The Unbearable Loudness of Cancellation: How the Court of Appeal failed free speech in Miller v College of Policing – Paul Wragg

Freedom of speech is quite the enigma. Just ask Laura Murray. When Rachel Riley retweeted, with embellishments, Owen Jones’s comment that if you don’t want to be pelted with eggs, you shouldn’t be a Nazi, moments after Corbyn had himself been pelted, his stakeholder manager lambasted Riley. Not unreasonably, Murray took Riley to be both condoning violence toward politicians and to be labelling Corbyn a Nazi.

I can’t speak for Murray’s motivation but anyone in politics, presently, must fear for their MP’s safety (and their own) given the death of Jo Cox and, subsequently, Sir David Amess. Any encouragement of violence toward politicians, of any kind, deserves vehement criticism. Sadly, the High Court saw things differently, awarding Riley substantial damages for the serious harm to reputation that Murray’s passionate response was said to have caused.

Notably, the High Court expresses no concern that this decision might have a chilling effect on political debate of public interest matters. There is no recognition that the right to freedom of speech extends protection even to provocative, offensive, and upsetting speech.

Riley v Murray, then, sits uncomfortably with the Court of Appeal’s decision in Miller v College of Policing [2021] EWCA Civ 1926,  which was handed down on the same day. Here, the Court made much of the supposed ‘chilling effect’ that the threat – even the distinct prospect – of state interference with Twitter debate has.

The facts of Miller are relatively complicated. Put simply, Mr Miller, a company director, regularly posts tweets/retweets that speak to his strong belief that trans women are not women. Matters came to a head when Mr Miller re-tweeted an upsetting limerick denigrating trans women. This was subsequently reported to the police for transphobia who, following investigation, recorded it as a ‘non-crime hate incident’ (NHCI – a classification I explain below). The investigation – which amounted to a visit and several phone calls – was sufficiently unsettling (undoubtedly so) that Mr Miller understood the police to be saying that if he continued tweeting about trans people in derogatory terms, criminal prosecution would follow. Mr Miller, though troubled by this, ‘decided to continue tweeting’ – with no adverse consequences or further visitations from the police. So far, so good. No chilling effect here.

This might have been the end of the story. Yet, troubled by the police’s heavy-handiness, Mr Miller pursued legal action against both Humberside Police (who visited him) and the College of Policing. His claim against the first defendant succeeded, as I think it should have done. The police ought not to have pursued this matter as vigorously as it did. His claim against the second defendant failed – and it was this that he appealed against on public interest grounds.

The College of Policing wrote the guidance on handling/recording NCHIs. Put simply, NCHIs are a logistical resource that helps identify localised patterns of hate. For example, the Northern Irish police use them to monitor sectarianism. This information has operational value since, as the guidance notes, at 6.1, ‘incidents can escalate to more serious actions if unchecked’. They have their origin in the MacPherson report concerning the chronic police failings in the Stephen Lawrence investigation.

As it is logistical rather than prosecutorial, an NCHI demands less investigative resources than a criminal investigation would. Accordingly, the hate element of an NCHI is perception-based, and the officer in charge need be satisfied only that the complaint of a hate motivated attack on protected characteristics is credible rather than indisputable.

The guidance states that the officer investigating an alleged hate-motivated complaint, if she decides that the aggressor’s actions do not breach the criminal law, must conclude her investigation by recording it instead as an NCHI. Any related information identifying the aggressor, for example, should such information exist, is stored locally rather than on the national database.

Mr Miller argued, but unsuccessfully in the High Court, that the mere existence of such a record had a chilling effect on his freedom of expression and, therefore, breached his rights under Article 10 of the European Convention on Human Rights. Accordingly, he said, the guidance was itself an undue interference with his free speech rights by its insistence that all such perception-based complaints must be recorded as an NCHI. Whereas the High Court thought Mr Miller’s complaint fanciful and dismissed it entirely – following thorough consideration of it – the Court of Appeal concluded that the High Court had erred so seriously that it was empowered to conduct the Article 10 analysis afresh. Consequently, it agreed with Mr Miller that the guidance represented both a serious and disproportionate interference with his free speech rights and, therefore, that the High Court was wrong on both these points. The guidance, it concluded, should be changed.

The Court of Appeal is wrong.

That it should ennoble Mr Miller’s unedifying contribution to epistemic advance with a dignity it barely deserved is bad enough, but, worse, it relied upon utter irrelevancies to reach its untenable conclusion that freedom of political expression is an absolute right for which even purely hypothetical interferences are intolerable. For, plainly, the Court of Appeal confused the spectre of hostility to hostile speech where intolerance is met with intolerance (and in doing so dignified right-wing ‘cancel culture’ paranoia), which was entirely irrelevant to these proceedings, with the actual consequences of an NCHI classification that are so far-fetched the High Court could barely conceive of circumstances in which Mr Miller could be affected by it in anything other than purely hypothetical terms. In doing so, the Court of Appeal lost sight of the real issue in these proceedings which was no more than this: 1) does the classification by the police, that a complaint is not a crime but an NCHI, constitute an interference with freedom of speech? 2) is that interference unjustified?

On the first question, the Court took its first misstep by categorising Mr Miller’s speech as an important contribution to a debate of public interest. In doing so it elevated both the speech itself (which the High Court had concluded were, collectively, ‘opaque, profane, or unsophisticated’) and the underlying debate – that trans women are not women – to the level of academic respectability by stating, repeatedly, that Mr Miller’s denigration of trans women is a credible contribution to public debate because it is shared by Kathleen Stock, amongst others, a former professor at the University of Sussex. It is deeply regrettable that the Court of Appeal would so readily embrace a position on a deeply divisive issue in favour of those that trumpet repressive politics and, worse, that it would imbue it with the imprimatur of academic respectability by proffering the views of an individual who is, at best, a controversial figure.

As far as these legal proceedings ought to have been concerned the specific of Mr Miller’s world view are utterly irrelevant. At no point was it necessary for the Court of Appeal to decide whether unvarnished hostility toward trans women is a respectable political position to adopt. The question was whether the guidance is capable of interfering with speech and it should have confined itself entirely to an impartial examination of the consequences that follow from being interviewed by the police as part of their investigation into the complaint. It did no such thing.

The Court then compounded this categorisation error by endorsing, uncritically, the existence of a so-called cancel culture in which open hostility toward protected characteristics is said to be met by permanent expulsion from public life. These ‘cancelled’ individuals include Stock herself, who has been anything but silent since she resigned her academic post and, of course, Piers Morgan, who seems pathologically incapable of silence.

Thus, for the Court of Appeal, the interference with Mr Miller’s speech, which it describes as ‘real and substantial’ results from ‘belong[ing] to a group of people [ie, those hostile to trans rights] who could easily be stigmatised for their opinions and be subject to complaints by those offended by his views’ (para 76). This is entirely likely – and also entirely irrelevant. We are back to the realm of Owen Jones’s good advice: if you don’t want to be pelted with eggs… Nevertheless, the guidance produced by the College of Policing has no bearing on this.

The only question that the Court of Appeal ought to have asked is whether the instruction in the guidance, that a perception-based complaint of a hate motivated attack on a protected characteristic must be recorded as an NCHI, constitutes an interference with speech.

The High Court saw this point clearly (and it is a shame that the Court of Appeal saw fit to interfere with the sensible findings of Julian Knowles J on this point). There are only two conceivable ways in which the NCHI record could constitute an interference relating either to the nature or consequences of that record. In Wood v Commissioner of Police for the Metropolis, the Court of Appeal accepted that the retention, by the police, of photographs of an activist constituted an actionable breach of the Article 8 right to privacy given the sinister overtures of state monitoring that this retention spoke to. Mr Miller’s situation, though, is radically different because a) the information relating to him, if it exists, lies dormant; and b) that information is collated as part of the investigation into the complainant’s complaint – regardless of whether or not it is recorded as an NCHI. To reiterate: an officer cannot designate the incident as an NCHI unless satisfied that it does not constitute a crime (for example, a public order offence). Consequently, whether it is classified as an NCHI or not, some record would exist, locally, that identified Miller as the object of a complaint. The guidance alters nothing.

The High Court also considered, at length, the personal consequences for Mr Miller of an NCHI classification – and could not readily identify any of significance. The only plausible consequence relates to employability prospects which, for Mr Miller, were both remote and entirely hypothetical. NCHIs do not appear on standard criminal record checks. Some evidence of police investigation could appear on an enhanced criminal record check if that information is deemed relevant to the (prospective) employee’s role. For the High Court the only plausible circumstance in which this might occur was if Mr Miller pursued a career caring for trans children or vulnerable trans adults.

Whereas it seems highly unlikely a prospective employer could even encounter this information, it could hardly constitute an interference unless that employer was affected by it. This presupposes that an adverse reaction to news of Mr Miller’s being contacted by the police, following a perception-based complaint, which led to no action being taken against him personally but which was recorded by the police, for logistical purposes, as an NCHI. It also presupposes that this same employer, whilst highly sensitive to this fact, is insensitive to Mr Miller’s openly expressed attitude toward trans rights which he broadcasts, freely, to the world at large via his twitter account. That the Court of Appeal should indulge this fantasy demonstrates something of a radical break with reality.

Yet, even if, the mere recording of an NCHI is an interference, it is not, without more, an undue interference, for the right to freedom of speech is qualified by the duties and responsibilities listed in Article 10(2). This includes safeguarding the public amongst other things. How then is the mere recording of an NCHI by the police, which has no conceivable consequences for the aggressor except in the most extreme circumstances, a disproportionate interference with freedom of speech? The answer is simple: it isn’t. The risk to freedom of speech posed by the guidance is vanishingly small, especially compared to the risk of a hostile reaction to online provocation, which is both serious and predictable.

Quite why the Court of Appeal should pander so obsequiously to the sensitivities of the insensitive is beyond me. For me, as a free speech lawyer, I am no more than disappointed by this hopeless mess of poorly executed Article 10 reasoning.

For those directly affected – the transitioned, the transitioning, and their loved ones – the daily grind of hostility towards them as they try to find themselves, to live their life, whilst asking for nothing more than compassion, is unbearable. To now find that the Court of Appeal – a supposed bastion of liberal principle –dignifies this senseless, abhorrent abuse and denigration of trans women and calls it ‘important’ legitimate debate on a matter of public interest is as devastating as it is inexcusable. Truly, the decision is a travesty of justice.

Paul Wragg is Professor of Media Law, at the University of Leeds and co-host of @medialawpodcast.