Riley & Anor v Heybroek

Ms Rachel Riley (“the First Claimant”) is a television presenter. Ms Tracy Ann Oberman (“the Second Claimant”) is a television, theatre and radio actress. Both are Jewish and take active positions against the rise of anti-Semitism in the UK over recent years, which rise they attribute in the main to individuals and groups on the left wing of politics. Both use Twitter in order to speak out against this phenomenon. Ms Myrna-Jane Heybroek (“the Defendant”), a member of the bar, is equally active on Twitter. Beyond being broadly supportive of Jeremy Corbyn it is pointed out on her behalf that there is no direct evidence of her political position.

On 18^th January 2019 Mr Shaun Lawson published an article on Medium.com headlined: “Beneath Contempt: How Tracy Ann Oberman and Rachel Riley harassed, dogpiled and slandered a 16-year-old child and her father” (“the Lawson article”). This is a lengthy piece which contains numerous hyperlinks to other articles and materials on which the Defendant places some reliance as valuable context. On the same day, Mr Lawson sent a Tweet from his Twitter account with an embedded hyperlink to the Lawson article, no doubt with the intention of securing a wider dissemination. This in turn was retweeted by the Defendant, also on the same day, with the same hyperlink facility.

Following pre-action correspondence in which the parties’ respective positions on some of the issues I have to decide are not the same as they are now, the claim for libel was filed on 12^th November 2019. The claim was not based on the retweet but on the Defendant’s alleged publication of the Lawson article. This is admitted for the purposes of this application. Before filing a Defence, the Defendant applied under CPR Part 23 for the determination of the following preliminary issues under the provisions of CPR PD53B, para 6.1, viz.:

The parties have filed detailed skeleton arguments in line with the Order of Warby J given on 15^th April 2020 (although in both cases the page limit was exceeded). A remote hearing took place by telephone. I am grateful to Mr Ian Helme and Mr John Stables for their written and oral submissions, the latter advanced in somewhat difficult circumstances. I regret that it has proved necessary to conduct a further remote hearing after a draft of my judgment was provided to the parties.

This extends to 115 paragraphs and informs readers at the outset that it is a 27minute read.

Para 3 of the Lawson article is part of the exordium:

“My comment was in relation to her indefensible behaviour towards [R], the 16-year-old girl whose deplorable treatment by Rachel Riley was at the heart of my weekend piece; and even

[R]’s father. This article sets out in detail exactly what that behaviour involved, the consequences for the poor child, as well as the latest depth-plumbing by, yet again, Riley herself.”

I should make a number of observations about this. First, the 16-year-old child is identified in the Lawson article by her first name. The parties sensibly decided that she should be anonymised as “R”. Secondly, the “weekend piece” – hyperlinked in the online version of the Lawson article – is a reference to a blog published on

Returning to the Lawson article, paras 4-24 provide an overview of certain aspects of contemporary anti-Semitism, the author explaining (not of course for the first time) that this form of racism is more prevalent on the political right, and is far from being an endemic problem on the left. Mr Lawson’s position is that the rightwing press have distorted the reality by labelling people as anti-Semites who are not. The parties agree that this section of the Lawson article amounts to comment or opinion and does not seek to purvey hard fact – although data are provided which are said to support the central thesis. They also agree that the rights and wrongs of this debate are irrelevant to the claim.

The Lawson article changes course at para 26 by returning to the experiences of R. It is said that when she challenged the First Claimant “a horrendous bullying dogpile was the result, which disgraced all involved”. As I have said, the factual basis for this assertion (I am continuing to use language which does not prejudge the fact/opinion issue) is provided in the First article on which the Defendant places reliance. Briefly, there were a series of exchanges over Twitter between the First Claimant and R over the anti-Semitism issue. Following these exchanges:

“108.

Wisely, [R] went on to block Rachel. Unfortunately though, the way Twitter works, if you block someone with a large following, you still receive replies from everyone else commenting on the same thread. Not only that, but Riley’s supporters were offended, so a pile-on began. Against a 16year-old child.

109.

This was not Riley’s responsibility exactly; not at this stage, at least. She never wrote directly to [R] again; she didn’t directly encourage the dogpiling. But as a public figure, the lack of responsibility or remorse she’s demonstrated for the horrendous bullying [R] has experienced has been horribly instructive. And as I noted above, she has no compunction with encouraging it towards … more or less the entire Labour support on Twitter either.”

These events occurred at the end of 2018. Moving on both literally and metaphorically to the Lawson article, on 11^th January 2019 R sent a tweet to the Second Claimant telling her that she had “no idea of the distress having a huge pile-on has caused”. The Second Claimant’s reaction was to send R the following tweet at 9:46am on the same day:

“But I’d love to meet you [R]. When you are next in London let’s go for a coffee. Maybe @charley_yorks [another young woman] could meet us too. It would be great to connect young people who could maybe open a dialogue.” (para 29 of the Lawson article)

The offer to meet was repeated by the Second Claimant on a number of occasions that same day in the terms set out verbatim in the Lawson article. For example:

“Aw [R] I would love you to meet a young tweeter called @charley_yorks. She tweeted me terrified at coming out as Jewish at Uni because of abuse. Like you she has been v brave.

Come & meet us for tea. The youth should do it better. Talking listening making connections. My treat.”

Later that same morning, a friend of R warned the Second Claimant that her invitations were “kind of creepy” and that R had bad anxiety. The Second Claimant’s reply was as follows:

“Oh no. We didn’t mean to creep R out. I would absolutely love to meet her and I think she love to meet Charlie. She can bring parents too. Genuine offer.”

The Lawson article then stated (I use a neutral verb at this stage): “Tracy: wake up and smell the restraining order!”.

Friends of R warned the Second Claimant to stop pestering her. This prompted a number of further tweets from the Second Claimant including the following:

“Oh dear @thelittleleftie I think you’ve inadvertently become a tool for tweeters like Robin. Be careful not to be used. I won’t dignify Robins maliciousness to me but I still genuinely wd love to take you & your parents to tea. Fortnum’s (as Robin has suggested) or anywhere,”

The Second Claimant’s tweets continued in similar vein. As before, supporters of

The Second Claimant’s tweets continued. At 2:13am on 12^th January she sent a tweet stating, “I blame the dad”. The Lawson article then states:

“When a 16-year-old child has spent the entire day being harassed, in public, by a celebrity, there’s only one thing that’s going to make her feel even worse. When her father is attacked too. Yet that’s what Oberman did. All [R’s] Dad had done was try to protect his daughter from this utterly revolting madness. We all blame the Dad, Tracy; just not quite in the way you might imagine.”

Meanwhile, R was posting screenshots of what was described as the abuse she had been receiving from the First Claimant’s followers.

It is unnecessary to refer to the entirety of the Second Claimant’s tweets. Mr Lawson’s summary of the position was as follows:

“Sending this child 16 unsolicited invitations to meet a complete stranger, around 60 unsolicited tweets in total, slandering anyone trying to protect her and even defaming her Dad is quite another. And on the subject of libel: telling the truth is an absolute defence. But hey ho, who cares about any of that? It’s only harassment of a terrified child by someone more than three times her age, after all.” (para 72)

Similarly, para 83 of the Lawson article describes the Second Claimant’s tweets as “relentlessly harassing a terrified child”.

Up to this point, the Lawson article has focused on the Second Claimant. Paras 85-

Mr Helme points out that the Claimants do not complain about anything that appears in the Lawson article before para 73. Furthermore, certain later disparagements of the Claimants (e.g. in para 93) are not part of the words complained of either. For the purposes of this litigation the Claimants rely on extracts from thirteen paragraphs of the Lawson article, and these appear in the Annex to this judgment.

Mr Stables did not seek to differentiate between the individual positions of the First Claimant and the Second Claimant, and so from his perspective their cases stand or fall together. Mr Helme submitted that it might be thought odd that according to the Claimants the words complained of mean the same in this context whereas what is actually said about the individual Claimants in the Lawson article is dissimilar. I have three observations. First, I do not read the Lawson article as suggesting that the Claimants have colluded with each other to harass R: it follows that, at least in principle, separate consideration should be given to each of them. Secondly, it is true that the evidence relied on by Mr Lawson to support the proposition that the Claimants or either of them have harassed R differs as between them and is more extensive in relation to the Second Claimant than it is as regards the First Claimant. Thirdly, and notwithstanding the foregoing, my overall assessment of how the reasonable reader would understand the Lawson article read as a whole is that no material difference falls to be drawn between the Claimants. This equation is borne out by the terms of the headline, the author’s conclusions, and the overall impression generated by the article itself.

These are not substantially in dispute. A valuable compendium has very recently been provided by Nicklin J in Riley v Murray [2020] EWHC 977 (QB) which I can gratefully adopt, recognising as I do that the subject-matter of that case was not a blog or an article but a tweet. In these circumstances, I consider that I should be attempting to provide a distillation rather than a full recitation of the governing legal principles tailored to what is germane to the instant case.

The general approach to determination of meaning is that expounded by Nicklin J in Koutsogiannis v The Random House Group Ltd [2019] EWHC 48 (QB). All of these principles are relevant, but in view of the parties’ submissions I consider that I should be paying particular regard to key principles (iv), (v), (viii), (ix) and (xii). In short, the court is enjoined to eschew an overly analytical approach (the default mode of the lawyer); the entirety of the article must be considered (even if the

Two issues arise on the parties’ submissions in connection with the question of context and mode of publication. The first is whether the fact that the Lawson article was hyperlinked to a tweet is a relevant consideration in determining the meaning of the latter. The second is whether the First article provides any relevant context.

On the first of these issues, I naturally bear in mind what Nicklin J said in Riley v Murray, at paras 13-16. However, in the present case the words complained of do not form part of a tweet and the fact that access to them is gained through a hyperlink in a tweet is, at best, of marginal relevance. The meaning would be the same if Mr Lawson had been sued directly. The article extends over many pages and the author has shown no sign of believing himself constrained by any prescribed word-count or, indeed, the virtues of brevity and concision. It must follow that the general exhortation to apply an impressionistic approach is applicable, but not the light-touch, fluid and hyper-impressionistic approach enjoined by the recent tweet-specific jurisprudence. This may be just a matter of fact and degree, but it is nonetheless of some importance.

As for the second of these issues, I have concluded that nothing really turns on it. Even if the First article is brought into consideration, it does not bear upon the meaning of the words complained of in the Lawson article, or indeed the article as a whole. Nor – and this is really the Defendant’s purpose in advancing the submission – does it impinge on the question of whether the words complained of amount to fact or opinion. I will be returning to this last point, but the real objection to the Defendant’s argument is that all relevant context is provided within the four corners of the Lawson article itself, and one need travel no further. In any event, applying the principles set out by Nicklin J in Falter v Atzmon [2018] EWHC 1728 (QB) cannot lead to the conclusion that the reasonable hypothetical reader, even one with a keen interest in this debate, would open the particular hyperlink to the First article and read all of it. I have reached that conclusion having regard to the large number of hyperlinks in the Lawson article and accepting Mr Stables’ submission that for the First article to be brought into account it would have to be regarded as at the very least probable that such a hypothetical individual would read the whole of it. This would have required over forty minutes’ reading time. It does not serve the Defendant’s purposes to show that the First article would have been skimmed or read only in part. Para 12 of Nicklin J’s judgment in Falter is particularly relevant here.

The second broad question for determination is whether the words complained of amount to fact or opinion. Again, the general principles may be taken from Koutsogiannis, paras 16 and 17, with the ultimate question being how the words would strike the ordinary reasonable reader. However, I draw attention to Nicklin J’s fourth point of general application:

“Some statements which are, by their nature and appearance opinion, are nevertheless treated as statements of fact where, for instance, the opinion implies that a claimant has done something but does not indicate what that something is, i.e. the statement is a bare comment.”

A good example of the converse of this is to be found in Nicklin J’s very recent judgment in Peck v Williams Trade Supplies [2020] EWHC 966 (QB), handed down after the first remote hearing in this case. I will be returning to this under para 30 below.

In the light of the parties’ submissions, I should deal at this juncture with a number of discrete points.

First, Warby J in Yeo v Times Newspapers [2015] 1 WLR 971 emphasised that political speech is particularly highly valued in the sense that “the court should be alert to the importance of giving free rein to comment and wary of interpreting a statement as factual in nature, especially where … it is made in the context of political issues” (para 97). If it had been relevant, I would certainly have applied this guidance to the opening section of the Lawson article (i.e. paras 4-24) which deals with the anti-Semitism debate generally. However, I cannot regard Mr Lawson’s treatment of the interactions by tweet between the Second Claimant and R as amounting to political speech or anything close to it. What I am calling the anti-Semitism debate generally provides relevant background and context, in the sense that the Second Claimant would presumably not have been sending these tweets to a supporter of hers, but it does no more than that.

Secondly, I agree with the Defendant that Nicklin J’s judgment in Greenstein v Campaign Against Anti-Semitism [2019] EWHC 281 (QB) is helpful, but it is not as highly relevant as is claimed. The issue in that case was whether an imputation that the claimant was a notorious anti-Semite was one of fact or opinion. Nicklin J held that it amounted to the latter, because (amongst other things) an explanation of the author’s basis for the criticism was given and the overall context was the highly contentious one of what amounts to anti-Semitism. In my judgment, reasonable people may disagree at the margins as to what amounts to harassment, but it cannot be characterised as a highly contentious issue. However, there may be greater force in the contention that the factual basis for Mr Lawson’s criticisms of the Claimants was fully set out in his piece.

Thirdly, and following on from the observation I have just made, it may be important for the court to identify whether the statement or statements under consideration relate to a clear, identifiable factual substratum such that it or they can properly be regarded as commentary. As Lord Porter explained in Kemsley v Foot [1952] AC 345:

“The question, therefore, in all cases is whether there is a sufficient substratum of fact stated or indicated in the words which are the subject-matter of the action, and I find my view well expressed in the remarks contained in Odgers on Libel and

Slander (6th ed, 1929), at p.166. ‘Sometimes, however,’ he says, ‘it is difficult to distinguish an allegation of fact from an expression of an opinion. It often depends on what is stated in the rest of the article. If the defendant accurately states what some public man has really done, and then asserts that ‘such conduct is disgraceful,’ this is merely the expression of his opinion, his comment on the plaintiff’s conduct. So, if without setting it out, he identifies the conduct on which he comments by a clear reference. In either case, the defendant enables his readers to judge for themselves how far his opinion is well founded; and, therefore, what would otherwise have been an allegation of fact becomes merely a comment. But if he asserts that the plaintiff has been guilty of disgraceful conduct, and does not state what that conduct was, this is an allegation of fact for which there is no defence but privilege or truth. The same considerations apply where a defendant has drawn from certain facts an inference derogatory to the plaintiff. If he states the bare inference without the facts on which it is based, such inference will be treated as an allegation of fact. But if he sets out the facts correctly, and then gives his inference, stating it as his inference from those facts, such inference will, as a rule, be deemed a comment. But even in this case the writer must be careful to state the inference as an inference, and not to assert it as a new and independent fact; otherwise, his inference will become something more than a comment, and he may be driven to justify it as an allegation of fact.” (at 356) [emphasis supplied]

Fourthly, Mr Helme submitted on the authority of Branson v Bower [2001] EMLR 32 that it is relevant that a significant portion of the words complained of appear in the section of the Lawson article headed “Conclusion”. I do not think that Branson goes that far. In my view, the fact that statements are made in the concluding section of an article of this nature is no more than weakly indicative that they may be opinion. Ultimately, though, there is no issue of legal principle here: everything depends on the facts and context of the individual case.

The third broad question is whether the statements complained of defamed the Claimants at common law. The test is that set out by Warby J in Allen v Times Newspapers Ltd [2019] EWHC 1235 (QB), at para 19(i):

“ At common law, a statement is defamatory of the claimant if, but only if, (a) it imputes conduct which would tend to lower the claimant in the estimation of right-thinking people generally, and (b) the imputation crosses the common law threshold of seriousness, which is that it "[substantially] affects in an adverse manner the attitude of other people towards him or has a tendency so to do": Thornton v Telegraph Media

Group Ltd[2010] EWHC 1414 (QB)[2011] 1 WLR 1985[96] (Tugendhat J).”

Further, as was explained by Warby J in Sube v NGN Ltd and another [2018] EWHC 1234 (QB):

“The second reason is that the imputations are very plainly expressions of opinion. If an article consists of a clearly stated non-defamatory account of the claimant's behaviour, coupled with the expression of a derogatory opinion about that behaviour, the fact that the opinion is clearly presented as such must mitigate its defamatory impact. The derogatory statement will be seen for what it is: someone's evaluation of the behaviour laid out for the reader's consideration. And if, as here, the opinion expressed is not particularly harsh, the impact of its publication may fall short of the s 1 threshold. That is the position here, in my judgment. The third and contributory reason is the source of the opinions expressed. Where explicit, the statements that convey the opinions complained of derive from neighbours, and officials. Where implicit, they are the insinuations of the publisher. None of these are authoritative sources, which the reader would take to be better able to judge the situation. The reality is that readers are likely to form their own assessment of the facts presented to them, perhaps influenced, but not determined, by the opinions expressed or implied by the articles.” (at para 42)

Finally, the authors of Gatleyon Libel and Slander, 12^th edn. observe:

“Whether the threshold of seriousness has been met is a multifactorial question, that must be viewed in light of the rights in article 8 and article 10, and that will require the court to consider matters such as the nature and inherent gravity of the allegation, whether the publication was oral or written, the status and number of publishees and whether the allegations were believed, the status of the publisher and whether this makes it more likely that the allegation will be believed, and the transience of the publication. The result in each case will depend on the particular facts…” (at para 2.4)

It is convenient at this juncture to set out the Claimants’ pleaded meaning, which is:

For the reasons explained below, I do not set out the Defendant’s version as particularised in Mr Helme’s skeleton argument.

I have already addressed a number of the parties’ submissions during the course of my review of the governing law. Given the detailed nature of the written arguments in this case, I am able to concentrate on certain key matters.

Mr Helme submitted that this is an unusual case where the second issue (sc. fact versus opinion) should be determined in advance of the first (meaning of the words complained of). He relied on the approach of the Court of Appeal in British Chiropractic Association v Singh [2011] 1 WLR 133. Applying that approach, he took me carefully through the Lawson article and submitted that the author at all material times had clearly identified the material – the specific tweets - that was the factual basis for Mr Lawson’s statements. Mr Helme contended that it was noteworthy that the Claimants were not suing on any of the numerous criticisms made of the Second Claimant in particular before para 73. As regards these criticisms, it is clear from the structure of the Lawson article that the author has set out the facts (i.e. the tweets) and then separately commented on them. Those comments are clearly identifiable as such, rather than as imputations of fact. The same applies to the subsequent commentary (para 73ff) because it relates back to the factual substratum which has been earlier identified. Mr Helme submitted that the Claimants’ approach wrongly treats the statements they have chosen to sue upon as somehow dissociated from any underlying factual foundation.

In his skeleton argument Mr Helme proposed a somehow elaborate and wordy version of the meaning conveyed by the Lawson article which he did not press particularly strongly in oral argument. In these circumstances, I have decided to proceed on the basis that Mr Stables’ competing text is a good starting-point, subject to refinement and adaptation in the light of the parties’ further oral submissions. Ultimately, Mr Helme focused his oral argument on Mr Stables’ version and made helpful submissions about it.

As for sub-para (1), Mr Helme did not dispute the Claimants’ formulation save as regards the epithet “serial” and was more concerned to contextualise the words complained of as the essential stepping-stone for his overarching submission that they amount to comment/opinion. As for sub-para (2), the issue between the parties turns on the use of the verb “encouraged”. In his skeleton argument Mr Helme submitted that an allegation that the Claimants have encouraged others to act in a certain way is not an actionable criticism of them without it also being said that they did so deliberately or recklessly. In oral argument he submitted that the Lawson article does not specifically allege encouragement. Finally, in his reply Mr Helme submitted that the verb “inspired” could be substituted for “encouraged”.

On the fact/opinion question, Mr Helme advanced a number of submissions in relation to “encouraged”. He submitted that the reasonable reader would recognise this part of the statement complained of in sub-para (2) (sc. on the Claimants’ original wording, “the Claimants’ campaign of bullying and abuse has encouraged others to follow their lead and resulted in …”) as an inferential conclusion rather than a statement of fact. He advanced four reasons in support which are addressed below.

As for sub-para (3), Mr Helme submitted that this is a Chase level 3 case because, at its highest, all that para 112 of the Lawson article was saying is that the “details” are “in the hands of the police”; it is not being alleged that a police investigation is being undertaken. Further, given that it was clear from the overall context that it was Mr Lawson or a third party known to him who had contacted the police, and would do so in future if there were any repetition, the reasonable reader would take this to be comment rather than a statement of fact.

Mr Stables submitted that I should recognise the very central significance of an approach based on impression and an absence of close analysis. A reasonable reader would conclude that the Claimants’ actions amounted in fact to serial harassment of R and nothing else: there are numerous references to this in the Lawson article, as well as references to abuse. A reader would not understand the references to harassment in particular as Mr Lawson’s interpretative gloss on what the Claimants’ did. Rather, they would understand what was being said as amounting to an allegation in categorical, Chase level 1 terms that the Claimants had as a matter of fact harassed R. Furthermore, the presentation of the evidence is entirely one-sided and there is no antidote of any sort. I admire the rather elegant way in which Mr Stables reinforces this last point:

“The text is consumed with accusation; there is no balance. Therefore the Cs are alleged to have actually committed the acts, the misconduct, alleged against them.”

As for sub-para (1), Mr Stables conceded that “acted obscenely” was opinion but strongly submitted that the remainder would be understood by the reasonable reader as amounting to a statement of fact.

Mr Stables made the same point in relation to the verb “encouraged” in sub-para (2): a reasonable reader would not understand this to be an expression of Mr Lawson’s opinion as to the quality of the Claimants’ conduct but a separate, further allegation of wrongdoing. In essence, what is being alleged is that the Claimants were pursuing conduct which they knew were capable of motivating others to perpetrate “terrible mischief”. He submitted that the facts of the misconduct encouraged make concrete the allegation of encouragement itself.

As for sub-paragraph (3), Mr Stables submitted that this was a Chase level 1 case. In the alternative, if I were not minded to hold that the level of culpability should be expressed at this pinnacle, because the imputation falls short of actual wrongdoing by the Claimants, I should hold that this is an example of what has on occasion been called a Chase level 1½ case: the true meaning is that there are strong grounds to suspect the Claimants of having committed criminal offences, including the encouragement of death threats.

I will reflect a number of the parties’ further submissions in the analysis which follows.

I begin with two introductory observations.

In my judgment, the instant case does not fall into that rare category of case where the issue of fact/opinion should be determined before the issue of meaning, out of concern that the determination of the latter will tend to “stifle” the analysis of the former. The approach in Singh is exceptional and applies only to cases where the allegations concern a subject about which different conclusions are inherently value laden. On the other hand, there is authority for the proposition that these two issues should not be taken separately, and that a “flexible and holistic approach” is to be preferred: see Peck’s case, at para 11(ii). Although my judgment is structured in a way which appears to take the meaning issue first, this does not imply any rigidly sequential thinking. I note that this flexibility underlay my own approach in para 22 of Triaster v Dun & Bradstreet Ltd [2019] EWHC 3433 (QB).

Nor is it helpful in my judgment to characterise the Lawson article as a paradigmatic opinion piece. A reasonable reader would no doubt understand it in that way when reading paras 4-24, but as I have pointed out the article rather changes course when it begins to set out what happened to R. I have observed that the Claimants and R appear to sit at opposite poles of the discourse about antiSemitism, but that does not somehow make the tweets under consideration a part or playing out of that debate. On the other hand, my conclusion that the Lawson article is not an opinion piece tout court does not mean that it does not contain opinions, as Mr Stables accepted.

I have set out the Claimants’ pleaded meaning at para 35 above. It is necessary to refine and adapt this in the light of the parties’ focused submissions.

As for sub-para (1), the issue between the parties is fairly narrow. Para 101 of the Lawson article does mention “obscene conduct”, but an examination of the language used overall might well lead the reader to prefer the adverb

“outrageously” (para 108) or “contemptibly” (para 109). Given the possible overtones conveyed by “obscenely”, which in my view it is to the purpose of the Lawson article to impart, I would prefer “outrageously”.

The phrase “serial bullying and abuse” is an attempt to encapsulate a number of nouns and descriptors whose meaning is broadly synonymous. My preferred formulation would be: “repeated harassment and abuse”. It is arguable that the Lawson article could be understood as stating that the Claimants have bullied R, but the noun “harassment” appears on several occasions and seems to me more accurately to capture the overall sense. Further, I see little force in the argument advanced in later written submissions that “repeated” or its synonym is effectively otiose because “harassment” necessarily connotes ongoing behaviour. It is correct that the legal definition of “harassment” requires persistence, but the reasonable reader would approach the matter less forensically. There is no unfairness in using

“repeated” because the Lawson article indicates on how many occasions this conduct is said to have occurred.

As for sub-para (2), it seems to me that the Defendant’s case has vacillated somewhat (see para 40 above).

It is correct that the word “encouraged” does not appear in para 101; there, it is said more neutrally that the matters itemised have been the consequence of the Claimants’ obscene conduct. The issue is whether a hypothetical reasonable reader would, as it were, read into this more than the bare causal connection. In my judgment, there are a number of factors which lead to the conclusion that the meaning of the Lawson article viewed both as a whole and impressionistically is that the Claimants appreciated or apprehended that these deleterious consequences would or might ensue. I have in mind:

In my judgment, the cumulative impact of these various citations, as well as the tone and tenor of the Lawson article read as a whole, would lead the hypothetical reasonable reader to conclude that the meaning intended to be conveyed was that the Claimants were in some way morally responsible for the consequences specified under para 101. The same reader would also no doubt bear in mind that the Claimants as experienced users of Twitter would tend to know how others can behave. The verb “encouraged” is appropriate to convey this element of responsibility because it connotes a measure of realisation or understanding that their actions could impact on the conduct of third parties, namely the Claimants’ alleged supporters.

As a separate matter, Mr Helme baulked at the parenthesis in sub-para (2) of the

Sub-para (3), as pleaded by Mr Stables, is not borne out by the language of the Lawson article, even read on an impressionistic basis. Paras 77, 102 and 112 mention police involvement, although only this last para clearly relates to the Claimants. According to para 112, these matters are in the hands of the police. The reasonable reader would conclude that the police were in some way seized of the matter, not that criminal offences have been committed or that the police had decided to charge such offences. It follows that any meaning at Chase level 1 cannot be sustained.

Nor can I accept Mr Stables’ alternative submission that the meaning is, “there are strong grounds to suspect the Claimants of having committed criminal offences, including the encouragement of death threats”. I have two difficulties with this. First, I do not consider that this is a Chase level 1½ case: see, for a statement of the principle, para 17 of Nicklin J’s judgment in Brown v Bower [2017] 4 WLR 197, applied by me in Travel Insurance Facilities Plc v The Times [2019] EWHC 1337 (QB). In this latter case, the matters in question were being actively investigated by regulators. Secondly, the Lawson article does not specify exactly what is in the hands of the police, and I do not consider that the reasonable reader would interpret para 112 as indicating that it would have gone further than the core allegation of harassment.

In submissions advanced on 7^th May it seemed to me that the issue between the parties eventually crystallised into whether this should be regarded as a Chase level 2 or level 3 case. The difference between an imputation of grounds to suspect someone of wrongdoing and an imputation of there being grounds to investigate what someone has done is often difficult to draw, although the authors of Gatley have observed that level 3 meanings are rarely found.

On this issue I agree with Mr Helme that paras 77 and 102 of the Lawson article cannot be deployed to increase the gravity of the meaning that arises from para 112 itself. Para 77 in particular refers to a police “investigation”, but I also agree with Mr Helme that this does not clearly relate to the Claimants. Conversely, I disagree with Mr Helme that a Chase level 2 finding would be inconsistent with my conclusion (see para 72 below) that sub-para (1) of the Lawson article is opinion. I consider that this raises a separate issue. In the present case, the dividing-line between Chase level 2 and level 3 has proved to be particularly difficult to detect, given the author’s use of metaphor. I have concluded that an allegation that the details are in the hands of the police should not amount from the viewpoint of the reasonable reader to an allegation that the matter is actively under police investigation. To reach such a conclusion the reader would be forced to speculate. It follows that I cannot accept Mr Stables’ submission, based on Gatley at para 3.28, that “in practice a statement that C is under investigation will almost always justify the inference that the police have some basis for suspicion”. The Lawson article falls just short of making an imputation to this effect.

Accordingly, I am not prepared to find that a Chase level 2 meaning is applicable to sub-para (3). I appreciate that Chase level 3 rarely applies, but in my judgment the reasonable reader would understand para 112 as meaning, or implying, that a police investigation would be appropriate. Thus, the third sub-meaning is: “there are grounds for the police conducting an investigation into the Claimants’ conduct”.

The fourth sub-meaning is said to be derived from a combination of paras 105, 106 and 108 of the Lawson article. The Defendant submits that this is an utterly strained meaning. I would not accept the hyperbole, but on balance I have concluded that a hypothetical reasonable reader would understand the statements in these paragraphs as being a reference back to the Claimants’ alleged conduct in relation to R rather than any overarching statement as to how they comport themselves more generally. Although this is a point which tends to cut both ways, I consider that the Lawson article fails to lay the evidential ground for any such sweeping statement.

In my judgment, therefore, the single meaning of the words complained of is as follows:

The ambit of the contest between the parties should be summarised. As I have said, they agree that “acted outrageously” in sub-para (1) is opinion but differ as to the clause, “subjected a vulnerable 16-year-old to repeated harassment and abuse”. As for sub-para (2), Mr Stables’ submission was that the whole of it is a statement of fact. Mr Helme did not dispute this in relation to the words “… further harassment and abuse” to the end of sub-para (2), but submitted that the opening clause, “the Claimants’ actions have encouraged others leading to …” is opinion. As for subpara (3), Mr Helme submitted that the words complained of amount to an expression of opinion.

I begin with sub-para (1). I do not think that the correct question is: does this look like a statement of fact? The reasonable reader would require further context and explanation. How the words would strike the ordinary reasonable reader is the ultimate issue, but I see force in Mr Helme’s formulation, “can the reader recognise that the statement complained of is, or can reasonably be inferred to be, a deduction, inference, conclusion, criticism etc.?”

In addressing that question, the principles set out in the authorities are no more than that; they are not rules of universal application. If it were clear from the overall context that “harassment” and “abuse” did not entail the making of any sort of value-judgment by the author, the answer to Mr Helme’s question would be in the negative. My impression of the Lawson article is that it is laden with such judgments, however tendentiously articulated. In these circumstances, one of the principles of particular relevance in a case such as this is whether the article read as a whole indicates with sufficient clarity and precision the evidential basis for the statement under consideration: in other words, whether there is a sufficiently clear setting out of what the individual allegedly defamed has done. In the absence of a stated evidential basis, the court should be more inclined to find that the statement under consideration is one of fact. Conversely, the existence of such an evidential basis may well lead to the contrary conclusion, because the reader would understand that the author was expressing an opinion about something specific.

If the Claimants had chosen to complain about those earlier sections of the Lawson article which clearly amounted to commentary on specific tweets (here, I accept, I am using a less neutral term than I did previously), I consider that the conclusion that these amounted to opinion and not fact would not be overly difficult to reach. The Claimants have not chosen to do that, but this self-denying ordinance makes no difference to the analysis. Focusing on later sections of the Lawson article makes it easier to contend that the author is making a series of unevidenced assertions, but in reality all that he is doing is repeating, by way of summation, the opinions he had earlier articulated. The reasonable reader would understand the words complained of in that way: as being a reference back to the tweets, the terms of which have been set out.

An important point here is that the reasonable reader is in a good position to assess for her or himself whether these tweets do amount to harassment. In this respect, the fact that Mr Lawson has chosen to use language that some might observe is loaded if not inflammatory is nothing to the point. The reader would not be oblivious to what the tweets actually say.

In oral argument at the first hearing, Mr Stables submitted that Mr Lawson has “served up what purports to be evidence for his allegations”. Without delving into the underlying merits of this case unnecessarily, I can see some force in that submission. However, I think that it travels a step too far. Putting the Claimants’ case at its highest, the evidential basis for the evaluative judgment – that these tweets amount to harassment – may not be particularly strong but it is not nonexistent; or, at least, it would not be right for me to reach that conclusion on an application of this sort. As I have already said, it is not the role of the court at this stage of the analysis to evaluate the strength of the parties’ cases on the merits. It is sufficient, as Lord Porter explained in Kemsley’s case, that the author has accurately explained what the Claimants have done. Mr Lawson has done that by setting out the terms of the tweets. If the reasonable reader is left wondering whether these tweets amounted to harassment, this is only because sufficient information has been provided to engender that scepticism.

It may be that Mr Lawson has been selective about which tweets he has chosen to reference. I do not think that matters, at least for the purposes of the fact/opinion discussion. Mr Stables also submitted that the headline is couched in the language of fact rather than opinion. I consider that it is neutral, and ultimately the issue must be resolved by assessing the Lawson article as a whole.

I must therefore conclude that the clause, “subjected a vulnerable 16-year-old to harassment and abuse”, is a statement of opinion and not of fact.

The next issue is whether the opening clause in sub-para (2), “the Claimants’ actions have encouraged others leading to …” are fact or opinion.

I have found this aspect of the case particularly difficult and confess that my mind has wavered. If the focus is merely on the causal connection between the Claimants’ actions and those of third parties, the reasonable reader would say, as Mr Helme concedes, that the statement is one of fact. If, on the other hand, the focus is on the Claimants’ motivation and state of mind, the authorities relied on by Mr Helme hove clearly into view. The verb “encouraged” imports both a causal connection and, on my earlier finding, a state of mind. It may be seen that this is somewhat of a hybrid situation.

Upon careful reflection, I have concluded that Mr Helme’s analysis is correct, but not for all the reasons he has given.

First, the word “encouraged” does not appear in para 101 of the Lawson article. My conclusion on meaning is that the reasonable reader would infer the attribution to the Claimants of knowledge of a causal connection from the overall context. I cannot accept Mr Stables’ submission that this militates in favour of the statement being one of fact as opposed to “some related re-expression or continuing expression of Mr Lawson’s opinion”. If anything, the need to draw out an inferential meaning is a factor militating against the verb “encouraged” (and other parts of the clause) containing a statement of fact.

Secondly, I accept Mr Helme’s submission that Mr Lawson should be understood as drawing an inference as to the Claimants’ state of mind and motivation. If, as I

It should be emphasised that these authorities do not state any principle of invariable application: see Duncan & Neill on Defamation, 4^th Edn., at 13.16, and Pena v Tameside Hospital NHS Foundation Trust [2011] EWHC 3027 (QB). In Branson (No 1) at first instance, Eady J expressed the principle in these cautious terms:

“Thus, if a journalist makes inferences as to someone’s motives, that may be treated as the expression of an opinion even though the inference drawn may be to the effect that there exists a certain state of affairs (including a state of mind): see

Gatley (9th Edition) paragraph 12.10 and Kemsley -v- Foot

[1952] AC 345, page 356.”

This dictum has been upheld by the Court of Appeal on two occasions, including in Keith Burstein. But as Keene LJ explained in that case, “the issue is whether the words complained of …, when seen in context, bear the unmistakable badge of comment” (at para 24).

Thirdly, I accept Mr Helme’s submission that paras 37-38 and 87-88 of the Lawson article in particular are a clear indication that the author at all material time is doing no more than proffering comment on the Claimants’ state of mind.

I am unable to accept Mr Helme’s submission that my holding in relation to subpara (1) (viz. statement of opinion) segues seamlessly into the opening words of sub-para (2). Sub-para (1) characterises the Claimants’ conduct in particular (opinionated) terms, but sub-para (2) makes a different and separate point as to the consequences of that conduct. Accordingly, I would regard my finding on sub-para

However, there is greater force in the submission that Mr Helme developed orally that the Lawson article does identify the factual basis for the inferential conclusion of encouragement. The relevant tweets have been set out as well as the action by third parties that is said to have been fomented by them. That said, the identification of a factual basis would lead nowhere if the statement at issue were clearly one of fact.

If the Lawson article meant, as distinct from merely said, that “the consequences of [the Claimants’] obscene conduct have been etc.”, then the statement would be understood as one of fact. It flows from my decision on the first issue that the words complained of mean more than that; or, put another way, that one should not stop at the purely literal meaning. The notional insertion of “encouraged” into the meaning is designed not merely to emphasise the causal connection, which is already clearly made by the use of the word “consequences”, but to draw attention to the moral quality of the Claimants’ actions. It is for this reason that Mr Helme is right in bringing the discussion round to the Claimants’ motivation.

For all these reasons, I would hold that the opening clause of sub-para (2), “the Claimants’ actions have encouraged others leading to …”, contains a statement of opinion and not one of fact.

In my view, on the basis that sub-para (3) means, “there are grounds for the police conducting an investigation into the Claimants’ conduct”, whether this statement would be understood by the reasonable reader as fact or opinion probably makes little practical difference. But I should address Mr Helme’s submission that, given that this reader would infer from the overall context that it was Mr Lawson or someone known to him who placed the “details” in “the hands of the police”, it would also be understood that it was no more than Mr Lawson’s opinion that such grounds existed.

I consider this is an overly subtle argument. The issue is how this would strike the reasonable reader. In my judgment, such a person would conclude that the Lawson article was stating that there is a grounding in fact for a police investigation and that is why the details have been placed in the hands of the police.

Ultimately, whether the common law threshold is met is very much a matter of impression in the context of the established principles that I have summarised. Turning to sub-paras (1) and (2), Mr Stables submitted that the threshold of seriousness was easily transcended. Mr Helme argued that Mr Lawson’s opinions, expressed at the conclusion of a long, polemical and even hyperbolic article were, in context, very unlikely to lower the Claimants’ reputation in the eyes of the relevant reasonable readership; that is followers of the Defendant’s Twitter account.Mr Helme also relied on the hurly-burly of Twitterspace (my formulation, not his) and a general understanding amongst users that polemic is inevitable.

I have reflected further on sub-paras (1) and (2) in the light of Mr Helme’s submissions. In my view, the issue should not be over-complicated. Although the defamatory sting is recognisable as opinion, I have reached the conclusion that the common law test is satisfied. The Lawson article is accusatory, vituperative and prescriptive. Seen in their context these imputations are sufficiently serious and are expressed in such a way that the Claimants’ reputation will have been lowered in the eyes of the likely readership.

As for sub-para (3), I have considered the parties’ further written submissions. Given my conclusions in relation to sub-paras (1) and (2), sub-para (3) adds very little. It follows that I can accept Mr Helme’s submission that sub-para (3) makes no substantive difference, but I would add that this is only in the light of my other findings. This sub-para is on the borderline, and I have not altered my provisional view that this issue should be left for future determination when ss.1 and 3 of the Defamation Act 2013 are also considered.

I have annotated my conclusions on meaning to reflect my decision on the three issues:

I invite the parties to draw up an Order that reflects my conclusions on the three issues.