassment is not, without more, a personal injury claim. The short point, again as the judge observed, is that the respondent made no claim that she has been caused a recognised psychiatric injury by the alleged harassment. Her claim is therefore not a claim for personal injury within the terms of section 5 of the SIA. It is simply a claim for distress, anxiety and depression (none of which, as pleaded, are recognised psychiatric conditions) caused by the alleged harassment.’

No submission was made to us that this case is binding upon us, but it seems to me that it is binding authority that a claim for distress and anxiety which does not amount to a recognised psychiatric injury is not a claim for personal injury within section 5, and is at the very least strong persuasive authority, approving Ogbonna, that a standalone claim to have suffered a recognised psychiatric injury is a claim for personal injury within section 5. If that is not so, the passage which I have set out makes little sense.

Professor Webb was able to show that even in 1978 ‘personal injury’ was regarded in English law as encompassing standalone psychiatric injury. She pointed to numerous statutes enacted between 1948 and 1980 which defined ‘personal injury’ as an impairment of a person’s physical or mental condition: the Law Reform (Personal Injuries) Act 1948, section 3; the Employer’s Liability (Defective Equipment) Act, section 1; the Administration of Justice Act 1970, section 33; the Defective Premises Act 1972, section 6; the Trade Union & Labour Relations Act 1974, section 14; the Unfair Contract Terms Act 1977, section 14; the Consumer Safety Act 1978, section 9; and the Limitation Act 1980, section 38. Case law also established that ‘nervous shock, or, to put it in medical terms, … any recognisable psychiatric illness caused by the breach of duty by the defendant’ was a form of personal injury (Hinz v Berry [1970] 2 QB 40, 42H, the famous ‘bluebell time in Kent’ case; McLoughlin v O’Brian [1983] AC 410; and, albeit somewhat later, Page v Smith [1996] AC 155).

It is therefore highly probable that when Parliament used the term ‘personal injury’ in the 1978 Act, that term was understood to include standalone psychiatric injury, at any rate in the absence of a settled contrary meaning in international law.

Mr Justice Underhill’s analysis of the international law materials in Ogbonna was as follows:

‘16. I start with article 11 of the European Convention. That reads simply as follows:

“A Contracting State cannot claim immunity from the jurisdiction of a court of another Contracting State in proceedings which relate to redress for injury to the person or damage to tangible property, if the facts which occasion the injury or damage occurred in the territory of the State of the forum, and if the author of the injury or damage was present in that territory at the time when those facts occurred.”

That provision by itself seems to me to cast no relevant light on the question before me. The phrase “injury to the person”, or “préjudice corporel” in the French text (which is of equal authority), seems to me perfectly apt to cover cases of injury to mental health, though I accept that it does not necessarily do so.

17.

I turn to the explanatory report on the European Convention promulgated by the Council of Europe. Article 11 is the subject of paragraphs 47-49 of the commentary. I need only quote paragraph 48, which reads as follows:

“Where there has been injury to the person or damage to property, the rule of non-immunity applies equally to any concomitant claims for non-material damage resulting from the same acts, provided of course that a claim for such damage lies under the applicable law (e.g. in respect of pretium doloris). Where there has been no physical injury and no damage to tangible property the Article does not apply. This is the case, for example, as regards unfair competition … or defamation.”

Mr Pipi relies on the statement that “where there has been no physical injury … the Article does not apply”, but I cannot place any real weight on that statement in the context in which it appears. I am ready to accept that the phrase “physical injury”, read literally, refers more naturally to bodily than mental harm, but it does not appear that the authors were concerned with the distinction between injury to physical and mental health. Rather, as is clear from the concluding sentence, they were concerned with the distinction between injury to the person on the one hand and such other forms of injury as damage to economic interests or to reputation on the other.

18.

The next item to which Mr Pipi refers consists of the Report of the International Law Commission on the work of its 43^rd session, which laid the foundations for the United Nations Convention. Article 12 of that Convention is in substantially the same terms as article 11 of the European Convention. In its commentary on article 12 the Report says at paragraph (5):

“Article 12 does not cover cases where there is no physical damage. Damage to reputation or defamation is not personal injury in the physical sense, nor is interference with contract rights or any rights including economic or social rights damage to tangible property.”

I would make the same observations about that passage as I do about the commentary on article 11 in the explanatory report on the European Convention: see above.

19.

Mr Pipi also relied on two passages from The Law of State Immunity by Lady Fox. He referred me to passages in the second edition at pages 281 and 577. At page 281 Lady Fox says, by reference to section 5 of the 1978 Act:

“The limitation to torts causing physical damage reflects the general reluctance of states to adjudicate on statements made by other states where and however published and whether malicious or negligent.”

The passage at page 577, commenting on article 12 of the United Nations Convention, says this:

“The tortious conduct covered by this exception is confined to acts causing physical damage to the person or property; damage resulting from words spoken or written remains immune.”

But, again, Lady Fox was concerned there to draw a distinction between physical damage on the one hand and damages, to other interests, in particular to reputation, on the other. She was not addressing the question of whether personal injury could include damage to mental health, and the use of the phrase “physical injury” cannot fairly be read to be expressing a view on that question.

…

21.

In sum, I find nothing in the international law materials which supports Mr Pipi’s submission that there is a recognised meaning in international law to the phrase “personal injury” which is more limited than the natural meaning of those words in domestic law.’

I respectfully agree with this analysis.

In Schreiber v Germany the claimant was arrested and spent eight days in prison in Canada pursuant to an extradition warrant issued by a court in the Federal Republic of Germany. He claimed damages for mental distress, denial of liberty and damage to reputation allegedly suffered as a result. The Canadian Supreme Court held that this did not constitute ‘personal injury’ within the meaning of section 6 of the Canadian State Immunity Act 1985, as the term ‘personal injury’ only extended to mental distress and emotional upset insofar as they were linked to a physical injury. Section 6 provided in its English text:

‘A foreign state is not immune from the jurisdiction of a court in any proceedings that relate to

(a)

any death or personal injury, or

(b)

any damage to or loss of property

that occurs in Canada.’

It is clear, however, that the decision was based on interpretation of this legislation in the light of domestic rather than international law principles, including the Canadian principles for interpreting bilingual statutes. As Justice LeBel explained:

‘51. The questions at stake fall within the purview of the domestic legislation. Indeed, it can be argued that the domestic legislation is more specific than the rules set out by the international legal principles and as such, there would be little utility in examining international legal principles in detail. In other cases, international law principles and the disposition of the matter might turn on their interpretation and application. In this appeal, the case turns on the interpretation of the bilingual versions of s.6(a) of State Immunity Act, discussed below, rather than the interpretation of international law principles.’

Justice LeBel accepted that the English text might be understood as including more than physical injury, but considered that the French text (‘dommages corporels’) was confined to injury related to bodily injury and that it was necessary to give a harmonious interpretation taking account of both language versions:

‘56. A principle of bilingual statutory interpretation holds that where one version is ambiguous and the other is clear and unequivocal, the common meaning of the two versions would a priori be preferred. … Furthermore, where one of the two versions is broader than the other, the common meaning would favour the more restricted or limited meaning.’

Thus:

‘65. Based on the provisions of the C.C.Q. [Civil Code of Quebec] and the relevant case law and doctrine set out above, I believe that the civil law concepts of “préjudice corporel – bodily injury”, despite their flexibility, incorporate an inner limitation to the potential ambit of s. 6(a) of the Act, requiring some form of interference with physical integrity. Although the terms “death” or “personal injury” found in the English version allow the possibility of non-physical injury to be captured within the s. 6(a) exception, the civil law concept of “dommages corporels” found in the French version of s. 6(a) does not. As the French version is the clearer and more restrictive version of the two, it best reflects the common intention of the legislator found in both versions. …’

I would add that the case is also consistent with English law in that the damage allegedly suffered did not consist of any recognised psychiatric injury, although this does not appear to have formed any part of the reasoning of the court.

In Kazemi v Iran the claimant’s mother was imprisoned in Iran, where she was beaten, sexually assaulted and tortured. She later died of a brain injury sustained while in the custody of Iranians officials. The claimant instituted civil proceedings against Iran on his own behalf and on behalf of his mother’s estate. The estate’s claim was for damages for physical, psychological and emotional pain and suffering. Iran was held entitled to immunity on the basis that (as in in Jones v Ministry of the Interior of the Kingdom of Saudi Arabia in this jurisdiction) there is no ‘torture’ exception to a state’s general immunity and the statutory exception in section 6 applies only where the tort causing the personal injury or death occurred in Canada. The claimant’s claim for his own psychological damage and emotional prejudice failed on the additional basis that the ‘personal or bodily injury’ exception in section 6 did not apply where the injury suffered did not stem from a physical breach of personal integrity.

Despite a challenge, the Canadian Supreme Court affirmed its earlier decision in Schreiber, holding that:

‘74. However, even if the alternative interpretation of s. 6(a) were accepted, Mr Hashemi’s circumstance would still not fall within the exception to state immunity. The “personal bodily injury” exception to state immunity does not apply where the alleged injury does not stem from a physical breach of personal integrity.

75.

In Schreiber, our Court confirmed that “the scope of the exception” in s. 6(a) is limited to instances where mental distress and emotional upset were linked to a physical injury” (para 42). Only when psychological distress manifests itself after a physical injury will be exception to state immunity be triggered. In other words, “some form of a breach of physical integrity must be made out” (para 62).’

Although Kazemi affirmed Schreiber, this does not alter the fact that both decisions on this issue are founded on Canadian principles for interpreting bilingual statutes. Although any decision of the Canadian Supreme Court is entitled to great respect, these cases do not purport, at least on this issue, to be based on principles of international law. Nor do they evidence any settled understanding in international law of the meaning of ‘personal injury’ in the context of state immunity.

For these reasons I would reject ground 3 of this appeal.

I should record that the claimants submitted that an interpretation of section 5 of the 1978 Act which confers immunity on Bahrain would breach their right of access to a court under Article 6 of the European Convention on Human Rights and that the section would therefore need, in accordance with section 3 of the Human Rights Act 1998, to be read down so as to preclude a grant of immunity to Bahrain in this case. Professor Webb (who also dealt with this issue) relied on the principles stated in Benkharbouche v Embassy of the Republic of Sudan that (1) state immunity does in principle deprive the claimant of access to a court under Article 6, but (2) this can be justified, but only where such immunity is in accordance with a rule of customary international law which denies the court jurisdiction.

As I have concluded that the appeal fails, it is unnecessary to consider this issue and I do not propose to do so.

I would dismiss the appeal.

A claim for damages for harassment by spying is unusual. At first sight it seems paradoxical. A course of conduct cannot amount to harassment of another unless it comes to their attention and has an impact upon them. Commonly, that is what the perpetrator intends. Spies, on the other hand, typically act surreptitiously, hoping and intending that their activities will go undetected by the target. But such a claim is not unprecedented: see Gerrard v Eurasian Natural Resources Corp Ltd [2020] EWHC 3241 (QB), [2021] EMLR 8 and Al-Masarir v Kingdom of Saudi Arabia [2022] EWHC 2199 (QB), [2023] QB 475 [12]-[20]. And this appeal is not about the viability of the claim but about the court’s jurisdiction over it.

I agree that the appeal on ground one fails for the reasons given by Lord Justice Males. In my view, the artificiality of the appellant’s argument is highlighted by the answer Professor Sarooshi gave to a question posed in argument. He submitted that when a person uses a pen to create a manuscript document the marks on the page are not part of the act of writing but only the effect of that act. I do not wish to add anything in relation to grounds two and three. I agree that the appeal on those grounds should also be dismissed for the reasons given by my Lord.

I also agree that the appeal should be dismissed for the reasons given by Lord Justice Males.