From: Laura
Hoyano <laura.hoyano@law.ox.ac.uk>
Sent: Wednesday
21 January 2026 17:29
To: Matthew
Dyson; obligations@uwo.ca
Subject: Re: UKSC
in Lewis-Ranwell v G4S
Matt, I agree that coherence remains lacking
in the English approach, and is only re-enforced by this latest UKSC judgment.
Best wishes
Laura
Laura Hoyano
Emeritus Professor of Law, University of
Oxford
Emeritus Fellow, Wadham College, Oxford
Barrister, Red Lion Chambers
From: Matt Dyson <matthew.dyson@law.ox.ac.uk>
Sent: Wednesday, January 21, 2026 10:59:19 AM
To: obligations@uwo.ca <obligations@uwo.ca>
Subject: UKSC in Lewis-Ranwell v G4S
Dear All,
A new case has been handed down concerning a person, criminally insane at the time of the events, suing in tort in England and Wales those connected to why he was able to act.
https://supremecourt.uk/cases/uksc-2024-0043
The appeal succeeded: a person insane for the purposes of the criminal law who killed and brings a claim against those involved in why he was able to, can have his claim struck out on the basis of illegality. There was previously some uncertainty about the level of wrong that would found the illegality defence, which the court here called the "threshold question". Criminal offences outside those not requiring mens rea generally did, and very serious non-criminal wrongs could do, but not, e.g., trademark infringement (see Apotex). The differences between insanity and diminished responsibility were downplayed, and then the Patel test applied. For that test, the killing was the focus, rather than any importance on the police and others following their obligations to care of those with mental health problems. This reverses first instance and a 2:1 majority at the Court of Appeal. It is an expansion of the 'illegality' defence. In practice there are not many insanity cases, but conceptually this is quite a move. It is the first case in England establishing the applicability of the defence where the claimant was insane, cf. the multiple cases where the claimant had had the defence of diminished responsibility (cf. Hunter v Presland, and some other perhaps less direct comparisons in Canada and NZ).
Some Obligations discussion group members were involved, including James Goudkamp as junior counsel for one group of appellants and me, as an academic adviser for the respondents.
I won't add much commentary right now. Killing others is a serious tort, and tort law does not have much of an insanity reason to prevent/remove liability (Dunnage v Randall). But it is not clear that this rule will only apply to killing, rather the issues will be weighed in the Patel three factors test. The court is clear that there was no illegality for the purposes of criminal law, only tort law. The result is that when C commits a tort, even after and causally linked to D's tort, C's claim enters the Patel criterial and could be denied in its entirety.
I am in print as not finding persuasive much of the illegality rules in England, so it will not be a surprise to know I have not yet understood the merits of this expansion. With a criminal lawyer's hat on as well, it's hard to see how the significant difference to the criminal law of not guilty by reason of insanity cf diminished responsibility (which amounts to no crime even for third parties cf there is a crime but the label is lower and the punishment can be too) is given so little relevance to tort law if the coherence of the law is so important. If the matter is coherence within tort law alone, or coherence with an earlier order regardless of whether it is criminal, I still as yet miss some of the merits of this approach.
All best wishes,
Matt
--
Professor of Civil and Criminal Law
Faculty of Law, University of Oxford
matthew.dyson@law.ox.ac.uk
Global Professor of Law, The University of Notre Dame (USA) in England
Associate Member of 6KBW College Hill