KB-2024-002315 - [2025] EWHC 1782 (KB)

The burden is on the defendant both to prove the defence of honest opinion at trial, and, on this summary judgment application, to establish that the claimant has no real prospect of defeating that defence at trial and that there is no other compelling reason for the case to go to trial.

The basic facts on which the defendant relies are the fact that the claimant posted the tweets, their content, and their known context at the time of the publication. There is no dispute about those facts. The claimant admits he posted the tweets. He admits their content. There is no dispute about what was known of the context at the time of publication.

It is important to focus on the agreed meaning of the email. The claimant’s arguments, in a number of places, engage with the question of whether an honest person could form the opinion that he is antisemitic. That is not, however, the opinion that the parties agree was conveyed by the email. That agreed opinion is not that the claimant is antisemitic, but that he had posted comments online that were antisemitic. There is a difference. The expressed opinion relates to what (it is agreed) the claimant posted (in the known context in which it was posted) rather than to the claimant’s mindset as a person.

The claimant points to what he says is the exculpatory context of his general social media discourse and the immediate context in which those three tweets were posted. The defendant does not dispute that (so far as that context was known by it) these are part of the underlying facts on which it relies for the purposes of section 3(4)(a) of the 2013 Act. That includes the context for the first tweet, the fact that the second tweet was a response to a tweet that has been deleted (so that the precise context is unknown), and that the third tweet was a response to the tweet from the Count Dankula account which was in turn a response to the tweet from Mr Corbyn.

Mr Hirst submits that it is necessary to consider the broader context, including that which was not known to the claimant. That may, at trial, include the content of tweets that were deleted and so were unknown to the defendant at the time of publication. It also includes the fact that the Count Dankula account is operated by Mark Meechan, and his antecedents, and it also includes the claimant’s explanation for each of the posts.

I accept the defendant’s submission that this broader context comprises truly extraneous matters that do not throw into doubt the basic facts on which the defendant relies. In his oral submissions, Mr Hirst relied on the decision of Nicklin J in Riley v Murray [2021] EWHC 3437 (QB) [2022] EMLR 8 at [73], [77] and [98] – [99]. He submitted that the defendant must prove as an underlying fact that these tweets are antisemitic. It is for this reason that he relies on the broader context because he says that shows that they were not. However, this approach treats the defendant’s email as a statement of fact and the applicable defence as a defence of truth under section 2 of the 2013 Act, such that the defendant would have to prove that it is substantially true that the claimant posted antisemitic tweets. That overlooks that the parties have agreed that the email conveys an opinion rather than expressing a fact. The issue is not whether the tweets are antisemitic. The issue is whether the defendant has proved a fact or facts on the basis of which an honest person could form the opinion that the tweets are antisemitic. The decision in Riley does not support Mr Hirst’s submission. In that case, the defendant relied on a section 3(4) defence in respect of a published opinion that the claimant was dangerous and stupid and risked inciting unlawful violence. In support of that defence, the defendant pleaded and relied upon an asserted fact that the claimant had publicly stated in a tweet that Jeremy Corbyn deserved to be violently attacked. The defendant failed to establish the truth of that asserted fact. It was for that reason that the defence could not be sustained.

Against that background, I consider each of the individual tweets. I do so in reverse order.

The third tweet, read in isolation and without any context, is obviously and grossly antisemitic. The context is that it was a response to a deleted tweet from the Count Dankula account and that subsequent replies expressed doubt as to whether somebody was a holocaust denier. That limited context does not greatly assist in the interpretation of the tweet. A particularly cautious person, recognising the limited known context, might conclude that it is not possible to form an opinion as to whether the third tweet, read in its true context, was antisemitic. However, not all honest people are particularly cautious. Some would (and certainly could) conclude that the tweet is antisemitic. I do not consider that the claimant has a real prospect of establishing otherwise.

For the reasons I have given, I do not consider that the broader context on which the claimant seeks to rely is relevant to the issues that arise. But even if it were, I would have reached the same conclusion. Even if it had been known that the third tweet was a response to Mark Meechan, and that he had achieved a degree of infamy, and that the language used by the claimant, in part, directly referenced the language used by Mr Meecham, that would not make a difference to the inevitable success of an honest opinion defence. Read in that context, the claimant’s account that he was simply seeking to lampoon Mr Meechan may well be capable of belief. It is possible that some people, fairly reading his tweet in its proper context, would form the opinion that it was not antisemitic at all (or that if it was, then it was not “deeply troubling” because that context shows that it was not intended to be taken in that way). At the time, it appears that nobody took objection to the tweet. If the email had amounted to a statement of fact that the claimant was an antisemite, and it was necessary for the defendant to prove that the claimant had no real prospect of defeating a truth defence, then an application for summary judgment might be difficult to sustain. However, the statutory honest opinion defence is, as I have explained, much broader and more permissive than a defence of truth. The defendant does not have to show that the tweet is antisemitic, or that a fair-minded reader would form the opinion that it is antisemitic, far less that the claimant is an antisemite. It only has to show that an honest reader, perhaps a grossly unfair or prejudiced, but honest, reader, could form the opinion that the tweet is antisemitic, deeply troubling, unacceptable, prejudicial, discriminatory, and incompatible with holding office for the defendant. There is a clear relationship between the content of the tweet and the charge that it has those characteristics. The context put forward by the claimant may raise an issue as to precisely what was intended by the claimant when posting the tweet, but it does not remove the scope for an opinion that, notwithstanding that context, it was an antisemitic and “deeply troubling” (etc) thing to write. It follows that an honest person could form the view that is expressed in the email, so far as this particular post is concerned. The defendant is bound to succeed on that issue.

The email refers to comments (in the plural). On the face of it, therefore, it might arguably not be enough for the defendant to succeed in respect of the third tweet alone. However, much the same analysis applies to the second tweet. Again, the latter part of that tweet is, on its face, read in isolation and without its context, obviously antisemitic. Again, the context may raise a question as to the claimant’s intention when posting the tweet, but it does not remove any scope for an honest opinion that it was antisemitic. The defendant is bound to succeed in showing that an honest person could form that opinion. For these purposes I have not considered it necessary to take account of the academic article on which the defendant relies, or on a detailed analysis of different rhetorical devices deployed in social media discourse. Honest people have differing levels of familiarity with such matters. Irrespective of their level of such familiarity, an honest person could form the opinion that the tweet is antisemitic.

It follows that the defendant is bound to succeed on its honest opinion defence, by reference to these two tweets alone. For completeness, however, I consider the same applies to the first tweet. Of course, there is conceptually a potential difference between criticism of the state of Israel and antisemitism. Objectively, the former may not necessarily amount to the latter, hence the reliance that Mr Hirst places on the decisions in Jewish Rights Watch and Hussain. But, at least in some contexts, it may do. And an honest person may hold that opinion whether or not it is objectively sound. It seems to me to be incontestable that a call to boycott Israel “out of spite” is capable, in isolation, of being understood by an honest person as antisemitic. The context, which at least in this case is complete, may give some pause for thought, but it does not remove the scope for an honest opinion that the tweet is antisemitic, particularly when read with the other two tweets (and bearing in mind that these are just 3 individual tweets amongst several thousand).

Once it is recognised that an honest person could form the opinion that the tweets were antisemitic and deeply troubling, it necessarily follows that an honest person could form the opinion that they were unacceptable, prejudicial and discriminatory. It also follows that they could form the opinion that they are incompatible with holding office with the defendant. For these purposes, it is not necessary for the defendant to prove anything about its internal policies or disciplinary processes, or anything beyond the accepted fact that the defendant is a large trade union representing the interests of thousands of doctors. An honest person who formed the opinion that the tweets were antisemitic and deeply troubling could likewise form the opinion that they are incompatible with holding office with the defendant. Again, I do not consider that there is scope for the contrary view that an honest person could not form that opinion.

The question of whether an honest person could hold the expressed opinion, based on the three tweets and their context as pleaded and proved by the defendant, is an objective issue that requires an evaluative assessment. Neither party has identified any further evidence that is likely to become available that would impact on that issue. The application of the first component of the summary judgment test (that is whether the claimant has a real prospect of success on that issue) is, likewise, a question for evaluative assessment. It does not involve conducting a mini-trial or making any finding of disputed fact. It is eminently capable of resolution on this interim application, and it is in accordance with the overriding objective (particularly saving costs) that it should be resolved.

Nor is there any other compelling reason why the case should go to trial. That would cause significant additional cost and delay, for no discernible benefit. Having concluded that the claim is bound to fail at trial, and in the absence of any other compelling consideration, the overriding objective requires that summary judgment is entered in the defendant’s favour.