Walter Tzvi Soriano v Forensic News LLC & Ors.

This is an appeal and a cross-appeal against a decision of Jay J by which he granted the claimant permission to serve five media defendants in their jurisdictions of domicile in the United States of America with proceedings for libel and limited claims for misuse of private information, but refused permission to serve a variety of other claims advanced by the claimant. The defendants’ appeal challenges the grant of permission to serve the libel claim. It raises issues about s 9 of the Defamation Act 2013, which contains a new test for jurisdiction over libel claims against those domiciled abroad. The claimant’s cross-appeal challenges the refusal to allow service of claims in data protection and malicious falsehood. It raises issues about the territorial scope of the General Data Protection Regulation (GDPR) and requires us to consider whether the Judge was right to find that the claimant’s malicious falsehood claims are untenable.

The claimant, Walter Soriano, is a businessman, who was naturalised as a British citizen in May 2009 after moving to the United Kingdom in 2003. He also has Israeli nationality. The first defendant is a California corporation that owns and operates Forensic News, a publication with the stated mission of delivering “original long-form investigative journalism that actually matters”. Forensic News operates via a website, a Twitter account, a Facebook page, and podcasts. Its journalistic output is made available free of charge to members of public, with funding drawn from a mix of sources, including subscriptions, donations, merchandise sales and advertising revenue. The second defendant is a journalist, and founder of Forensic News. He and the other individual defendants are all domiciled in the United States. Three of them are journalists, and contributors to Forensic News. The sixth defendant appears to be a blogger, independent of the other defendants.

Between 5 June 2019 and 16 June 2020, a series of eight publications appeared on Forensic News outlets which referred to the claimant in unflattering terms. There was a podcast and seven articles. Some of these included photographs of him which he claims were stolen from social media accounts. There were two further publications making reference to the claimant on a website said to be operated by the sixth defendant. In addition, there were some 432 tweets and Facebook posts that referred to the various publications complained of, to which hyperlinks were provided.

The gist and substance of the publications complained of was summarised in paragraph [20] of the judgment below, where Jay J set out this “preliminary assessment”:-

“… the Publications appear to make extremely serious allegations against the Claimant at various Chase levels (including level one) asserting, for example, that he is the “thug” of the current Prime Minister of Israel, has close and corrupt links to the Russian State and various individuals of note, is guilty of multiple homicide, has received illegal “kickbacks”, has been convicted of corruption in Monaco, is part of a money laundering operation and makes illegal arrangements for corrupt oligarchs and public figures. Mr Price’s skeleton argument describes these publications as “very far from hit pieces”. I am not sure quite what he means by that, but on any view they amount to a sustained assault on the Claimant and his reputation.”

The reference to Chase levels is to the three main levels of gravity of a defamatory imputation, identified by this court in Chase v News Group Newspapers Ltd [2002] EWCA Civ 1772, [2003] EMLR 11 namely (1) guilt; (2) reasonable grounds to suspect guilt; and (3) reasonable grounds to investigate whether there is guilt.

On 14 July 2020, the claimant issued proceedings against all six defendants in the High Court of England and Wales in respect of the ten publications, the tweets, and the Facebook posts. He made claims under the laws of libel, misuse of private information (“MOPI”), data protection, malicious falsehood, and harassment.

The libel claim was brought against all six defendants in respect of publications 2 to 8; the MOPI claims were made against all six defendants in respect of all ten publications; the data protection claim was against all six defendants in connection with all the publications, and additional processing; the malicious falsehood claim was brought in respect of publications 1 to 8, but against the first and second defendants only; and the harassment claim was brought against all the defendants in respect of all the publications, and what was dubbed “the torrent of social media posting”. All the claims were confined to tortious conduct causing damage in this jurisdiction except for the data protection claim, which complained of breach of duty without geographical limitation.

As all the defendants are domiciled abroad, none of these claims could be brought against any of them unless (a) they agreed to it, which none of them did, or (b) the Court gave permission to serve them with the claim outside the jurisdiction, pursuant to Part 6 of the Civil Procedure Rules. The claimant issued an application for permission. Such applications are normally determined without notice to the respondent. If permission is granted, the respondent may apply to set it aside. On 30 September 2020, Nicklin J ordered that the application notice and supporting evidence be served on the defendants. He gave directions for the service of evidence by the defendants, and the exchange of skeleton arguments. He reasoned that this was not a straightforward application and if permission were granted on an ex parte basis there was a real prospect that the respondents (or some of them) would apply to set it aside. It would save time and costs “to direct that the Application Notice be served on the Respondents so that they can, if they wish, oppose the application.” The defendants did serve evidence and instructed solicitors and Counsel to oppose the application.

On 21 January 2021, after a contested hearing, Jay J granted the claimant’s application for permission to serve the first five defendants with his claims in libel and some of the claims in misuse of private information. He refused permission to serve those defendants with the bulk of the claims in misuse of private information and refused him permission to serve them with any claim in data protection, malicious falsehood, or harassment.

The first to fifth defendants now appeal against that decision, with the permission of Arnold LJ, contending that the Judge was wrong to grant permission for the libel claims, and any claim in misuse of private information. The claimant supports the Judge’s decision on those two causes of action for the reasons the Judge gave, and additional reasons. With the permission of Arnold LJ, the claimant cross-appeals, arguing that he should also have been allowed to pursue the claims in data protection and malicious falsehood.

The claimant does not challenge the Judge’s decision to refuse permission for the rest of his claim in MOPI, or the refusal of permission to bring a claim in harassment, about which no more need be said. Nor are we concerned with the application for permission to serve the sixth defendant. That was adjourned by Jay J and dealt with in April 2021 by Johnson J. He granted permission to serve the sixth defendant with claims in libel but gave permission to appeal that decision. The claim was later settled, and the appeal dismissed by consent. The judgment of Johnson J is before us, but only passing reference was made to it in oral argument and it too can be put to one side.

This is well-established. For present purposes, it can be adequately distilled as follows. The court can only give permission to serve a claim on a defendant outside the jurisdiction if it meets three conditions.

A claimant seeking permission to serve outside the jurisdiction always bears the legal burden of proof on all these issues. That is so whether the matter is being considered on an application by the claimant at the initial, without-notice stage, or at the hearing of a subsequent application by the defendant to set aside an order permitting service outside the jurisdiction. But a defendant challenging such an order needs to identify some other forum which does have jurisdiction; and even the initial application requires there to be another candidate with the requisite jurisdiction: UnwiredPlanet International Ltd v Huawei Technologies (UK) Co Ltd [2020] UKSC 37, [2021] 1 All ER 1141 [96-97]. Where the claimant’s contention that the case is a proper one for service out is disputed by the defendant on a specific ground the defendant bears an evidential burden in relation to that ground: see AstraZeneca (above) [33-39] (Hamblen J, as he then was).

Issues about the appropriate forum for the trial of a claim are not confined to cases of service outside the jurisdiction. A defendant may raise such an issue even if he has been served here as of right - for instance, on the straightforward basis that he is present here and can physically be served with proceedings. CPR Part 11 provides a mechanism for making a forum conveniens challenge in such a case. A defendant who fails to make such a challenge within 14 days of acknowledging service is “treated as having accepted that the court has jurisdiction to try the claim”: CPR 11(5). It is clear law that a defendant who does make such an application bears the legal burden of establishing that there is another available forum which is more suitable than England: VTB Capital plc v Nutritek International Corporation [2013] 2 AC 337 [44] (Lord Mance).

It is against this background of general principle that I turn to consider each of the four causes of action with which we are concerned.

At common law, a cause of action for libel is made out by proof that the defendant was responsible for the publication to one or more third parties of a written statement that bore a defamatory meaning about the claimant. Statute has added a requirement that publication caused serious harm to the claimant’s reputation or is likely to do so: Defamation Act 2013, s 1(1). If this much is established, the burden shifts to the defendant to raise a defence. So defamation remains a relatively simple tort to prove. English law holds that mass publication gives rise to a series of separate causes of action, one for each individual to whom the words are communicated. And a claimant is permitted to select the publications in respect of which he brings an action.

These features of the law, coupled with the application of the classic forum conveniens principles to claims in defamation, led to results which some found unpalatable. Perhaps the best-known illustration of this is the decision of the House of Lords in Berezovsky v Michaels [2000] 1 WLR 1004 where the claimant, a prominent Russian businessman who also held a senior post in the Russian government, was granted permission to serve proceedings for libel on the publisher and editor of the well-known American magazine “Forbes” and its editor in respect of the publication in this jurisdiction of 1,900 copies. The magazine sold 785,000 copies in the United States and Canada. But the House rejected a submission that in such a situation the court ought to proceed as if the various publications gave rise to a single cause of action and ask where a claim in respect of that cause of action should be tried. It applied the established approach, by which each publication of a defamatory statement is a separate tort. The claim was confined to publication here. The question was whether the English court was clearly the appropriate forum in which to try that claim. The House held that in answering that question regard must be had to the principle that the jurisdiction in which a tort is committed is prima facie the natural forum for the trial. The evidence did not displace that presumption. Disparity between the scale of publication here and abroad was a relevant factor, but so were the claimant’s connections with this jurisdiction and the existence of a reputation to protect in this jurisdiction. Mr Berezovsky had both.

In King v Lewis [2004] EWCA Civ 1329,[2005] EMLR 4 this Court applied these principles in the context of a claim by an American claimant against defendants domiciled in the US, over publication on the internet. The claim was limited to publication here, as is generally taken to be mandated by the decision in Berezovsky. The court noted that the fact that each publication amounted to a separate tort means that in theory at least a defendant posting on the world wide web might be exposed to claims in multiple jurisdictions. But it observed that people do this knowing that what they publish will be available to all and sundry without geographic restriction. The global reach of an internet publication meant that in such a case, the court’s discretion would tend to be more open-textured than otherwise. But the decision of the courts below that permission should be granted to serve proceedings on the defendants outside the jurisdiction was not wrong.

Over the years, libel claims brought in England by and against foreigners who had, or were thought to have, relatively slender connections with this jurisdiction acquired the disapproving label, “libel tourism”. Section 9 of the Defamation Act 2013 was aimed at this perceived problem. It applies to defamation claims against persons such as these defendants, who are domiciled in the USA, though it has broader effects than that. The section has been amended following Brexit. But the version that applies to the present case is the original as enacted, which provided so far as material as follows:

“9 Action against a person not domiciled in the UK or a Member State etc

(1)

This section applies to an action for defamation against a person who is not domiciled – (a) in the United Kingdom; (b) in another Member State; or (c) in a state which is for the time being a contracting party to the Lugano Convention.

(2)

A court does not have jurisdiction to hear and determine an action to which this section applies unless the court is satisfied that, of all the places in which the statement complained of has been published, England and Wales is clearly the most appropriate place in which to bring an action in respect of the statement.

(3)

The references in subsection (2) to the statement complained of include references to any statement which conveys the same, or substantially the same, imputation as the statement complained of.”

In some ways this language resembles that of the common law test of forum conveniens, but it is plainly intended to establish a different approach. At a minimum, it modifies the common law position in two respects: (a) by requiring the court to answer the question of which jurisdiction is “clearly the most appropriate” by considering “all the places in which the statement complained of has been published” and (b) by treating any statement that conveys substantially the same imputation as if it were a “statement complained of”.

Section 9 has been considered in a handful of cases to date: Ahuja v Politika Novine I Magazini D.O.O [2015] EWHC 3380 (QB), [2016] 1 WLR 1414 (Sir Michael Tugendhat); Huda v Wells [2017] EWHC 2553 (QB), [2018] EMLR 7 (Nicklin J); Wright v Ver [2019] EWHC 2094 (QB) (Nicklin J), affirmed [2020] EWCA Civ 673, [2020] 1 WLR 3913; Al Sadik v Al Sadik [2019] EWHC 2717 (QB), [2020] EMLR 7 (Julian Knowles J); and Kim v Lee [2020] EWHC 2162 (QB) (Julian Knowles J).

Some uncontroversial propositions emerge from these cases:-

In Ahuja and in Wright v Ver the court considered the impact of s 9 on a claimant with a “global reputation”.Sir Michael Tugendhat addressed the issue in Ahuja at [31-32] and [71]. He observed that it would be unsurprising if claimants resident in England and Wales were to surmount the new threshold more readily than foreign claimants. But he said there are many in business, sport and other areas of activity who reside and own properties in London and spend time there but have reputations in one or more foreign countries. Section 9 applies as much to claimants of that character as it does to claimants who are non-residents. He considered that it might be that “in addressing the complaint that non-residents with a tenuous connection to London have sought to vindicate their reputations in England, Parliament has so framed s 9 as to make it unduly difficult for some people who are resident or domiciled here to vindicate their reputations”. But the only way the court could do anything about that was if “claimants put before the court the evidence of the harm to their global reputations which enables the court to be satisfied” that the statutory test is met. In Wright v Ver Dingemans LJ observed at [58] that:-

“It is plain that a person domiciled in England and Wales may find it easier to show that the jurisdiction of England and Wales is clearly the most appropriate jurisdiction to bring the claim, but the position may not be so straightforward with persons who have moved to this jurisdiction or who have a global reputation.”

It was conceded that the libel claims satisfied the Gateway Requirement, as claims within PD6B 3.1(2) (claims for injunctions restraining conduct in the jurisdiction) and 3.1(9) (claims in tort where damage is sustained within the jurisdiction). It was also accepted that they met the Merits Test. The issue for the Judge was whether the courts of this jurisdiction are clearly the most appropriate forum for the trial of the claims, applying s 9 of the 2013 Act and the general Forum Test.

The claimant’s case on this issue was set out in his Particulars of Claim, duly verified by a statement of truth, and in two witness statements of his solicitor, Schlomo Rechtshaffen. The defendants relied on a witness statement of their solicitor, Patrick Doris.

Paragraph 1 of the Particulars of Claim said this:

“The Claimant is a British citizen, naturalized in May 2009 after moving to the United Kingdom in September 2003 (although he has also kept dual Israeli citizenship). He is both domiciled and habitually-resident in London, and is domiciled in the UK for tax purposes. He is the director of seven English companies, all of which are domiciled in London, and his professional life is centred in London. His immediate family - eight of his nine children, and all of his seven grandchildren -are based in the UK, with his ninth child due to move to the UK from Israel in the near future. The UK is the 'Member State' which is the centre of his interests for the purposes of the Brussels Recast Regulation …”

Mr Rechtshaffen dealt with the reasons for believing that the courts of England and Wales were the forum conveniens at paragraphs 33 to 38 of his first statement, where it was said that this was a claim “by an English citizen and resident seeking to recover damages suffered in England in respect of civil wrongdoing committed in the UK.” There had been substantial publication in England. The claims were governed by English law. Although the defendants were based in various US jurisdictions, the majority of the other witnesses and documents were likely to be located in England. Although some of the defendants’ conduct occurred in the US “the centre of gravity of the dispute plainly points in favour of England being the proper place to bring the claim.”

Elaborating on these points, the evidence said that the claimant was not only a citizen and resident of the UK, he was also domiciled in the UK; save for one child living temporarily in Israel, the claimant, his wife, eight children and all their grandchildren lived in England; he spent his time in the UK, save for family vacations and business trips; apart from a few passive investments abroad, his business was in England; and the businesses and many of the matters written about in the offending publications were in England. It was asserted that the distress and harm to his business and professional reputation caused by the offending publications occurred in England.

In a section of his statement headed “full and frank disclosure”, Mr Rechtshaffen gave brief details of some foreign investments of the claimant in the US and Israel, and of several libel actions he had brought in Israel, Ireland and the UK. Among these were two claims brought in Israel against one Raviv Drucker. The claimant sued Mr Drucker in 2018 over allegations that the claimant had gathered information about police officers involved in the investigation of the Israeli Prime Minister Mr Netanyahu, and the families of such officers. In 2019, there was a further claim against Mr Drucker and others over publications on a news website. A third claim was brought in this jurisdiction in July 2019 over reports in the Daily Mail of statements made by Mr Drucker. I shall have cause to mention Mr Drucker again later.

The statement of Mr Doris was made “solely for the purposes of contesting the claimant’s application for permission to serve proceedings on the defendants outside of England and Wales.” It described the proceedings, and then gave details of the foundation, personnel, funding, and activities of the first defendant and Forensic News. Mr Doris identified the intended audience of Forensic News. He gave details of the second to fifth defendants, including their jurisdictions of domicile, and an account of their involvement in matters concerning the claimant. He provided audience statistics for the Forensic News website and publications 1 to 8, as well as the Twitter and Facebook accounts of the individual defendants.

From paragraph 103 onwards, Mr Doris dealt with the “requirements for permission to serve out”. In doing so he set out what he described as “evidence which demonstrates that the claimant has no reasonable prospects of success with any of the claims, and that England and Wales should not be considered the appropriate forum to determine them.” Paragraphs 104 to 157 addressed the merits of the various claims advanced by the claimant, asserting that none of them was tenable. Paragraph 158, running to three pages, gave reasons in support of an assertion that even if there were a pleaded case with a reasonable prospect of success there was “no basis on which it can reasonably be suggested that England and Wales is the appropriate forum to determine any such claim”.

It was said that the claimant had produced no evidence “that the source of his business derived predominantly from individuals and/or companies based in the UK”. It was further alleged that “In fact, the claimant is an international businessman with clients all over the world”. Factual assertions were made in support of this allegation, relating to the business partners and business activities of the claimant’s company USG Security Ltd. Reliance was placed on the claimant’s pleaded assertion that he was a “family friend” of Mr Netanyahu, and an “understanding” that the claimant “owns or controls a network of companies across multiple jurisdictions”, and “holds investments outside the UK”. These companies and investments were identified.

In reply, the second statement of Mr Rechtshaffen addressed issues about the extent of publication, analysing in more detail than Mr Doris the number of individual publications via the website and social media. The statement also addressed the topic of “Business interests”. It denied that the corporations identified by Mr Doris were owned or controlled by the claimant and re-asserted that “Mr Soriano’s business is in England, save for a few passive investments abroad.” It reconfirmed that the Particulars of Claim, which bore a statement of truth signed by Mr Rechtshaffen, were true to the best of his knowledge and belief.

In the opening paragraphs of this section of his judgment the Judge observed that s 9 clearly applied and “replaces classic forum conveniens in the context of a libel claim”, requiring “a comparative analysis of publications in all relevant jurisdictions”. He approached the matter on the footing that the claimant bore the burden of proving his case on the balance of probabilities. He rejected the submission of Mr Callus on behalf of the claimant, that the applicable standard of proof was the “good arguable case” required in classic forum conveniens disputes. He accepted the submission of Mr Price for the defendants that “s 9 is a distinct parallel regime whose requirements must be met by a claimant independently of issues as to service and jurisdiction”. He recorded the parties’ competing submissions on the burden of proof. Mr Callus argued that there was an evidential burden on the defendants to show that there was a US state that would accept jurisdiction in respect of a claim brought by the claimant; Mr Price contended that the burden lay on the claimant throughout.

The Judge’s key reasoning on the s 9 issue is to be found between [135-164]. It is dense and detailed. The full text is readily available to any interested reader. I shall do no more than summarise some key features. The Judge took as his “point of departure” a consideration and assessment of the nature and extent of the claimant’s reputation in England and Wales. He identified the founding in 2010 of USG Security Ltd with offices in Central London. He considered the defendants’ case about the alleged worldwide nature of the claimant’s reputation and, in that context, observed that the claimant had “been far from forthcoming about his business interests here and overseas”. The Judge said that the claimant “could have afforded me much greater assistance”, and identified some matters about which it “would have been useful” to know more. He accepted that the claimant’s secrecy added support to the defendants’ argument that “even if he is not necessarily hiding something, he has not been open with the court.” But he noted that the claimant’s evidence was that “his personal and business life is centred in this jurisdiction, even if a significant number if not proportion of his clients are based overseas” and concluded that the defendants had not done enough to contradict that evidence.

The Judge regarded the instant case as “somewhat sui generis”. He did not think the claimant’s reputation could fairly be described as “global”. England and Wales could be identified as the place where his reputation was most obviously centred. The case was unlike Berezovsky and Wright where it was not possible to say that the claimants’ reputations were centred anywhere in particular. The proposition that the claimant’s reputation might be centred in Israel or have a significant foundation there was “belied by the evidence”. The proposition that he had a significant reputation in the US was “difficult to accept”. Given the extent of publication in England and Wales, the claimant had “a convincing argument that he has suffered serious harm to his reputation” here.

The Judge went on to consider the extent of publication in other jurisdictions, and where else the claimant might sue. The defendants had identified California as an appropriate candidate jurisdiction, and the Judge proceeded on the footing that California would accept jurisdiction in respect of all publications in the US. The absence of expert evidence on the topic was not fatal to the defendants’ case on that issue, held the Judge. The burden was then on the claimant to satisfy him that the courts of England and Wales were clearly the most appropriate place to bring the action. Having considered various factors relating to convenience, he concluded that the claimant had discharged that burden. He found that if compelled to litigate in California the claimant would not receive an adequate remedy for reputational damage sustained in England and Wales.

Ultimately the Judge held that the claimant had discharged the burden imposed on him by s 9 by showing that he was a British citizen whose personal and business interests lie principally in this jurisdiction, and whose reputation is centred here, seeking remedies for harm caused to that reputation by publication in this jurisdiction and nowhere else. He was “not a libel tourist”. The defendants had failed to discharge “the evidential burden of showing whether a Californian court would countenance a claim for reputational harm suffered in England and Wales, or as to whether a remedy in the US would be adequate in these circumstances.”

There are two grounds of appeal: (1) the Judge was wrong to hold that when a court is considering s 9 there is an evidential burden on a defendant to show that another forum is available and appropriate; and (2) having found that the claimant had been “far from forthcoming about his business interests”, the Judge had no safe basis on which to conclude that England and Wales was clearly the most appropriate place in which to bring the libel action, and was wrong to so conclude. The essential contention, which underlies both grounds of appeal, is that the burden always lay on the claimant to satisfy the court, by adducing sufficient evidence of fact and/or law, that England and Wales is the most appropriate place to bring the claim.

By his Respondent’s Notice, the claimant seeks to uphold the Judge’s decision on the further and alternative grounds that the Judge was wrong (a) to hold that the standard of proof is the balance of probabilities rather than a good arguable case; and (b) to reject the submission that expert evidence on the availability and appropriateness of an alternative forum is a necessary condition for the comparison required by s 9.

The effect of the above is that four inter-related points about s 9 are controversial on the defendants’ appeal:-

For the reasons that follow, I have come to the view that on points (1) to (3) the claimant’s submissions are to be preferred. This means that on points (1) and (3) I disagree with the Judge. But on those points, I take a view more favourable to the claimant, and as I am satisfied that the Judge was right on the other points I conclude that his decision to grant permission was correct.

As Lord Sumption observed in Lachaux v Independent Print Ltd [2019] UKSC 27, [2020] AC 612 [2]: “Broadly speaking, [the 2013 Act] seeks to modify some of the common law rules which were seen unduly to favour the protection of reputation at the expense of freedom of expression”. As Lord Sumption also pointed out at [13], two principles of interpretation are relevant when considering the nature and extent of the changes made to the common law rules. The first is that Parliament is taken to have known what the law was prior to the enactment, and to have legislated against that background. The second is that Parliament “can be presumed not to have altered the common law further than was necessary”.

In my judgment, s 9 can be construed as bringing about, for claims within its scope, the two changes to the forum conveniens regime that I have identified above, and nothing more; and that is how it should be construed. The language of the threshold test for jurisdiction (“… the court is satisfied that … England and Wales is clearly the most appropriate place in which to bring an action …”) reflects the wording of CPR 6.37(3) and the terminology of the cases on forum conveniens. The key differences are the introduction in subsection (2) of the words “of all the places in which the statement has been published” and the expanded notion of “the statement complained of” provided for by subsection (3). The most natural conclusion is that the draftsman has taken the classic forum conveniens test as a starting point, introducing these modifications in order to alter the balance by removing or watering down the presumption, referred to in Berezovsky, that England and Wales is the natural forum for a claim in respect of publication here. In addition, although I would not place undue weight on this, I accept the submission of Mr Callus that support for this interpretation is provided by the Explanatory Notes which state, at paragraph 67, that “It is the intention that this new rule will be capable of being applied within the existing procedural framework for defamation claims.”

These are important modifications. They mean that claimants who are better known outside this jurisdiction, or who have global reputations, are likely to find it hard to show that this is “clearly the most appropriate” jurisdiction in which to sue for an international libel. But it does not follow from the significance of the provisions that they should be treated as creating an entirely novel and revolutionary regime. The better view is the one expressed by the authors of Duncan & Neill on Defamation (5^th edition) at paragraph 9.09:

“Formerly, when considering whether ‘England and Wales is the proper place in which to bring the claim’, the court … would apply the forum conveniens test. … the court will now consider instead, under CPR 6.37(3), whether the statutory test in s 9(2) of the 2013 Act is satisfied [as occurred in Ahuja…]”.

The contrary view, accepted by the Judge, is that s 9 is a substantive provision which sets a “jurisdictional bar which the claimant must overcome”, concerned with “the subject-matter of the suit and not with personal jurisdiction over the defendant”. Those are the words of Julian Knowles J in Al Sadik [54-55]. But this was in effect the approach of Jay J. I acknowledge that this view gains support from two sources. There are the stark opening words of s 9(2): “A court does not have jurisdiction to hear and determine a claim…”. And there are the opening words of s 10 of the 2013 Act, which are almost identical. This wording, described by Mr Price as “mandatory”, would be apt to impose a jurisdictional bar. There are also some authorities about the application of s 10, which the Judge thought were important. But I think the “subject-matter” interpretation should be rejected for several reasons.

First, I think it is an unnecessarily radical interpretation of the section. The less revolutionary construction that I would adopt is consistent with the interpretative principles I have identified. I do not think it is inconsistent with the opening words of s 9(2). In classic forum conveniens, the court “has no jurisdiction” to try a claim against someone who cannot be served within the jurisdiction, unless and until the court gives permission for service abroad, having been “satisfied” that this is the proper place to bring the claim. As a general rule, service of originating process is the act by which the defendant is subjected to the court’s jurisdiction: Barton v Wright Hassall LLP [2018] UKSC 12, [2018] 1 WLR 1119 [8]. Although a case in which service is set aside on forum non conveniens grounds following an application under Part 11 is not so easily categorised as one where the court has “no jurisdiction”, that is not an entirely inapt description.

Secondly, I find it hard to see the force of the argument that the section is concerned with the “subject-matter” of the claim. The first threshold condition is about the defendant’s domicile, which is a personal characteristic. The second condition is that the claim is brought in defamation. The third is that this is clearly the most appropriate jurisdiction in which to bring the claim, which is an evaluative matter requiring a multi-factorial assessment. Looked at in the round, it seems to me that s 9 is designed as a safeguard for defendants with a defined characteristic. It protects them from the exercise of jurisdiction in respect of a specified cause of action, namely defamation but is not otherwise about the subject-matter of the suit.

Thirdly, this alternative approach to the nature of s 9 has at least one striking and unfortunate procedural consequence. On this view, a defendant served within the jurisdiction can in principle raise s 9 as a jurisdictional bar at any time, and the issue may have to be addressed at a trial. That is what Julian Knowles J decided in Al Sadik, where the claimant sued the defendantfor libel in certain WhatsApp messages sent to 34 family members, all resident outside England and Wales. After acknowledging service and serving a Defence, the defendant applied to strike out the claim on the basis, among others, that when the claim was issued she had been domiciled in Kuwait and by virtue of s 9 the court had no jurisdiction. Julian Knowles J held that the defendant’s case on domicile was not plain and obvious, but it was arguable, and it was no answer for the claimant to say that the defendant had waived her right to take the point by failing to follow the procedure for contesting jurisdiction under CPR 11. As he put it at [55], “jurisdiction under s 9 cannot be conferred by waiver, submission or consent”.

It is paradoxical that a trial should be required in order to determine whether a defendant can take advantage of a provision intended to spare non-domiciled defendants from being sued for libel in England and Wales. It is not easy to think of any other provision that can be invoked at any time as a complete bar to a claim, regardless of the procedural consequences. This approach would seem to make s 9 a stronger tool than a substantive defence to the claim. It does not appear that the attention of Julian Knowles J was drawn to what Nicklin J had said on this topic in Wright v Ver at first instance, at [58(vii)]. Nicklin J pointed to “the clear undesirability of establishing an entirely new regime for objecting to jurisdiction in defamation claims” and held that the mechanism for taking a point under s 9 was provided by Part 11 (this aspect of his decision was unaffected by the later appeal). I doubt this is the only awkward procedural consequence of the “subject-matter jurisdiction” analysis. I do not consider that we should impute to Parliament an intention to create such difficulties. The passage I have quoted from the Explanatory Notes seems to indicate that the Government did not intend or foresee them.

Fourthly, I do not believe that s 10 of the Act is as helpful in resolving the issues about the juridical nature of s 9 as the Judge evidently thought. Section 10 provides:

“10

Action against a person who was not the author, editor etc

(1)

A court does not have jurisdiction to hear and determine an action for defamation brought against a person who was not the author, editor or publisher of the statement complained of unless the court is satisfied that it is not reasonably practicable for an action to be brought against the author, editor or publisher.

(2)

In this section “author”, “editor” and “publisher” have the same meaning as in section 1 of the Defamation Act 1996.”

(Section 1 of the 1996 Act creates a statutory defence of innocent dissemination for those who are not authors, editors or publishers). It will be immediately obvious that the two sections have similarities of language and of structure.

At [120], the Judge addressed compendiously the juridical nature of s 9 and the standard of proof. He confessed that on the standard of proof, “initially I was of the view that there should be no difference between forum conveniens and s 9”. He explained that he had ultimately accepted the submission of Mr Price, for the defendants, that s 9 should be interpreted as a provision about subject-matter jurisdiction, distinct from the forum conveniens regime:-

“… largely because there is no distinction between subject-matter jurisdiction cases arising under s.10 of the 2013 Act and under s.9. Both provisions require the court to be “satisfied”. In Brett Wilson LLP v Persons Unknown [2015] EWHC 2628 (QB), [2016] 4 WLR 69, paras 20-23, and Pirtek (UK) Ltd v Jackson [2017] EWHC 2834 (QB), paras 27-38, Warby J applied the ordinary civil standard of proof to s.10.”

The similarity of the two sections does not in itself point one way or the other, of course. There is no obvious reason why s 10 should not be construed as a provision about personal jurisdiction, to be applied when an application is made for permission to serve outside the jurisdiction, or on an application to set aside service under Part 11. The fact that the jurisdiction under s 10 turns on the role an individual played in respect of a publication is perhaps more consistent with the view that the section is a provision about personal jurisdiction. Further, in my judgment, on a proper analysis, the two decisions cited lend no real support to the Judge’s conclusion.

These were applications for judgment in default of acknowledgment of service. In Brett Wilson, the defendants were persons unknown responsible for the operation and publication of a website called solicitorsfromhelluk.com. In Pirtek, the claim was in respect of anonymous or pseudonymous publications on a website, Twitter, and Facebook. On such an application, the court enters “such judgment as it appears to the court that the claimant is entitled to on his statement of case”: CPR 23.11(1). This enables the court to proceed on the basis of the claimant’s unchallenged Particulars of Claim, which is normally the right approach: Brett Wilson [18], Pirtek [26]. This seemed to me to call for consideration of the jurisdiction provisions of s 10, which I raised of my own motion. My primary approach was to analyse the pleaded allegations to see if the defendant was alleged to be the “editor, author, or publisher” of the offending words, within the meaning given to those terms by s 1 of the Defamation Act 1996: Brett Wilson [22-23], Pirtek [29-30].

I do not think the issue I am now addressing should turn on the fact that in these two cases where no acknowledgment of service had been filed, and the defendants were not participating, I addressed, of my own motion, the question of whether on the claimant’s own case s 10 applied. Nor do I think the exercise in which I engaged is one that involves any standard of proof. Indeed, in Pirtek at [32] I expressly doubted that the claimant bore any evidential burden when the test was laid down by CPR 12.11(1).

My conclusion on this issue also follows logically from what I have just said. The defendants’ case is that the Judge was correct to hold that s 9 is a distinct parallel regime from that which governs personal jurisdiction and that he was correct for the reason he gave (namely that s 9 jurisdiction is qualitatively the same as s 10 jurisdiction). But if that is wrong, and s 9 is a modification of the well-established forum conveniens regime, we should not without good reason treat the section as requiring or implying a departure from any aspect of that regime. I do not think the wording of the section indicates that Parliament intended a departure on this issue, and we have been given no other good reason to conclude that it had that intention. Indeed, it seems to me that the argument in favour of an evidential burden may be stronger in cases of worldwide publication than in some other cases. A claimant may not know, and may have no way of discovering, the reach and extent of online publication. The defendant publisher is very likely to have the details to hand.

It seems to me that the debate on this issue has been somewhat clouded by the fact that this case did not follow the standard procedural route. It might have been better if it had. In that event, the court would either have refused or granted the claimant’s without notice application. Had permission been granted, the defendants would doubtless have applied to set aside. They surely would have made a case that there was an alternative jurisdiction available, and that this was California. The effect of the unusual procedure adopted in this case might perhaps have been open to some debate. But the defendants never sought to appeal against Nicklin J’s order, or to vary it or to clarify its effect. It is clear that Nicklin J intended to bring about a combination of the two stages. In practice, that is how the parties behaved. The defendants chose to file evidence, to attend the hearing and present argument.

The witness statement of Mr Doris made clear that the defendants “do not accept” that this court has jurisdiction, and over 51 pages it set out in considerable detail factual material that was considered relevant to the court’s decision-making, as well as extensive argument on the merits of the claims including the assertions of fact that “the claimant is an international businessman with clients all over the world” who “owns or controls a network of companies across multiple jurisdictions”.

The defendants have not suggested that, having attended the hearing and actively opposed the application, they still have a further opportunity to contest the court’s jurisdiction under s 9. Instead, they have brought this appeal. In substance and reality, the position adopted by the defendants at the hearing before Jay J was that California - the domicile of the first and second defendants - was an available jurisdiction that was at least as appropriate as England and Wales. I consider the Judge was right to conclude that they bore, and in my view they in fact assumed, an evidential burden on that issue.

I have already noted that in Wright v Ver Dingemans LJ said that the standard was the balance of probabilities. But as the Judge rightly held at [117], this observation was obiter, because the contrary was not fully argued. Before Nicklin J the balance of probabilities standard had been common ground. Permission to argue on appeal that Nicklin J had wrongly applied a higher standard was refused, so the standard of proof was not an issue on the appeal. Nobody formally argued that the claimant need only establish a “good arguable case” that the threshold test was met.

Jay J observed that no “lengthy disquisition” on this issue was required, as on his view of the answer to the s 9 question the upshot would be the same. I take the same view. It is strictly unnecessary to decide the point. But we have had full argument, and it may be of value for future contests under s 9 to express an opinion. Mine will come as no surprise. It is that the standard of proof which a claimant must meet on an issue under s 9 is the well-established standard for forum conveniens disputes, of a good arguable case. That is because, as I have explained, s 9 should not be treated as a fresh stand-alone provision of unique character but rather as a tailored modification of the established regime, and it does not purport to alter the standard of proof.

I see no good reason for adopting any rigid rule about the nature of the evidence that either party will be required to adduce on a contest under s 9. It is sufficient to say that the court must be satisfied of the matters specified in the section, that the legal burden of doing so rests on the claimant, and that the claimant has a duty of full and frank disclosure at the without notice stage. Whether the evidence adduced in a given case is enough to meet these requirements will depend on the circumstances of the case. This will ordinarily be a matter for the assessment of the Judge, and not apt for review on an appeal. Of course, there could be a challenge on grounds of perversity, but that is not alleged here. I certainly do not think it possible to say that an application under s 9 must fail unless the claimant provides full details of his “business interests here and overseas”, as suggested on behalf of the defendants.

As for expert evidence, it is true it is that on conventional principles an English court cannot take judicial notice of foreign law; where disputed, the content of the foreign law must normally be proved as a fact, by expert evidence: Dicey, Conflict of Laws Rule 25. In the absence of evidence, the court may apply English law, adopting what is sometimes called “the presumption” that the foreign law is no different: Dicey Rule 25(2), para 9-025. But there are cases in which courts have interpreted and applied foreign law without expert evidence: Dicey para 9-009. And “the recent practice of the English courts suggests that the default application of English law, where foreign law is not proved, is not unqualified”: ibid., 9-026. The modern cases show much greater practicality and flexibility in their approach to these principles.

There are cases in which no evidence of foreign law has been adduced, but the court has declined to apply the so-called “presumption”. This is a point that comes up from time to time in libel claims about foreign publication, which are still subject to the common law requirement of “double actionability”: Private International Law (Miscellaneous Provisions) Act 1995, s 13. As long ago as 1991, the Supreme Court Procedure Committee referred to the presumption as “quite unrealistic and curiously egocentric in the post-imperial age”. In Al Misnad v Azzaman Ltd [2003] EWHC 1783 (QB) [37] Gray J declined to order summary disposal of a libel claim that was mainly concerned with publication in North Africa and the Middle East on the basis of the “presumption”. In like manner, Jay J held in this case that it would be “absurd” to assume, for instance, that Californian law offers the same remedies as exist under the Defamation Act 2013 and has the same costs regime. Again, this seems to me legitimate. Since the hearing before us the Supreme Court has analysed this topic in some depth in FS Cairo (Nile Plaza) LLC v Brownlie [2021] UKSC 45, [2021] 3 WLR 1011. As Lord Leggatt put it at [149], “[t]he essential point is that the presumption of similarity is only ever a basis for drawing inferences about the probable content of foreign law in the absence of better evidence.”

On the other hand, the reality is that English lawyers know or can safely infer quite a lot about some foreign laws. As Sir Michael Tugendhat observed in Ahuja [28], “the law in the USA is, in its broad outline, very well known to English libel lawyers and others…”. An example of this is the principle of federal law, that a public figure suing in defamation must prove express malice (New York Times v Sullivan [1964] 376 US 254). That proposition is pleaded in the Particulars of Claim here, as part of the claimant’s case of malice. It is so familiar that before the Judge Mr Price said that “if backed into a corner” he would accept it. Although there was no evidence about US procedural law, Jay J felt able to rely on “common sense” and “sensible inferences” to conclude that California would accept jurisdiction over claims by this claimant against these defendants in respect of all US publication. I can see nothing wrong with that. Again, the recent decision in Brownlie assists. At [148] Lord Leggatt said this:

“… it should not be assumed that the only alternative to relying on the presumption of similarity is necessarily to tender evidence from an expert in the foreign system of law. The old notion that foreign legal materials can only ever be brought before the court as part of the evidence of an expert witness is outdated. Whether the court will require evidence from an expert witness should depend on the nature of the issue and of the relevant foreign law.”

Lord Leggatt (with whom the other members of the Court all agreed) referred by way of example to the ready availability of legal texts and translations online. But he did not suggest this was exhaustive of the sources of evidence on which a court might draw.

I do not think there can be a single answer to the question of whether either party to a dispute under s 9 needs to adduce expert evidence about the law of one or more foreign jurisdictions. There may be many cases in which that is at least desirable, and some in which it is essential. But there may be cases in which the law can be proved by other evidence, or by inference, or the presumption of similarity can be relied on. All will depend on the facts and circumstances of the particular case.

It is possible, moreover, to think of cases in which the s 9 comparison could be carried out with confidence without resort to any evidence, presumption, or inferences as to foreign law. This observation applies, it seems to me, whatever view is taken about where the burden of proof lies.

Applying these conclusions to the facts and circumstances of the present case, there is no good reason to disturb the Judge’s decision. The Judge was mildly dissatisfied with the extent of the claimant’s disclosure on this topic, but he concluded that there was enough to enable him to reach a decision. I do not think we can or should hold that this conclusion was wrong. It is not alleged that there was a failure of full and frank disclosure. The defendants’ contention that in this case the claimant’s approach impaired their ability to adduce factual or expert evidence to show that California was a suitable alternative jurisdiction is not persuasive. They are investigative journalists who deliberately targeted serious allegations at the claimant, and the principal defendants are domiciled in California. In my view, the Judge’s approach to the claimant’s reticence was open to him both as a matter of law, and as a matter of judicial assessment in the circumstances of the case.

According to Jay J’s findings on the facts, which cannot be impugned, the claimant is not an “international businessman” of the kind the defendants sought to suggest; the present case is rather closer to the hypothetical example I have just given. The defendants published here, on a substantial scale, some seriously defamatory allegations about a British citizen who was domiciled and resident here, and whose reputation was centred here, not in Israel or anywhere else. The only real candidate alternative jurisdiction was the US, where the claimant had no significant reputation. For my part, I think these considerations might well have been sufficient on their own to justify the Judge’s conclusion that the claimant in this case was not a “libel tourist” and that England and Wales was clearly the most appropriate place in which to bring the claim. As Sir Michael Tugendhat pointed out in Ahuja [31], “the extent of publication in different jurisdictions may have little bearing on where the claimant’s reputation mainly lies, and on where that reputation has been most seriously damaged”.

It is worth adding, however, that this was a long way from the paradigm case, in which a foreigner is sued in this jurisdiction even though (as, for instance, in Berezovsky) the scale of foreign publication dwarfs that in this jurisdiction. According to the evidence which the defendants placed before the Judge, Forensic News’ contributors write for a predominantly US-based audience. The cumulative visitor statistics for all visits to the Forensic News website for the period from 20 May 2019 to 23 November 2020 showed some 680,000 visits of which 75% were from the US. Visitors from the UK made up 5.17% of the whole, so about 35,000. But a more detailed analysis of the defendants’ figures conducted on behalf of the claimant was before the Judge. This gave the number of page views and unique page views of each of the Forensic News publications and compared the extent of publication in England and Wales with that in California, New York and Pennsylvania. This is relevant because these are the defendants’ jurisdictions of domicile and the only places ever identified as alternative venues for the claimant’s libel claims. On this basis, as the Judge put it “England and Wales is in first place on four occasions and California in first place on four others (with England and Wales coming in either second or third).” The absolute number of unique page views of all 8 publications in England and Wales is 2,479. The total for California is 4,169. The total figures for the publications complained of as libels are 2,180 and 3,594.

Expert evidence was, as I see it, not essential as a matter of principle. But once the defendants had identified California as an alternative jurisdiction available to the claimant, they bore a corresponding evidential burden. The Judge was entitled to place weight on their failure to address by expert evidence or any other form of evidence the question of whether the claimant could obtain vindication by means of an action in California, where he had little reputation, in respect of publication here, where his reputation was centred.

For all these reasons I would dismiss the defendants’ appeal in respect of the Judge’s decision to allow the claimant to serve the defendants with his claims in libel.

By concession, the MOPI claims satisfied the Gateway Requirement by virtue of PD6B 3.1(2) and 3.1(21) (breach of confidence or misuse of private information where detriment suffered in the jurisdiction). The Judge held that the main MOPI claim failed the Merits Test, but that the claimant had a tenable claim for misuse in respect of the four photographs of which he complained. Taken in isolation, that claim would have been too trivial to justify the grant of permission to serve outside the jurisdiction, but taken in conjunction with the libel claims it was appropriate to exercise his residual discretion by granting permission. The grant of permission was therefore parasitic on permission to serve the claims in libel. There is no independent challenge to the Judge’s decision to grant permission to sue in respect of the photographs. It follows that his order in that respect will stand.

The GDPR is an EU regulation. It applies to this case because the claim was brought before 31 December 2020, when the post-Brexit transition period came to an end and the GDPR ceased to apply directly. Our decision is not of merely historic interest, however. That is for at least three related reasons: the claim extends to processing after 31 December 2020; some of the continued processing complained of has allegedly taken place in the EU; and the content of the GDPR has been adopted, with appropriate amendments, as part of English law in the form of “the UKGDPR”. This is the effect of the Data Protection Act 2018 and SI 2019/419: see R (Open Rights Group) v Secretary of State [2021] EWCA Civ 800, [2021] 1 WLR 3611 [5], [12-13]. For ease of reference I shall refer simply to the GDPR.

The original Particulars of Claim alleged that the defendants as data controllers or data processors of his personal data acted in breach of GDPR Article 5(1)(a) (processing must be fair, lawful and satisfy a condition under Article 6), Article 5(1)(d) (data must be accurate), Article 10 (processing of special category data must have a lawful basis), and Article 44 (international transfers of personal data). Although these claims are broad, there is some overlap with the claims in libel. For example, the Particulars of Claim allege that publication 2 bore the defamatory imputation that the claimant “is guilty of receiving illegal kickbacks and other corrupt payments from the Russian state under the façade of security consultancy work for Sochi Airport during the Olympic Games held in 2014”. The claimant’s case under Article 5(1)(d) of the GDPR alleged that publication 2 was inaccurate because he “did not receive any kickbacks or illegal payments from the Russian state in relation to the Sochi airport contract” (my emphasis).

English data protection law contains nothing corresponding to s 9 of the Defamation Act 2013. The court’s jurisdiction to entertain a data protection claim against a defendant based outside England and Wales is determined by reference to the GDPR, and the general principles identified in section III above. The claimant’s claim was brought on the basis that the processing of which he complains falls within the ambit of Article 3 of the GDPR, which provides so far as relevant:-

“Territorial Scope

(1)

This Regulation applies to the processing of personal data in the context of activities of an establishment of a controller or a processor in the Union, regardless of whether the processing takes place in the Union or not.

(2)

This Regulation applies to the processing of personal data of data subjects who are in the Union by a controller or processor not established in the Union, where the processing activities are related to:

(a)

the offering of goods or services, irrespective of whether a payment of the data subject is required, to such data subjects in the Union; or

(b)

the monitoring of their behaviour as far as their behaviour takes place within the Union.”

Article 3(1) is in similar terms to those of Article 4(1)(a) of the GDPR’s predecessor, the Data Protection Directive 95/46 (“the Directive”). And Recital (22) to the GDPR contains relevant wording, that also appeared in Recital (19) of the Directive: “[e]stablishment implies the effective and real exercise of activities through stable arrangements. The legal form of such arrangements, whether through a branch or a subsidiary with a legal personality, is not the determining factor in that respect.”

Although there appears to be no authority as yet on the meaning of Article 3(1), either in the CJEU or in any Member State, there are three CJEU decisions on the meaning of the term “establishment” in that context: Google Spain SL v Agencia Espanola de Proteccion de Datos (AEPD) (Case C-131/12) [2014] QB 1022 (“Google Spain”); Weltimmo sro v Nemzeti Adatvedelmi es Informacioszabadsag Hatosag (Case C-230/14) [2016] 1 Wlr 863 (“Weltimmo”); and Verein fur Konsumerenteninformation v Amazon EU Sarl (Case 191/15) [2017] QB 252(“Amazon”). Principles to be derived from these cases include the following.

The jurisprudence relating to this aspect of the Directive is also summarised in the European Data Protection Board’s Guidelines 3/2018 on the Territorial Scope of the GDPR, which are not binding but relevant. They refer to Article 3(1) as the “establishment criterion”. They state that the presence of a single employee may be sufficient to satisfy the “stable arrangement” threshold, if there is sufficient stability; but mere presence to that extent will not trigger the GDPR if the processing in question is not carried out in the context of the activities of the EU-based employee. They emphasise that “The threshold for ‘stable arrangement’ can actually be quite low when the centre of activities concerns the provision of services online.”

The Directive contained nothing resembling Article 3(2) of the GDPR, on which no authority appears to exist so far; but Recitals (23) and (24) to the GDPR throw light on Articles 3(2)(a) and (b) respectively. They state as follows:-

“(23)

In order to ensure that natural persons are not deprived of the protection to which they are entitled under this Regulation, the processing of personal data of data subjects who are in the Union by a controller or a processor not established in the Union should be subject to this Regulation where the processing activities are related to offering goods or services to such data subjects irrespective of whether connected to a payment. In order to determine whether such a controller or processor is offering goods or services to data subjects who are in the Union, it should be ascertained whether it is apparent that the controller or processor envisages offering services to data subjects in one or more Member States in the Union. Whereas the mere accessibility of the controller’s, processor’s or an intermediary’s website in the Union, of an email address or of other contact details, or the use of a language generally used in the third country where the controller is established, is insufficient to ascertain such intention, factors such as the use of a language or a currency generally used in one or more Member States with the possibility of ordering goods and services in that other language, or the mentioning of customers or users who are in the Union, may make it apparent that the controller envisages offering goods or services to data subjects in the Union.

(24)

The processing of personal data of data subjects who are in the Union by a controller or processor not established in the Union should also be subject to this Regulation when it is related to the monitoring of the behaviour of such data subjects in so far as their behaviour takes place within the Union. In order to determine whether a processing activity can be considered to monitor the behaviour of data subjects, it should be ascertained whether natural persons are tracked on the internet including potential subsequent use of personal data processing techniques which consist of profiling a natural person, particularly in order to take decisions concerning her or him or for analysing or predicting her or his personal preferences, behaviours and attitudes.”

The EDPB Guidelines are of some assistance in relation to Article 3(2), to which they refer as “the targeting criterion”. The Guidelines point out that Article 3(2) focuses on what the “processing activities” are “related to”, which can be considered on a case-by-case basis. They emphasise that “the fact of processing personal data of an individual in the Union alone is not sufficient … The element of ‘targeting’ individuals in the EU must always be present in addition.”

In relation to Article 3(2)(a), the Guidelines set out a non-exhaustive list of nine factors which could be taken into consideration, possibly in combination with one another, when considering whether a data controller is “offering…goods or services” in a relevant fashion. The list, with numbering added by me, is this:

“[1] The EU or at least one Member State is designated by name with reference to the good or service offered;

[2] The data controller or data processor pays a search engine operator for an internet referencing service in order to facilitate access to its site by consumers in the Union; or the controller or processor has launched marketing and advertisement campaigns directed at an EU country audience;

[3] The international nature of the activity at issue, such as certain tourist activities;

[4] The mention of dedicated addresses or phone numbers to be reached from an EU country;

[5] The use of a top-level domain name other than that of the third country in which the controller or processor is established…;

[6] The description of travel instructions from one or more other EU Member States to the place where the service is provided;

[7] The mention of an international clientele…;

[8] The use of a language or currency other than that generally used in the trader’s country…;

[9] The data controller offers the delivery of goods in the EU Member States.”

The EDPB distinguishes cases exhibiting indicia of this kind from those in which “goods or services are inadvertently or incidentally provided to a person on the territory of the Union”, when “the related processing of personal data would not fall within the territorial scope of the GDPR.”

Turning to Article 3(2)(b), the Guidelines express the following views:

“The EDPB does not consider that any online collection or analysis of personal data of individuals in the EU would automatically count as “monitoring”. It will be necessary to consider the controller’s purpose for processing the data and, in particular, any subsequent behavioural analysis or profiling techniques …

The application of article 3.2(b) … could therefore encompass a broad range of monitoring activities, including in particular:

-Behavioural advertisement

-Geo-location activities, in particular for marketing purposes

-Online tracking through the use of cookies …

-Personalised diet and health analytics services online

-CCTV

-Market surveys and other behavioural studies based onindividual profiles

-Monitoring or regular reporting on an individual’s healthstatus.”

Article 79 of the GDPR is headed “Right to an effective judicial remedy against a controller or processor”. Article 79(1) confers on each data subject a right to an effective judicial remedy “where he or she considers that his or her rights under this Regulation have been infringed …”. Article 79(2) provides that proceedings against a controller or processor shall be brought before the courts of the Member State where the controller or processor has an establishment, or (subject to immaterial exceptions) where the data subject has his or her habitual residence.

Relevant aspects of the evidence included those summarised at [69] above. It was also in evidence that 8.9% of readers of the defendants’ website whose location could be identified were in the UK or EU, with a further 8.95% of unascertained location. The website was made available free of charge, but relied on various sources of income to sustain itself. In round terms, its income from inception in May 2019 to November 2020 was US$50,000. Of this, more than 80% came from paid subscriptions via a platform called Patreon, which gave early access to stories and various other benefits. Another 10% came from one-off donations, about 5% from sales of branded merchandise. About 1% came from Google advertisements and sponsorship.

For the first 14 months of the website’s operation, subscriptions and donations via Patreon were denominated in US$. But this changed from 7 August 2020, when the first defendant tweeted, and the second defendant re-tweeted

“Everyone in the UK or EU can now pledge to Patreon in their local currency of Euros or Pounds. This will help prevent you paying extra conversion fees from your bank.”

Mr Doris’ statement told the court that after this date there were three Patreon subscriptions in Euros and three in Sterling. There was no evidence as to the geographical breakdown of those who subscribed in Dollars before 7 August 2020.

The claimant’s case on jurisdiction was that his data protection claims satisfied the Gateway Requirement by virtue of GDPR Article 79 and PD6B 3.1(2) and 3.1(20) (claim under an enactment, not covered by any other ground).

His case on Article 3(1) was that the publications were in English; the website solicited donations in Sterling and in Euros; the website included a “store” with its own branded merchandising, accepting shipping addresses in the UK; and a tweet sent on 7August 2020 invited pledges to Patreon, a subscription platform, from readers in the UK and the EU. It was contended that the Patreon subscriptions were “stable” arrangements, inasmuch as they remained in place until cancelled.

The case advanced in respect of Article 3(2)(a) was that the defendants, as data controllers, offered goods and services to readers in the UK irrespective of payment. As for Article 3(2)(b), it was argued that the website placed cookies on readers’ devices and processed their personal data using Facebook and Google analytics for the purpose of targeting advertisements, with Facebook Ireland Ltd and Google Ireland Ltd operating as the registered joint data controller. Further, Mr Callus submitted that the defendants were collecting and obtaining data about the claimant, and were monitoring his behaviour within the UK and the EU with a view to making publishing decisions.

The Judge accepted the claimant’s case on the Gateway Requirement. He would also have held that these claims satisfied the general Forum Test: the Courts of England and Wales were clearly the right place to bring any such claim. But he held that the claims failed the Merits Test. He declined Mr Price’s invitation to decide the jurisdiction issue definitively rather than determine whether the claimant had a real prospect of success. Instead, he asked himself whether the claimant had demonstrated a real prospect of showing that his claims fell within the territorial scope of the GDPR, as defined by Article 3. He held that the claimant had fallen short of doing so.

The Judge considered the facts that the defendant had no branch or subsidiary, employees or representatives in the UK to be relevant but not determinative. The presence of a more than minimal readership was of marginal relevance only. He identified the real question under Article 3(1) as “whether, taking the claimant’s case at its reasonable pinnacle … he has the sufficient makings of an argument on ‘stable arrangements’ to enable him to pass through the merits portal”. The case on Patreon amounted to “less than a handful of UK subscriptions to a platform which solicits payment for services on an entirely generic basis and which in any event can be cancelled at any time”. There was nothing else of substance, and therefore no tenable case that the defendants had an “establishment” in the EU (Article 3(1)). It was unnecessary to consider whether the processing at issue was “in the context of” any activities of such an establishment.

Nor was it realistic, held the Judge, to suppose that the claimant could prove that the processing activities complained of were “related to … the offering of goods or services … to such data subjects in the Union” within the meaning of Article 3(2)(a). There was nothing to suggest that the first defendant was targeting the UK for the provision of goods and services. The fact that this country was a potential shipping destination for merchandise was not enough. A cursory examination of the EDPB’s list of indicia served to demonstrate how far short the claimant fell of meeting the quoted criterion. What he had to show was that the offer of goods and services relied on was “related to” the defendants’ core activity. That was a narrower and stricter requirement than the phrase “in the context of”, and it was not satisfied.

Turning to Article 3(2)(b), the Judge accepted that the claimant had an arguable case that the defendant used “cookies etc.” for the purpose of behavioural profiling or monitoring, but he held that this was “purely in the context of directing advertisement content”. There was no evidence that such monitoring was “related to” the processing complained of, namely the defendants’ use of the internet as an investigative tool to support their journalism.

The claimant contends that, contrary to the Judge’s conclusions, he had an arguable case that the GDPR applied to the conduct complained of. He puts forward what, on analysis, are three grounds of challenge.

Mr Price appeared at one stage to submit that we should decide these questions definitively – “grasp the nettle”, as it is commonly said. That is a submission rejected by the Judge, and I do not think the defendants were entitled to advance it on appeal, having filed no respondent’s notice. For his part, Mr Callus submitted that we should hesitate to decide a point in a developing area of the law. That is a facet of the Part 24 jurisprudence. We were shown no authority to support the view that it is part of the approach under CPR 6.37. At any rate, that is not how the grounds of appeal are framed. I approach this part of the appeal on the basis that the issue is whether it is shown that the Judge was wrong to hold that the claimant’s case on the three grounds I have identified failed the Merits Test. That said, it is certainly relevant to note that there is no authority that assists directly on the meaning and effect of Article 3(1), and nothing other than the EDPB Guidelines to help us with Article 3(2). According to Mr Callus, this is understood to be the first case anywhere in the EU on the territorial applicability of the GDPR under that provision.

After careful consideration, I find myself in disagreement with the Judge. I have concluded, for the reasons and to the extent that follow, that each of the three lines of argument outlined at [93] above satisfies the Merits Test. I would therefore allow the cross-appeal on the data protection issue.

It is very likely that the CJEU authorities on the meaning of “establishment” in Article 4 of the Directive would guide a decision on the meaning of the same word in Article 3(1) of the GDPR. I consider it arguable that the defendants had an “establishment” in the EU within that meaning at the material times. I share the doubts expressed by Arnold LJ when granting permission to appeal on this point. However, as Mr Callus has emphasised, the authorities set a low bar, and all the claimant has to do at this stage is to persuade the court that it is not fanciful to think that his case crosses that low bar.

Recital (23) suggests that a critical consideration here is whether the data controller or processor “envisages offering services to data subjects in one or more Member States in the Union”. On the evidence, the defendants did more than merely making their journalism accessible over the world wide web. They intended to make their output available in the UK and EU, and succeeded in attracting a more than minimal readership. In due course they expressly solicited subscriptions from within the UK and EU, via the Patreon platform. They succeeded in securing three subscriptions in sterling and three in Euros. This may be “minimal” activity but nothing more is required, according to the authorities. The offer and acceptance of subscriptions in these local currencies is arguably a “real and effective” activity that is “oriented” towards the UK and EU. Jay J said that the journalistic endeavour was not so oriented but that, as it seems to me, goes more to the separate question of whether the journalistic processing complained of arguably took place “in the context of” the activities of an “establishment”. That is not an issue on this appeal. The Judge did not decide it, and the defendants have not filed a respondent’s notice seeking to uphold his decision on alternative grounds.

The key issue under Article 3(1), as I see it, is whether the creation and use of the Patreon subscription facility demonstrates “stable arrangements”. Jay J considered it could not be said to do so, because the subscriptions could be cancelled at any time. I do not think the answer is so clear. There does seem to be a set of “arrangements” here. The CJEU authorities provide no criteria for deciding whether a given arrangement is “stable”. Unlike the Judge, I do not think the answer is necessarily dependent on what subscribers can do in principle, as opposed to what happens in fact. I see force in Mr Callus’ submission that, in the context of this online media publication, subscription arrangements of this kind should be viewed as stable in nature. The defendants’ evidence as to the funding arrangements for Forensic News shows that subscriptions are a major income stream, contributing some 85% of total income. I strongly suspect that once established a subscription will tend to be maintained. Certainly, there is no evidence to the contrary.

The Judge considered it important that the tweet of 7 August 2020 post-dated the publications sued upon. I do not consider that is a sufficient basis on which to take a different view, for two reasons. First, the tweet is merely an item of evidence as to the orientation of the defendants’ activity towards readers in the UK and EU. We know from other evidence that some 9% of readers or more came from this territory. There is no way, on current evidence, of excluding the inference that many of these subscribed in dollars when that was the only means of doing so. The language of the tweet implies that this may well have been the case. Secondly, this is an artificial distinction to draw when the publications complained of were and are continuing. According to the principles identified above, each fresh publication gives rise to a separate and distinct cause of action.

On a superficial reading, there could be a short answer to the claimant’s arguments on Article 3(2)(a). This limb of the Article refers to the processing of “personal data of data subjects” which is related to “the offering of goods or services … to such data subjects” (the emphasis is mine). The language could be read as indicating a legislative intention to ensure that the Regulation should apply to processing of an individual’s personal data that has some relationship with offering goods or services tothat individual. That is not the position here. But the case has never been argued on that basis, so I approach the issue - as everyone appears to have done so far - on the footing that Article 3(2)(a) applies to the processing of personal data of data subjects who are in the Union whether or not they are the same individuals as those to whom the goods or services are offered, providing the two activities are “related to” one another.

Even on that basis, the Judge was right in my opinion to say that on no reasonable view is the journalistic processing of which the claimant complains “related to” the offer of goods to data subjects within the Union. Not only do the two activities relate to different personal data they are also, in substance, separate and distinct functions. No doubt they have a commercial relationship with one another, but I do not accept that is even arguably enough. This approach to the interpretation of Article 3(2)(a) would give it a breadth and scope beyond what is indicated by the words used and the Guidelines. I do however consider it arguable that the journalistic processing complained of is “related to” an offer made by the defendants to data subjects in the Union to provide them with services in the form of journalistic output.

In my view, and for similar reasons, it is also arguable that the case falls within Article 3(2)(b). I would reject Mr Callus’ argument that the publication of articles containing the claimant’s personal data is itself a form of “monitoring” within Article 3(2)(b), such that any processing “related to” the publication is within scope. This is artificial. It distorts the meaning of the word “monitoring” – at least as it is used in this context - and belongs in the realms of the fanciful. But there is a more compelling case under Article 3(2), which runs as follows. Someone who uses the internet to collect information about the behaviour in the EU of an individual who is in the EU, and then assembles, analyses and orders that information for the purposes of writing and publishing an article about that behaviour in (among other places) the EU is thereby engaging in “… the monitoring of [the data subject’s] behaviour … within the Union” within Article 3(2)(b). The publication of personal data clearly is a form of “processing”. The preparatory activities are plainly integral to that processing. It follows that the GDPR applies in such a case on the footing that publication amounts to a “processing of personal data of [the data subject]” which is “related to” the monitoring.

Having regard to what is said in Recital (24) and the EDPB Guidelines, I do not think this is fanciful. The mere fact that the defendants created a collection of personal data relating to the claimant’s behaviour in the EU might not be enough. But what they are alleged to have done is to assemble, analyse, sort, and reconfigure such data, and then publish the result in articles including (among others) one entitled “The Walter Soriano files”. I think it is arguable that those activities fall within the meaning of “monitoring”, and within the scope of the EDPB’s notions of “behavioural analysis and profiling”. It may be debatable whether the notions of targeting or orientation towards the EU apply in this context. Mr Callus submits that the EDPB is wrong about that. But if they do they are arguably satisfied by “stable arrangements” as already discussed, which enable EU citizens to subscribe in their local currency to receive Forensic News’ journalistic output.

In conclusion, I add this. As these issues will, so it seems, need further and definitive consideration in this case it seems to me that the Information Commissioner should be invited to consider intervening to assist the court.

At common law, a claim in malicious falsehood has four ingredients: (1) a publication by the defendant about the claimant which (2) is false (3) was published maliciously and (4) caused special damage. Ingredients (2), (3) and (4) make this a more demanding tort to establish than defamation. Malice will be inferred “if it can be proved that the words were calculated to produce damage and that the defendant knew when he published the words that they were false or was reckless as to whether they were false or not”: Kaye v Robertson [1991] FSR 62, 67 (Glidewell LJ). Proof of malice is notoriously difficult. Nevertheless, it is quite common for a claimant to do as Mr Soriano has done and sue for both libel and malicious falsehood in respect of the same publications. Often, a claimant will rely on s 3(1) of the Defamation Act 1952 which exempts a person suing for malicious falsehood from the common law requirement to allege and prove special damage if the words on which the action is founded are

“(a)

calculated to cause pecuniary damage to the plaintiff and are published in writing or other permanent form; or (b) … calculated to cause pecuniary damage to the plaintiff in respect of any office, profession, calling, trade or business held or carried on by him at the time of the publication.”

The word “calculated” in this context means more likely than not: Tesla Motors Ltd v BBC [2011] EWHC 2760 (QB) [7].

The Defamation Act 2013 does not apply to this cause of action, and there is no equivalent of s 9 for malicious falsehood. So the court’s jurisdiction to try such a claim depends on the rather less demanding common law principles identified in section III above.

The malicious falsehood claim was brought against the first and second defendants only. The claimant’s case on their responsibility for publication was pleaded as part of his libel claim. His case on falsity was set out as part of his data protection claim. It is unnecessary to examine either. For reasons that will become clear, the focus of attention was on the issues of malice and damage.

The claimant’s case on malice had two limbs. The first was that from the outset the second defendant (for whose conduct the first defendant was vicariously responsible) had known the words complained of were false or was reckless as to their truth or falsity. The allegation - developed at some length in the Particulars of Claim - was that these defendants had conspired with Mr Drucker to exact retribution for the claimant’s refusal to discontinue the Israeli libel proceedings I have mentioned, by conducting a “campaign” of defamation, knowing that their allegations were false, or with reckless disregard for their truth or falsity.

The second limb of the claimant’s case on malice was that at some point after the first publication these defendants came by knowledge that rendered the original publication unsustainable, such that continued uncorrected publication was malicious. This category of case has been described in argument as one of “Loutchansky malice”, after the decision of this court in Loutchansky v Times Newspapers Ltd (No. 2) [2001] EWCA Civ 1805, [2002] QB 783, although the concept was well-established before that. In support of this limb of his case, the claimant relied on correspondence passing between his lawyer and the second defendant which was said to contain “definitive answers” to questions raised by the second defendant and to make the latter “unambiguously aware of the falsity of the allegation”. He relied, further, on the fact that the defendants continued publishing even after a report of the US Senate Intelligence Committee, which the defendants had seemingly believed would implicate the claimant in wrongdoing, turned out to have very little to say about him. The court was invited to assess these points in the context of the sheer number of falsities perpetrated, failures to check facts and the tone of the pieces.

The claimant’s Particulars of Claim went on to allege that he had suffered damage of various kinds. Paragraph 42 alleged that the allegations complained of were “calculated to have a pecuniary effect on his businesses and that of his companies” and “calculated to cause pecuniary damage to the claimant in respect of his role as a businessman.” It was said that the alleged malicious falsehoods targeted and predominantly concerned the claimant’s businesses, which were named and identified; and that “receipt of the allegations complained of would make it highly unlikely that the recipient would enter into business with the claimant or use his services.” Paragraph 57.3 alleged that the claimant had incurred “considerable expenses” in meeting the requirement of the compliance departments of Banks that he “rebut the false and defamatory allegations made in the publications”. But he has not asserted that he has suffered special damage. He relies on s 3(1) of the 1952 Act.

There was no dispute that the malicious falsehood claims satisfied the Gateway Requirement by virtue of PD3.1B(2) and (9). The Judge would have held that the courts of England and Wales were the most appropriate place to try these claims. There was no dispute that the claimant had a real prospect of proving publication and falsehood. But the claims were held to fail the Merits Test on the twin bases that the claimant had no real prospect of establishing malice and had failed to plead a tenable case under s 3 of the 1952 Act.

The Judge dismissed the first limb of the case on malice holding that:-

“90.… the Claimant’s primary case on malice is a synthetic artifice which has no basis in substance. …

93.

Overall, there is no evidence to support the proposition that the First Defendant was set up, effectively at the instance of Raviv Drucker … to launder the Claimant’s information into the public domain.”

The Judge regarded the case on Loutchanksy malice as “equally thin”. He observed that although the claimant’s lawyer had categorically denied a number of allegations on his behalf, he had “provided no detail and no evidential fortification”. It would not be right to treat the second defendant’s failure to reply to a subsequent letter as the basis for an inference that he knew his stories lacked evidential foundation. Although there was marginally greater force in Mr Callus’s point about the Senate Intelligence Committee’s report, the Judge had not been asked to consider the scope of the evidence it received, and its terms of reference were much narrower than the scope of the publications complained of.

At [98], the Judge adopted what Tugendhat J had said in Tesla Motors [66] about pleading a case under s 3(1): “…the claimant must … give particulars of the nature of the allegedly probable damage and the grounds for saying that it is more likely than not.” At [99], the Judge held that paragraph 57.3 of the Particulars of Claim, which represented the “high point” of the claimant’s case on loss, was “too vague an averment to amount to the giving of particulars” in accordance with the law as stated in Tesla Motors.

The first ground of appeal is a challenge to the dismissal of the claimant’s case on malice. Mr Callus does not seek to quarrel with the Judge’s conclusion on the first and main limb of that case. He accepts that conclusion was open to the Judge on the evidence. He contends, however that the Judge was wrong to hold that his case of Loutchansky malice was untenable. Mr Callus submits that the claimant has a real prospect of proving that category of malice in circumstances where there is unamended and continuing publication of all the articles online despite the correspondence relied on, and the provision of detailed draft Particulars of Claim setting out the inaccuracies and falsehoods. He adds that although the defendants have intimated an intention to plead a defence of publication in the public interest under s 4 of the 2013 Act, they have made no intimation that they would plead truth under s 2 of that Act.

The second ground of cross-appeal is that “there was an arguable case that the words complained of were calculated to disparage the claimant in his business held at the time of publication”, such that s 3 of the 1952 Act relieved him of the burden of pleading and proving special damage. It is submitted that the Judge’s application of the Tesla Motors test was wrong on the facts; the pleaded case met that standard and raised an arguable case. Alternatively, we are invited to hold that Tugendhat J was wrong in law, as the requirement he specified effectively re-imposes the common law duty to allege special damage, which is precisely what s 3 was aimed at avoiding.

I would reject the cross-appeal in respect of the malicious falsehood claim. Mr Callus’s skilful submissions have not persuaded me that the Judge erred on this issue.

It is trite law that malice is tantamount to dishonesty and not to be equated with carelessness, impulsiveness, or irrationality in arriving at a belief that the matter published is true; nor is partiality or even “gross and unreasoning prejudice” enough: Gatley on Libel and Slander, 12th ed, para 17.17 and cases there cited. And a party alleging malice must plead and prove facts which are more consistent with the presence of malice than with its absence; allegations of malice must go beyond that which is equivocal or merely neutral: see Alexander v Arts Council of Wales [2001] 1 WLR 1840 and the many other cases cited in Gatley (op. cit.) para 28.6 notes 16 and 17 of the Main Work and of the second Cumulative Supplement. The fact that a defendant continues to publish an allegation about the claimant after he has been told by someone acting for the claimant that it is untrue cannot of itself be sufficient to raise a probability of malice. Still less can malice be inferred from the mere fact of silence in the face of complaint or allegation of falsity.

Equally, although continued publication after action brought but without a plea of truth might in appropriate circumstances be evidence of malice, it does not in and of itself provide a ground for inferring a probability of malice. A defendant may fervently and sincerely believe that what he is publishing is true, and that its publication is in the public interest, without being in a position to plead and prove the former to the standards required in our courts. In those circumstances a public interest defence under s 4 of the 2013 Act would succeed if the belief was shown to be reasonable. If not, the defence would fail; but even then, it would not follow that the defendant would be found malicious.

Whether a claimant has a case of malice that passes the threshold tests I have mentioned and has a real prospect of success at trial is primarily a matter for the evaluation of the Judge at first instance, having regard to the particular facts alleged, any evidence put before the court, and the arguments advanced. The question in this Court is whether the Judge erred in law or made an assessment that can be seen to be wrong. In my judgment, Jay J’s reasoning and conclusions cannot be impugned. Mr Callus has argued in this Court that the claimant was “entitled” to have his denials featured in the continuing publication, and to “a fair presentation” of any material said to be published in the public interest. Mr Callus has also submitted that account should be taken of the failure of the second defendant to put in a witness statement of his own, rebutting the allegation of malice, on this interim application. I do not believe that either of these points is capable of making good the case of Loutchansky malice. The first falls into the trap of confusing unfair or irresponsible publication with malice. The second is far too frail and tenuous a basis on which to allow a serious allegation of this kind to proceed.

It follows that the cross-appeal on malicious falsehood must fail, and it is unnecessary to discuss the claimant’s second ground of appeal on this part of the case. I do not believe it would be useful to do so, beyond the following brief observations. There are some obvious difficulties with the way the case on damage is pleaded. These include the fact that the averments in paragraph 42 of the Particulars of Claim are mainly if not exclusively allegations that the publications were likely to cause financial harm to the claimant’s businesses. Those – on his own pleading and evidence - are corporate entities, and they are not claimants. To the extent that the Particulars of Claim alleged a likelihood of financial harm to the claimant himself, they were vague in the extreme and to my mind fell well short of the explanatory requirement set out in Tesla Motors.

For my part, I am not immediately attracted by the argument that the imposition of this requirement is tantamount to undoing s 3 but that, and the wider question of whether the Tesla Motors test is too exacting, are issues that will have to be resolved, if at all, on another occasion. If and when that takes place, account will need to be taken of what was said about the issue when the Tesla Motors case reached this court: [2013] EWCA Civ 152 [37] and [45] (Moore-Bick LJ, with whom Rimer and Maurice Kay LJJ agreed).

For the reasons I have given, I would dismiss the appeal against the Judge’s decision on the libel claim and the cross-appeal against his decision on the malicious falsehood claim. I would allow the cross-appeal against the Judge’s decision on the GDPR claim.

I agree.

I also agree.