From: Christine
Beuermann <Christine.Beuermann@newcastle.ac.uk>
Sent: Friday 12
December 2025 09:39
To: Robert
Stevens; Peter Watts; Lionel Smith; ODG
Subject: Re: X v
Lord Advocate [2025] UKSC 44
This is an extract taken from the appealed judgment:
The second alleged incident
[4] The pursuer avers that on 5 July 2018 when she was at
court in the course of her work the first defender communicated via his bar
officer that he wished to see her in his chambers. The bar officer led her
there. On their arrival the first defender told the bar officer to leave. He
hugged the pursuer without her consent. He used inappropriate phrases such as
"your pretty face". She considered she was unable to leave given the
first defender's status as a sheriff and because she was in a secure part of the
building.
Without engaging in the servants tort/masters tort quandary, I
don't think occupiers is the appropriate relationship for determining an
'assumption of responsibility' or any other basis for strict liability.
Jason Allen mentioned police is his response to this thread on the position of
officers. The conferral of authority on police and judges by the state
and the potential for abuse of that authority would provide a defendable basis
for the imposition of strict liability on the state without having to engage in
questions of employment.
This is an extract taken from the appealed judgment:
⚠ External sender. Take care when opening links or
attachments. Do not provide your login details.
“Plainly
if the principal procured the tort the principal will be liable, but that
is true of anyone who procures a tort (i.e. irrespective of an agency
relationship).”
I hesitate to
disagree with Peter about agency law, but I do not agree with this short
statement. “Procuring” requires “inciting, inducing or persuading”. By contrast
if P authorises A to do an action on his behalf, and A does that action on P’s
behalf, if that action is a tort against X, P has committed a tort against X
jointly with A. There is no need to satisfy the narrower test of procurement.
Although rare, you can also have situations where P ratifies the unauthorised
actions A took on P’s behalf, which will also render P a joint tortfeasor. The
ratification cases are more obviously not ones of “procurement”, time’s arrow
only going one way.
As Jason says,
on “vicarious liability” more generally, I remain unrepentantly of the view
that it was explained, and can still only be explained, on the basis of the
“Master’s Tort” theory, and that the basket of policy concerns that the UKSC
(and others) invoke to try and explain it as a species of “strict liability”
are (politely) unconvincing.
Under the
“Master’s Tort” explanation, the employee’s action is attributed to the
employer, so that if that action constitutes a tort, each is committing the
same tort jointly (ie “vicarious liability” is a species of agency law). Under
the “servant’s tort” theory, the master isn’t a tortfeasor at all, instead
we’re holding them strictly liable for the torts of someone else for an array
of policy reasons (loss spreading, enterprise risk, deterrence etc etc).
The master’s
tort theory, unlike the servant’s tort theory, explains the central features of
the doctrine. For example
- Why it is vicarious. None of the policy explanations, either
singly or in combination with any of the others, explains why liability is
dependent upon a tort committed by someone else
- Why there is no causal requirement. The employer’s liability
is not dependent upon any causal connection between the employment and the
tort. Indeed, liability cannot be avoided by showing beyond peradventure
that the employee would have committed the same tort regardless of having
been employed. Exactly the same is true of express authorisation. If I
authorise you to punch X, and you do so on my behalf, I am a tortfeasor
even if you would have punched X yourself anyway.
Once we accept the
Master’s Tort theory must be the truth (and there are several other rules we
all accept, such as how contributory fault works, that I consider to be
proofs) the old Salmond test of scope of employment (“an unauthorised mode of
an authorised act”) made perfect sense. This is an extension of, and part of
the same rule as, express authorisation.
Partnership works in the
same way, and the rules there also can’t be explained using the basket of
policy concerns that are invoked to explain vicarious liability.
Partnership isn’t a species of employment, nor is it akin to employment. An
employee is someone the employer has control over. One partner has no control
over another partner.
But, partners agree to act
together. The acts of one partner within the scope of the partnership agreement
are attributed to all other partners. If the acts of one is a tort, each
jointly commits that tort. Agreement, like authorisation, is a way in which we
attribute one person’s actions to another. It is loose to say that this is an
intimate relationship. I am in an intimate relationship with my children but,
thank God, that doesn’t mean under English law that their actions are
attributed to me.
And this account also
explains the Crown Proceedings Act, which the UKSC cannot do. The Act says
“Subject to the provisions of this Act, the Crown shall be
subject to all those liabilities in tort to which, if it were a private person
of full age and capacity, it would be subject:—
(a)in respect of torts
committed by its servants or agents;
….
Provided that no
proceedings shall lie against the Crown by virtue of paragraph (a) of this
subsection in respect of any act or omission of a servant or agent of the Crown
unless the act or omission would apart from the provisions of this Act have
given rise to a cause of action in tort against that servant or agent or his
estate.”
On the “servant’s
tort” theory, that proviso makes no sense (as the UKSC admits). Under the
servant’s tort theory, the employer can only be liable if the employee is
liable, so there is no need for any proviso. But as Glanville Williams
explained, sometimes the employee may not be liable because they have an
immunity that is personal to them but the employer can still be liable. An
example is the immunity a husband had for torts to his wife. If the husband was
acting in the course of his employment in injuring his wife, his employer was
still liable because the *acts* of the husband are attributable to his
employer: Broom v Morgan [1953] 1 QB 597. Under the “modern” view that is of
course inexplicable, but much more central features of the doctrine than that
are also inexplicable.
This became very
important when Crown immunity was abolished in 1947, because many agents of the
Crown (eg judges) themselves have personal immunities from committing various
torts (eg defamation when giving a court judgment). If the proviso did not exist,
these immunities could be bypassed by suing the Crown.
Notice that the
entire section makes no sense on the “servant’s tort” theory. On that theory
the liability of the employer/Crown isn’t for a tort at all. Compare the
opening words of the section (“in tort”).
Is the case wrong
under the law as it was understood in 1947 (which made perfect sense and has
been mucked up since)? Indeed, how should it have been decided today?
First, judges aren’t
just acting for themselves, they’re acting for the State. If they commit a tort
in the course of acting for the State, the State should be liable. It was that
immunity for the Crown that the 1947 Act was intended to remove.
But, second, for
those acts that judges carry out on behalf of the State (eg judges giving
judgments) they themselves always(?) have an immunity. So, you can’t sue the
Crown either.
Here the alleged
assault that was carried out was not an act done as a judge. It wasn’t an
“unauthorised mode of an authorised act” so you can’t sue the Crown for that
either.
But, two of the three
alleged assaults occurred in the court building. Under the law as it was in
1947 (in England I don’t know about Scotland) occupiers who invited people on
to premises assumed a duty that care would be taken of them. (ie there was a non-delegable
duty to invitees). So, the Crown should have been liable in 1947 after the
passing of the Crown Proceedings Act like any other occupier who has invited
another on to the premises. It shouldn’t matter a damn that the assault was
deliberate, that is a fortiori from their having been carelessly injured (cf
the HCA’s appalling decision in NSW v Lepore, which is currently being sought
to be reversed).
But, alas, in 1957
the Occupiers Liability Act was passed, and that legislation foolishly
abolished the distinction between invitees (to whom responsibility was assumed)
and licensees (to whom responsibility was not assumed and who had only their
vanilla Donoghue v Stevenson rights). I assume the Occupiers Liability
Act (Scotland) Act 1960 leads to the same position.
So, after 1957 that
claim doesn’t work. It should. The Crown ought to be liable for the assaults of
people to whom it has assumed responsibility by inviting them into court rooms.
RS
.
I have no contribution to make at this point on
the possible vicarious liability of the Crown for acts of judges and other
state officials. But in response to Lionel’s assumption that vicarious
liability applies to the principals of agents, I think this is wrong. Agency is
not a status for the purpose of the operation of vicarious liability (the car
cases are anomalous and mis-reasoned). An agent is either an employee or
an independent contractor (putting aside gratuitous actors). With most torts (but
cf negligent misstatement) a principal is vicariously liable only for the acts
of employees or those akin thereto. The principal is not so liable for the acts
of agents who are independent contractors. Plainly if the principal procured
the tort the principal will be liable, but that is true of anyone who
procures a tort (i.e. irrespective of an agency relationship).
As for the vicarious liability of partners, I
think that this turns not on agency but on the fact that partners, as co-owners
of the business, are in an even more intimate relationship than employment. It
might be thought odd to be vicariously liable for the acts of employees but not
those of co-managers of the business. Peter.
Thank you Haim, Jason, and Neil.
I am no tort lawyer, nor a legal historian,
but this might be interesting for those who are. I was recently looking into
C19 decisions on public trusts whose trustees were performing functions of
government but which were not charitable trusts inasmuch as the property held
in trust was not given, but rather raised under statutory authority. The
majority view, as I understand it, following the resolution of some earlier
debates by the decision in Attorney-General v Dublin (City) (1827)
I Bligh NS 312, 4 ER 888 (HL), was that these were public non-charitable trusts
(although by the turn of the C20 some judges were already starting to use the
expressions ‘public trust’ and ‘charitable trust’ as synonymous, as they had at
the start of the C19 and perhaps before). (For those who find all this
surprising, as I did, see: J Barratt, ‘Public Trusts’ (2006) 69 MLR 514; C
Mitchell, ‘Public Trusts, 1750-1850’ in D Foster and C Mitchell (eds), Essays on the History of
Equity (Hart 2026) forthcoming but available at https://ssrn.com/abstract=4495395;
and, for a recent example of a public non-charitable trust, R (on the
application of Day) v Shropshire Council [2023] UKSC 8.)
In relation to these public trusts performing
functions of government, some cases held that the trustees were immune from
vicarious liability, but this seems to have been rejected, with contrary cases
overruled, in Mersey Docks Trustees v Gibbs (1866) LR 1 HL 93. When
I read that case, I assumed that the earlier idea that there should be no tort
liability for such trustees (who were always acting under statutory authority)
was a kind of read-over from Crown immunity. But if that was right, the
immunity described by Jason in his point (3) should not have survived the Crown
liability legislation. Indeed, with the greatest respect for a great judge and
not having read the decision, I find the passage quoted by Jason from Dixon J
hard to follow. In every case, there will surely be someone ‘between’ the
tortfeasor and the Crown who was exercising a ‘discretion and responsibility
which rests upon him in virtue of his office or of some designation under the
law’, even if it is a Minister of the Crown. Is the doctrine that Jason
describes only applicable when the office-holder him- or herself commits the
tort (and if instead it is their subordinate, the Crown can be liable)? That
would be a bit strange.
As for Neil’s point, surely an agency
relationship attracts vicarious liability independently of an employment
relationship, and partners are mutual agents in relation to the business of the
partnership?
Cheers,
Lionel
It is an interesting decision.
Three observations:
- I have some sympathy for
the plaintiff’s argument that servant of the Crown might have a different
and wider meaning than the modern law of vicarious liability stage 1
(especially in light of the independent discretion doctrine that had
developed in response to this liability, more on this in 3).
- Discussion of the
Proviso to Section 2(1)(a) at [55]: The court can’t seem to find any
reason for the proviso, but it seems fairly obvious to me that it was
inserted because the juridical foundation of vicarious liability (was it
the attribution of torts or acts, ie servant’s tort vs master’s tort
theory?) was still at the time (and might still, see Stevens, be) up for
grabs.
- I found it interesting
that there was no discussion of the (very) traditional independent
discretion doctrine in relation to vicarious liability and the Crown:
As Dixon J. noted in Field v. Nott (1939), 62 C.L.R. 660
(H.C.A.) at 675:
When a public officer, although a servant of the
Crown, is executing an independent duty which the law casts on him, the Crown
is not liable for the wrongful acts he may commit in the course of his
execution. As the law charges him with a discretion and responsibility which
rests upon him in virtue of his office or of some designation under the law, he
alone is liable for any breach of duty.
The rule had been applied in many cases, Little
v. The Commonwealth (1947), 75 C.L.R. 94 (H.C.A.); Fowles v. Eastern and
Australian Steamship Co., [1916] 2 A.C. 556 (P.C.); Esso Petroleum Co.
v. Hall Russell & Co., [1989] 1 A.C. 643 (H.L.), see Kneebone, “The
Independent Discretionary Function Principle and Public Officers” (1990) 16
Monash Univ. L. Rev. 185. Perhaps there is some UK-specific reason (such
as a statutory provision) why it no longer applies, or perhaps because the
court concluded that Judges are never Crown servants it was thought totally
irrelevant?

Jason Neyers
Professor
of Law
Faculty of Law
Western University
Law Building Rm 26
e. jneyers@uwo.ca
t. 519.661.2111 (x88435)
Dear all,
I hope this email finds you well.
The UK Supreme Court handed down a decision
in X v Lord Advocate [2025] UKSC 44, in which it provides a detailed
analysis of the vicarious liability of the Crown for alleged battery and
(sexual) harassment committed by a Scottish judge against a solicitor.
The key paragraphs of Lord Reed and Lord
Burrow’s judgment are, in my view, 41-63, and I include them below.
Both the interpretation of ‘the Crown’ for
the purposes of the Crown Proceeding Act, and approach Stage 1 of the vicarious
liability analysis, are narrow.
The ‘five incidents’ test hasn’t been clearly
applied fully, and instead the focus was on control and separation of powers,
resulting in a decision that Stage 1 couldn’t be established in cases such as
this.
In my view, other factors in this case could
lead to the opposite conclusion, particularly the source of judicial powers,
the provision of means to exercise these powers, and the risks created on and
off premises given the nature of the dynamics.
Some of the key paragraphs:
41. That description of the Crown, and the
characterisation of it as a corporation sole, indicate that in some contexts,
at least, the Crown simply means the Sovereign in his official capacity…
42. ... The effect of that provision was
that, unless a contrary intention appeared (and there is none), references to
the Crown in the 1947 Act were to be construed as referring to the Sovereign
for the time being…
43. However, if it is correct that the Crown
is referring to the Sovereign acting in his official rather than personal
capacity, one still has to recognise that the Sovereign’s official capacities
are wide-ranging. It is at this point in the analysis that the parameters of
the 1947 Act, as laid down in section 40(2)(b), are crucial. Section 40(2)(b)
lays down that the liability of the Crown must arise in respect of His
Majesty’s Government in the United Kingdom or the Scottish Administration. This
is of critical importance because the Crown possesses functions and powers
which do not concern His Majesty’s Government in the United Kingdom or the
Scottish Administration.
44. Some powers and functions of the Crown
fall outside those categories for territorial reasons. So, for example, the
Crown could not be sued under the 1947 Act in respect of a nuisance allegedly
arising out of the use of a military base in West Germany: Trawnik v Lennox
[1985] 1 WLR 532. Other illustrations include R v Secretary of State for the
Home Department, Ex p Bhurosah [1968] 1 QB 266, concerned with the Crown in
right of Mauritius, Franklin v Attorney General [1974] QB 185 and Mutasa v
Attorney General [1980] QB 114, concerned with Southern Rhodesia, R v Secretary
of State for Foreign and Commonwealth Affairs, Ex p Indian Association of
Alberta [1982] QB 892, where the relevant liabilities were held to be in
respect of the Crown in right of Canada, and R (Quark Fishing Ltd) v Secretary
of State for Foreign and Commonwealth Affairs [2005] UKHL 57; [2006] 1 AC 529,
where it was held that the Secretary of State had acted on behalf of the Queen
in right of South Georgia and the South Sandwich Islands. This territorial
limitation was even more important in 1947, when the Sovereign was the head of
an extensive empire owing a single allegiance.
45. Other powers and functions of the Crown
fall outside those categories because, although they relate to the United
Kingdom, they are not of an executive character (and hence do not arise “in
respect of His Majesty’s Government in the United Kingdom or the Scottish
Administration”). An example is the Crown’s function as a legislator: every Act
of Parliament states that it was “enacted by the King’s most Excellent Majesty,
by and with the advice and consent of the Lords Spiritual and temporal, and
Commons, in this present Parliament assembled”. Constitutionally, the Crown’s
commission is also the source of all judicial authority, although the monarch
cannot personally execute any judicial office: Prohibitions del Roy (1607) 12
Co Rep 63; 77 ER 1342. Section 40(1) and section 40(2)(b) reflect the
constitutional role of the judiciary in relation to the monarch in his or her
private capacity and as the head of the executive.
…
48. Accordingly, in essential agreement with
the views of the lower courts on this issue – although it will be clear that we
have regarded it as important to explore this issue in much greater depth than
was possible in the lower courts – the vicarious liability imposed on the
Crown, that is the Sovereign in his official capacity, by section 2(1)(a) of
the 1947 Act is a liability arising in respect of the Government of the United
Kingdom or the Scottish Administration. As we have explained at para 18, the Scottish
Administration includes the Scottish Government but also includes other
office-holders. Vicarious liability can therefore arise under the 1947 Act in
respect of the acts or omissions of the Scottish Ministers (who form the
Scottish Government), junior ministers (who do not form part of the Scottish
Government, but are part of the Scottish Administration), and other
office-holders, together with the staff of all of the foregoing.
49. In the present proceedings, the focus
must be on the Scottish Administration, and more specifically on the Scottish
Government, as a constituent part of the Scottish Administration. That is
because it is the Scottish Government that has responsibility for the payment
of the salaries of judicial office-holders and, through the First Minister, has
responsibility for the appointment and removal of judicial office-holders. In
other words, as regards the relationship with judicial office-holders, the Sovereign
in his official capacity acts through the Scottish Government. We therefore
consider that it is the relationship between the Scottish Government and
judicial office-holders that needs to be focussed on in considering the
vicarious liability of the Crown in this case.
…
51. Returning to the primary submission
advanced on behalf of the pursuer, while not necessarily disagreeing with what
Mr McBrearty has argued on the meaning of “Crown servant”, we fundamentally
disagree with his suggested interpretation of section 2(1)(a) (see para 30
above). On the correct interpretation, section 2(1)(a) is not self-contained.
Rather the common law, and hence the modern development in the common law, of
vicarious liability is relevant and important.
…
53. Focusing on section 2(1)(a), therefore,
the words effect the purpose of imposing on the Crown the law on vicarious
liability in delict or tort. The purpose is not, as the pursuer submits, to
impose on the Crown only a limited part of that law and nor is it to deem
aspects of that law (ie the first stage of vicarious liability) as being
automatically satisfied where the wrongdoer is a Crown servant.
54. It follows from this correct
interpretation that all the law on vicarious liability, that would apply if the
Crown were a private person, applies to the Crown; and that includes the
development of the common law at stage 1 to embrace a relationship that is akin
to employment. That one should include that development is supported by the
principle of statutory interpretation that a statute is “always speaking” (see
para 23 above). In so far as the words permit it, an interpretation that
embraces developments in the common law is in general to be preferred to one
that excludes such a development by treating the interpretation of the statute
as frozen in time.
…
58. That alternative submission also fails.
There are two compelling and linked reasons why the relationship between a
sheriff and the Scottish Government is not akin to employment. 59. First, there
is no control by the Scottish Government over the performance by sheriffs of
their judicial functions. The judiciary itself determines listing matters (ie
where and when a sheriff sits and which cases he or she hears). At the extreme,
there is a statutory procedure, laid down in sections 21-25 of the Courts Reform
(Scotland) Act 2014, for the removal from office of a sheriff by the order of
the First Minister as being unfit for office. But that requires a report from a
specially convened tribunal recommending removal from office, and the order is
subject to the negative procedure before the Scottish Parliament.
60. Secondly, and most crucially, it is a
constitutional principle, resting on the separation of powers, that the
judiciary is independent of government. A sheriff must be free to decide a case
without any interference or the fear of interference by the Scottish Government
... Accordingly, the Scottish Government can tell a sheriff neither what to do
nor how to do it.
…
62 … Although the question whether a person
is a “worker” within the meaning of the Employment Rights Act 1996 is different
from the question whether the stage 1 test of vicarious liability is met, as
Lady Hale made clear in Various Claimants v Barclays Bank plc [2020] UKSC 13;
[2020] AC 973, para 29, the constitutional context to which Lady Hale referred
in Gilham v Ministry of Justice is important in both contexts.
Best,
Haim
Dr Haim Abraham
Assistant Professor
UCL Faculty of Laws, University College London
E: haim.abraham@ucl.ac.uk W: https://www.ucl.ac.uk/laws/people/haim-abraham Twitter:
@HaimAbraham
UCL Faculty of Laws | Bentham House| 4-8 Endsleigh Gardens |
London | WC1H 0EG
New Book: Tort Liability in Warfare: States' Wrongs and
Civilians' Rights (Oxford University Press, 2024)