Eyes, optics and fraud in the libel courts: Judge takes troll down
29 May 2026
Rodoy v Optical Express Ltd EWHC 1219 (KB)
This is a substantial defamation judgment – at over four hundred paragraphs, quite a long read – but worth it.
In short, Griffiths J held that the claimant’s allegations about the defendant caused serious harm, but were substantially true and protected by qualified privilege, so the claim failed.
The case does not create any precedent in libel law but will be of general interest because the court engaged in unusually detailed factual findings about both the claimant Ms Rodoy’s long-running anti-refractive-surgery campaigning and the defendant Optical Express’s internal complaint-handling practices.
Apart from his findings on defamation law, those of a certain age might be grateful to Griffiths J for providing us with with a very clear understanding of what that much over-used word “trolling” means. Read on.
Short summary
The dispute arose from four materially identical letters sent by Optical Express to dissatisfied former patients in 2020 and 2021, each including a paragraph describing Ms Rodoy as “a vexatious individual”, “a self-confessed and known fraudster”, and someone who trolls Optical Express and its staff online, with police involvement said to have been required for employee safety. Prior to trial, Lewis HHJ had determined the ordinary and natural meanings of the words and ruled that they were defamatory statements of fact. The issues at trial were therefore serious harm, truth, qualified privilege, malice, and relief.
Background facts
Ms Rodoy had become a campaigner against refractive eye surgery after her own adverse experience with Optimax in 2011, following which she created the “Optimax Ruined My Life” website and later, after settling with Optimax, the “Optical Express Ruined My Life” site and related social media channels. Optical Express is a major private provider of refractive eye surgery, and after settlement the claimant turned her attention to them. The evidence showed a long-running conflict between Ms Rodoy and the company, including public criticism, site visits, social media posts, and communications with patients and staff. The judge found that she was not a conventional neutral commentator: she ran a campaign, but also repeatedly engaged in personalised online abuse, baiting and humiliation of individual staff members.
A central factual strand was an episode from the 1990s, in which the court found that Ms Rodoy had sold false stories to the media for money about a fictitious detective agency, recruiting others to support the deception and knowingly creating an elaborate hoax for payment. The judge treated that history as highly material to the issue whether she was accurately described as a fraudster [see the “Decoy Dolls” story in paras 144 et seq.]
Legal background
The dispute centred on a series of letters making strong and confident allegations against the claimant. First, the court applied section 1 of the Defamation Act 2013 to determine whether the claimant had suffered “serious harm” as a result of the defendant’s statements and activities. This test has been modified by Lachaux v Independent Print Ltd [2019] UKSC 27 [2020]AC 612 at para 13). It requires proof of some historic harm, which is shown to have actually occurred (para 14). Proof of this will depend on a combination of the inherent tendency of the words and their actual impact on those to whom they were communicated (para 14). The application of section 1 has to be determined by reference to the actual facts about its impact and not just to the meaning of the words (para 12). Because there were four separate publications to four different recipients, the judge assessed harm recipient by recipient rather than aggregating the effects. That approach mattered, because each recipient’s pre-existing attitude to Ms Rodoy and her website differed, as did the extent to which they credited her before receiving the letters.
Serious harm
The evidence of Colin Hind, Robert Beaton and Phoebe Arthur was especially important. Each had some level of engagement with Ms Rodoy’s website or advice, and each accepted that the letters from Optical Express caused them to doubt her credibility, lose confidence in her, and in varying degrees cease or curtail contact [paras 33 – 34]. The judge accepted their evidence that the allegations had immediate and lasting reputational impact, and he found that the statements “stuck” because they came from a large, apparently credible business and directly targeted the trust they had placed in Ms Rodoy. Even in relation to James Marshall, who did not give evidence, the documentary record and the surrounding evidence persuaded the judge that the allegations seriously damaged her reputation in his eyes.
The serious-harm finding is significant for practitioners because it shows that, even with a small readership, serious harm can be established where the audience is highly relevant and the allegation lands with force. The judgment is also a reminder that reputation can be harmed even where the publishee does not know the claimant personally but comes to know them through the claimant’s online presence and associated advocacy.
Truth defence
The main battleground was the section 2 defence of substantial truth. Section 2(4) of the 2013 Act replaces the common law defence of justification. However, the common law principles applied to the former defence of justification before the coming into force of the Act provide guidance which may be applied to the section 2 defence of truth: Dhir v Saddler [2018] 4 WLR 1; [2017] EWHC 3155 (QB), at paras 67-68 per Nicklin J.
The court approached each of the pleaded imputations separately and reiterated orthodox principles: the defendant need not prove literal truth of every detail, only the substantial truth of the sting [paras 109-119]. The first imputation was that Ms Rodoy was a fraudster and that her dishonesty extended to using aliases or different names.
On that issue, the judge made extensive findings about the Decoy Dolls material. He held that Ms Rodoy had, as a mature adult, deliberately fabricated stories for media payment, used false names in some contexts, and orchestrated a long-running deception that yielded substantial income. His findings included that Ms Rodoy used deception to obtain information from a man in a divorce context, sold fictitious media stories about a fake detective agency, arranged for false corroboration, and repeatedly received payment for knowingly untrue material [paras 121–167]. He rejected her attempt to characterise the affair as a harmless prank, saying it was an “out-and-out fraud” done for money [paras 167–176]. He rejected her attempts to minimise the episode as a prank or media-driven exaggeration, finding instead that it was a cynical and dishonest commercial fraud. Importantly, the judge went further and held that the description of her as a fraudster remained substantially true in 2020 and 2021 because she had not shown remorse, repudiation or reformation; on his findings, she remained “the same person” morally and in terms of her own character [paras 177- 187]. That is a striking aspect of the judgment, because it allowed a past fraud to ground a present-tense imputation.
The “aliases” limb was more nuanced. The judge accepted some examples where Ms Rodoy had used or suggested false names, including in relation to the cross-dresser sting and the media hoax, but he rejected the proposition that her true names themselves were aliases used dishonestly. He nevertheless held that any residual inaccuracy on that limb did not add to the sting of the allegation that she was a fraudster, so section 2(3) preserved the defence. He also treated the use of “Mrs More” and “Suriya” in certain Optical Express interactions as tactical subterfuge, but not enough to transform the essence of the allegation.
The second imputation, that she trolls Optical Express and its staff online, also succeeded on truth [see paras 213 – 366]. This term is of relevantly recent coinage and therefore Griffith J’s observation is most welcome:
“. “trolls” and “trolling”, in a context which includes the word “online”, are relatively new words. In different contexts they might bear a variety of meanings. They derive from identical or similar words in older contexts which are themselves quite complicated and diffuse, and even their etymology is ambiguous.” [para 216]
He examines the trolling evidence convincingly largely because he does not rely on labels alone. Instead, he works through a large number of concrete incidents, sets them in chronological and interpersonal context, and asks whether each episode was genuinely campaign-related or was just meant to annoy, humiliate, or unsettle [paras 227–365]. He distinguishes between the Ms Rodoy’s hard-edged advocacy and conduct that had no discernible informational or campaigning purpose, which makes the findings feel grounded rather than merely impressionistic:
“In making this judgment, I am not applying any legitimacy test to what counts as campaigning. Campaigning may be high-minded or go low; it may use conventional tactics or hit below the belt; it may sometimes (although not usually) involve breaking the law. None of that concerns me for the purposes of this defamation action. We live in a free country and a person can be an aggressive, even wrong-headed, campaigner, without being a troll. But I am satisfied that these incidents were not campaigning at all. They were deliberate trolling. [para 361]
Griffiths J goes on to adopt a contextual understanding of “trolling” as malicious or disingenuous online conduct intended to provoke, upset or intimidate, rather than legitimate campaigning. In his long factual survey of the claimant’s website, Facebook, YouTube, direct emails and chat logs, the judge found repeated instances of gratuitous personal abuse, goading and harassment: naming individual staff, inviting readers to contact them, posting personal photos and addresses, and using demeaning nicknames such as “Tweedledum” and “Wee Shugster”. In the court’s view, the campaign was often a vehicle for cruelty and baiting, not just advocacy.
Then there was the third imputation, that at times the trolling was such that there were reasonable grounds to suspect she was placing staff safety at risk. The judge found objective grounds for such suspicion, especially from the escalation involving Dr and Mrs Hannan: posting their telephone numbers, identifying or signalling their home address, trawling personal Facebook photographs, making a threatening pig’s-head email, and then republishing the episode online after police involvement. Griffiths J accepted that the police took the matter seriously and that Dr Hannan’s fears, and his wife’s distress, were objectively reasonable. The result was that even the most serious part of the allegation was substantially true.
Qualified privilege and malice
Although unnecessary to the outcome once truth had been established, the judge also dealt with qualified privilege and malice. He held that Optical Express and the recipients of the letters had an existing relationship giving rise to a common interest: the recipients were patients or former patients seeking redress, while the company had an interest in protecting them and itself from what it reasonably saw as a disruptive and unreliable intermediary [paras 108, 390, 398]. For practitioners, the judgment is a reminder that “fraudster” can be justified by old conduct if the court concludes the claimant has not reformed, and that “trolling” can be proved by a sustained pattern of online behaviour extending beyond ordinary campaigning (see paragraphs above and in the judgment, paras 360–366]. On that basis, the communications fell within the scope of common interest privilege.
The malice argument failed. The court accepted that David Moulsdale, chairman of Optical Express, was the effective publisher behind the standardised paragraphs, but found that he honestly believed the allegations to be true and was motivated by a genuine concern about Ms Rodoy’s influence on patients and staff, not by dominant spite. The omission of background attachments over time, and the imperfect disclosure response, did not persuade the judge that the letters were sent with a primary purpose of defaming rather than informing. In short, even if the claim had not failed on truth, privilege would likely have defeated it.
Significance
This is a highly unusual defamation case because the court was required to make sweeping findings about a claimant’s historic character and online conduct over decades. It is of interest beyond the libel bar for a number of reasons, but mainly for its application of the Defamation Act 2013 to all sorts of activities online. First, it is a forceful illustration of serious harm in a narrow-audience publication where the audience is highly relevant and the allegation “sticks”. Second, it shows how an apparently pejorative label like “fraudster” can be justified by detailed historical evidence, including long-past conduct if the court concludes the claimant’s current character remains unchanged. Third, it is a strong example of how online campaigning can tip into actionable trolling, and how the context of repeated, targeted abuse can justify serious concerns about safety.
It is a long and entertaining judgment, and the judge’s reflections on “trolling” are worth dwelling upon. His definition is obviously tailored to the facts; is it serviceable in a wider context? As mentioned above, Griffiths J treats trolling as malicious or disingenuous online conduct aimed at provoking, upsetting, or deceiving someone, rather than legitimate campaigning or robust criticism [see paras 222–226]. He also makes clear that the ordinary and natural meaning in this case includes an element of bad faith and a purpose beyond mere expression of a view [paras 222–226]. This a sensible legal definition because it captures why the conduct is objectionable without requiring the court to police political or consumer campaigning more generally [para 361].
It remains to be seen whether “trolling” as online baiting can be applied to sustained harassment online, without evidence of personal abuse, demeaning nicknames, and police involvement [paras 371–385]. So the judgment gives a convincing working definition for this dispute, but perhaps not a universal one.
The practical takeaway is that, in defamation litigation, the court will look at the whole factual relationship between the parties, not just the text of the impugned words. Here, that broader factual canvas decisively favoured the defendant.
