On 10 November 2026, the Guardian and ITV News released details of an investigation into David Walliams that took more than three years to carry out and a High Court trial to publish. Several women say that Walliams used his celebrity to pursue them for sex when they were teenagers or young women, and one alleges a sexual assault when she was 15.
Walliams denies that he is a sexual predator or that he abused his power, and says that he has broken no law. He accepts that in casual sexual relationships he “did not always take people’s feelings properly into account”. At the time of writing, no court has determined where the truth of the published allegations lies, and nothing I discuss depends on it.
A week before publication, Walliams had been “PPT”, whose claim for an injunction to restrain that journalism was dismissed by Nicklin J ([2026] EWHC 2818 (KB)). Professor Paul Wragg has written on this blog about the Article 8 reasoning, leaving Articles 2 and 3 aside, although he found it “counter-intuitive that the Court should think itself so powerless”. Wragg’s instinct is exactly what I want to explore further.
It takes great courage to give an account of this kind about a famous and powerful man. The journalism that supported it was, in the judge’s words, “prolonged and extensive” ([319]). Nothing below questions the decision to publish or the refusal of an injunction. My difficulty, expressed with respect, is with part of the reasoning by which the court arrived at that outcome. Nicklin J himself gave permission to appeal, but time expired on 9 November 2026 with no appellant’s notice filed.
The decision
PPT sought a final injunction by a Part 8 claim, tried on an expedited basis ([1], [9]), and deliberately did not rely on a cause of action ([302]). He did not say that the allegations were private, and the court was not asked whether they were true ([33], [293]). His case was that publication would put his life at real and immediate risk, and that Articles 2 and 3 obliged the court to prevent it.
The judge found, on unchallenged expert evidence, that PPT’s life would be put at “a real and immediate risk” by publication, by the anticipation of it and by an adverse judgment itself, a risk which treatment in hospital would reduce but not remove ([159], [166]).
The claim nevertheless failed, for three reasons. First, the court owed no operational duty, because the State had assumed responsibility for the risk through its psychiatric services and a duty on the court to restrain private publishers would be a new principle of Convention law ([246]-[258]). Article 3 failed for much the same reasons ([259]-[263]). Second, without a cause of action there was no juridical basis for an injunction ([264]-[271]). Third, treatment and containment in hospital were in any event the reasonable response ([272]-[279]).
Whose duty?
The first reason treats the operational duty as owed institution by institution. The State had assumed responsibility for PPT’s safety through its psychiatric services ([247]) and the court had not ([249]), so that whether the court “has assumed, or otherwise bears” responsibility became an “anterior” question ([251]), and the answer was no.
With respect, I do not think that the authorities go so far. Morahan and Maguire, on which the judge relies ([188]-[189]), are about which risks the duty covers: it is “targeted on the specific risk of which the authorities were aware” (Maguire v HM Coroner [2023] UKSC 20, at [189]). They do not decide which arm of the State must act once the State has assumed responsibility for the very risk in question.
The Supreme Court has spoken of the duty arising where the risk is known to “the High Court as a public authority” (Re T (A Child) [2021] UKSC 35, at [175]). In D & F v Persons Unknown [2021] EWHC 157 (QB), Tipples J held that, where identifying a mentally disordered person in the State’s care would put that person at real and immediate risk of suicide or serious self-harm, the court is “under a duty to take all reasonable steps to exercise the Venables jurisdiction” (at [69]). Nicklin J distinguishes those cases as ones in which the State had already assumed responsibility for the person at risk ([198], [250]). So it had here, however, and the court had assumed no more responsibility for the two prisoners in D & F than it had for PPT.
The judge’s concern, that a grave enough risk would otherwise generate a duty whenever a public authority had power to remove its trigger ([254]), is a serious one. I wonder, though, whether the limits were not already to hand. On the judge’s own account of the law, a duty to prevent suicide arises only in particular circumstances, of which detention is the paradigm ([182]-[187]), and it extends only to what is reasonable. What the “anterior” question adds is the proposition that a court may give a judgment which, on its own findings, will itself put a life at real and immediate risk ([166]), and yet treat Article 2 as the concern of another institution. Chamberlain J, in observations made without full argument, put it differently and I think better: “the duty is owed by the state”, and what may be unclear is which of its organs should take the lead in discharging it (Attorney General v BBC [2022] EWHC 826 (QB), at [44(d)]). On that view, the hospital’s role goes to what can reasonably be asked of the court, and not to whether the court owes anything at all.
A new principle?
The judge’s further step is that no Strasbourg authority recognises such an obligation, so that to recognise one would create “a new category of positive duty”, contrary to Ullah and AB ([257]). Much depends on the level of generality. As a duty on courts to restrain lawful publication, it would certainly be new. But as a duty on the State to take reasonable steps within its powers to protect a person whom it has detained from a known risk of suicide, it dates back to Keenan v United Kingdom [2001] 33 EHRR 913. AB itself says that, where Strasbourg has not spoken, domestic courts “can and should aim to anticipate” how it would decide (R (AB) v Secretary of State for Justice [2021] UKSC 28, at [59]).
The nearest Strasbourg case to PPT may be Mikayil Mammadov v Azerbaijan, (2014) 58 EHRR 18 which the judgment does not cite. A woman took her own life when officials and police came to evict her family. The Court held that State agents who learn in time of a threat of suicide, made in reaction to what they themselves are doing, must prevent it “by any means which are reasonable and feasible” (at [115]). The means it had in mind were protective ones, not the abandonment of the eviction (at [116]), and on the evidence no substantive violation was established. The duty was recognised all the same, and the question was what it reasonably required.
I think that Strasbourg would approach PPT in that way, with Article 10 and the margin of appreciation doing much of the work, and would probably find no violation. That is, in substance, the route which Nicklin J took, but only in the alternative. This matters because if the next claimant’s risk cannot be contained by treatment, the alternative reasoning lets the court look at the facts, while the primary reasoning requires the claim to fail however certain the outcome.
Venables after Abbasi
The second reason PPT’s claim failed was that, without a cause of action, an injunction had no proper juridical basis even if a duty existed ([264]-[271]). Venables itself was founded on confidence ([200]), and RXG v Ministry of Justice [2019] EWHC 2026 (QB), so far as it may be understood as treating the equitable power and section 6 of the Human Rights Act as enough, “might require reconsideration” after Abbasi v Newcastle upon Tyne Hospitals [2025] UKSC 15 ([215]).
There is force in this. Section 6 confers no power which the court does not otherwise have (Abbasi, at [85]), and a claimant with an effective cause of action must use it (at [99]). But Abbasi was an Article 8 case, and does not mention the Venables line. It also accepts that a court of equity has power in principle to grant an injunction “when its failure to do so would be incompatible with Convention rights” (at [86]), and reserves cases in which protection would otherwise be theoretical and illusory (at [98]). Nicklin J would require “an identified equitable principle” first ([270]). If Article 2 did oblige the court to act, however, a refusal for want of a cause of action would itself be incompatible with the Convention, and that is the very case for which Abbasi keeps equity available as an option. If it did not, the question does not arise. The second reason seems to me to add little to the first.
The other first-instance cases
As Olivia O’Kane has recently pointed out, the question in PPT had come before the High Court in Northern Ireland thirteen years ago, and had then been answered the other way. In ZY v Higgins [2013] NIQB 8, in which she acted for the defendant court reporter, a prisoner who had pleaded guilty to sexual offences obtained an injunction against being named, because publicity would put him at risk of suicide. McCloskey J upheld his claim that the Crown Court, in lifting his anonymity, had acted incompatibly with Article 2, and restrained the reporter by treating his identity as confidential (at [8], [27], [29]).
In her comparison of the two decisions, Ms O’Kane observes that they differ on their facts (ZY concerned the naming of a man convicted in open court) and that in ZY the right to life prevailed as soon as the threshold of risk was crossed; in PPT, she argues, crossing it was only one stage of the inquiry. She finds it “hard to see how the ZY approach would survive a challenge in England today” and, there having been no appeal, regards it as an open question whether the courts of Northern Ireland will follow PPT.
I would add two points of my own. Ms O’Kane describes as novel McCloskey J’s use of confidence, which he framed by reference to Article 2 itself (at [29]). I would go further: it shows how little work a cause of action does once a life is at stake. She also reports that criminal defendants have since relied on ZY in hundreds of applications. McCloskey J had foreseen “no litigation bandwagon” (at [30]), and the contrast seems to me to lend support to Nicklin J’s concern about an unconfined duty. The answer, though, may be his own alternative reasoning, which limits the duty to what is reasonable, rather than the denial of any duty.
The English cases are divided too. In RXG, a Divisional Court said that a real and immediate risk cannot be balanced against Article 10, “no matter how weighty” (at [35(vii)]). Sir Geoffrey Vos C (as he then was) and Sir Andrew McFarlane P preferred a balance, the President leaving the conflict for “another case on another day” (Re Al M [2020] EWHC 702 (Fam), at [11]-[19]). PPT holds, in the alternative, that Article 10 cannot balance an established duty away, but bears on what the duty “reasonably requires” ([278]). I think that is right, but the judgment does not say where it leaves the earlier cases.
Comment
None of this means that the result was wrong; in fact, I think it was right. An indefinite injunction, granted to the subject of public-interest journalism to stop its publication, would have been a remarkable order, and the alternative reasons for refusing one ([272]-[279]) are persuasive. My difficulty is with the route. A case which could have been decided on the ground that the State was already doing what was reasonable was instead decided first on the ground that the court owed the claimant no duty at all, and then on the ground that it could not have acted if it did.
It would have been better for the law if a higher court had been asked. It would not have been better for the people concerned, and they come first. The evidence was that delay was itself harming the claimant, and that publication would be safer while he remained in hospital ([145], [276]). Those who had come forward, some to describe events more than twenty years old, had already waited months for publication, and the judge accepted that an injunction would interfere with their own right to tell their stories ([315], [323]). Nobody could have asked either to wait until the law was tidied up.
The case for an authoritative answer is nonetheless strong, and Nicklin J plainly contemplated one, finding the facts needed for the case to be considered “upon any legal basis” ([36]). If a court owes no duty whatever the evidence, such a claim can be struck out and dismissed in short order, without a trial of the psychiatric evidence and without months of waiting on either side; if everything depends on what is reasonable, it cannot. Those next in PPT’s position, and in the position of these women, are entitled to know which it is.
As Ms O’Kane points out, it remains to be seen whether the courts of Northern Ireland will follow PPT. If they do not, the same Convention right under the same Act will mean different things in two parts of the United Kingdom, and only the Supreme Court can put that right. Until it does, a judge with a few days in which to decide whether a publication must be stopped to protect a life will have two careful first-instance judgments which pull in different directions, and nothing from any court whose word would be final.
Harry Camp is a trainee barrister at Matrix Chambers


Leave a Reply