MEDIA & COMMUNICATIONS LIST
Royal Courts of Justice
Strand, London, WC2A 2LL
Before :
THE HONOURABLE MR JUSTICE NICKLIN
Between :
(1) Richard (Raziel) Davidoff (2) Hanni (Hannah) Davidoff (3) Tamara Davidoff (4) Debby Davidoff (5) ABC Block Management Limited (6) ABC Hendon Limited | Claimants |
- and - | |
Google LLC | Defendant |
William Bennett KC (instructed by Patron Law) for the Claimants
The Defendant did not attend and was not represented
Hearing date: 25 May 2023
Approved Judgment
This judgment was handed down by the Judge remotely by circulation to the parties and their representatives by email and by release to The National Archives. The date and time for hand-down is deemed to be 2pm on 28 July 2023.
The Honourable Mr Justice Nicklin :
By Part 8 Claim Form issued on 15 February 2023, the Claimants seek Norwich Pharmacal relief against the Defendant (“the Application”).
A: The parties and the Trustpilot Reviews
The Fifth and Sixth Claimants are companies that operate a property management/estate agency business under the name ABC Estates in North London. They claim to have been the victims of what they call “fake reviews” posted by several people on Trustpilot (“the Reviews”). Previously, the Claimants have been granted a Norwich Pharmacal order against Trustpilot which has enabled them to identify the email addresses associated with the accounts that had posted the Reviews. Each email account was registered with the Defendant’s Gmail platform. These email addresses were used by the people who posted the Reviews to register with Trustpilot. The purpose of this further Norwich Pharmacal application is to attempt to identify the individuals who control or operate these email addresses and, ultimately, to seek to demonstrate that these are the individuals who posted the Reviews.
As part of the evidence, the Claimants have provided copies of the Trustpilot page on which the Reviews were posted. Overall, ABC Estates has received a two-star rating of “poor”. This score is based on a total of 28 reviews. 93% of the reviews (i.e. 26 of the 28) gave ABC Estates a one-star rating. Two reviews gave ABC Estates a five-star rating. These summary statistics are provided at the top of the Trustpilot review page, with individual reviews being displayed below. Visitors to the website can choose to have the reviews displayed based on relevance or date. The default setting is relevance. A visitor to the website would have to scroll down to read the reviews that are displayed.
I have set out, in a table in the Annex to this judgment, details of each Review relied upon by the Claimants. The Claimants have identified, through previous applications, the IP address from which each Review was posted. The geographic location associated with an IP address can readily be established online. This may indicate the location of the person when s/he posted the relevant Review. However, if the person uses a virtual private network (“VPN”), this can generate a different IP address (and therefore different location). The table shows only those Reviews which, by the time of the hearing, were still pursued in the Application. The Claimants have removed some targets since issuing the Application.
B: Previous claims for defamation brought in respect of anonymous reviews
In 2021, the Claimants brought a claim for libel against two individuals, Dhir Doshi and Thomas Govan, in relation to anonymous online reviews alleging fraud and dishonesty. The proceedings were resolved by agreement with the payment of damages and costs and with the defendants joining in a statement in open court apologising to the Claimants. The Claimants can therefore point to previous instances where they have been the target of anonymous online defamatory publications. Mr Doshi and Mr Govan have assured the Claimants that they are not responsible for the Reviews.
C: Evidence in support of the Application
The Application was originally supported by the witness statement of the Claimants’ solicitors, Mark Lewis, dated 15 February 2023. Mr Lewis set out the following information that the Claimants had obtained in respect of the Reviews:
“The [Reviews] may be grouped by themes – and indeed the strikingly similar language – which is used in each case. There are allegations of (a) rudeness, aggression and incompetence; (b) aggression to a female relative; (c) references to undertaking research about [the Claimants] (in particular Richard Davidoff) online. (In one case, the review gives the wrong first name as ‘David’, which would be unlikely is the person had in fact undertaken the Google search suggested.) The majority of these reviews are from accounts who have had only… one activity ever, which is to publish the single defamatory review. Secondly, where the accounts have posted statements about other businesses, on each occasion the posts are also one-star reviews ostensibly calculated to destroy a business activity. The account of ‘Lisa Mathieson’ has in fact posted two such reviews, both about the Claimants.
…
The Claimants intend to commence proceedings for libel and/or malicious falsehood, in order to obtain damages and an injunction to prevent the ongoing campaign. To that end the Claimants seek the relief sought in order to identify the individual(s) who have posted the [Reviews].”
Mr Lewis gave the following further details under a heading “Causes of action”:
“The Reviews are defamatory of both the Companies and my individual clients. Richard Davidoff is named in the review:
It is the claimants’ case that each of the individual claimants is closely identified with the company through their work: this is a family business, which largely operates in and around Jewish communities in North London. Each of the individual claimants drives an ABC Estates branded car. Moreover, these reviews are on sites designed to be looked at by those seeking to do business with the claimants.
The reviews contain a number of serious and highly defamatory allegations repeatedly published to relevant large audiences and are likely to cause the claimants serious harm and serious financial loss.
The Fake Reviews also constitute malicious falsehoods. They are false, fabricated statements which Unknown person(s) know are untrue, but which are calculated to cause pecuniary damage to the Companies and to my individual clients…”
No further evidence was provided with the Application Notice. The Claimants provided no actual evidence of any serious harm to their reputation being caused by the Reviews (as required by s.1 Defamation Act 2013). Mr Lewis’ statement on this point was nothing beyond assertion in the most general terms.
In a second witness statement, dated 19 May 2023, Mr Lewis provided the following further information:
Three of the Reviews had, since their original publication been removed (or ‘filtered’) by Trustpilot (Reviews 3, 4 and 6), suggesting that Trustpilot considered that there was something suspicious about the Reviews.
As to identification of the Claimants in the Reviews (an essential ingredient of a cause of action for defamation – see further [48] below), Mr Lewis added:
“The [Reviews] concern ‘ABC Estates’ and are directly linked to the company website ‘www.abcestates.co.uk’, this is a trading name for ABC Block Management Limited and ABC Hendon Limited, the fifth and sixth claimants respectively. Similarly, the first through the fourth claimants are, as stated, members of the same family who run and manage the Companies which are known to be family businesses. They each drive ‘ABC Estates’ branded vehicles, and are known to the Jewish communities in North London. The claimants are therefore all synonymous with one another”
As to serious financial harm (required to be demonstrated by a company that trades for profit in a claim for defamation – see [51]-[53] below), Mr Lewis added:
“The Fifth and Sixth Claimants cannot point to any specific business which they have lost as a result of the publications. Their case is that potential clients would visit Trustpilot in order to read reviews about them before deciding whether to engage them. If a potential client read any one of the reviews in issue, it is likely that the potential client would therefore not instruct the Fifth or Sixth Claimants. It is inferred that this must have occurred on several occasions. The purpose of Trustpilot is to enable such decisions to be made with the knowledge provided by the website. The consequential loss of contracts would cause serious financial loss.”
Finally, Mr Lewis exhibited the terms and conditions of the Trustpilot website which prohibit creating false accounts.
The evidence that the Claimants have obtained, concerning IP addresses connected to the Reviews does, at least in part, support their case that some of the Reviews have been posted by the same person. The IP addresses used by the users named Lisa Matherson, Michelle Stonefield and Sophie Adler are all connected, being the same IP address for different posts or IP addresses that are geographically located very close to each other. The IP addresses used by the users named Anthony Redfield and Stephen Michaels are the same and these users and those named Sharon Macenzie and Andrea Luckovic all share IP addresses linked to Tamworth.
On 18 May 2023, the Claimants’ solicitors sent a letter by email to each of the target Gmail addresses, identifying the Review to which the Claimants took objection and asking whether each individual would identify him/herself. Mr Bennett KC told me at the hearing that there had been no response to these emails.
D: Directions for a hearing of the Application
Ordinarily, Norwich Pharmacal applications are heard and determined by the Assigned Master. Upon reviewing the Application, I directed that a Judge of the Media & Communications List would hear the Application because of the importance of some of the issues that arise. In my Order of 24 February 2023, I explained:
The use of the Norwich Pharmacal jurisdiction to obtain information to identify those who have published material online anonymously engages the Article 8/10 rights of the anonymous posters (and, depending upon the circumstances, potentially in addition the Article 10 rights of the respondent): see Standard Verlagsgesellschaft mbH -v- Austria (No.3) (7 December 2021 Application No. 39378/15). It is arguable that the exercise of the Norwich Pharmacal jurisdiction in such cases by the English Court needs to be reviewed to ensure that the Court has properly considered the engaged Convention rights before making any order.
Pending further argument, it appears to me arguable that the Court needs first, to assess the interference with the Article 8 (reputational) rights of the Claimants said to be occasioned by the publications complained of, second to assess the interference with the Article 10 rights of the anonymous posters in making a Norwich Pharmacal order, and then carry out the conventional parallel analysis and intense focus on any competing convention rights. Under English defamation law, assessment of the extent of any interference with the Article 8 rights of the claimant, that might require the putative defamation claimant to demonstrate that there has been a real and substantial tort. As part of that, the claimant might be required to demonstrate that it has a real prospect of satisfying, inter alia, s.1 Defamation Act 2013 (as that section applies to individual and corporate claimants).”
The Order of 24 February 2023 did provide the Defendant with the opportunity to file any evidence in answer to the Application. It has not done so. Indeed, the Defendant has taken an entirely neutral position and has not engaged with the Application. It has confirmed that it will comply with any order the Court makes.
At the hearing, Mr Bennett KC provided a draft of an undertaking offered by the Claimants in the following terms:
“The information and documents disclosed pursuant to… this Order may be used by the Applicants solely and exclusively for the purpose of these and any connected legal proceedings.”
An undertaking in these – or similar – terms is usually required to be provided by the applicant for a Norwich Pharmacal order. Its purpose is to ensure that any information obtained, as a result of any order the Court grants, is not misused. Mr Bennett KC accepts, for example, that it would not be permissible for the Claimants to seek to obtain details of those who posted the Reviews and then, for example, serve on them a notice to quit under a tenancy agreement.
E: Legal Principles
Norwich Pharmacal Applications
Requirements for Norwich Pharmacal relief
The basic requirements before a Norwich Pharmacal order can be made are:
a wrong must have been carried out, or arguably carried out, by an ultimate wrongdoer;
there must be the need for an order to enable action to be brought against the ultimate wrongdoer; and
the person against whom the order is sought must: (a) be mixed up in, so as to have facilitated, the wrongdoing; and (b) be able or likely to be able to provide the information necessary to enable the ultimate wrongdoer to be pursued.
- Mitsui & Co Ltd -v- Nexen Petroleum UK Ltd [2005] 3 All ER 511 [21]; Collier -v- Bennett [2020] 4 WLR 116 [35].
The “wrong” relied upon
The wrong alleged in Norwich Pharmacal -v- Customs and Excise Commissioners [1974] 1 AC 133 was a tort, but it has been established that any type of wrong may be sufficient, whether civil or criminal. However, the applicant must be the alleged victim of the crime. A third-party cannot rely upon detection of crime as a justification for a Norwich Pharmacal order if s/he is not the victim of it: Ashworth Security Hospital -v- MGN Ltd [2002] 1 WLR 2033 [54].
The need for the order
It is not necessary for the applicant for a Norwich Pharmacal order to intend to bring civil proceedings. The information may be sought for other avenues of redress, for example a disciplinary action against an employee: British Steel Corporation -v- Granada Television Ltd [1981] AC 1096, 1200.
The applicant must demonstrate that an order for the information is necessary. This is a threshold condition, not a question of discretion: R (Omar) -v- Secretary of State for Foreign and Commonwealth Affairs [2014] QB 112 [30]. As such, if the applicant could obtain the information through other practicable means, the relief will be refused: Mitsui [24].
Mixed up in the wrongdoing
It remains important for an applicant to demonstrate that the respondent to a Norwich Pharmacal application has been involved, in some way, in the alleged wrongdoing. In the original Norwich Pharmacal decision, the status of the respondent as beyond something than a ‘spectator’, ‘mere witness’, or ‘bystander’ was a recognised limit of the jurisdiction: see Lord Reid, 174F; Lord Morris, 180D-E; and Lord Kilbrandon, 188A-C. Similarly, and applying these principles, in Ashworth [35], Lord Woolf CJ drew a distinction between a person who was “involved” in the wrong and someone who was simply a “mereonlooker or witness”. Facilitation (as opposed to participation) in the wrongdoing has been held to be sufficient: R (Mohammed) -v- Secretary of State for Foreign and Commonwealth Affairs [2009] 1 WLR 2579 [71]. Indeed, facilitation is what usually satisfies this element for Norwich Pharmacal orders against the operators of websites on which material has been posted by (anonymous) third parties. In such cases, the websites have (at least) facilitated the publication that is the arguable wrong.
In NML Capital -v- Chapman Freeborn Holdings Ltd [2013] EWCA Civ 589 [2013] 1 CLC 968, Tomlinson LJ held:
[25] … it is in my judgment clear that if the Norwich Pharmacal jurisdiction is not to become wholly unprincipled, the third party must be involved in the furtherance of the transaction identified as the relevant wrongdoing. King J put it well in Campaign Against Arms Trade -v- BAE [2007] EWHC 330 (QB) [12] when he said:
“The third party has to have some connection with the circumstances of the wrong which enables the purpose of the wrongdoing to be furthered.”
[26] It follows that it is important to analyse with some care in what precisely lies the alleged wrongdoing. There is nothing inherently wrong in chartering an aircraft, unless it be said that any trading by a judgment debtor which involves using his assets for that purpose rather than satisfying a judgment debt is in itself wrongdoing. However I reject that proposition. It would lead to a jurisdiction of absurd width. It is no answer to that objection that the exercise of the jurisdiction would be subject to discretionary considerations. It would be absurd and exorbitant if parties were exposed to the risk of having to defend applications for discovery on the basis of no more than having traded with a person who turns out to have been at the relevant time a judgment debtor. It would encourage speculative litigation.”
In EUI Limited -v- UK Vodaphone Limited [2021] EWCA Civ 1771, the claimant insurers sought a Norwich Pharmacal order against the mobile phone service provider to disclose the telephone and data account of the mother of the policy holder. This disclosure was sought to establish whether the policyholder’s parents had vacated their home when the policy holder moved in and, as such, whether displacement costs covered by the insurers had been obtained fraudulently. The claimant submitted that the defendant had become involved in the wrongdoing because mobile phones have “enabled people to live in one place and conduct their affairs as if they are living somewhere else” and mobile telephone providers had “enabled” this activity ([17]). Baker LJ rejected this argument [18]:
“In my judgment, [this] argument is misconceived. If the claimant is right in thinking that the policy holder has fraudulently asserted that his parents moved out of their home for a period to allow him and his family to occupy the house exclusively, it is arguable that his parents were involved in the wrongdoing. But I can see no basis on which it could be said that his mother’s mobile phone service provider was more than a mere witness or, in Mann J’s phrase from Various Claimants -v- News Group Newspapers Ltd [2014] Ch 400, engaged with the wrong. The fact that the phone account holder would have been able to pretend she was somewhere she was not does not draw the phone company into her wrongdoing. It is true that the phone records may assist in establishing the truth of the parents’ whereabouts. But in that regard the phone company is manifestly a mere witness. Its position is no different from anyone else who may be able to provide evidence about that issue – for example, the nephew living in Milton Keynes, or the neighbours to the parents’ property, or, as Lewis LJ helpfully suggested in the course of the hearing, the milkman. The phone company’s position seems to me to be analogous to that of a security company which installs CCTV cameras at a property. Such cameras are also a feature of modern life. The purpose of the cameras is to detect or deter burglars who have no right to be at the property, but they may also incidentally detect the presence of the householders who have every right to be there. The security company would therefore be a witness to any unlawful activity engaged in by the householders but it would not be drawn into that activity in any way.”
In the News Group case, cited by Baker LJ, a Norwich Pharmacal order was granted against the Metropolitan Police to provide material held as a result of the police investigation into alleged phone-hacking at the defendant’s newspaper. The police investigation long post-dated the alleged phone-hacking, and was conducted in furtherance of its public duties to investigate alleged criminal activities. Having reviewed these authorities, in Hayden -v- Associated Newspapers Ltd [2022] EWHC 2693 (KB) [50]-[56], I held that the ratio of News Group was that [76]:
“… a Norwich Pharmacal order may be made against the police if the police have, using their statutory powers, carried out an investigation into alleged wrongdoing and, as a result, now possess information that would assist a claimant in bringing a civil claim for that (or related) wrongdoing, and that the claimant has no other practicable way of obtaining the information.”
In Hayden, the claimant sought an order for disclosure against His Majesty’s Courts and Tribunal Service (“HMCTS”) of documents that would identify the person who had obtained a copy of Court order that had subsequently been posted, anonymously, on a website, KiwiFarms. She based her application, in part, on the Norwich Pharmacal jurisdiction. The claimant’s contention was that the person who obtained the order was likely to be the same person as the person who had posted it on the website. I refused to make the order sought under the Norwich Pharmacal jurisdiction.
[73] … [HMCTS] has not in any way participated in or facilitated the publication of the KiwiFarms Post. [Its] provision of a copy of the [Order] – in discharge of the duty under CPR 5.4C(1) – no more “facilitated” the KiwiFarms Post than would a stationery shop selling someone a pen and paper “facilitate” the sending of a defamatory letter.
[74] If the well-established principles of the extent of the Norwich Pharmacal jurisdiction are applied and not allowed to become “wholly unprincipled”, then the Respondent’s involvement is simply insufficient to justify a Norwich Pharmacal order against it.
...
[76] … [HMCTS] has not carried out any investigation into the KiwiFarms Post. Its involvement is limited to having provided X with a copy of the Order. On the established authorities, that is insufficient to sustain a Norwich Pharmacal order and the Claimant cannot bring her claim within the ratio of the News Group decision. [HMCTS] is a mere witness, par excellence. Moreover, it is a witness to an event which would go only some of the way towards unmasking the person who has posted as “Notso jolly Halliday” on KiwiFarms.
Discretion whether to make an order
If the requirements for a Norwich Pharmacal order are satisfied, the Court has a discretion whether to grant the relief. The leading authority on the factors relevant to the exercise of this discretion is The Rugby Football Union -v- Consolidated Information Services Limited (formerly Viagogo Limited) [2012] 1 WLR 3333 (“Viagogo”). It establishes the following overarching principles:
The jurisdiction exists to allow a prospective claimant to obtain information in order to seek redress for an arguable wrong: [14].
The Court should exercise flexibility and discretion when considering whether the remedy should be granted: [15].
It is not a pre-requisite that the applicant intends to bring legal proceedings in respect of the arguable wrong; any form of redress (for example disciplinary action or the dismissal of an employee) will suffice to ground an application for the order: [15].
An order for disclosure will be made only if it is a “necessary and proportionate response in all the circumstances”. But the test of necessity does not require the remedy to be one of last resort: [16].
The essential purpose of the remedy is to do justice. This involves the exercise of discretion by a careful and fair weighing of all relevant factors: [17].
Lord Kerr identified the following factors that would be included in the Court’s consideration ([17] and [25]) (“the Viagogo factors”):
the strength of the possible cause of action contemplated by the applicant for the order;
the strong public interest in allowing an applicant to vindicate his legal rights;
whether the making of the order will deter similar wrongdoing in the future;
whether the information could be obtained from another source;
whether the respondent to the application knew or ought to have known that he was facilitating arguable wrongdoing;
whether the order might reveal the names of innocent persons as well as wrongdoers, and if so whether such innocent persons will suffer any harm as a result;
the degree of confidentiality of the information sought;
the privacy rights under article 8 of the European Convention for the Protection of Human Rights and Fundamental Freedoms of the individuals whose identity is to be disclosed;
the rights and freedoms under the EU data protection regime of the individuals whose identity is to be disclosed. Before a Court makes an order requiring disclosure of personal data, it must first take into account and weigh in the balance the right to privacy with respect to the processing of personal data; and
the public interest in maintaining the confidentiality of journalistic sources, as recognised in section 10 of the Contempt of Court Act 1981 and article 10 ECHR.
The weight to be attached to each of these factors is likely to vary on a case-by-case basis.
Engaged Convention rights
Viagogo factors (2), (6), (8)-(10) require consideration of any relevant engaged Convention rights. Where, by a Norwich Pharmacal action, the applicant seeks to obtain disclosure of information for the purposes of identifying a person (“the target”) where the alleged wrongdoing involves the exercise a right of freedom of expression, then it is likely that the Convention rights of the target (and, in most cases, also the respondent) will be engaged. The target’s engaged rights are likely to include Article 8 and/or Article 10: Totalise PLC -v- The Motley Fool Ltd [2002] 1 WLR 1233 [25].
The need, before granting a Norwich Pharmacal order, for the Court to balance any engaged competing rights has been recognised by the ECtHR: Standard Verlagsgesellschaft mbH -v- Austria (No.3) (Application No. 39378/15); (2021) 53 BHRC 319 [92]-[96]. The Court observed [92]:
“…a potential victim of a defamatory statement must be awarded effective access to a court in order to assert his or her claims before that court. In the Court’s view this means that the domestic courts will have to examine the alleged claim and weigh – in accordance with their positive obligations under Articles 8 and 10 of the Convention – the conflicting interests at stake, before deciding whether the data relating to the author’s identity are to be disclosed. In the instant case, those conflicting interests do not only comprise the plaintiffs’ right to protect their reputation and the applicant company’s right to freedom of press, but also its role in protecting the personal data of the comment’s authors and the freedom to express their opinions publicly...”
As the Court of Appeal noted in Motley Fool, Article 10 protects both speech by an identified individual and anonymous speech. Whilst anonymity on the Internet can be used as a cloak behind which to harm others by unlawful acts, not all anonymous speech is of this character. Such speech, particularly in a political context, as a dimension of freedom of expression, can have a real value and importance. It also has a long pedigree both in the United Kingdom and the United States. As Lord Neuberger noted (Footnote: 1), extra judicially:
“It is unsurprising that the most robust protection of anonymous speech is to be found in US law. In McIntyre -v- Ohio Elections Commission (1995)514 US 334, a case on a statute prohibiting anonymous political literature, it was famously said by Justice Stevens that:
‘Under our Constitution, anonymous pamphleteering is not a pernicious, fraudulent practice, but an honourable tradition of advocacy and of dissent. Anonymity is a shield from the tyranny of the majority.’”
The modern equivalent of the anonymous pamphleteers of 200 years ago are anonymous online commentators, such as “The Secret Barrister”, for whom anonymity is an important dimension of the exercise of their rights of freedom of expression.
As a starting point, therefore, where a Norwich Pharmacal order is sought to unmask an anonymous online poster, the terms of that order are likely to interfere with the privacy interests of the target. Depending on the nature of the speech, for example if anonymity is (or maybe) being used to avoid recrimination/retribution/punishment (e.g. a whistle-blower), it may also interfere with the Article 10 rights of the target (and the respondent), see e.g. Standard Verlagsgesellschaft mbH.
As both Article 8 and Article 10 are qualified rights, where they are engaged the Court must consider whether there is a sufficient justification for the interference with the right and, if so, ensure that any Norwich Pharmacal order the Court makes is necessary and proportionate. In the case of online publications that are alleged to constitute an actionable civil wrong, the likely engaged countervailing rights that may be engaged are Article 6 (fair trial), 8 (privacy/reputation) and 13 (effective remedy). The Article 6/13 rights are reflected in Viagogo factors (2) and (6) and Article 8/10 rights in factors (8)-(10). When balancing the competing rights, the Court will apply the familiar parallel analysis: In re S (a child) [2005] 1 AC 593 [17]:
“… First, neither article has as such precedence over the other. Secondly, where the values under the two articles are in conflict, an intense focus on the comparative importance of the specific rights being claimed in the individual case is necessary. Thirdly, the justifications for interfering with or restricting each right must be taken into account. Finally, the proportionality test must be applied to each. For convenience I will call this the ultimate balancing test…”
An intense focus on the comparative importance of the specific rights being claimed requires an applicant for a Norwich Pharmacal order to demonstrate more than simply an arguable case that s/he has been the subject of a civil wrong. S/he must show that a claim that has sufficient weight or substance to outweigh the countervailing rights of the target. Viagogo factor (1) requires, an assessment of the strength of the underlying claim relied upon, which is consistent with the obligation to examine the claim articulated in Standard Verlagsgesellschaft mbH. For practical purposes, this means that an applicant applying for Norwich Pharmacal relief must demonstrate, in the evidence in support of application, that s/he has, at least, a claim with a real prospect of success.
What a claimant is required to demonstrate to establish a cause of action with a real prospect of success will depend on the nature of the underlying claim upon which the applicant relies. I have set out below the requirements for claims in defamation and malicious falsehood ([47]-[62] below). For speech-based harassment claims, the law is summarised in Hayden -v- Dickenson [2020] EWHC 3291 (QB) [44].
The concept of a threshold of seriousness in the context of Norwich Pharmacal applications in this area is not new. The authors of Disclosure (§3.19, 5th edition, Sweet & Maxwell, 2017) observed:
“… in the context of website postings, the Court has held in a number of cases it would be unjustifiably intrusive and disproportionate to order the disclosure of the identities behind online postings which were barely defamatory, little more than abuse or ‘saloon-bar moanings’ rather than serious allegations.”
The authorities cited are Sheffield Wednesday -v- Hargreaves [2007] EWHC 2375 (QB), Clift -v- Clarke [2011] EWHC 1164 (QB) and the Canadian case of Warman -v- Wilkins-Fournier (2010) 319 DLR (4th) 268 (Ontario Superior Court of Justice).
In Sheffield Wednesday,HHJ Parkes QC explained ([9]):
“… The proposed order will, if granted, disclose to the Claimants the identities, or at least the e-mail addresses, of users of the Defendant's website who must have expected, given their use of anonymous pseudonyms, that their privacy would be respected. As the Court of Appeal observed in Totalise PLC -v- The Motley Fool Ltd [2002] 1 WLR 1233 [25], in a case where the proposed order will result in the identification of website users who expected their identities to be kept hidden, the court must be careful not to make an order which unjustifiably invades the right of an individual to respect for his private life, especially when that individual is in the nature of things not before the court. Equally, it is clear that no order should be made for the disclosure of the identity of a data subject, whether under the Norwich Pharmacal doctrine or otherwise, unless the court has first considered whether the disclosure is warranted having regard to the rights and freedoms or the legitimate interests of the data subject (see paragraph 6 of schedule 2 of the Data Protection Act 1998). As the Court of Appeal pointed out (at paragraph 26 of the judgment) it is difficult for the court to carry out this task if it is refereeing a contest between two parties neither of whom is the person most concerned, that is to say the data subject…”
The Judge granted a Norwich Pharmacal order in respect of some of the online postings complained of, but refused to make an order in respect of those that were “barely defamatory or little more than abusive or likely to be understood as jokes” and little more than “saloon-bar moanings about the way in which the club is managed”: [17]-[18].
In Clift -v- Clarke, the Claimant sought a Norwich Pharmacal against a journalist employed by Associated Newspapers to identify individuals who had posted comments in response to an online article published in Mail Online. Refusing the order, Sharp J held:
[32] In my view, the postings are clearly one or two-liners, in effect posted anonymously by random members of the public who do not purport, either by their identity or in what they say, to have any actual knowledge of the matters in issue. It is difficult to see in the context, and having regard to their content, how any reasonable, sensible reader could take either of them seriously, or indeed how they could conceivably have caused any damage to the Claimant's reputation.
…
[36] The postings are in reality, it seems to me, no more than “pub talk”, as it has sometimes been described, and I consider it fanciful to suggest any reasonable sensible reader would construe them in any other way. See for example what was said in Smith -v- ADVN Plc [2008] EWHC 1797 (QB) [17] where it was said by Eady J of what was posted on a bulletin board that “It is often obvious to casual observers that people are just saying the first things that come into their head and reacting in the heat of the moment. The remarks are often not intended, or to be taken, as serious.” (See in particular [14]-[17]).
Sheffield Wednesday and Clift were both decided before the introduction of the serious harm requirement in s.1 Defamation Act 2013 (see [51]-[53] below).
In Warman -v- Wilkins-Fournier, the Ontario Divisional Court allowed an appeal against an order requiring the administrators and moderators of a message board to disclose the email and IP addresses of individuals who had posted allegedly defamatory comments under pseudonyms (the Norwich Pharmacal jurisdiction also being a part of Canadian law). The Court held that the order interfered with both the right to freedom of expression, guaranteed by the Canadian Charter of Rights and Freedoms, and Charter-recognised privacy interests. The order for disclosure was set aside because the Judge had not properly considered the issue of freedom of expression or whether the plaintiff had established a prima facie case of defamation before ordering disclosure of the documents sought by the plaintiff. Wilton-Siegel J, giving the judgment of the Court, held:
[42] … because this proceeding engages a freedom of expression interest, as well as a privacy interest, a more robust standard is required to address the chilling effect on freedom of expression that will result from disclosure. It is also consistent with the recent pronouncements of the Supreme Court that establish the relative weight that must be accorded the interest in freedom of expression. In the circumstances of a website promoting political discussion, the possibility of a defence of fair comment reinforces the need to establish the elements of defamation on a prima facie basis in order to have due consideration to the interest in freedom of expression. On the other hand, there is no compelling public interest in allowing someone to libel and destroy the reputation of another while hiding behind a cloak of anonymity. The requirement to demonstrate a prima facie case of defamation furthers the objective of establishing an appropriate balance between the public interest in favour of disclosure and legitimate interests of privacy and freedom of expression.”
In most cases, proper respect for (and protection of) any engaged Article 10/8 rights is likely to be achieved by the Court making a careful assessment of whether there has been an arguable wrong and the strength of the identified cause(s) of action, and whether the public interest in allowing an applicant to vindicate his legal rights is outweighed by any countervailing interests of the target. Norwich Pharmacal orders will not be granted, speculatively to strip away online anonymity, unless the Court is satisfied that justice requires it. The danger of too lax an approach is obvious. The subject of an unfavourable publication may have many reasons for wanting to identify his/her online critic, not all of which would provide a justification for a Norwich Pharmacal order. The jurisdiction is not to be used to satisfy curiosity or to enable any form of revenge or retribution. It exists to do justice by enabling someone who can demonstrate that s/he has been the victim of an arguable wrong, for which s/he wishes to seek legitimate redress, to obtain an order from the Court that will assist him/her to do so by assisting in the identification of the wrongdoer.
The duties of a party making a Norwich Pharmacal application
On any ex parte application, the applicant has an obligation to ensure that the Court’s attention is drawn to any matter known to him/her which might affect the decision whether to grant relief or what relief to grant: Fitzgerald -v- Williams [1996] QB 657, 667. In Memory Corp plc -v- Sidhu (No.2) [2000] 1 WLR 1443, 1460A-B, Mummery LJ held:
“… there is a high duty to make full, fair and accurate disclosure of material information to the court and to draw the court's attention to significant factual, legal and procedural aspects of the case. It is the particular duty of the advocate to see that the correct legal procedures and forms are used; that a written skeleton argument and a properly drafted order are prepared by him personally and lodged with the court before the oral hearing; and that at the hearing the court's attention is drawn by him to unusual features of the evidence adduced, to the applicable law and to the formalities and procedure to be observed.”
This duty is particularly important in this type of Norwich Pharmacal application. The nature of the relief sought means that, if the Court later finds that the order should not have been made, setting aside the original order is likely to be a wholly inadequate remedy. Once the anonymous poster has been identified, any harm that has been caused by an order wrongly granted cannot be undone.
It is especially important that in this type of Norwich Pharmacal application the Court is made fully aware, both in the evidence and in any written submissions or skeleton argument, of all factors that could fairly be raised by the absent target in opposition to the order that is sought (particularly any matters that would be relevant to the assessment of the strength or viability of the potential claim upon which the applicant relies).
As noted in Sheffield Wednesday,the very nature of the Norwich Pharmacal jurisdiction means that the target of the application – although having a direct and obvious interest in it – cannot be a respondent to it. The burden of full and frank disclosure that falls on the applicant is not lessened by the fact that the application is made on notice to the person against whom the Norwich Pharmacal order is sought. Without a reasonable basis on which to conclude otherwise, the applicant cannot assume that arguments that could be raised by the absent target will be made by the respondent to the application. Typically, in applications concerning online postings, most respondents to a Norwich Pharmacal application do not actively oppose the grant of relief (some may even consent to it); and most do not attend any hearing. And even if the respondent does engage with the application, in most cases, the interests of the respondent to the application are unlikely to align with the interests of the absent target of the application. His/her interests are therefore unlikely to be represented by the respondent, whether adequately or at all. The fact that the respondent either consents to or does not oppose the grant of relief does not release the applicant or the Court from properly considering the absent target’s position. The applicant retains the burden of ensuring that the Court receives full and frank disclosure and Courts dealing with this type of Norwich Pharmacal application must be especially vigilant to ensure that the engaged Convention rights of the absent target are properly identified and considered.
It is up to the individual Judge, to whom a Norwich Pharmacal application is made, whether to deal with the application without a hearing, under CPR 23.8. The experience in this case perhaps demonstrates the value of a hearing in identifying the relevant issues and evidence. Nevertheless, any party who seeks to have an application dealt with without a hearing must recognise that the duty of full and frank disclosure is in no way diminished if the application is dealt with on the papers; indeed arguably the burden is likely to be more onerous as the applicant will need to ensure that the written submissions clearly and effectively identify to the Judge any matter(s) which might affect the decision whether to grant the application and, if so, in what terms: see Masri -v- Consolidated Contractors International Co SAL [2011] EWHC 1780 (Comm) [58].
Relevant principles of the law of defamation
Before the Defamation Act 2013 came into force, it was relatively easy for an applicant in a Norwich Pharmacal claim to demonstrate that s/he had a prima facie case for defamation if s/he could point to a statement, that referred to him/her, that had been published to a third party and which was defamatory of him/her at common law.
To establish that the statement complained of referred to (or identified) the claimant, s/he must prove that the words complained were published “of and concerning” him/her: Knupffer -v- London Express [1944] AC 116, 121. It is not necessary for the claimant to be named. There may be some other way in which the hypothetical ordinary reasonable reader would identify him/her: Economou -v- de Freitas [2017] EMLR 4 [9].
The test whether a statement is defamatory at common law was stated by Warby LJ in Millett -v- Corbyn [2021] EMLR 19[9]:
“At common law, a meaning is defamatory and therefore actionable if it satisfies two requirements. The first, known as ‘the consensus requirement’, is that the meaning must be one that ‘tends to lower the claimant in the estimation of right-thinking people generally.’ The Judge has to determine ‘whether the behaviour or views that the offending statement attributes to a claimant are contrary to common, shared values of our society’: Monroe -v- Hopkins [2017] 4 WLR 68 [51]. The second requirement is known as the ‘threshold of seriousness’. To be defamatory, the imputation must be one that would tend to have a ‘substantially adverse effect’ on the way that people would treat the claimant: Thornton -v- Telegraph Media Group Ltd [2011] 1 WLR 1985 [98]...”
At common law, once these matters had been established, falsity, malice and damage were presumed in a claimant’s favour, meaning that the applicant needed to demonstrate nothing further in respect of the underlying claim in support of a Norwich Pharmacal application. As the cases of Sheffield Wednesday and Clift demonstrate, the Court nevertheless retained a discretion to refuse Norwich Pharmacal relief on the grounds that the underlying claim was trivial. The common law threshold of seriousness before a statement would be regarded as defamatory had been raised in Thornton -v- Telegraph Media Group [2011] 1 WLR 1985,and a jurisdiction to strike out trivial and pointless claims had been recognised in Jameel -v- Dow Jones & Co Inc[2005] QB 946.
However, following the enactment of the Defamation Act 2013, a claimant in a defamation claim must now also satisfy the requirements of s.1. That section provides:
A statement is not defamatory unless its publication has caused or is likely to cause serious harm to the reputation of the claimant.
For the purposes of this section, harm to the reputation of a body that trades for profit is not ‘serious harm’ unless it has caused or is likely to cause the body serious financial loss.”
Whether the publication of the statement has caused or is likely to cause serious reputational harm is a matter of fact, “which can be established only by reference to the impact which the statement is shown actually to have had. It depends on a combination of the inherent tendency of the words and their actual impact on those to whom they were communicated”: Lachaux -v- Independent Print Ltd[2020] AC 612 [14] per Lord Sumption. Overall, s.1 “not only raises the threshold of seriousness above that envisaged in Jameel and Thornton, but requires its application to be determined by reference to the actual facts about its impact and not just to the meaning of the words”: ibid [12]. For a claimant that is a body that trades for profit, it must prove that publication of the statement has caused (or is likely to cause) serious financial loss.
If a claimant alleges that s/he has been defamed by a publication, then s/he must show that the publication of the statement has caused serious harm to his/her reputation (or is likely to do so). The scope for, and the Court’s willingness ultimately to draw, an inference that such harm has been (or is likely to have been) caused will depend on the particular facts of the case. The basic rules of causation continue to apply: Sivananthan -v- Vasikaran [2023] EMLR 7 [43]-[46]; Amersi -v- Leslie [2023] EWHC 1368 (KB) [157]-[159]. As I observed in Amersi ([158]): “The impact of Lachauxis that such reputational harm must be proved... Drawing inferences is not a process of optimistic guesswork; it is a process whereby the court concludes that the evidence adduced enables a further inference of fact to be drawn.” The same applies if the claimant is required, by s.1(2), to prove serious financial loss.
In the context of Norwich Pharmacal applications, where the arguable wrong relied upon is defamation, the applicant must demonstrate, by evidence, a case of serious harm to reputation/serious financial loss under s.1 that has a real prospect of success. Without that, the applicant will be unable to demonstrate the first requirement for a Norwich Pharmacal order.
Relevant principles of the law of malicious falsehood
At common law, a claimant in a malicious falsehood claim must prove publication to a third party of a statement referring to him, his property or his business which (1) is false; (2) was published maliciously; and (3) has caused special damage: Ratcliffe -v- Evans [1892] 2 QB 524, 527.
The claimant must show that the statement complained of is false. Where the statement conveys only an expression of opinion, a claimant may struggle to demonstrate that it is “false”. The delineation between allegations of fact and expressions of opinion is frequently required to be drawn in defamation claims. The well-established principles are summarised in Koutsogiannis -v- The Random House Group Ltd [2020] 4 WLR 25 [16] and would apply equally where the question of fact/comment arises in malicious falsehood claims (subject possibly to an argument as to the inapplicability of the single meaning rule to malicious falsehood).
In Tinkler -v- Ferguson [2018] EWHC 3563 (QB) [16], I summarised the law as follows:
In general, an unverifiable statement of opinion cannot be complained of as a falsehood for the purposes of a claim in malicious falsehood: Euromoney Institutional Investor plc -v- Aviation News Ltd [2013] EWHC 1505 [102].
This may be subject of qualification. One such qualification was… identified by Tugendhat J in [103]: where it can be shown, as a fact, that the commentator does not hold the expressed opinion: see Gatley §21.7 and footnote 55 (12th edition, 2013).
It is perhaps also important to bear in mind that, in defamation cases, it was always open to a defendant to seek to prove that the expressed opinion was true (Gatley §11.20). If, in context, a defamatory publication conveyed the meaning, as an expression of opinion, that the claimant was dishonest, a defendant could seek to prove that the claimant had been dishonest, as a matter of fact. Largely the ability to do so would be dependent upon the meaning conveyed and whether the expressed opinion was verifiable or capable of being proved true. In the example given by Gatley of a theatre critic stating that the claimant’s play was a bad play and not worth seeing is not capable of being proved objectively true, whereas the opinion that someone was dishonest could be.
It seems to me, therefore, that a statement of opinion that is capable of being proved true is, in principle, capable of founding an action for malicious falsehood where the opinion can be proved to be false and the claimant takes on the burden of doing so. In such circumstances, however, there may be substantial obstacles in the path of establishing that it was published maliciously.”
Malice is an essential ingredient of the tort, which the claimant must prove. Malice means publishing a statement that the defendant knew was false, or was reckless (in the sense of complete indifference) as to its truth or falsity. Alternatively, malice can be established by proving that the defendant published the statement with the dominant intention of injuring the claimant. A plea of malice is tantamount to dishonesty: Alexander –v- Arts Council of Wales [2001] 1 WLR 1840 [18]. The classic exposition of malice is from the speech of Lord Diplock in Horrocks -v- Lowe[1975] AC 135, 149-150. “It is that state of mind that justifies depriving a defendant of a defence of qualified privilege or makes it just to allow recovery for the publication of a falsehood”: Huda-v- Wells [2018] EMLR 7 [70].
I summarised the requirements for a plea of malice in Hudaas follows:
[72] As malice is a serious allegation – the equivalent of fraud – “it must be pleaded with scrupulous care and specificity. … [I]t is quite inappropriate to proceed on the basis that something may turn up (whether on disclosure of documents or at trial)”: Henderson –v- The London Borough of Hackney [2010] EWHC 1651 (QB) [40] per Eady J.
[73] Each of the particulars relied upon by the Claimant is required to be indicative of this dishonest state of mind order to be sustainable. Each particular has to raise a “probability of malice” and each particular has to be “more consistent with the existence (of malice), than with its non-existence”: Turner -v- MGM [1950] 1 All ER 449, 455a-eper Lord Porter; Telnikoff -v- Matusevitch [1991] 1 QB 102 at 120per Lloyd LJ. As made clear in Turner “each piece of evidence must be regarded separately... [I]f the result is to leave the mind in doubt, then that piece of evidence is valueless as an instance of malice whether it stands alone or is combined with a number of similar instances” (455b-c).
[74] The Court will scrutinise the statement of case in order to discern whether the malice plea has any prospects of success: Branson –v- Bower [2002] QB 737 [16] per Eady J.
[75] … A plea of malice against those who are passing on information that they have received or reporting concerns arising from such disclosures has an unpromising foundation. It will be an unusual case in which an individual in such a position will know that the allegations made by the complainant are false.
A claimant must prove the elements of both falsity and malice. S/he cannot make up for a failure to demonstrate malice by proving an abundance of falsity (or vice versa).
A claimant can be relieved of the obligation to prove that special damage was caused by the publication of the falsehood if s/he can rely upon s.3(1) Defamation Act 1952, which provides:
“In an action for … malicious falsehood, it shall not be necessary to allege or prove special damage -
if the words upon which the action is founded are calculated to cause pecuniary damage to the plaintiff and are published in writing or other permanent form, or
if the said words are 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.
In George -v- Cannell [2023] QB 117, the Court of Appeal held that “calculated to cause pecuniary damage” in s.3(1) means the publication by the defendant of a false and malicious statement of such a nature that, viewed objectively in context at the time of publication, financial loss was an inherently probable consequence or, putting it another way, financial loss was something that would probably follow naturally in the ordinary course of events: [27].
F: Submissions
The underlying claims justifying Norwich Pharmacal relief
Mr Bennett KC submitted that the Claimants have good arguable cases in defamation and malicious falsehood in respect of the individuals who posted each of the Reviews.
As to reference, the entity the subject of the Trustpilot reviews was identified as “ABC Estates”. Based on the evidence of Mr Lewis, Mr Bennett KC contends that this would be recognised, by readers of the Reviews, as the trading name of the Fifth and Sixth Claimants. As to the First to Fourth Claimants, each of these, he submits, would be identified by people as members of the family that operated the business of the Fifth and Sixth Claimants. The First Claimant was also named in several of the Reviews (items 1, 5, 8 and 9 in the Annex).
Mr Bennett KC argued that each of the Reviews was defamatory of each of the Claimants at common law, particularly the Fifth and Sixth Claimants. He submitted, “anyone who read them would not want to do business with them”. In respect of the company Claimants, Mr Bennett KC accepted that they could not point to “defined financial loss” caused by publication of any of the Reviews, but that there was a “strong inference” that, at least some, potential customers of the Fifth and Sixth Claimants would go to Trustpilot as part of research about the business and would be put off doing business with them as a result of the Reviews. That, he suggested, was “manifestly the purpose” of posting the Reviews. Mr Bennett KC suggested that the resulting financial loss would be “serious”.
Mr Bennett KC rightly acknowledged that, in respect of most of the Reviews, a potential limitation defence would be available for claims in defamation. and malicious falsehood: s.4A Limitation Act 1980. He nevertheless argues Norwich Pharmacal relief should not be refused on this basis because (a) it is a defence to be raised by a defendant in answer to the claim; and (b) were it to be raised as a defence, the Claimants would have a realistic prospect of defeating the defence under s.32 Limitation Act 1980 and/or the Court disapplying the period of limitation under s.32A Limitation Act 1980.
In relation to the malicious falsehood claim, Mr Bennett KC argued that “on the Claimants’ evidence, the allegations are false”. That submission was based on Mr Lewis’ evidence (set out in [7] above). In his skeleton argument, he submitted that it was possible that one or two “bad actors” were responsible for all of the Reviews. When I asked Mr Bennett KC to identify the evidence of malice upon which the Claimants relied, he accepted that the inability to identify the individuals meant that the claim of malice had to be advanced in fairly general terms. He relied upon the commonality of IP addresses to demonstrate that some of the Reviews were probably published by the same person, using different identities. That, he submitted, was consistent with a lack of honest belief in the contents of each Review. Overall, Mr Bennett KC submitted that the Court should be slow to shut out someone in the position of the Claimants from obtaining Norwich Pharmacal relief as it would represent a “defamers’ charter”.
Defendant mixed up in the wrongdoing
In their skeleton argument for the hearing, the Claimants contended that the Defendant provided the email addresses connected with the accounts that posted the Reviews on Trustpilot and that, therefore, the Defendant “enabled” them to set up the Trustpilot accounts from which the Reviews were published.
This part of the test did not receive the attention that it perhaps needed and deserved at the hearing. It is only when I started writing this judgment that it became clear to me that this part of the test was not straightforward. As a result, I gave Mr Bennett KC the opportunity to make any further submissions that he wished on behalf of the Claimants.
In these further submissions, the Claimants contended that, if the authors of the Reviews had acted honestly and in accordance with the Trustpilot conditions of use, they would have provided their true identities on registration. If they had done so, the earlier Norwich Pharmacal order they obtained against Trustpilot would have provided them with all the information they needed to assess whether to bring claims against the publishers. However, prior to registering with Trustpilot, the Claimants contend that the publishers created false Gmail identities. This enabled them to create accounts on Trustpilot in false names. The Claimants submit that the inference should be drawn that the use of false names, with matching Gmail addresses, was done in order not to rouse suspicion on the part of Trustpilot.
Mr Bennett KC argued that the facts of the present case are therefore distinguishable from those in EUI and Hayden -v- Associated Newspapers.
In EUI, Mr Bennett KC contends, the identity of the potential defendant was already known. EUI’s case against him depended partly on proving that he was living in the UK with his parents during the relevant time. The potential defendant claimed that his parents had been in India at the relevant time. Mobile phone data for the relevant period would have revealed the location of the potential defendant's mother during the relevant period and therefore have been probative of this issue. However, disclosure was refused because Vodafone had been no more than a mere witness. Mr Bennett KC argues that the present case is distinguishable. Whereas Vodafone, by merely providing its services, was not “engaged with the wrong” and was not “drawn into” the alleged wrongdoing, Google was drawn into the wrongdoing because, to publish the relevant material on Trustpilot, the publishers used the false Gmail addresses to register with it (which were chosen to match the false names used to register). The use of the false Gmail addresses therefore facilitated/enabled the posting of the relevant material on Trustpilot.
In Hayden -v- Associated Newspapers Ltd, Mr Bennett KC argues, the case for refusing Norwich Pharmacal relief was clear-cut. The performance of the duty to provide a copy of a court order did not cause it to be anything other than a mere witness. HMCTS was not engaged with or drawn into the alleged wrong. By contrast, and relying on Tomlinson LJ’s judgment in NML Capital [25] (see [21] above), Mr Bennett KC submits that the use of the false Gmail addresses furthered the wrongdoing. There is, he argues, something “inherently wrong” ([26] of NML) with creating false Gmail addresses in order to deceive Trustpilot into believing that the identities linked to those Gmail addresses were real (and to breach Trustpilot's conditions of use).
After this judgment was provided, in the usual way, in draft to the parties, Mr Bennett KC sent further submissions. Normally, the Court does not afford a dissatisfied litigant an opportunity to advance yet further arguments once s/he has seen the draft judgment. Nevertheless, I have considered these further submissions because, as I have noted, very little attention was paid to this aspect in the Claimants’ skeleton argument and at the hearing. As it is the point that has proved decisive in this application, justice requires that I consider all the arguments that the Claimants wish to advance.
Mr Bennett KC referred me to pages of the Trustpilot website, which had been exhibited to Mr Lewis’ second witness statement. In a section headed “Guidelines for Reviewers”. It included the following (the underlined words represented hyperlinks):
“Trustpilot is here to help you shape and improve the world. We do it by giving you a powerful, open-to-all-review platform where you can share and discover experiences, and connect with businesses to help them improve.
The only thing we ask is that you respect and follow these guidelines (together with our Terms of Use to help keep Trustpilot a collaborative and trustworthy place for everyone to enjoy…”
Links are then provided to various further sections, one of which is “Your user account”, in which the following appears:
“You need a user account to post a review. Just like your first kiss, you only get one user account and it should involve a real person. Your username, profile description and picture must reflect who you are (don’t go impersonating other people, thanks) and can’t be harmful, hateful, discriminatory, defamatory or obscene – because everyone can see your profile! If you do include something you shouldn’t, or create more than one account, we can delete your account/s.
“Your user account needs to be connected to a valid, permanent email address in case we need to contact you…”
Although the Claimants accept that there is no evidence of whether (and if so for what purpose) Trustpilot used the Gmail addresses associated with any of the accounts that posted the Reviews, Mr Bennett KC argues that continued maintenance of the email address was necessary in order to post a review and keep it posted. If a user deactivated the registered email account, that would entitle Trustpilot to suspend the relevant account. As such, the Claimants contend that the Defendant continues to be “mixed up” in the wrongdoing after the Gmail addresses were used to set up the relevant accounts on Trustpilot.
Defendant likely to have information to enable the ultimate wrongdoer to be pursued
The information the Claimants seek is the subscriber registration information. In correspondence with the Claimants, the Defendant has stated that an order in these terms (if the Court made one) would be “acceptable”.
Balancing of the competing rights
Mr Bennett KC submitted that it was relevant that the Defendant had not asserted any opposition to the order being sought by the Claimants either on its own behalf or on behalf of the targets. He sought to distinguish Standard Verlagsgesellschaft mbH on two grounds. First, Trustpilot stipulates in its terms and conditions that users should use their true identity when posting a review. The authors of the Reviews did not do so. Fake reviews undermine the value of a platform like Trustpilot. Second, there is a qualitative difference between false allegations of fact concerning a consumer service, most probably published with knowledge of their falsity in order to cause unwarranted harm, and the expression of political opinions.
G: Discussion and decision
The “wrong” relied upon
The Claimants have based their Norwich Pharmacal application on the contention that publication of each Review amounts to an actionable libel and/or malicious falsehood. Mr Bennett KC’s submissions did not descend into any real analysis of each Review. Each Review is relied upon to justify a Norwich Pharmacal order requiring the Defendant to provide the information it has that may shed light on the identity of the person operating the relevant Gmail account.
I will make some observations about the individual Reviews, but there are some points that apply across the board.
First, apart from the Reviews that name the First Claimant, for claim in malicious falsehood and/or defamation to have a real prospect of success, the Claimants must demonstrate that each Review would have been understood to refer to them. The Claimants have provided no evidence that any reader of any of the Reviews did so. Beyond the evidence that “ABC Estates” is a family business, and that each of the individual claimants drives an “ABC Estates” branded car, this important element of the underlying claims is advanced merely as assertion. For the malicious falsehood claim, the Fifth and Sixth Claimants can point to the fact that the Reviews of “ABC Estates” would have been understood to refer to them, as the companies that operate the business of “ABC Estates”. In this respect, the element of reference in malicious falsehood claim can be easier to establish than in a claim for defamation: see observations in Dyson -v- Channel Four [2023] EMLR 5 [30].
Second, insofar as the Claimants base their Norwich Pharmacal application on potential claims for defamation, then they have produced no evidence that any Review has caused (or is likely to cause) either serious harm to their individual reputations or, for the Fifth and Sixth Claimants, serious financial harm. This point is closely allied to the point on reference. Mr Lewis’ limited general evidence (see [9(3)] above) is nothing more than assertion. Even the most basic evidential building blocks upon which an inferential case might be based, for example a proven drop off in business for the Fifth and Sixth Claimants following a Review, are absent. There are also obvious potential causation issues.
Third, insofar as the Norwich Pharmacal application is based upon potential claims in malicious falsehood, in addition to the issue of reference, there are serious problems with the Claimants’ case of both malice and falsity. Subject to what the Claimants have been able to demonstrate in relation to some of the Reviews (see [91]-[93] below), the evidence of malice is sparse. That is the unavoidable consequence of the Claimants’ inability to identify the person(s) who posted the Reviews. I recognise that this presents a problem for the Claimants, but this does not to relieve them of the obligation to demonstrate this important ingredient of the tort if it is to be relied upon as a basis for a Norwich Pharmacal application.
In relation to falsity, the Claimants confront the problem that, for the most part, the Reviews contain unverifiable expressions of opinion (e.g. that staff are “rude”, “aggressive”, “lack respect” or have a “bad attitude”; that customer service is “poor”; that the business is “useless”, “the worst”, “horrible”, “horrific”, “no good” and “to be avoided”). The evidence the Claimants have provided as to falsity is perfunctory, even desultory; contained in a single general paragraph of Mr Lewis’ witness statement, which descends to no detail (see [7] above). No attempt has been made to identify what, in any Review, is alleged to be false. For a claimant advancing a claim for malicious falsehood as a basis for a Norwich Pharmacal order it is simply not good enough to state: “[The Reviews] are false, fabricated statements which Unknown person(s) know are untrue…” Neither Mr Lewis, nor any of the Claimants, is in a position to state that the publishers of the Reviews have no genuine belief in the truth of what they have published.
Further, even this “evidence” of falsity has been undermined by being shown to be unreliable in respect of Review 9. In that Review, posted under the name of Stuart Conway, the poster said that he had done some online research about the First Claimant and had discovered that he had been expelled from ARMA (the Association of Residential Managing Agents). Before the hearing, and following Mr Conway’s lead, I carried out some basic online research. Without difficulty, I found articles which did suggest that the First Claimant (or one of his companies) had been expelled from ARMA in November 2021. These articles referred to decisions of the First-Tier Tribunal Property Chamber (Residential Property) (“FTT”), of 12 August 2021 and 13 October 2022, in which the First Claimant was a party. In the latter decision (LON/00AW/LVM/2022/0005), Judge Nicol included the following in his judgment:
“[12] [Richard Davidoff] used to be a member of IRPM (Institute of Residential Property Management) and Propertymark. His company, ABC Estates Ltd, used to be a member of ARMA (Association of Residential Managing Agents). They are no longer members of these organisation because they were expelled…
…
[18] [Mr Davidoff] … identified the root cause of his expulsion from IRPM, ARMA and Propertymark as being the Tribunal’s decision dated 12 August 2021 … (LON/00AL/LSC/2020/0111 and LON/00AL/LAM/2018.0012)… The Tribunal made findings as to various defaults by [Mr Davidoff]…
It is not necessary for me to set out the defaults found by the FTT in the 12 August 2021 decision. They are set out in the judgment.
At the hearing, I provided copies of the two FTT decisions to Mr Bennett KC and asked him whether it was correct that the First Claimant and one of his companies had been expelled from these bodies. Mr Bennett KC confirmed that this was correct. After taking instructions, he told me that the Claimants wanted to withdraw their Norwich Pharmacal application in relation to Review 9.
As I indicated at the hearing, the statement in Mr Lewis’ witness statement that the Reviews “are false, fabricated statements which Unknown person(s) know are untrue…” was not correct in relation to Review 9. On the contrary, and as had been confirmed by Mr Bennett KC, an important factual allegation in Review 9 was not “false and fabricated” but true. It is a matter of very real concern that the Claimants put evidence before the Court, on an ex parte application, that was not true.
Mr Lewis has not provided an explanation to the Court for this error (I have not required him to provide one). Nevertheless, I accept that Mr Lewis would not have knowingly misled the Court. It is likely that this error occurred because he had simply failed to carry out sufficient (or any) research or to take adequate instructions from his clients. Nevertheless, as a result, he included a statement in his witness statement that was seriously in error. That error was not detected (or corrected) by the Claimants. The same lack of rigour meant he also failed to provide the Court with evidence – in the form of relevant articles and FTT decisions – that the Claimants should have provided as part of their general obligation of full and frank disclosure (see [42]-[46] above). There was a significant failure to comply with that duty. It is perhaps fortunate for the Claimants that I had found the relevant material, through my own inquiries before the hearing, and so this error has not affected the outcome of the application. This incident does, however, serve to demonstrate the importance of applicants for Norwich Pharmacal orders ensuring that they put before the Court full and accurate information relevant to the Court’s decision.
Overall, even if I were to assume in their favour that the Claimants could advance a meaning for each of the Reviews that is defamatory at common law (as to which I have significant reservations), the underlying claims in defamation are hopelessly weak. Without evidence of serious harm to reputation/serious financial loss caused by the publication of each Review, the defamation causes of action have no real prospect of success. Insofar as the Norwich Pharmacal application is based upon a claim for defamation, the Claimants have failed to demonstrate that a wrong has been carried out.
When assessing the Claimants’ case of malicious falsehood, it is necessary to look at two groups of Reviews: Reviews 3, 4, 5 and 6 and Reviews 2, 7, 8 and 11.
The evidence relating to Reviews 3, 4, 5, and 6 shows that one IP address was used to set up the two Trustpilot accounts under the name “Lisa Matherson”. The same IP address was used to post Review 3. A second, but closely linked, IP address was used to set up the Trustpilot accounts under the names “Michelle Stonefiled” and “Sophie Adler”. That second IP address was used to post Reviews 4, 5 and 6. That evidence, and in particular the overlap in usage between the two IP addresses, tends to demonstrate that a single person was also behind all this online activity. On that basis, there is a solid basis on which to infer that some or all of these names were not genuine.
In the second group, the strength of the case comes, largely (if not exclusively), from the fact that Reviews 7 and 8 – posted under the names of Anthony Redfield and Stephen Michaels – were posted from the same IP address, and the other Reviews were posted from the same geographic area. On the evidence as it stands, and in absence of a plausible explanation, there is a strong suspicion that a single person has been responsible for publishing all these Reviews. There is also other evidence – from the email addresses used – that suggests that the registered names are not genuine.
Overall, the evidence relating to these two groups of Reviews raises a strong prima facie case that these Reviews are not genuine. I accept Mr Bennett KC’s submission that the evidence supports a conclusion, at this stage, that each group of Reviews was authored by a single person who, by using false identities, made it appear that each Review came from a distinct individual (thereby, arguably, augmenting its impact). Although a final determination would have to await further investigation and evidence (including any evidence presented by the person responsible for posting the relevant Review), at this stage the use of multiple (false) online identities to post negative reviews is more consistent with a dishonest attempt to cause damage than the expression of honest criticism.
Had these Reviews in these two groups of suspicious postings been looked at simply in isolation, they would not have been promising candidates for a claim for malicious falsehood, for the reasons that I have identified (see [81]-[84] above). For example, Reviews 2, 3, 4, 7 and 11 would appear, on their face, simply to be complaints about poor customer service in terms that would be likely to be found to be an expression of unverifiable opinion. However, the Claimants have demonstrated a credible case that these reviews are not genuine. If it turns out, once the identity of the author of each Review is established, that there was never any consumer engagement with the ABC Estates – then the relevant opinion would not have been expressed honestly. If that were the evidence, the Claimants might well be able to demonstrate both malice and falsity.
As to s.3 Defamation Act 1952, viewed objectively at the time of publication, I am satisfied, at this stage, that the Fifth and Sixth Claimants have a real prospect of demonstrating that financial loss was an inherently probable consequence of the publication of the Reviews in these two groups. If these Claimants are successful in demonstrating both falsity and malice in respect of each Review, then there is a credible basis on which they will be able to invite the Court also to conclude that the purpose of publication was to cause financial loss to the Fifth and Sixth Claimants. I do not consider that the same can be said for a malicious falsehood claim brought by the individual Claimants. The Claimants in their evidence have not explained the mechanism by which publication of any (false and malicious) Review was inherently likely to cause them financial loss. More evidence would be required as to their status and the basis of which they benefit financially from, or are remunerated by, the Fifth and Sixth Claimants.
Although the terms of each Review caused me initially to be sceptical that any of them amounted to an arguable malicious falsehood, I have been persuaded that, on the basis of the evidence as it stands, the Fifth and Sixth Claimants have demonstrated a claim for malicious falsehood with a real prospect of success (i.e. a claim that is not fanciful) in relation to Reviews 2, 3, 4, 5, 6, 7, 8 and 11. sufficient to satisfy me that a wrong has arguably been carried out by at least one wrongdoer. The First to Fourth Claimants have not satisfied this requirement.
The Claimants have rightly identified the potential limitation issue that might be raised against any claims that they might bring for libel/malicious falsehood. The short point is that limitation is a defence. It needs to be raised by a defendant. If it is, it would be for the Court to determine the extent to which that defence would provide an answer to the claim having regard to ss.32 and 32A Limitation Act 1980.
The claim in respect of Review 9 has been withdrawn. In respect of those Reviews 1 and 10, unlike the two groups of Reviews that I have identified, the Claimants can point to no evidence that raises a suspicion that the relevant Review is not genuine. Without that, any claim for malicious falsehood has no real prospect of success (for want of evidence of both malice and falsity).
For these reasons, only the Fifth and Sixth Claimants have demonstrated that a wrong has arguably been carried out by an ultimate wrongdoer.
The need for the order
I am satisfied that there is a need for an order if the Fifth and Sixth Claimants are to be able to bring a claim against the alleged wrongdoer(s). There is no other readily available avenue that they could explore to identify the alleged wrongdoer(s).
Mixed up in the wrongdoing
As noted above (see [68]), the Claimants’ case is that the authors of the Reviews used their Gmail email addresses to set up accounts on Trustpilot from which they published the Reviews. This use of the email address, they contend, “enabled” them to post the Reviews on Trustpilot. Further, Trustpilot required users to maintain that email address so that they could be contacted.
Conventionally, respondents to Norwich Pharmacal applications in cases like these tend to be the operators of the websites on which the alleged wrongdoer has posted. In such cases, there is little doubt that the website has become mixed up in the wrongdoing because it has, through its platform or service, enabled and facilitated the publication complained of.
Here, the Claimants originally relied only upon the use of an email address provided by the Defendant to sign up for a Trustpilot account, an account that was then used to post the relevant Review. Once registration for the Trustpilot account has been completed, there is no evidence that posting a review requires the further use of any email account.
The authorities on what amounts to being “mixed up” in the wrongdoing are not perhaps as clear as they might be. I considered this issue in Hayden -v- Associated [50]-[56]. The Court of Appeal decision in NML was not (apparently) cited either to Mann J in News Group, or later to the Court of Appeal in EUI. The NML decision was argued on an inter partes basis, whereas in EUI only the claimant was represented.
In my judgment, NMLstands as clear and binding Court of Appeal authority that the defendant in a Norwich Pharmacal claim/application must be “involved in the furtherance of the transaction identified as the relevant wrongdoing”; having “some connection with the circumstances of the wrong which enables the purpose of the wrongdoing to be furthered” [25]. When making this assessment, the Court must “analyse with some care what precisely lies in the alleged wrongdoing” [26].
The Court of Appeal’s judgment in EUI does not contradict or undermine the ratio from NML. I would regard, and read, the phrases “engaged with the wrong” and “[being] drawn into [the] activity” [18] as not stating the test in materially different terms.
Carefully analysing the nature of the Defendant’s involvement, whilst it might be argued that the Defendant, by providing a Gmail address, has “facilitated” the relevant individuals signing up for a Trustpilot account, that is not the wrongdoing upon which the Norwich Pharmacal application is based. The alleged wrongdoing is the subsequent use of the Trustpilot account to post the Review. In that second phase, the Defendant, in its provision of a Gmail account, has played no role; it has neither engaged in nor facilitated the alleged wrongdoing; nor has it furthered the posting of the Review. It can be tested this way. Once a person has registered for a Trustpilot account, even if the email account used for registration were to be deactivated (whether by that person or the Defendant), it would not prevent that person from posting a review on Trustpilot. If a Trustpilot account holder deactivated the registered email account, it might lead him/her to be in breach of Trustpilot’s requirement to maintain a valid email address, but again that does not involve or concern the Defendant. Once the Trustpilot account has been registered, the Defendant is, for all practical purposes, not involved in (and is powerless to control) what is posted using it. At the stage when each Review was posted, the Defendant was and is a mere witness to the anterior event of initial registration.
Following circulation of the draft judgment, the Claimants advanced the further argument that the Defendant remained engaged in the wrongdoing of posting the Review because the authors of the Reviews were expected by Trustpilot to maintain a valid email address. In my judgment, this makes no difference for the reasons I have given.
An analogy might perhaps be drawn of a case of a road traffic collision where the victim of the collision seeks Norwich Pharmacal relief against the owner of a garage from which the driver of the car had purchased the car. In a very general sense, by selling the car to the driver, the garage might be said to have “facilitated” the driving that led to the collision, but the sale of the car is not the wrong that is alleged. The garage simply played no role in the subsequent collision. The information it has about the driver, it holds as a witness.
I do not accept Mr Bennett KC’s submissions that it makes any difference that the relevant individuals signed up for a Gmail account using a false name. First, the Defendant probably has no way of telling whether the selected name has been chosen to practise, or further, a deception. Even if Mr Bennett KC’s submission, that there is something “inherently wrong” with creating false Gmail addresses, were to be accepted – and I can see contrary arguments – the Defendant is not aware of this wrongdoing. Second, and more generally, the Defendant has no way of knowing for what purpose the person signing up for the Gmail account will use that email address. Of course, if the person does then send a defamatory email, using the Gmail account, applying the principles I have identified, it would be likely to be found to have been involved in (or facilitated) the transmission of that message, and potentially liable as a target for Norwich Pharmacal relief.
Discretion
The conclusions I have reached on the preceding issues mean that the question of discretion does not arise. Had the Fifth and Sixth Claimants demonstrated that the Defendant was mixed up in the arguable wrongdoing of those who had published Reviews 2, 3, 4, 5, 6, 7, 8 and 11, then I would have granted a Norwich Pharmacal order against the Defendant. Consideration of the Viagogo factors would have led me to conclude that the order ought to be made.
As recognised, some anonymous speech is deserving of protection on Article 10/8 grounds. However, where there is a real basis on which to suspect or conclude that a person has used anonymity merely as an expedient by which s/he hopes to avoid identification and the potential consequences of their online activity the Court is likely to attach limited weight to the need to protect that anonymity. The Fifth and Sixth Claimants have satisfied me that they are the victims of an arguable wrong. It would have been in the public interest that they be given the opportunity to establish that in a civil claim. Although I consider that the Article 8 and 10 rights of the target(s) are engaged, this is not a case where the Court can apprehend that there will be potentially serious consequences for the target if his/her identity is revealed to the Fifth and Sixth Claimants. I would have attached some (albeit limited) weight to the fact that posting on Trustpilot using a false identity is against the terms of conditions of the website, but I make clear that that factor would not have been decisive on its own.
For these reasons, the Claimants’ Norwich Pharmacal application is refused.
Annex
Date | Reviewer | Text of Review | IP info | |
1. | 4 Nov 2021 | Lewis P | HORRIFIC!! Horrific estate agent. They made my mother’s life a misery for almost 2 and a half years. Strongly urge anyone considering ABC Estates to check the negative/1 star reviews on their Google page. Alternatively, a simple Google search of Richard Davidoff (the director) will do. | Finchley, UK |
2. | 4 Mar 2022 | Andrea Luckovic | Very poor service. Very poor service. Not a good company to deal with and staff are very rude. Have decided to use another company who I am very happy with. Do not use ABC Estates. They do not care about their customers. | Tamworth, UK |
3. | 30 Mar 2022 | Lisa Matherson | Poor customer service and rude Experience with this company is very bad. They offer poor customer service, not helpful and quite rude. Not recommended at all. | Edgware, UK (similar IP address to Reviews 4, 5 & 6) |
4. | 10 Apr 2022 | Lisa Matherson | They make lives miserable Very poor service and lack of respect towards customers. They are extremely bad in responding to customer enquiries and complaints and they do not know how to manage properties. This company should not be allowed to trade. They make peoples (sic) lives miserable. | Edgware, UK (same IP address as Reviews 5 & 6, similar IP address to Review 3) |
5. | 13 Apr 2022 | Michelle Stonefield | The worst of the worst The worst of the worse. They cannot manage properties to save their lives. They are useless and never keep their word. My husband and I have called so many times for several weeks and they do not help. Our block is a mess and so many issues need fixing. They don’t seem to care unless they are chasing you for money. Then they are very quick to contact you and start threatening. Never have I ever been so rudely spoken to or dealt with in such an aggressive way. Richard Davidoff and his worker [name given – not a Claimant] are the most arrogant and aggressive people I have ever met. They seem to forget that they work for you. Not the other way round. Do not deal with this company. Horrible and nasty people. | Edgware, UK (same IP address as Review 4 & 6, similar IP address to Review 3) |
6. | 25 Apr 2022 | Sophie Adler | Horrible and rude We tried to contact ABC Estates for help with our property needs. After several phone calls and also having visited their offices in Hendon, we were treated in a horrible way. They did not care about our enquiry or our needs and they make you feel like they are doing you a favour by even talking to you. They seem to think they own the property market and that they have no competitors that we could use. So very wrong. We were appalled by their rude response especially by a woman called [same name as given in Reviews 5 and 8] which we found to be even more rude as it was from a woman. We would never have expected this. We found out who the owner was and when calling to speak to him (Richard Davidoff), we were told he is not interested in discussing this with us. All we needed was some simple help with a property matter related to an elderly persons home we help look after which they manage. They should be ashamed of themselves. Do not use this company as they are horrible people. | Edgware, UK (same IP address as Review 4 & 5, similar IP address to Review 3) |
7. | 8 May 2022 | Anthony Redfield | Very poor customer service Very poor customer service. Not helpful at all and I found them quite rude when speaking to them by phone. A simple enquiry was too much for them to handle. Not a company you want to use as they just don’t seem to care about their customers. | Tamworth, UK (same IP address as Review 8) |
8. | 19 May 2022 | Stephen Michaels | By far the worst estate agent By far the worst estate agent and customer service I have ever dealt with. We deal with agents all the time on behalf of the elderly or people who have difficulty with the English language or other disadvantages. Dealing with ABC Estates was a nightmare. They are rude and inconsiderate and do not listen or care about your problems. One particular staff member called [same name as given in Reviews 5 and 6] was extremely rude. When asking to speak to the owner who I believe is Richard Davidoff, I was told in these exact words “Mr Davidoff has no time to deal with these matters.” I was shocked that the owner of a business who I wanted to speak with and make a complaint about how I was already being treated by his staff, was not at all interested in taking my call or in trying to help. After doing some research online and finding out more about Richard Davidoff and ABC Estates, I am not surprised. Stay away from this company. They are the worst! | Tamworth, UK (same IP address as Review 7) |
9. | 2 Jun 2022 | Stuart Conway | Rude and a waste of time My wife and I were warned not to use ABC Estates but we gave them the benefit of the doubt. Sadly, we were proven as their attitude towards customers is extremely bad. Their staff are unhelpful and aggressive in their tone and you can sense that they do not care about helping you. After doing some online research, I found dozens of recent articles and links about the company and the owner David Davidoff who was fined and actually expelled from ARMA as an estate agent. It seems that they have a history of problems and have made enemies with many people. Shame we did not listen to the advice not to use them as it was a complete waste of time. Stay away from ABC Estates. They are no good and a waste of time. | Preston, UK |
10. | 22 Jun 2022 | Natalie Schaffer | Aggressive to an elderly woman My grandmother contacted ABC Estates as they manage her block. When she got off the phone she was very upset and tearful. I asked her what was wrong and she told me how they spoke to her and told her to stop complaining about minor things. An on-going mould issue is not minor especially for an elderly person with difficulty breathing. I called ABC and was also spoken to very rudely after asking to be put through to the manager. They said he has not time to deal with these matters and that it is not their problem, and that we should just paint the room to get rid of the mould. Anyone with a brain can tell you that painting over mould does not solve the problem. In fact, it makes it worse. I have now contacted the local council and other organisations including the health authority to take this matter further. ABC Estates is a very bad company with bad staff. They are rude and aggressive. A disgusting way in which to speak to anyone especially an elderly woman. | Portugal |
11. | 30 Jul 2022 | Sharon Macenzie | Avoid at all costs! We contacted ABC Estates about a property they manage which we were interested in renting. It was extremely difficult to try and get any answers to our questions. They are very poor with customer service and avoided answering any of our questions. Seems they obviously had something to hide. Not a good estate agent to use. Avoid at all costs! | Tamworth, UK |