Daniel Miller & Anor v Luke Turner

This case is all about freedom of expression – and its limits in law. Freedom of expression – the freedom to hold opinions and to receive and impart information and ideas without interference by the state – is recognised in this country as a fundamental human right, and protected as such by law. And that may matter most where the opinions and ideas in question may be strongly contested, and/or outside the mainstream.

It is of course not an absolute or unqualified right. The exercise of freedom of expression carries with it legal duties and responsibilities. It is subject to legal limits, but only because those are necessary in a democratic society. In our democratic society, Parliament has given a general steer to courts considering whether to grant any relief which might affect freedom of expression: they are to have particular regard to the importance of that right (Human Rights Act 1998, section 12). At the same time, words and images are powerful, and can unfairly cause harm and even ruin lives. Parliament has recognised that too, and made important interventions in the criminal and civil law to set necessary boundaries accordingly. But where it does pass laws abridging freedom of expression, Parliament has done so in careful terms which courts must respect. The legal boundaries of freedom of expression are patrolled with great care.

Each of the parties in this case has been active in the world of art and ideas in ways which have attracted high controversy and strong opinion.

This case arises out of events in the art world, including various exhibitions and art events, and a high-profile durational work by LRT, each of which radically polarised opinion and ignited heated debate. More than that, these opinions expressed themselves in vociferous protest demonstrations outside art galleries, in the disruption of art events and damage to artworks, in claims and counterclaims of censorship and no-platforming, and in narratives of personal abuse, intimidation and violence. Each of the parties in the case has their own account to give of the distress, hurt and harm they have sustained along the way.

In particular, this case arises out of the parties’ experience of encountering each other online, on the social media platform that at the relevant time called itself Twitter. Mr Miller told me ruefully that in retrospect he had come to the conclusion that social media platforms of this sort are ‘very bad for discussing ideas’ and tend to ‘derange’ those who attempt to do so. The well-known features of Twitter – its character limits, its pace and evanescence, its offer of anonymity, its multivalency and its constant contextual flux – may well not be conducive to contests of ideas which are sustained, reflective or nuanced. Nor does it naturally foster a respectful, engaged and empathetic approach to differences of perspective. It has a reputation for fostering disinhibition and the consolidation of polarised and antagonistic group-thinking. It functions as much as a mode of (often impulsive) self-expression as a platform for dialogue. Its discourse is necessarily epigrammatic – that of the quip or slogan, of the emoticon or emoji, and of the ‘like’, link, quote, retweet or image by way of comment. It tends to be thought of as largely consequence-free for Real Life, and so is sometimes described as a playground. But it has given a new meaning to the word ‘troll’ in the English language. And online behaviour can and does have real life consequences.

The phenomenon of trolling has itself spawned a whole new vocabulary of ‘cyberbullying’. It has given us a lexicon which includes gaslighting (undermining someone’s confidence in their own beliefs and self-belief, even in their own rationality and sanity); sealioning (the relentless pursuit of demands for evidence, including under the guise of a faux-sincere interest in understanding more); doxxing (the public exposure of personal data and private identifying facts about someone, with a view to undermining, humiliation or intimidation); flaming (posting crude and/or obscene personal insults and invective); concern trolling (feigning concern, often for someone’s mental or emotional state, as a means of slurring that state) and dogpiling (joining or inciting a hostile group response designed to overwhelm someone and stifle their voice).

The parties in this case all describe themselves as passionate advocates and defenders of freedom of expression, and point to a track record of having actively done so at some personal cost. By this litigation, however, each side seeks to deploy the weaponry of the law to restrain and exact compensation for the other’s online behaviour. Mr Miller and Ms Power bring a libel action against Mr Turner. They complain of 16 tweets and a webpage, published by Mr Turner between 18^th October 2018 and 18^th August 2019 (the webpage republished one of the tweets and included links to an ‘archive’ of online material, including about Mr Miller, under a content warning). And Mr Turner brings a counterclaim in harassment, complaining of a course of online conduct by Mr Miller and Ms Power between July 2018 and the autumn of 2019.

Both sides urge upon me that ‘context is everything’ in this case, as it often is in legal matters. The potentially relevant background is vast – ranging across contemporary art, society, politics and public discourse. I also received a great deal of written and oral evidence from each of the parties – a plainly stressful and sometimes painful and emotional experience for them all. But as I said at the hearing, no-one’s ideas, works or world view is on trial in these proceedings and, as I explain below, the matters I am required to decide are in the end relatively limited.

Events in the art world

LD50 was a small independent contemporary art gallery in London. It had, for example, hosted a major exhibition by Jake and Dinos Chapman. From around 2016, it began to take an interest in art associated with a certain contemporary milieu of thought which has attracted (and resisted) a number of labels, and around which there is considerable controversy. I call it a ‘milieu of thought’ as a compendious term to embrace ideas in philosophy, aesthetics and politics, ideas which themselves are diverse and span a conspicuously broad spectrum of positions both consistent and inconsistent. By doing so I recognise that some who feel generally positively about this milieu would feel strongly that not everyone whom they might encounter there was recognisable as their fellow traveller. And I am going to call it, for the purposes of this judgment, ‘New Right’. There is no uncontentious and at the same time commodious identifier for this milieu in everyday usage. Some would recognise it as including ideas variously self-labelled as the Neo-Reactionary movement (NRx), the Dark Enlightenment, or Accelerationism. Others find labels such as Alt-Right or Far Right come to mind. Its opponents call at least some of this milieu neo-fascist or even neo-Nazi.

The New Right milieu embraces both the committed and the simply interested. It embraces aesthetic, spiritual and esoteric ideas. It embraces philosophical and moral ideas about what does and what does not constitute a good life. It embraces ideas about ‘population groups’, the ordering of society, the family, gender and sexuality. And it embraces political ideas, particularly about the limitations of contemporary liberal democracy with some of its assumptions about equalities and rights – and about what might lie beyond or after societies of this sort. It claims kinship with some great thinkers, artists and leaders of the past. Its opponents claim its kinship with some of history’s darkest chapters. And it includes a minority element in whose speech and actions the agents of criminal law enforcement and national security take an active interest from the perspective of protecting the democratic society in which we live. As I say, it is a compendious term.

The discourse within the New Right milieu has, on the extensive materials provided in connection with this case, some distinctive cultural features. Its natural medium is the internet. It eludes clear categorisations and labels. It moves fluidly between high ideals and low humour; it oscillates between the serious on the one hand, and the playful, performative, provocative or disruptive on the other; it enjoys irony and satire. It is dismissive of its critics and opponents (‘Antifa’, ‘Normies’). It privileges the subjective. Some have labelled these qualities post-Enlightenment or post-truth discourse. It is certainly polarising. For some, the New Right milieu, its paradigms and its discourse, are exciting, futuristic and liberating. For others, it can seem a terrifying arena of unreason, inhumanity and moral vacuity. There is little visible neutral ground in between, little space perhaps for constructive engagement.

At the end of 2016, LD50 hosted a symposium called ‘Neo Reaction’. Speakers included Nick Land, Brett Stevens, Iben Thranholm and Peter Brimelow. It was associated with an exhibition called ‘71822666’ (a reference to a thread on the website 4chan associated with the US presidential candidacy of Mr Donald Trump) and with a ‘triskele’ added (an ancient symbol, adopted in the last century by the South African pro-apartheid AWB). The show was advertised on LD50’s website with a quotation from Adolf Hitler (‘It must be thoroughly understood that the lost land will never be won back by solemn appeals to God, nor by hopes in any League of Nations, but only by the force of arms’)in German gothic font, next to a picture of singer Taylor Swift and above her signature. It featured the works of internet New Right pseudonymous figures Kantbot2000, Mencius Moldbug, Melchizekek, Outsideness, BronzeAgePervert, HBDNRX, and E-Terrorist, along with notorious spree killer Elliot Rodger.

LD50 attracted robust local, public and indeed national and international criticism in these circumstances. Mr Turner later joined in the criticism – he called it an alt-right or alt-Nazi gallery. So did many others. By February 2017, vociferous public protests were taking place outside LD50, with some seeking to have the exhibition shut down. But others were concerned that this was an exercise in censorship. On 25^th February 2017, Mr Miller went along to counter-protest. He was (nearly) alone. He held up a placard. On one side was ‘the right to openly discuss ideas must be defended’ and on the other ‘stand up to violence and intimidation’. He says he was thereupon himself treated to violence and intimidation by the protestors. He says he subsequently suffered further violence at the hands people associated with the LD50 protest. It seems that the gallery’s landlord asked LD50 to quit the premises not long after, in view of the public order issues. It has ceased to operate. Mr Miller published an article in April 2017 (and still online) called ‘The True Story of LD50’. He describes his experience, critiques the gallery’s critics, and discusses his concerns about censorship and no-platforming.

Mr Donald Trump had been sworn in as President of the USA in January 2017. In response, on 20^th January 2017, LRT initiated a durational artwork they called HEWILLNOTDIVIDE.US (‘HWNDU’). It consisted of a fixed camera on a wall outside the Museum of the Moving Image (MoMI) in New York. It invited members of the public to deliver the words ‘he will not divide us’ into the camera, and the result was livestreamed online. It achieved considerable prominence – and considerable opposition. Both at the site itself and online, the project, and LRT personally, were subjected to vociferous criticism, and abuse, much of it identifiably from within the New Right milieu, and some of it explicitly antisemitic. On 10^th February 2017, MoMI took down the installations because of concerns about disorder and violence.

HWNDU relocated to another venue within the US, but was forced to abandon that after a few days following more disorder and abuse, some of it antisemitic. LRT reinvented the project as a flag with the words ‘HE WILL NOT DIVIDE US’ on it, flown from a flagpole in a new US location. The flying flag was livestreamed. Within days, the flag was stolen by two members of the US Traditionalist Worker Party, an organisation of the New Right milieu. The field where it had been flying was set alight.

The HWNDU project relocated to Europe, in both its original format and by flying the flag. The flag flew briefly in Liverpool in March 2017, but had to be abandoned on police advice as protesters were found making a dangerous ascent of the roof towards it. The project continued on to other European art institutions.

Meanwhile, LRT were undertaking a new performance art project in remote Finland in April 2017. They recount how they were evacuated from that when the museum organising the project received intelligence that three armed men, one wearing a New Right badge, were travelling towards it.

Some of the abuse directed towards HWNDU deployed the counter-phrase ‘you will not replace us’. At the Unite the Right rally at Charlottesville on 12^th August 2017, at which a counter-protester was killed, slogans of ‘Jews will not replace us’ were reported.

Mr Turner reports that around this time he spent almost a month in his London flat without leaving it, because he was in terror of the volume and virulence of online abuse directed towards him, and because he was being actively monitored and stalked there. Cameras were put through his letterbox, and detailed pictures of the inside and outside of his home posted online. There was an incident with an intruder, and a sinister masked figure took up a presence outside. A swastika marked 88 (code for ‘heil Hitler’) was posted through his door. Mr Turner asked for help from the police but was advised that the sheer number of anonymous online abusers left him little choice but to change his identity and move house. He began instead to monitor the online campaign himself for signs of physical danger.

In October 2017, at a location in France, the HWNDU flag was targeted by a drone with a petrol-soaked rag which crashed into the building. HWNDU continued, in peripatetic form, until it completed at the end of the Trump Presidency.

LRT had been due to appear at the Athens Arts Biennale in October 2018. Mr Turner wrote to the organisers in August 2018 to say he was not prepared to share a platform there with US artist and commentator Daniel Keller because of his association with New Right, pro-LD50 or anti-HWNDU positions, and indeed with Mr Miller. The organisers rebuffed him. Mr Turner withdrew from the event, publishing an open letter explaining his reasons. The organisers responded by publishing their own letter in September 2018, saying they regarded Mr Turner as having resorted to ‘the epitome of online hate speech’ in what he had said about Mr Keller and Mr Miller.

Mr Miller was in Athens at this time. On 17^th September he delivered a performative lecture as part of the University of Muri project, in the persona of Julius Evola, self-styled ‘superfascist’ and thinker associated with the Mussolini and Hitler regimes.

Mr Turner had been due to lead a panel discussion at a Berlin art gallery in early 2019 on Metamodernism. It was cancelled by the organisers out of concerns about hate speech and threats having been made to Mr Turner.

In the early months of 2019, an exhibition by French-Canadian artist Mathieu Malouf opened in Los Angeles. Its title was ‘#luketurnerisretarded/#luketurnerisgay’. It included a sculpture which Mr Turner identified as being a grotesque caricature of himself and as drawing on ‘Judeo-Bolshevik’ stereotypes. Ms Power appeared alongside Mr Malouf on a panel at an art event in Berlin in May 2019 at which she declined to condemn the exhibition. In the spring of 2021, Mr Malouf sent Mr Turner an Instagram message, an image of a cake, bearing the face of the sculpture, and the message ‘happy birthday to this sculpture’ with three hearts.

Events online

The total online discourse accompanying all of this can only be imagined. HWNDU was, I am told, conceived by LRT as a humane, hopeful and unifying project. It was nevertheless a plain public statement at the centre of a highly polarised political arena, and unsurprisingly received as such. Art-critic Ms Power described it as interesting in the way it referenced the anonymity and accessibility of online speech, and as provocative, piggybacking on the energy of the political New Right and positively intended to provoke a hostile response. Other unsympathetic critics considered it an exercise in self-publicity, one satirist positing the idea that LRT had itself arranged to have the HWNDU flag stolen for that very purpose. But there can be no doubt whatever that the frightening episodes and hate speech experienced by Mr Turner in real life were the merest glint on the tip of an iceberg of personal criticism, hate and terror that congealed around him online. I was shown a thankfully small, but unforgettable, sample of some of the worst of this material. It was grossly antisemitic, not excluding the superimposition of Mr Turner’s face on Holocaust images. And it was explicitly violent.

Not all of the discourse, of course, was at that level of baseness, nor anywhere near it. Mr Turner’s online critics included an array of artists, thinkers and commentators, partisans in the LD50 affair and/or associated in some way or other with the New Right milieu. That does not, of course, mean that it was conducted according to the etiquettes of an academic symposium. Twitter is not like that.

There is another reason why the total online discourse – and the total discourse among the parties to this case in particular – can only be imagined. Mr Miller has entirely deleted his Twitter history. He told me that is something he did systematically as he went along, because he was not interested in the preservation of a record of his activity. And Ms Power deleted the majority of her own pseudonymous account as late as December 2020, after the commencement of these proceedings. So the immediate contextual material I have been given to work with continues to exist at all only because of Mr Turner’s disclosure of his own account and his preservation of other material he thought he had reason at the time to be concerned about. This is a very incomplete account of events indeed. There is much in this immediate context which is missing. And ‘context is everything’.

The parties first encountered each other on Twitter in July 2018. Mr Turner used @Luke_Turner as his handle; Mr Miller used @dcxtv, and Mr Turner did not know at the time who that was. He did note, however, the association and interactions of that account with a number of accounts associated with the New Right milieu. Mr Turner had recently tweeted that LD50 was an ‘alt-right (aka alt-Nazi) gallery’, but had then deleted that tweet. He told me that was so he could do some more research into the gallery. Mr Miller responded with a series of abusive tweets, calling Mr Turner a ‘dumb-as-fuck, dishonest propaganda hack cocksucker’ and a ‘pathologically dishonest careerist retard shitbag’. He told me he was not proud of this outburst, but it was a sign of his anger over the LD50 affair, in which he had been personally involved. Mr Turner reported the account to Twitter and it was temporarily suspended for violating Twitter’s conduct code.

A few days later, on 30^th July 2018, there was an online incident of which much is made in these proceedings. There are a year’s worth of exchanges featuring in this case, and I deal with them in detail below only to the extent that that is essential to the decisions I have to make. But I do set out this early incident in full, because it sets the scene for much of what followed, and helps explain the rival narratives contended for in this litigation.

US artist Deanna Havas (Mr Miller described her as ‘a funny girl’ and ‘the world’s greatest Twitter artist’; LD50 had put on a solo show of her work in the late summer of 2016, immediately before the show which had sparked so much controversy) joined the continuing Twitter thread between Mr Miller and Mr Turner, asking what had happened to HWNDU. Mr Turner explained it was ‘still going strongand resisting the normalisation of the kind of bigoted behaviour dcxtv has been targeting me with on here’. Another contributor then posted a cartoon showing a band of Pepe-the-Frog figures gleefully running away with the HWNDU flag, a distraught Shia LaBeouf in the background and drones overhead.

(Pepe-the-Frog is an internet meme with a chequered history. Infinitely adaptable, he gained a hold on the online imagination by being reproduced in any number of cute, funny or satirical contexts and to make any number of points. But he was adopted as something of a mascot in the New Right milieu. One of the gunmen who terrified LRT in Finland was said to have worn a badge showing Pepe as President Trump – itself a much-reproduced motif. The US Anti-Defamation League put Pepe on its database of online hate symbols as long ago as 2016 because of his use by antisemites. The history of Pepe’s ‘cousin’, the Groyper – a similar figure with a distinctive chin-on-hands pose – is also chequered. Deployed in all sorts of guises, he too was particularly enjoyed in the New Right milieu. Both Mr Miller and Ms Power had ‘groyper’ Twitter accounts – they told me each was operated in the other’s persona, as a playful private game between them.)

Ms Havas ‘liked’ the cartoon. Mr Turner said he was curious as to why, and wondered whether she was unaware that ‘the neo-Nazi Traditionalist Worker Party’ was responsible for the theft of the flag. She replied that the cartoon was funny. The contributor who posted it commented ‘omg everyone who doesn’t agree with me is a nazi!’. Mr Turner responded that the TWP was self-described as ‘National Socialist (neo-Nazi)’. He said to Ms Havas the cartoon was a depiction of an antisemitic campaign in which actual crimes had been committed, and asked whether it was genuinely funny to her. She replied, ‘chill out bro, it was just a meme’. Mr Turner responded, ‘try telling that to the people whose property was set fire to, who could easily have been killed as a result. Don’t tell me to chill out. You need to rethink finding antisemitic hate crimes and abuse ‘funny’ or ‘just a meme’. Fast’. She responded, ‘don’t tell me what to think, but do have a nice day, bye’. And he replied, ‘not good enough’ and ‘I’m afraid your stance makes you complicit in antisemitic violence – no matter whether you are Jewish or not. Perhaps you cannot relate, but that is your privilege’.

The contributor who posted the cartoon commented that ‘Luke Turner is a Nazi sh*t’. Mr Miller responded with this: ‘Luke Turner is, specifically, a Normie Fascist. An institutionally-supported propagandist, his limited intelligence and narrow ethical horizon is circumscribed by a conformist-narcissistic ideology that prevents him [from] participating in any truth, alas at the present time.’ Ms Havas responded to Mr Turner with this: ‘do spare me the ‘woke’ lingo and do me a favor – rather than scrutinising every minute action (such as faving a replied tweet of a cartoon) as an act of symbolic violence, go outside and enjoy life for a change’. Mr Turner replied, ‘It’s a bit harder when you have to have security because you have violent neo-Nazi groups posing credible threats to your safety on a daily basis across several countries, but yeah’.

Mr Miller then contributed the following (it is an unreferenced quotation from William Burrough’s The Naked Lunch): ‘A low pressure area, a sucking emptiness. He will be portentiously anonymous, faceless, colourless. He will – probably – be born with smooth disks of skin instead of eyes. He always knows where he is going like a virus knows. He doesn’t need eyes.’The contributor who posted the cartoon tweeted to Mr Turner ‘you are no victim of nazis, you are one’.A follower of LRT tweeted: ‘the defense of violence, harassment and threats as ‘just a meme’ seems always to come from those who wish to hide behind those memes and weaponise irony as an excuse for their true ideologies…’. Ms Havas, apparently under the impression that this tweet came from Mr Turner himself, replied ‘Go fuck yourself’ along with symbols of a star of David, a hammer and sickle, an iron cross and a swastika. That tweet was ‘liked’ by Mr Miller.

The narrative contended for by Mr Miller and Ms Power, and which they say played out in the later tweets they complain of as being libellous, is, they say, demonstrated in microcosm here. In this narrative, Mr Turner has a propensity to label anyone at all in the New Right milieu as an antisemite, or a violent antisemite, on the slenderest of bases or on no basis at all. Here, he labels Ms Havas as ‘complicit in antisemitic violence’ on no better ground than that she laughed at a cartoon. And, they say, he did much the same to them. These are labels which stick, and which ruin reputations. Ms Havas, and they themselves, went on to find themselves increasingly stigmatised, shunned and no-platformed. That is censorship, and Mr Turner’s casual, or spiteful, name-calling and bullying is to blame. It degrades the true evil of antisemitism. It punches down from a privileged position at less prominent artists and thinkers and hurts their livelihoods.

The narrative contended for by Mr Turner, and which he says played out in a subsequent year-long campaign of online harassment by Mr Miller and Ms Power, is also, he says, demonstrated in microcosm in this incident. In the context of a virulent and violent antisemitic campaign against a Jewish artist, New Right milieu thinkers and artists online have no excuse for actively or passive-aggressively aligning themselves with it, and if they do so they reveal themselves as consorts of hateful ideologies themselves. Such people do not engage, with any empathy or intelligence, with the patent acuteness of his predicament either as an artist or as a Jew. They trivialise and ridicule. They are mindlessly, relentlessly abusive. They gang up and bully. And worst of all, he says, there is an insidious miasma of implied antisemitism and thuggishness in the tone and content: denial of his experience, accusations of lacking perspective, imagining antisemitism or bringing it upon himself, sniggering over Nazism, the use of well-worn tropes. Context is everything. These people, he says, are themselves part of the problem of censorship and oppression of art and artists like himself, and need to be called out as such.

The claim

The publications Mr Miller and Ms Power complain of are set out in full in Miller & Power v Turner [2021] EWC 2135 (QB). That judgment followed a preliminary issues trial in this case to determine, as the law requires, the ‘single ordinary and natural meaning’ of the publications in question and whether they were allegations of fact or statements of opinion. That single meaning, and resolving the fact/opinion question, define the legal case Mr Turner has to meet and the legal defences potentially available to him. Mr Turner did not dispute that all of the publications were of ‘defamatory tendency at common law’, that is to say they ‘substantially affect in an adverse manner the attitude of other people towards a claimant, or have a tendency to do so’ (Triplark v Northwood Hall [2019] EWHC 3494 (QB) at [11]).

In the first place, complaint is made of the fact that Mr Turner had compiled and made available online an ‘archive’ of tweets he considered to be a cause of concern (including Mr Miller’s, Ms Havas’s, and others’) on a page with a content warning. The meaning determined for this was that:

The remaining publications complained of are 16 individual tweets. Their meaning, and imputations, are repetitious. They include the following factual allegations:

These are the imputations of defamatory tendency sued upon. The tweets complained of also included statements of opinion that (a) Mr Miller was a fascist, a superfascist, and a neo-Nazi; (b) Ms Power supported his views, and had herself demonstrated a fascist turn; and (c) both parties were fascist-friendly. The Claimants do not, however, choose to sue on these imputations. That decision is of course entirely a matter for them, and they are not bound to explain it. Mr Miller gave an indication that he considered ‘fascist’ a debatable term (and the fact that he himself had called Mr Turner a (‘Normie’) fascist on more than one occasion may have something to do with the matter). But neither do the Claimants acknowledge or own these labels. Indeed, in their oral evidence, they complained with some vehemence about them. Their position, in these circumstances, has consequences, which I consider below.

The law

As Ms Grossman, junior Counsel for the Claimants confirmed to me, there is no material dispute in this case as to the applicable law. It is the application of the law to the facts which is principally in issue.

Once the preliminary issues of meaning, fact/opinion and defamatory tendency have been settled, a defamation claimant’s principal remaining task is to demonstrate satisfaction of the test set out in section 1 of the Defamation Act 2013; and a defendant is then put to the task of making out any of the applicable statutory defences – truth (section 2), honest opinion (section 3), publication on a matter of public interest (section 4) – or any applicable common law defence or privilege.

Section 1(1) of the Act provides that ‘a statement is not defamatory unless its publication has caused or is likely to cause serious harm to the reputation of the claimant’. In the decade since that provision was enacted, the courts have consistently acknowledged the introduction of a ‘serious harm’ test in defamation proceedings to have been a significant intervention by Parliament, recalibrating libel law in favour of greater freedom of expression.

That was put beyond doubt by the Supreme Court in Lachaux v Independent Print Ltd [2020] AC 612. Lord Sumption’s judgment makes clear ([12]-[14]) that s.1 imposed an entirely new threshold test, and that its application was to be determined by reference to the actual facts about the impact of a publication, and not just to the meaning of the words. The statutory term ‘has caused’ points to some historic harm, which is shown actually to have occurred; and ‘is likely to cause’ points to probable, actual, future harm. ‘Likely to cause’ does not mean liable to cause; it means something more than the inherent tendency of the imputation.

The serious harm test is a question of fact, and facts must be established by evidence. Facts and evidence are matters which are entirely case-specific. Lachauxitself confirmed that there is no hard and fast rule as to how serious harm is to be evidenced. That is partly because of the nature of the harm in question: the ‘harm’ of defamation is the effect of a publication in the mind of a third-party publishee, and not any action they may take as a result (nor is it the direct effect of a publication on a claimant reading it themselves). And it is partly because of simple practical considerations: particularly in cases of mass newspaper publications, the minds of the publishees are effectively unreachable. In such cases, Lachaux confirmed ([21]) that the evidential process may be able to be discharged by establishing, and combining, the meaning of the words, the situation of the claimant, the circumstances of publication and the inherent probabilities. This is sometimes referred to as a ‘Lachaux inferential case’, based on the ‘Lachauxfactors’. But the Lachaux decision itself was at pains to emphasise it was not setting out any special standalone rule of law; it was illustrating the essential point that serious harm is a matter of fact and evidence. As I have said elsewhere, an inferential case is not an alternative to an evidential process; it has to be an evidential process. That is essential to give effect to the intention of Parliament to impose a meaningful test in addition to the common law defamatory tendency of words, and thereby to recalibrate defamation law in favour of greater freedom of expression.

Since Lachaux, the serious harm test has been given close attention in a series of High Court and Court of Appeal decisions. This jurisprudence was recently summarised fully and clearly by Nicklin J in Amersi v Leslie[2023] EWHC 1368 (KB) at [143]-[163], a passage to which I have addressed myself carefully and which at the start of the present trial I encouraged Counsel to do likewise. I do not need to replicate that passage in full here, but I do refer to and rely on some of the jurisprudence set out there in the analysis which follows.

The Claimants’ case on serious harm

There is no evidence from any publishee in this case – that is, from someone who said they read the items complained of, other than the Claimants. Ms Grossman refers to the authorities’ recognition that it may be difficult for defamation claimants to find witnesses to testify that they thought worse of them as a result of the publications complained of. She relies on the inherent seriousness of the imputations sued upon – death threats, police involvement, advocating Hitlerian euthanasia policies, antisemitism. She relies on Mr Turner’s Twitter followership of around 12,000 as an indicator of extent of publication. She relies on the fact that Mr Turner’s account was not anonymous; he was a renowned artist with an international profile, and his comments would attract attention. She relies on the fact that the parties all operate within the art world, and so Mr Turner’s views of Mr Miller and Ms Power might be taken to be informed. She points out that whatever controversial views the Claimants might express or be known to hold, there almost always remains room for additional damage to be inflicted by specific allegations - here, particularised acts, the allegation of harassment, and the distinctive slur of antisemitism.

In relation to the Claimants’ own evidence, Ms Grossman points to indications that they have been shunned and avoided, and indeed that Ms Power (and others) have been shunned and avoided merely by being associated with Mr Miller. Ms Power has lost positions and appointments, and been dropped by her agent. A publisher has told Mr Miller they are unwilling to commit to publishing his book about surrealist André Breton until this trial is over. Ms Grossman suggests I can at least take all of this into account.

Ms Grossman also points out that most of the material complained of continues to be published and accessible online (the Hitlerian article allegation being a notable exception). She says that both persistence and percolation should be taken into account.

Ms Grossman accepts that the authorities require the serious harm test to be satisfied in relation to each individual statement or imputation complained of (if not indeed each individual publication complained of) and that their impact cannot simply be aggregated or cumulated to address the test. But she submits that the same Lachaux factors can nevertheless be applied individually to each statement, and the test passed in that manner.

Analysis

I begin, as invited by Ms Grossman, with the Lachaux factors. The allegations complained of are indeed inherently grave. A bit of care is needed with the followership figure. The Twitter analytics I have been provided with indicate that Mr Turner’s Twitter account had 3,200 followers in the UK, and it is UK readership I have to be concerned with (extraterritorial defamation was not pleaded, argued before me, or evidenced). Total figures for ‘likes’, retweets and replies in relation to the individual tweets complained of are very low – ten or a dozen at most, and generally less than half of that. Some of these are visible in the immediate context of the tweets themselves. I bear in mind the extent to which the tweets complained of have not been taken down, but also that they are by now quite stale and inaccessible otherwise than on an active research basis. I am told, however, that Mr Turner’s ‘archive’ of preserved tweets from Mr Miller and others, with its content warning, remains online. The ‘percolation effect’ is about the propensity for publishees of an item to pass it on, so these figures and facts do have some bearing on that. Mr Miller’s own careful practice of deleting his tweets in real time must also be a factor limiting the probability that the other half of these amputated dialogues would survive in sufficiently graspable form to be readily shared onwards.

I agree Ms Grossman makes a potentially telling point about Mr Turner’s high profile, both by itself and by association with Mr LaBeouf, and also about the art world context within which all the parties, and their respective reputations, subsist. I take this into account. And I accept the Claimants’ evidence that they may have accumulated an adverse reputation in that world, which has caused them to lose opportunities, and to be distanced, dropped or no-platformed.

The question to which the law requires an answer is, however, whether any of that is to be attributed to anyone reading any of the publications complained of. Whether by inference from inherent probability, or by evidence from publishees, or by any other evidential route, the section 1 test requires the publication of a statement to be the cause of (past) serious harm or the likely cause of (future) serious harm. Without evidence from any publishee, or indeed evidence that any identified or identifiable publishee actually read the publications themselves and/or shared them and that anyone thought any the worse of the Claimants as a result, that requires an examination of the evidence to consider whether other possible causes suggest themselves, and if so whether they might be more probable.

The case Ms Grossman puts relies on (a) the Lachaux factors and the inferences to be drawn from them in all the circumstances of this case, (b) there being consistent evidence of actual reputational damage, and (c) there being no other evidence showing that the effects complained of could be attributable to other factors (in contrast, for example, to Lee v Brown [2022] EWHC 1699 (QB), where the claimant’s inference of social ostracism at his rugby club had a more obvious and probable explanation in the impact of the covid pandemic than in any reputational harm caused by any words of the defendant). I have to part company with Ms Grossman, however, when she suggests there is no such evidence in the present case. There is a great deal of contrary evidence, and it is principally from the Claimants themselves.

In the first place, there is the actual causative power, clearly attested to in the Claimants’ evidence, of those imputations contained in the publications complained of, but on which they have chosen not to sue. These include the imputations that both Claimants are ‘fascist’, and that Mr Miller is a ‘neo-Nazi’. In his trial witness statement, and in his oral evidence, Mr Miller repeatedly focused on the power of these ‘invidious epithets’ and what he said was their purpose and effect to attract ostracism, violence and hatred towards him. He said they were very serious, powerful and stigmatising allegations. Ms Power too testified to these being powerful and ‘demonising’ terms having a profound effect on the way people viewed Mr Miller and indeed herself. They do not sue on these imputations. But that does not mean I can ignore them. They are part of the overall evidential matrix.

I agree with Ms Grossman that that does not prevent the imputations which are sued upon adding to the reputational harm. But both of these terms come ready-freighted with at least some of the imputations of which they complain in these proceedings. ‘Neo-Nazi’ surely overlaps with ‘Hitlerian’ and ‘antisemitic’, if not with violence also. And while a more fastidious approach might be wished for by the Claimants to the semantics of ‘fascist’, to the extent that their own evidence is that it is an alienating and damaging epithet, that cannot easily be separated from its historic associations with bigotry, antisemitism and indeed violence. If people recoil from an imputed reputation for fascism, as the Claimants testify, it is probably not because of peaceable differences of opinion about philosophy and aesthetics. I have no evidence for any distinct shades of reputational harm added by the publications complained of to their imputations of fascism and neo-Nazism, and I have to reflect hard on the inherent probability that such shading in and of itself caused (further) serious reputational harm.

In the second place, the Claimants’ evidence also references the serious reputational impact of imputations not made by Mr Turner in the publications complained of at all (at least some of which were not even indexed to any publication he is evidenced to have made in any event). That includes Mr Miller’s written evidence that he was ‘particularly disgusted to be accused of Satanism’ and that a publisher of his work lost his position on the basis that Mr Miller was ‘a potentially dangerous alt-right philosopher/agitator and supporter of conspiracy theory’. Ms Power complained of the effect of being called a neo-Nazi, a transphobe, and a ‘terf’ (‘trans-excluding radical feminist’). But these are not meanings determined for the publications in issue here. None of this is in issue in these defamation proceedings and none of it is even relevantly and evidentially attributed to Mr Turner. But all of it, on the Claimants’ evidence, was materially destructive of their reputations.

In the third place is the issue of the extent to which any reputational harm occasioned by Mr Turner’s publication of his archive under its content warning was caused by the content warning rather than the content. A content warning is designed to alert publishees to the nature of the material contained within – and, of course, it may have deterred some readers altogether. Mr Miller’s reputation may have been impacted in the minds of deterred readers by the content warning, but I have no basis for inferring substantial (or perhaps any) levels of deterrence. A content warning like this is primarily intended to draw attention to content, not to achieve a low readership. Undeterred readers were capable of making up their own minds by looking at Mr Miller’s own words.

Mr Miller complains of decontextualisation here. Twitter is indeed a quintessentially conversational medium. But it is Mr Turner’s (unchallenged) evidence that both Mr Miller and Ms Power had a significant presence on Twitter – both in their own tweets and in their endorsement of others’ – of a nature which spoke for itself. Indeed, he says his entire online engagement with them was by way of commentary on their own publications and their own expression of their own views. That is largely what ‘calling out’ is. A claimant cannot complain in defamation of reputational harm caused by their own publications (including, in the present case, their own republication of some of the actual material they do complain of). But of course, as I observed at the trial, I have next to no unmediated evidence of these Claimants’ online presence because of the steps they have taken to remove it all. If their self-curated blank online profile is impacted because Mr Turner has rendered it incomplete, I have to be careful about the extent to which any resulting reputational harm can properly be said to have been caused by him in preserving and commenting on this material – or decontextualising it – rather than them first creating it and then themselves actively decontextualising it.

In the fourth place, it is notable that Mr Turner’s open letter to the Athens Biennale organisers about Mr Miller (and Mr Keller) was published on 4^th September 2018. That was some weeks before the first publication complained of in these proceedings. It contained many of the imputations Mr Miller complains of – ‘dozens of abusive and bigoted tweets’ (with a link to the archive with the content warning), ‘sinister threats of violence’ and antisemitic abuse which were the subject of police investigation. This publication may well have been more widely read and reacted to – and perhaps by a less resolutely partisan audience – than any of Mr Turner’s conversational tweets. But the Claimants have chosen not to sue upon it.

This is a state of affairs going beyond the technical issue of whether a defendant who makes repetitive defamatory publications can properly put a claimant to the task of establishing that each individual publication sued upon caused or was likely to cause serious harm. The Claimants here have elected not to sue on what was undoubtedly a high-profile attempt by Mr Turner to ‘call out’ Mr Miller in precisely the wider artworld context in which the Claimants say their reputation was especially important and vulnerable. That may be because of the Biennale organisers’ riposte to Mr Turner, calling him out in return. It may be that no-one’s mind was much changed about Mr Miller as a result of Mr Turner’s open letter. But I have to ask in those circumstances how probable it was that serious harm to Mr Miller’s reputation was subsequently caused by a handful of tweets saying the same sort of things as were already in circulation in the public domain on an unlitigated basis. It is a matter for a claimant what publications they choose to sue on. But the section 1 test must be passed on those publications.

In the fifth place, the Claimants make no bones in their evidence that they complain of the cumulative harm caused to their reputations by Mr Turner’s publications and those of many others. They complain of what has been said by his ‘associates’, his ‘mob’, other people who think like him, a ‘large network dedicated to defaming you’. They complain of a general ‘horribly defamatory atmosphere’. They blame Mr Turner for all of it. But this is a problematic approach to the serious harm test. I have no evidence that Mr Turner acted in concert with others or caused or incited any third parties to publish defamatory material (and if so, what). That is not the case the Claimants bring to court. Mr Miller described his critics as a network of which Mr Turner was ‘a part’.Ms Grossman relies on his prominence and authoritativeness in the art world. But the LD50 affair and the history of HWNDU are eloquent testimony to the reality that the whole debate around the New Right milieu in general and its associated art in particular (on the one hand) and those protesting it in art and otherwise (on the other) was long-standing, widespread, vehement and partisan. It may have played out for these three individuals acutely in their own lives. There was certainly a particular flashpoint for them in their social media interactions with each other, which brought them all into court. But it is no more objectively probable from first principles that Mr Turner (much less the tweets sued upon) is the source of what the whole of the rest of the world – or even the small sample of supportive thought visible in the online records in this case – thinks and says about Mr Miller and Ms Power than that they in turn are responsible, by virtue of their own tweets, for what the world thinks of him.

Mr Miller and Ms Power did and said controversial things in public. Many more people than Mr Turner were able to form and express adverse views by direct observation of this, and did so: they did not need to read or know about Mr Turner’s tweets to tell them what to think. The Claimants and Mr Turner belonged in populous, strongly opposed and deeply entrenched different camps – on HWNDU, LD50 and the public responses in each case – and each had given public evidence of their views. Each camp says what it thinks about the other, on any and every platform, online and otherwise, vocally and vehemently. People get hurt in the crossfire. But a published statement is not defamatory, however defamatory its meaning, unless its publication – rather than anything else – has caused or is likely to cause serious harm to the reputation of the claimant. Otherwise, freedom of expression takes its course.

There is a further problem of evidence of alternative explanations for the reputational harm sustained by the Claimants. That is that their evidence for such specific examples of real-world consequences of reputational harm as they mention is that it was positively not caused by publishees reading the material complained of. Such specific examples are not essential to making out serious harm, but Ms Grossman asked me to look at them. I have done so. And I see this is not a case where no-one can be found to say they thought any the worse of the Claimants. The issue in this case is not whether they took a dim view of the Claimants but why. Without any evidence from any of them, I have no basis for speculating on a potential link to the effects of Mr Turner’s tweets. And such evidence as the Claimants gave themselves does not support the inference they invite me to draw.

Apart from Mr Miller’s publisher (who is, after all, apparently keeping an open mind about his reputation, at least for the present), mention is made of an anonymous open letter ‘concerning Nina Power’ (but also aimed at Mr Miller, described as an ‘esoteric fascist’) which was published before any of Mr Turner’s tweets about Ms Power now complained of. A faint attempt was made to suggest Mr Turner was its author, but I have not been given anything like a sufficient basis for making any such finding of fact. Ms Power complained more generally of anonymous or pseudonymous accounts ‘repeating defamation’ and trying to get her cancelled from events. There is also complaint of an anonymous letter to Penguin Books. The same applies to these. Any action taken by third parties on these publications is not within the purview of the present claim alleging the causation of serious harm unless it can be linked back to the posting and reading of the publications complained of.

I have no evidence for that. It is not more inherently probable that all of this material and these outcomes can be attributed to someone reading Mr Turner’s tweets and being influenced by them than that, for example, they read someone else’s opinions, or Mr Miller’s and Ms Power’s own opinions, or that they formed their views independently by judging the Claimants on the basis of their real-world activities in controversial contexts. Mr Turner’s tweets drew attention to, and were comments on, Mr Miller’s and Ms Power’s own comments, actions and associations, and all of those had public lives of their own.

Ms Power appears also to have made a number of public statements in which she took ownership of personal decisions to step down from roles, including by deciding for her own reasons to leave her academic post. She accepts that some of the ‘cancelling’ or ‘no-platforming’ of which she complains happened before the publication of any of the material complained of. She is on record as further attributing a significant proportion of it to her own public statements about gender and trans people – a notoriously tense, sensitive and heated area of recent and polarised public discourse in which other women describing themselves as feminists have experienced negative reactions and no-platforming. There is no discernible causal connection between this and the publications complained of.

Conclusions

In its original form, Mr Miller and Ms Power’s claim had been made on a basis of both harassment and defamation. They discontinued the harassment claim. But some of their pleading and evidence appears to be still angled towards the legal targets of harassment claims (discussed below) rather than defamation claims. (Their litigation fundraising page is also still headed ‘targeted, harassed and falsely labelled a fascist’ – none of which features in the claim they actually pursued.)The ‘serious harm’ of the s.1 test is not constituted by aggregate campaigns or cumulative courses of conduct. It is not constituted by a claimant’s experience of oppression, or by their own personal distress at name-calling and accusation. It is constituted by a claimant’s demonstration – by direct evidence or by laying evidential groundwork for probable inferences of fact – that thepublications complained of were read, and thence led by a legally relevant chain of causation to third party publishees changing what they otherwise thought about the claimant to an extent describable as serious reputational harm.

That is a factual, and highly fact-sensitive, test. It does not necessarily require objective proof of the impact of any individual publication in the mind of any individual publishee. But it does require more than a defamatory publication on the one hand and a claimant with an impacted reputation on the other. It requires establishing a legally relevant cause-and-effect link between the two on the balance of probabilities. That is an essentially evidential process, with the burden on a claimant. It does require critical and contextual engagement with the objective probabilities on a case by case basis. It is sensitive to evidence supporting the causal power of factors other than those to be inferred from the meaning of the words. That is what Lachaux meant by an evidential process deriving inferences of fact based on a combination of meaning, the situation of a claimant, the circumstances of publication and the inherent probabilities. Those inherent probabilities must be considered in context.

Here, I have accepted that the meanings of the words complained of identify a number of imputations of real gravity. The publications were not to a limited number of identifiable individuals, but nor is this a mass publication case directly comparable to Lachaux itself. The evanescence of tweets, the supersaturation of Twitter with information flows, and the multivalency – the omnipresence, persistence and extensive partisan engagement – of the sorts of debates with which this case is concerned do have to be factored in, and not just the unlimited possibility of access to undeleted public posts online. I take the Twitter analytics of the individual publications complained of into account, especially with regard to engagement within the jurisdiction: these do not in my judgment encourage a ready inference of high or lasting impact. I allow for some percolation, and for Mr Turner’s profile. These are all relevant to the circumstances of publication.

I bear in mind the evidence I have of the class of immediate publishees of this material. The context of the relevant threads was extremely partisan. There is some strong evidence of a direct and powerful counter-reaction or backlash to Mr Turner’s tweets from among those who evidently knew or came to know the Claimants (at least online). This vocal response clearly thought none the worse of the Claimants – they did not accept the import or materiality of what Mr Turner said about them. On the contrary, they clearly thought a great deal the worse of Mr Turner for making these allegations, and said so with considerable force. Such responsive support as is visible for what Mr Turner said appears largely to be referable to those already polarised in the HWNDU and LD50 affairs. There is little or no evidence for adverse reputational impact (or for propensity to onward percolation) from the Twitter threads themselves. There is no sign of anyone’s mind being changed; minds showed every sign of being and staying firmly made up. And any neutral observers stumbling across the relevant threads and exchanges in real time would no doubt have been able to make their own minds up not just by reference to Mr Turner’s accusations but by reference to the Claimants’ own conduct in response – the immediate context for the publications. I consider that response in more detail below.

Then – and importantly – two further factors fall to be considered on the authorities set out in Amersi. First, there is no support in those authorities for drawing inferences for the causation of serious harm by the publications sued upon by means of an evidential process amounting to the indiscriminate aggregation of all the imputations complained of, other seriously damaging imputations not complained of, other publications not sued on, and a range of publications by third parties with similar content. Ms Grossman is technically right to say that making the same inferential case in relation to each publication or imputation sued on by reference to the same factors is something different from indiscriminate aggregation. Conceptually, it is. But in this case it is hard to discern any evidential case being made for the distinctive impact of the imputed meanings in the publications sued on, beyond what is referable to the meaning of the words in isolation. The case for any inference of causation is not effectively raised.

Accusing anyone, however otherwise polarising or besieged their reputation, of antisemitism, threatening violence or publishing Hitlerian euthanasia theories is a serious matter, and of undoubted defamatory tendency. But section 1 requires a clear articulation, and an evidential basis, for what difference the publications and imputations complained of made or were likely to make in real life. I do not in this case have that clear articulation or that evidential basis. ‘Drawing inferences is not a process of optimistic’ – or rather, it might be said, pessimistic – ‘guesswork; it is a process whereby the court concludes that the evidence adduced enables a further inference of fact to be drawn.’ (Amersi [158]).

The other factor emphasised in Amersi ([157]) is that section 1 is a threshold issue, and in applying it, it is necessary not to lose sight of the basic tort rules of causation. Evidence contrary to the imputation of causal responsibility is no less potentially important than evidence tending to favour it.

In all of these circumstances, I have not been able to conclude that Mr Miller and Ms Power have sufficiently discharged the burden Parliament has placed upon them of demonstrating that it is more probable than not that the imputations of which they complain, in the publications of which they complain, caused or were likely to cause serious harm to their reputations. That does not mean I have concluded that their reputations have not been seriously – and perhaps unfairly – harmed. It means that they have not sufficiently attributed such harm to Mr Turner’s publications, so that his freedom of expression would fall to be curtailed in law accordingly.

In these circumstances, the law does not require Mr Turner to defend his publications further, whether by reference to the statutory defences or otherwise. So it is neither necessary nor appropriate for me to reach any final conclusions about them. In the circumstances, I confine myself to the following observations.

Mr Miller did not ‘publish a pseudonymously authored blogpost entitled ‘Toward a Hitlerian Disability Politics’, in which he called for the state-mandated euthanasia of disabled people’ (Mr Turner’s tweet of 14^th February 2019). That was an allegation of fact, which was untrue. Mr Turner accepts he made a mistake about this, and removed the tweet accordingly (albeit after it had been posted a number of months). He indicated to me through Counsel that he was prepared to apologise for his mistake.

Mr Miller told me he had taken particular exception to this allegation. He said ‘I don’t know if we would be in these libel proceedings if Mr Turner had not claimed that I had written this article’.The Twitter analytics show a single retweet, by a US university professor, an individual outside the jurisdiction. It is not otherwise pleaded or evidenced that anyone outside the immediate (and partisan) online conversation read or reacted to it, or believed Mr Miller had published this article or thought he might have done. There has been no trace of any such article online. Mr Miller’s own immediate online response to this tweet was ‘LMAO’ (laughing my arse off). He also posted a tweet stating: ‘Luke Turner is tweeting as a resident of a lunatic asylum’ and a screenshot of a page from Günter Grass’s The Tin Drum. I have no evidence of any step it occurred to Mr Miller advisable to take to set the record straight, at the time or subsequently.

The imputation of fact about a police investigation relates to what Mr Turner said was the Claimants’ online harassment of him, its antisemitic character and its threats of violence. It is to Mr Miller’s and Ms Power’s alleged engagement in such a campaign that I turn next.

The dispute

Mr Turner alleges a campaign of online harassment of him by both Mr Miller and Ms Power lasting around a year. This is extensively particularised by reference to their own online tweets and their interactions with those of others. More than 80 such incidents are specifically referred to, each in its own context of other interactions. Ms Evans KC, leading Counsel for Mr Turner, suggests they indicated a repeating pattern of bursts or waves of harassment.

Mr Miller and Ms Power respond in corresponding detail. But their overarching pleading is that ‘The interactions between the Claimants and the Defendant formed part of an online dialogue about art, free expression and the extent to which the Defendant’s conduct in relation to LD50 and individuals or institutions of which he disapproved was proportionate to their alleged wrongdoing or amounted to criticisms made in good faith’.

Context is of course everything, and Mr Turner is clear that these interactions were at the same time very much ‘about’ his own experience of HWNDU and the hostile personal onslaught he was subjected to online and in real life – including gross antisemitic abuse, intimidation and threats of violence – and the extent to which others, particularly in the art world, had actively or passively supported or given comfort to the perpetrators.

There is, again, little dispute about the relevant law, and I set it out briefly below. Key factual issues between the parties are the extent, if any, to which Mr Miller’s and Ms Power’s online behaviour was (a) antisemitic and (b) threatening of violence, and these raise issues of interpretation of what they said and how they said it. Mr Turner’s online accusations about this of course provoked their defamation claim. In his own harassment claim, Mr Turner relies on these features, among others, as establishing that their behaviour had a harassing quality.

The law

Harassment is a statutory tort of relatively recent origin. It was created by Parliament in section 1 of the Protection from Harassment Act 1997, along with a criminal offence of harassment in section 2. It is defined in section 1(1) as ‘a course of conduct which amounts to harassment of another, and which [the perpetrator] knows or ought to know amounts to harassment of the other’. It is a fully objective test: what a complainant subjectively thinks about the course of conduct is not the legal test. Importantly, Parliament provided expressly that harassment may be constituted by speech (section 7(4)). A course of conduct will not count as harassment if a defendant shows that, in the particular circumstances, the pursuit of the course of conduct was reasonable (section 1(3)(c)).

This is an outline, or framework, and plainly fact-sensitive, definition. Its application to various factual circumstances has been considered in some detail by the courts in the intervening years. Again, this jurisprudence was helpfully surveyed and summarised by Nicklin J. In Hayden v Dickinson [2020] EWHC 3291 (QB) at [44], he set out a summary approved and expanded by the Divisional Court inScottow v Crown Prosecution Service [2021] 1 WLR 1828 at [25]. Nicklin J characterised harassment as ‘a persistent and deliberate course of unreasonable and oppressive conduct, targeted at another person, which is calculated to and does cause that person alarm, fear or distress’.

Of particular significance in the survey of the jurisprudence is the decision of the House of Lords in Majrowski v Guy’s & St Thomas’s NHS Trust [2007]1 AC 224. At [30], Lord Nicholls highlighted the ‘boundary between conduct which is unattractive, even unreasonable, and conduct which is oppressive and unacceptable. To cross the boundary from the regrettable to the unacceptable the gravity of the misconduct must be of an order which would sustain criminal liability under section 2’[of the 1997 Act]).

Nicklin J also noted the authorities’ indications that where, as here, the complaint is of harassment by publication, the claim will usually engage a defendant’s right to freedom of expression (Article 10 ECHR) and therefore a court’s duties under the Human Rights Act, including section 12. So the 1997 Act has to be interpreted and applied accordingly, with particular reference to the ‘gravity’ qualification, and the objective test for the quality of harassment. There is a real tension to be dealt with. Section 7(2) of the Act provides that harassment includes elements of causing alarm or distress to a claimant, but at the same time Art.10 is capable of protecting speech that offends, shocks and disturbs.

He further noted that cases of alleged harassment can also engage a complainant’s rights of personal autonomy and privacy (Article 8 ECHR). So there may need to be a final calibration or balancing of competing fundamental rights in cases of harassment by speech.

Analysis

Harassing course of conduct

There was little real dispute before me, and little apparent in the evidence, about whether Mr Miller’s and Ms Power’s conduct amounted to a ‘course of conduct’. They did dispute that it was (altogether) ‘targeted’ or ‘directed’ at Mr Turner – or, if it was, they said that he invited it and participated in it. But I have little hesitation in concluding that (enough of) it was targeted on Mr Turner, in the sense the law should recognise. The tweets specifically complained of were either addressed to him, or plainly about him and contained within threads with which he had engaged or which he was otherwise likely to see.

The next question is whether the course of conduct had the necessary objective quality of being harassing. On that objective basis, it is at least apparent that it was more than a vehement contest of ideas. It cannot easily be described as part of a consciously chosen and freely participative dialogue. Of course, there is always an option to leave Twitter or not to engage with certain accounts. Mr Turner told me he did not wish to hand that sort of trophy to his critics: it was a form of censorship in its own right. That is in the end a choice. Nevertheless, I bear in mind that in his particular circumstances he did have some reason to monitor, if not actively to participate in, this sort of material for signs of linkage to the pattern of real-life threat and stalking that he had already been experiencing. That is relevant context.

Where Mr Miller and Ms Power responded to Mr Turner directly, or referred to him by name, the full lexicon of Twitter trolling is objectively recognisable. There is the crude personal invective of which a sample from @dcxtv has already been cited (this account was in due course shut down altogether for breach of the Twitter rules on ‘hateful conduct’). There is material of this sort from Ms Power too. Then there are, from both, the arch demands for evidence or proof of both his own history and the opinions he criticises, and faux-naïve insistence that he is constantly mistaken in his understanding of their views. There is more than disagreement – there is denial of any possible legitimacy, rationality or tenability of Mr Turner’s opinions, and a plain imputation of malice.

There is, in particular, a relentless pursuit of the theme of his mental deficiency or mental illness (in addition to the casual use of ‘retard’) – ‘low IQ’, ‘sociopathic’, ‘paranoid’, ‘deranged’, ‘obsessive’, ‘paranoid schizophrenic’, ‘dangerously mentally ill’ and so on and on – sometimes accompanied by messages to ‘get help’. (Ms Power told me from the witness box that the latter proceeded from genuine and empathetic concern. I was unconvinced.) There was also mutually-reinforcing group participation in all this conduct, involving a range of other participants sympathetic to Mr Miller’s and Ms Power’s views, who shouted Mr Turner down by saying the same sorts of things to or about him in the same sort of way (or worse).

This could, on an objective basis, fairly be described as bullying. It was asymmetrical. Mr Miller and Ms Power found Mr Turner’s views offensive, and his labelling of them personally offensive, but he did not express himself in an offensive manner. They considered his viewpoint unreasonable, but he expressed it moderately, rationally and on a reasoned basis, with a view to engagement and eliciting answers to questions or challenges. As already noted, Mr Turner’s side of the interactions is recognisable as an exercise in ‘calling out’ – that is, an exercise in labelling Mr Miller’s and Ms Power’s words and deeds (and others’), on the basis of what is entirely apparent was his genuine and articulated belief that they merited those labels. Their response was to attack him. They did so persistently. (I observed a little of the flavour of this myself in their conduct of this litigation, in their oral evidence, and in the choices made, presumably on instruction, about the cross-examination of Mr Turner. He responded, and assisted the court, with presence of mind and composure.)

Context is everything. This was Twitter. This sort of register of discourse is ubiquitous there. That does not render it otherwise than something which I am satisfied could objectively and reasonably be described as an exercise in bullying. It does not make it reasonable. It does not exonerate it from having something of an oppressive quality. However, before the law will interfere and abridge this sort of speech, it demands consideration of whether it amounts to misconduct of a sufficient order of gravity to sustain criminal liability. It demands a court have particular regard to the importance of the fundamental right of freedom of expression. And, if a countervailing fundamental right is engaged, it demands a careful balancing exercise in proportionality.

There are two features of Mr Miller’s and Ms Power’s speech that Mr Turner particularly relies on as justifying abridgment. They parallel those of which they in turn complained in their defamation claim. He alleges their speech, in context, was antisemitic, and threatening of violence.

Gravity

The exchanges set out above in which Mr Turner challenged Ms Havas’s ‘liking’ the Pepe/HWNDU cartoon continued for a few days. Mr Miller and Ms Power were involved, and a few others joined in, on both sides of the debate. On 1^st August 2018, a friend of Mr Turner questioned Ms Havas’s wisdom in tweeting swastika images from Berlin. Mr Miller responded with a series of tweets, moments apart: (a) ‘You’re on the list too now…’; (b) an image of the Voudou ‘Kalfu’ symbol with ‘what they don’t know, but will know…’; (c)‘They probably think I’m joking’; (d) ‘BLOOD’; (e) ‘Watch your steps…’.

Mr Miller explained to me that these were not threats of violence towards Mr Turner. They were general musings on karma, and the propensity of those who treat others negatively to encounter negative consequences in their own lives. They reference the nine gates of hell in Haitian Voudou. But Mr Turner looked up the symbol and noted that some connected Kalfu with Satan. He was – in the context of his past experiences – frightened by these tweets. He went to the police.

He made a statement, setting out his experience of being stalked, harassed and threatened in the wake of the anti-HWNDU campaign. He showed the police the history of Mr Miller’s recent interactions with him. The police took the matter seriously. They accepted Mr Turner was genuinely fearful of violence. They opened a criminal investigation, noting the potential that antisemitic hate crime and/or offences under the Malicious Communications Acts had been committed (criminal harassment, as such, was not mentioned). But they did not in the end pursue the investigation, concluding that the material probably fell into a grey area of unpleasantness short of criminality. That was largely on the basis that any threat was insufficiently specific, direct and active. They also had other priorities (the hate crime unit in question also dealt with domestic violence).

A few days after the Kalfu tweets, on 4^th August 2018, Mr Miller tweeted: ‘@Luke_Turner I’ve had enough of your disgusting behaviour and lies. I challenge you to a physical fight to take place in London one month from now – are you man enough to accept?’ And then, later that day and in the absence of a response from Mr Turner: ‘I suppose it’s not surprising that this invertebrate homunculus, who’s already shown he has no manly virtues, would refuse the challenge of a duel’. Mr Miller said this was obviously satirical in intent. But it was plainly not a message of goodwill.

I was shown a number of examples in the course of conduct complained of where Mr Miller and Ms Power were said to have used or approved violent imagery in the context of their course of conduct towards Mr Turner. Ms Power quoted at some length, and without attribution, André Breton’s proposition that the simplest surrealist act consists of dashing down into the street, pistol in hand and firing blindly into the crowd. Pictures of mass murderers make an appearance, under the Nike slogan ‘just do it’. Ms Power sent Mr Turner a still image from a film showing a heretic about to be burned at the stake.

These examples are limited and sporadic, but undoubtedly uncomfortable. There may be some objectively discernible undertow of menace in all of this. Mr Miller’s and Ms Power’s course of conduct was certainly aggressive, in register, language and imagery. Mr Turner’s experience had taught him not to make the mistake of assuming that, within the broader New Right milieu, words and deeds, or irony and intent, necessarily belong in entirely discrete spheres of discourse. For their part, Mr Miller and Ms Power advanced from the witness box interpretations and understandings of their speech to distance it from any possible taint of a literal mindset of violence towards Mr Turner. From an objective and contextualised perspective, however, it is not too difficult to see the basis on which Mr Turner concluded he had reason to fear the contrary.

Then there is the question of antisemitism. Mr Miller protests his innocence especially vehemently on this issue, not least from the perspective that he is himself a man of Jewish heritage. I say at once that nothing in the course of conduct of which Mr Turner complains begins to approach the gross antisemitism to which he had been subjected by others. There is no unambiguous or overt antisemitism. Mr Turner relies instead on context – the context of the company Mr Miller and Ms Power kept online including those with whom they sided or who sided with them (including Ms Havas), and the context of the history of the clearly antisemitic component of some of the anti-HWNDU activism which they refused to condemn – and on an undertow of antisemitic tropes in their online speech.

Taking that latter first, Mr Turner draws attention to a number of recurring features. There is the theme of the subhuman (of which a couple of examples have already been cited) and the soulless. There is a theme of cringing cowardice. There is the enthusiasm with which Mr Miller and Ms Power at one point promoted a rumour that Mr Turner was a fabulously wealthy heir to an unscrupulous family textile business. There is Mr Miller’s use on one occasion of ‘Entartete Kunst’ as a put-down (Hitler’s notorious condemnation of ‘degenerate art’, much of it by Jewish artists). And perhaps most insidiously, there is a theme of Mr Turner’s paranoia and solipsism on the subject of antisemitism itself: his declared experiences and diagnoses of antisemitism are said to be false, deluded and malicious. Perhaps one of the oldest and most persistent of all tropes of antisemitism is that antisemitism itself is a figment of the corrupt Jewish imagination, a Jewish conspiracy. So again, while these individual examples may be oblique, limited and sporadic, taken as a whole, and especially in context, it is understandable that Mr Turner felt they created an uncomfortable atmosphere.

In the materials available in this trial there is visible, within the New Right milieu, what might be described as some diversity of thought about antisemitism. The milieu embraces some simple and unabashed antisemites, and others who find their views repugnant. There is a recognisable strand of strongly pro-Israel sentiment. There are some who do not think the issue of antisemitism important, or consider it overstated, for modern conditions – and who firmly resist the critique that that is in itself an antisemitic position. There are those who are interested in exploring antisemitism as a phenomenon, including by engaging with its taboos, not excluding Holocaust denial, through art and humour. Some of the critics, or adversaries, of the New Right milieu, however, say it is all one – that antisemitism is not to be finessed or gradated, mocked or fetishised, and that those interested in doing so are themselves fundamentally antisemitic in outlook.

Mr Miller talked to me about the history of antisemitism, the ‘complexity’ of the phenomenon, and his interest in the culturally ‘forbidden status’ of the Holocaust. And Mr Turner talked to me about the paradoxes of Jewish antisemitism. My task in all of this is however strictly limited. I am not called on to adjudicate anything apart from the gravity of Mr Miller’s and Ms Power’s online course of conduct, taken as a whole and considered on an objective basis.

Looking as objectively as possible at the course of conduct before me in this harassment claim, I can see the Claimants’ lack of enthusiasm for the cause of assiduously calling out and challenging antisemitism, and their insistence that Mr Turner was wrong, or at the very least subjective, in his labelling of their conduct as such. There is also at least some evidence of what might fairly be called a casualness, or perhaps (put at its highest) a knowingness or artfulness, around antisemitic tropes. Mr Miller is proud to be called an intellectual; his enjoyment of the nuances of word and the semiotics of image is plain. Ms Power agreed with me that as an art-critic she was a sophisticated appraiser and user of images. Both are students and exponents of societal and cultural issues. There may perhaps be inferred into their conduct towards Mr Turner something of an aware relish for the frisson of danger of their discourse, and that of those proximate to it who joined in. In the particular context of the well-understood history of HWNDU, the personal consequences for Mr Turner which he was at pains to set out in his own tweets, and an aggressive and unpleasant course of bullying behaviour, it is hard not to conclude that the risk of creating an experience of antisemitism was one which they were especially well placed to understand and which they were not equivalently scrupulous to avoid. This was not a thought experiment. It was a ‘dabbling in the soul’ of a real-life Jewish man who had been subjected to antisemitic outrage on account of an artwork. It had some quality of oppression in those circumstances.

The question the law asks, however, is not only whether this course of conduct had oppressive, alarming and offensive qualities, but whether it amounted to misconduct of a quasi-criminal nature. That is the test laid down in Majrowski and consistently applied by the courts since. Ms Evans KC put to me that while the full force of that test must be brought to bear where mainstream journalism is concerned, the bar is not so high in cases involving individuals bullying others online. But I was taken to no authority in support of that proposition. Majrowski was not itself a journalism case. And in a claim for harassment by publication I have to factor in the particular importance of freedom of expression.

Ms Evans KC relied on the High Court decision in Suttle v Walker [2019] EWHC 396 (QB) as an illustrative example of harassment by online speech. But two points must be recognised about that case. First, the facts were very different. There, the defendant posted her anonymous, untrue and defamatory allegations out of the blue on a bespoke animal cruelty Facebook page, and persistently repeated them, a course of conduct held to have been calculated to produce a terrorising frenzy and direct and credible threats to the claimant’s safety, and to have been the sole cause of producing exactly that result in fact. Those facts are entirely different from the present case, where the issue goes effectively to the register of an online argument about art and protest and indeed antisemitism itself. And second, the case was undefended, and hence produced a default judgment on the pleadings. The application of the Majrowski test was uncontested and unaddressed.

The Majrowski test is unambiguous and binding authority, and I must apply it. I agree that the fact the police did not pursue a sample of the course of conduct complained of in the present case is not determinative of the question of quasi-criminal gravity of the whole course of conduct (although neither, perhaps, is it entirely irrelevant). Prosecutability is not the test. Criminality by reference to offences other than harassment is not the test. It is simply a fact sensitive test of gravity, going to the quality of the course of conduct complained of. Assessing gravity is a holistic and evaluative exercise, to be undertaken in as full a context as possible. Where harassment by speech is concerned it has to be undertaken with great care. And for the reasons which follow, I have concluded that Mr Miller’s and Ms Power’s course of conduct, while it may fairly be said to be richly deserving of the epithet ‘regrettable’, fell short of the high bar set.

Conclusions

Mr Turner’s counterclaim, as pleaded and confirmed at trial, is against Mr Miller and Ms Power alone. It is their words and conduct I have to consider. The enormous wider debate is relevant context, but they cannot be held responsible for it all, or coloured with its very worst aspects and participants, or blamed for the aggregate effect it had on Mr Turner’s life. Their interests, views and world-view, and their online friends and tastes, are not themselves on trial in this claim. I have to focus on what they did to Mr Turner.

The high watermark of Mr Turner’s case on gravity relates to the queasy undertows of menace and antisemitism alleged in this course of conduct. I agree that this is not a fanciful imputation, in context. I accept that this was Mr Turner’s subjective experience and understand why. But I cannot agree that, objectively, the course of conduct sued on advances far enough to cross the boundary between the regrettable and the unlawful, in this respect or in total. I do not reach that view on any basis of clever parsing of their speech and imagery for lack of literalism. It is just that the whiff of threat and antisemitism which they themselves can fairly be held to have created is too occasional, intangible, diffuse and unconsolidated to impart those distinctive and characteristic flavours to this course of conduct as a whole.

I do not underestimate the effect it may have had on Mr Turner. I do not imply that his ‘calling out’ of their behaviour in these terms was baseless. Mr Turner evidently had been harassed to a quasi-criminal (or indeed criminal) degree by other people. But what Mr Miller and Ms Power did was not of that order. Mr Turner’s labelling of Mr Miller’s certainly disturbing tweets as death threats does not quite survive sober objective analysis (including, as invited by Ms Evans KC, by reference to the criminal offence of threats to kill (Criminal Law Act 1977 Sch. 12) – an offence with a complex mental element for which I cannot find a correspondence in the present facts: the evidence does not support any conclusion that Mr Miller intended Mr Turner to believe he would kill him). Otherwise, the radio frequencies of macabre imagery and casualness around antisemitic tropes into which Mr Turner tuned do yield a discernible signal but not one strong enough to be recognised objectively as antisemitic hate speech or a resolved engagement with physical violence sufficient to render this conduct quasi-criminal. That is not to say that harassment by batsqueak or dogwhistle can never amount to an actionable tort, just that I am not persuaded on the facts that there was sufficient of it to do so in this case.

Debating art and ideas, contesting the nature of the New Right milieu and its critics or opponents, arguing about free speech itself – about protest and counterprotest, about cancel culture and no-platforming, about calling out and smearing – all this is of the essence of the exercise of the fundamental right to freedom of expression which is protected by law. The law rightly hesitates long before trying to regulate the register of such debates – or indeed of debates involving participants with a shared Jewish heritage disagreeing about what constitutes antisemitism in the art world. None of this has to be done well to earn the law’s protection. None of it has by law to be done moderately, civilly, fairly, respectfully or kindly. On Twitter, it rarely is. The law protects speech which shocks, distresses, alarms and offends unless, so far as harassment in online exchanges are concerned, it is of an order of gravity which raises it above the general melee and into the sphere of that with which the criminal law could concern itself.

Mr Turner did not explicitly advance a case based on his own Article 8 rights. Again, I have noted that those rights appear to have been clearly violated by other people, in attacking his home life physically and by direct threats. I can see equally clearly that Mr Turner’s well-being and peace of mind have been seriously interfered with by the whole of his experience of the reaction to HWNDU. (His father spoke articulately and beautifully of the lights that had gone out in his life. Mr Turner has had more than his fair share of misfortune in the respect, but he is surely fortunate to have so steadfast, appreciative and eloquent a supporter by his side.) I cannot, however, on the materials and evidence with which I have been provided, discern a material and grave contribution to that impact specifically attributable Mr Miller’s and Ms Power’s course of conduct as pleaded, as distinct, that is, from the irreducible minimum of strain, distress, distraction and depletion subsequently attendant on being involved in protracted litigation of the present sort.

All the parties to this case, as I have observed, have passionately, and perhaps bravely, championed freedom of expression, particularly the freedom of art and art-criticism to explore or express ideas to which others take, and then themselves express, strong exception. All the parties have told me, at times with great emotion, how they have suffered as a result. Mr Miller and Ms Power told me how they have been shunned, dropped and no-platformed in the world of art and ideas in which they had looked to thrive on their own merits. Mr Turner told me how his work has been persecuted and he has been subjected to violent and antisemitic threats and abuse, including in the violation of his peaceful home life, leading to being cancelled himself.

This litigation arises because of the manner and degree in which each side blames the other, to a material but at the same time specific and limited extent, for their predicament. I have explained why I have not been able to agree that, on the evidence brought to court, there is enough of a cause-and-effect factor established to support that attribution in law. In the statutes it passes, Parliament has been careful to ensure that the law does not interfere in freedom of expression (in this case on Twitter) unless demanding threshold tests are passed on the facts. In the case of defamation, that is a test that the specific publications complained of have themselves caused serious reputational harm. In the case of harassment, that is a test that the aggregate discourse complained of must be of an order of gravity which could be described as quasi-criminal. These are high bars to clear, and I have not been able to be satisfied on their pleading and evidence that any party has succeeded in doing so.

All the parties have been caught up in, and are suffering the consequences of, much wider and deeper events and currents than those they brought before this court in their claims and counterclaims. Although each side told me that all they wanted was for the other side to leave them alone, they can have had no realistic expectation that this litigation by itself could have put an end to their respective wider troubles. And although my decisions may be disappointing in equal measure, the law’s default to freedom of expression is one which is at the same time concerned to respect and defend the personal autonomy of each of the parties and their rights to impart and receive information on their own terms without state interference. It may be that the conclusion of this litigation will itself enable the parties to turn to a future which is not consumed by rehearsal of the minutiae of their grievances against each other and the tweets of years ago, and in which Mr Miller’s book about André Breton can be considered by publishers on its own merits, Ms Power’s opinions on art and gender can be understood and critically assessed on their own terms, and Mr Turner can return to making artworks and playing Bach partitas on his violin to his father.

Mr Miller’s and Ms Power’s claim in defamation is dismissed. Mr Turner’s counterclaim in harassment is also dismissed.