The kind of reasoning that is the source of everything that has gone wrong with vicarious liability over the last 25 years is found in paragraph 1 of Lord Burrows judgment for the court. It (probably, still) reflects academic orthodoxy and is wrong. It states:

 

“Vicarious liability in tort is an unusual form of liability. This is because the vicariously liable defendant is held liable (and treated as a joint tortfeasor) not because it has itself committed a tort against the claimant but because a third party has committed a tort against the claimant. Vicarious liability has often been treated as imposing strict liability because it is not dependent on proving the fault of the defendant. But it differs from strict liability torts. They impose personal liability on a defendant irrespective of fault whereas vicarious liability is precisely not a personal liability. Vicarious liability therefore does not rest on the defendant having owed a duty, whether strict or of reasonable care, to the claimant. It was the third party (who I shall refer to as the tortfeasor) who owed that duty to the claimant.”

 

 

If this were true, it would be possible for an employer to be held liable for the breach of a duty that only his employee owes. If an employee owed a statutory or contractual duty, but not his employer there could be vicarious liability.

 

If this were true, if an employer owed a statutory duty that employees did not owe, an employer could not be vicarious liable for acts that are not wrongs by the employees.

 

If this were true, then where an employer is vicariously liable for a serious wrong, jointly with others, for purposes of contribution they wouldn’t be treated as the serious wrongdoer.

 

If this were true, for purposes of contributory fault the employer would be treated as an innocent person who has committed no wrong.

 

If this were true, if the employer were a public body such as the Metropolitan Police held vicariously liable for some grotesque wrong by a police officer, no punitive damages would be available because the employer would have committed no wrong.

 

None of those above propositions is correct as a matter of law. The employer isn’t just “treated” as a joint tortfeasor. They *are* a joint tortfeasor. The actions of the employee are attributed to the employer so that each are liable for the same wrong where they both owe the same duty not to injure another.

 

Which is why the old Salmond test made sense. If "vicarious liability” (sic) is an outgrowth of, or the outer margins of, agency law then the correct test is whether the conduct is an "unauthorised mode of an authorised act."

 

Unfortunately, that perfectly sensible test that we'd had for over a century and that is still the law in many other jurisdictions was abandoned in Lister v Hesley Hall, following the Canadian Supreme Court in Bazley v Curry (which Lord Burrows rightly identifies as the source of the chaos of the last 25 years see [30] onwards).

 

But that case (and the other cases concerning child abuse at schools) should have been nothing to do with vicarious liability.

 

If I bail my car to you, you owe me a non-delegable duty that care is taken of it because of your assumption of responsibility to me. If a stranger destroys it, when care would have prevented that, you're liable. That isn't vicarious liability.

 

Similarly, a school assumes responsibility for the children in its care. If a stranger abuses them, when care would have ensured that they weren't, the school is liable. You don't need to expand vicarious liability to explain that, anymore than you need vicarious liability to explain the liability for breach of strict contractual duties to achieve results. They’re duties that care will be taken, not duties of care.

 

The court in Lister could intuit the right result, but because of the way the case was argued thought they needed to abandon the Salmond test and instead we had the "close connection" test (which on its face is empty) and an attempt to explain vicarious liability according to a basket of policy concerns. Egged on by academics.

 

These attempts to invoke policies such as enterprise risk, deep pockets, deterrence etc are a dead end in my view. None of them explain the central feature of vicarious liability: that it is vicarious, dependent upon the wrong of another. It isn’t possible, in my opinion, to come up with any intellectually satisfying further explanation for that central feature of vicarious liability if you go down this path (and the courts haven’t even tried). It is that path that the UKHL went down in Lister that Lord Burrows is still traipsing down in this judgment.

 

But the good judges can still intuit the right answer. They can see what in Jehovah's Witnesses v BXB the right result is. There was no non-delegable duty. This wasn’t an unauthorised mode of an authorised action. The result is right, no liability. But we get what Burrows tries to do here, an attempt to firm up and narrow the "close connection" test a bit. But we’re still left with an area of law that on its face makes no sense. Why this test and not another one? Which is why we keep on getting appeals on it, because it can’t be rationalised in the way the courts are attempting.

 

Also notice that the current enthusiasm for “enterprise risk” ([47]) face a further insuperable hurdle. It would require that the employment caused the harm in a “but for” sense. But this has never been the law, the employer cannot escape liability for the wrongs committed by their employee in the course of their employment by saying “they would have committed the same wrong even if I had never employed them.” Causation is irrelevant, just as it is in agency law more generally.

 

In a sense, the old law is still there, ticking away in the background, shaping the outcome of the cases. It is just the outward form that no longer makes sense.

 

 

 

Dear Colleagues,

 

Today has seen judgment in the latest UK Supreme Court decision on vicarious liability, Trustees of the Barry Congregation of Jehovah's Witnesses v BXB [2023] UKSC 15 https://www.supremecourt.uk/cases/docs/uksc-2021-0089-judgment.pdf. Lord Burrows gives the judgment for the Court, and cites several ODGers (congratulations all!).

 

The case concerned a rape committed against the claimant by a man, when they were both members of a congregation of Jehovah’s Witnesses. The wrongdoer was a Mark Sewell, who had congregational responsibilities first as a ministerial servant and then as an elder. He and his wife became friends through the congregation with Mr and Mrs BXB. Having grown concerned about Sewell’s behaviour, Mrs B suggested that she and Sewell’s wife speak to another elder, who was Sewell’s father Tony. Tony asked that they both support Mark. Sewell continued to pursue Mrs BXB romantically, and she rejected him. Sewell perpetrated a rape against Mrs BXB at his home. Mrs B had initially reported the rape to the Congregation, who found the allegations unproven. She later reported the crime to the police and he was subsequently convicted.

 

Mrs BXB suffered psychiatric harm following the trauma of the rape, and later sued the Jehovah’s Witnesses’ organisation(s) alleging that they were vicarious liable for the rape (as well as alleging negligence on the part of the other elders in failing to investigate). (I say organistions(s) because there was a threshold point during the hearing about the correct defendant but it was not ultimately determinative – see [59]-[64]).

 

The appeal to the UKSC required consideration of stages 1 and 2 of vicarious liability (perhaps “stage zero” being that a tort has been committed by a tortfeasor): whether there was a relevant relationship between Sewell and the Congregation capable of giving rise to vicarious liability as being akin to employment; and then whether the close connection between that relationship and the tort was satisfied. Lord Burrows holds that stage 1 is satisfied, but that stage 2 is not. The claim therefore fails.

 

Lord Burrows traces over two decades’ worth of primarily House of Lords and UK Supreme Court authority on vicarious liability, including the expansion and contraction of liability with which members of this list will be familiar. His Lordship provides a summary of the present position at para [58] (which is too long to quote here, but is a reaffirmation of the restatement of the law in Barclays and Morrisons from 2020).

 

His Lordship states that,

 

“[58](v) The same two stages, and the same two tests, apply to cases of sexual abuse as they do to other cases on vicarious liability. Although one can reasonably interpret some judicial comments as supporting special rules for sexual abuse, this was rejected by Lord Reed in Cox. The idea that the law still needs tailoring to deal with sexual abuse cases is misleading. The necessary tailoring is already reflected in, and embraced by, the modern tests.”

 

(One might interject that one set of judicial comments that had perhaps given the impression that there was room to argue about whether such cases were different were from Lord Reed himself in Morrisons [2020] UKSC 12 at [23] and [36])

 

On stage 1, Lord Burrows analyses the nature of the role of an elder within the organisation. Sewell had duties assigned to him, pursued those duties in furtherance of the aims of the organisation, was appointed (and later removed) through prescribed process, and he fitted within an institutional hierarchy.

 

Lord Burrows agrees with most of the reasoning of the courts below on this point, but says “it was a mistake for them to drift into talking about creating the risk of rape by the elder being assigned the activities he was given. In relying for this on Lord Reed’s judgment in Cox at para 30, they were incorrectly confusing the criteria for satisfying the first stage test with the underlying policy justification for vicarious liability. The creation of the risk of rape should not have been included within the criteria for deciding whether the relationship was akin to employment.” (at [69]). This issue had caused some confusion at the hearing, and at least some of the submissions on the part of the claimant did indeed seem to conflate the general idea of risk creation as a policy justification with a specific judgement about responsibility for risk on a particular set of facts (which might shade more towards a fault-based approach).

 

On stage 2, Lord Burrows disagrees with the courts below, holding that they applied the wrong test and took into account irrelevant aspects of the facts. There was no close connection between the relationship and the rape. This was, finds his Lordship, because the rape was not committed while Sewell was carrying out activities in his role as an elder. “The lack of direct connection to the role assigned to him as an elder makes these facts significantly different from the institutional sex abuse cases” (at [74]). Lord Burrows gives five more reasons, questioning whether on these facts there were material similarities with cases in which vicarious liability had been found in sexual abuse cases.

 

[81] “In my view, therefore, the close connection test is not satisfied. The rape was not so closely connected with acts that Mark Sewell was authorised to do that it can fairly and properly be regarded as committed by him while acting in the course of his quasi-employment as an elder.”

 

Lord Burrows also judges (at [82]) that this is justified (as “a final check”) by underlying policy concerns.

 

Overall, from an initial reading, it seems to me that Lord Burrows’s judgment affirms the modern approach on stage 2, but the exact limits of the close connection test at stage 3 remain somewhat elusive, beyond it being clear that the Supreme Court feels that there should be some point at which liability should stop. There was difficulty drawing the line in argument before the Court, Lord Burrows notes

 

76. “Thirdly, James Counsell KC submitted that Mark Sewell never took off his “metaphorical uniform” as an elder. It was put to him by the court that that would mean that there would be vicarious liability even if he committed the tort of negligence, injuring a customer, while carrying on his cleaning business. He accepted that that would not be so and qualified his submission by saying that the metaphorical uniform was never taken off in his dealings with members of Barry Congregation such as Mrs B. But that is also an unrealistic submission. It cannot seriously be suggested that there would be vicarious liability if, for example, Mark Sewell was driving Mr and Mrs B and their children in his own car to the airport for their holidays and Mrs B was injured in an accident caused by his negligent driving. In my view, Mark Sewell was not wearing his metaphorical uniform as an elder at the time the tort was committed.”

 

However, one may respectfully question whether the cleaning business or airport driving examples are closer to the facts here, when compared to cases which have established liability such as Mohamud, arguably. It is also far from clear that previous (but still recent) more expansive authorities at Court of Appeal level in clergy sexual abuse cases would be decided in the same way now. On the other hand, the more restrictive approach seen in cases involving football clubs are likely in line with the latest UKSC position (and indeed permission to appeal was refused by the UKSC last August in one such case Blackpool Football Club v DSN [2021] EWCA Civ 1532)

 

Best wishes,

James

 

--

James Lee

Professor of English Law

The Dickson Poon School of Law

Somerset House East Wing, room SW1.12
King's College London
Strand
London WC2R 2LS

 

E-mail: james.lee@kcl.ac.uk

  

Profile: https://www.kcl.ac.uk/people/james-lee

 

“Not Time to Make a Change? Reviewing the Rhetoric of Law Reform” Current Legal Problems Lecture Recording: https://youtu.be/XM9ri1dYM98