THORMAN REVISITED

A paper based on the joint first prize entry in the

Hudson Prize essay competition 2014

Michael Regan

May 2015

194

www.scl.org.uk
 THORMAN REVISITED

Michael Regan

Introduction
A key feature of professional indemnity insurance is that it is written on a
claims made basis, allowing the insured an indemnity in respect of claims
made against it during the policy period.1 However, a problem which often
arises in a construction setting is whether there is one claim or more than one
claim being made against the insured. This is an important issue in practice
because it affects how many deductibles are payable by the insured, the extent
to which the primary layer insurers are on risk for the claim or claims which
are being made, and the extent, if at all, that any excess layers are exposed.

There is no single party to an insurance programme who invariably benefits

from a finding that there exists a single claim or multiple claims. Depending
upon the terms of the cover which is in place, multiple claims will involve an
insured being responsible for bearing more than one deductible, and perhaps
the entirety of the overall loss. Equally, a single claim may well be
advantageous to insurers when the exposure over the various layers of
insurance is taken into account.

Professional indemnity policies frequently provide for multiple claims to be

treated as a single claim through the operation of aggregation clauses. The
precise terms of aggregation clauses vary, but the usual formula is to link the
claims by a unifying factor, such as claims arising from the same ‘original
cause’, ‘event’ or ‘occurrence’. Therefore, if there is a common unifying
feature which has given rise to more than one claim, it can be treated as a
single claim for policy purposes.

Aggregation clauses serve a useful purpose on many occasions, but they tend
not to be particularly relevant in construction cases. This is because a single
project can give rise to a variety of unrelated claims which are not capable of
aggregation under the terms of the policy. Clearly, the fact that such claims
may arise from work carried out on the same project would not of itself
provide a basis to treat them as one under the usual aggregation wording
which is used in a professional indemnity policy. In any event, before there
can be any determination whether claims can be aggregated under the policy
in question, it first has to be decided whether there is in fact more than one
claim which is being made against the insured.

1 Sometimes the insured is also required to make a notification to its insurers during the
policy period.

1
The Thorman decision
There are surprisingly few construction cases which have considered the
number of claims which are being made, and the ‘default’ decision remains
that of the Court of Appeal in Thorman v New Hampshire.2 However, this is a
decision which gives rise to its own problems and perhaps leaves as many
questions unanswered as it seeks to address.

The facts of the Thorman case are relatively straightforward. The claimants
were architects who had been engaged in connection with a housing
development in Exeter, which was completed in 1977. Between August 1976
and July 1979, the owners made complaints relating to the brickwork. These
were minor in nature and the remedial works were paid for by the engineers,
with only a small contribution being made by the architects.

The architects thought that this was an end to the matter, but in May 1982 they
received a letter from the owners’ solicitors alleging problems which had
arisen in connection with the development and stating that they would have to
issue protective proceedings to preserve their client’s position under the
Limitation Act. The solicitors then apparently realised that there was an
arbitration clause in the contract between the architects and the owners.
Accordingly, they wrote again in June 1982, giving notice of arbitration and
stating that:
‘… serious problems have arisen in this development, inter alia, with
regard to cracking and defective brickwork, for which we hold you
responsible.’

A writ had originally been issued in June 1982 but was not served until
December 1983. A statement of claim was served in January 1984 and this
was followed in the course of the proceedings by a Scott Schedule. The
statement of claim and subsequent Scott Schedule revealed that allegations
were being made in relation to defects beyond the brickwork deficiencies.3

The architects had kept their insurers informed at all times. First, when the
original claim in relation to the brickwork had been made, and later when the
letters from the owners’ solicitors were received. Thereafter, insurers
engaged solicitors to represent the architects.

The problem which had to be considered by the court arose from the fact that
the architects’ professional indemnity cover had transferred from New
Hampshire Insurance Company to Home Insurance on 1st October 1983. New
Hampshire argued that it was only liable under its policy for the matters which
related to the defective brickwork, and not the other defects.

The Court of Appeal unanimously decided that all the claims which were
being pursued by the owners against the architects were covered by the New

2 Thorman v New Hampshire Insurance Co (UK) Ltd 39 BLR 41; also [1988] 1 Lloyd’s
Rep 7 (CA).
3 Clearly, the parties had decided to proceed by way of litigation in preference to
arbitration.

2
Hampshire policy. The leading judgment was given by Sir John Donaldson.
He referred to the correspondence which had been sent by the owners’
solicitors during the currency of the New Hampshire policy, and in particular
the use of the words ‘inter alia’ in the second letter, which made it clear that
the complaints were not confined to brickwork. Therefore, the claim which
was being made against the architects for the purposes of the policy extended
to all the matters which were subsequently particularised in the statement of
claim and Scott Schedule. He also thought that the issuing of the generally
endorsed writ had the same effect.4

On the facts, this was enough to dispose of the appeal. However, during the
course of his judgment, Sir John Donaldson gave a number of examples of
situations in which there might be one or more claims:
‘An architect has separate contracts with separate building owners. The
architect makes the same negligent mistake in relation to each. The
claims have a factor in common, namely the same negligent mistake,
and to this extent are related, but clearly they are separate claims.
Bringing the claims a little closer together, let us suppose that the
architect has a single contract in relation to two separate houses to be
built on quite separate sites in different parts of the country. If one
claim is in respect of a failure to specify windows of the requisite quality
and the other is in respect of failure to supervise the laying of the
foundations, I think that once again the claims would be separate. But it
would be otherwise if the complaint was the same in relation to both
houses. Then take the present example of a single contract for
professional services in relation to a number of houses in a single
development. A single complaint that they suffered from a wide range
of unrelated defects and a demand for compensation would, I think, be
regarded as a single claim. But if the defects manifested themselves
seriatim and each gave rise to a separate complaint, what then? They
might be regarded as separate claims. Alternatively, later complaints
could be regarded as enlargements of the original claims that the
architect had been professionally negligent in his execution of his
contract. It would, I think, very much depend upon the facts.’5

The background to this passage is Sir John Donaldson’s consideration of

Steyn J’s first instance decision against which the appeal had been brought, in
which he also gave examples of what might be one claim or more than one
claim. Sir John Donaldson agreed that there would be one claim if an owner
asserted loss due to faulty workmanship in the floors and the roof of a
building, but he was less certain that Steyn J was right when holding that if the
owner subsequently added a new and unrelated assertion of damage to the
windows, that would be a new claim. If it really was unrelated, then he would
agree, but in his view that would depend what ‘unrelated’ meant.

4 It is difficult to understand how merely issuing a writ could constitute a claim, but the
architects had been informed in correspondence that a writ had been issued and therefore
there was a communication of a claim.
5 Thorman, note 2, pages 51 and 52.

3
What is a ‘claim’ and is its formulation relevant?
Before analysing these examples in further depth, it is important to understand
that the meaning of ‘claim’ had previously been considered in West Wake
Price & Co v Ching by Devlin J.6 He had to decide whether a QC clause in a
professional indemnity policy could be invoked in circumstances where a
claim against a firm of accountants for the misappropriation of a client’s
monies by a clerk employed by the accountants was framed both in negligence
and on the grounds of dishonesty.

The policy would only cover the liability in negligence, and therefore Devlin J
had to determine whether there was one claim or more than one claim being
pursued against the accountants in the sense in which the word ‘claim’ was
being used in the policy. If it was only one claim, that would then lead to an
enquiry as to the true nature of the claim for the purposes of the policy.7

Devlin J referred to the definition of the word ‘claim’ in the Oxford Dictionary
and concluded that it is a demand for something which attaches to the object
being claimed, such as a sum of money; by contrast, it is not the cause of
action by which the claim is supported. He said the grounds upon which the
claim is being made, for example in negligence, may give the claim some
colour or character, but not its entity. Its entity can only be the object of the
claim itself.

However, does the way in which a claim is presented have any bearing on
whether it is, in truth, only one claim or more than one claim?

This is an issue which had to be considered by the Privy Council in Haydon v

Lo & Lo.8 In that case, a solicitor’s clerk stole HK$50m from one estate on 43
different occasions by four different methods. He also stole 31 parcels of
shares valued at HK$11m from another estate on eight different occasions.
The Privy Council unanimously decided that there was one claim in respect of
the theft of the monies from each of the estates, and therefore two in all for the
purposes of the solicitor’s professional indemnity policy, which provided an
indemnity on the primary layer for ‘any one claim hereunder’. As a result, the
excess layer insurers were on risk beyond the limit of the primary layer cover
for each of the thefts from the two estates.

In the Haydon case, Lord Lloyd said that if a claim was said to arise on each
occasion when something happened, which created a liability on the part of
the insured to a third party, that would equate a claim with a cause of action.
Of course, that was an approach which has been rejected by Devlin J in West
Wake Price & Co v Ching. Therefore, there could not be a separate claim
each time the clerk stole from an estate. However, the second estate had
decided not to make its claim against the firm of solicitors, but to bring

6 West Wake Price & Co v Ching [1956] 3 All ER 821, [1957] 1 WLR 45, [1956] 2
Lloyd’s Rep 618 (QBD).
7 Devlin J found that the one claim which was being pursued was, in truth, for dishonesty
and therefore not covered by the QC clause.
8 Haydon v Lo & Lo [1997] 1 WLR 198, [1997] 1 Lloyd’s Rep 336, [1997] CLC 626 (PC
– Hong Kong).

4
proceedings against the various companies for rectification of the share
registers. Through a circuitous route, the firm of solicitors was joined as a
third party to each of the 14 sets of proceedings. As such, there were 14
claimants in relation to the thefts from the second estate.

Despite this, Lord Lloyd concluded that there was in reality only one claim
being made in relation to the second estate. He said that the way in which the
claim was formulated was a good starting point for determining whether there
were one claim or more, but it cannot be decisive and the underlying facts are
paramount.

The Thorman examples

Sir John Donaldson’s first example of the same negligent mistake by the
architect in the case of separate contracts with separate owners is
uncontroversial. Unlike the Haydon case, there was no underlying single
claimant, but instead different claims being made by different parties, and
therefore two claims in all.9

A similar situation arose in Mabey & Johnson v Ecclesiastical Insurance

Office where the same error had replicated itself in the design of a number of
bridges, which had been supplied to various parties in different countries.10
The issue in that case was whether there were two separate contracts in the
case of bridges supplied to the Government of Ghana in two phases, but it was
common ground that in relation to those which had been supplied to separate
parties in Ethiopia, Peru and the Dominican Republic, there were separate
claims for the purposes of the insured’s professional indemnity policy. Had
there been an aggregation clause in the policy, the separate claims would have
been treated as one, as would the claims in Sir John Donaldson’s example.
This is one of the less usual situations in which an aggregation clause in a
policy would in fact assist in a construction setting.

However, difficulties arise when Sir John Donaldson moves on to his second
and third examples, both of which contemplate a single contract between the
architect and the owner and, in the first instance, a single complaint. In the
first limb of the second example of two separate houses to be built on different
sites in different parts of the country, but with different defaults by the
architect – a failure to specify the correct windows and a failure properly to
supervise the laying of the foundations – he believes that the claims would be
separate.

By contrast, in the second limb of the second example, he said the position
would be different if the same complaint was made in relation to both houses,
and therefore there would only be one claim. In the first limb of his third
example, he contemplates a situation in which there is a single development
with a number of houses in it, and the owner’s complaint is that they suffer
from a wide range of unrelated defects and a demand is made for

9 Haydon: note 8.
10 Mabey & Johnson Ltd v Ecclesiastical Insurance Office Plc [2003] EWHC 1523, [2004]
Lloyd’s Rep IR 10 (Comm).

5
compensation. In this case, Sir John Donaldson thought that this would be
regarded as a single claim.

The only material difference between the first limb of the second example and
the first limb of the third example is the geographical location of the houses in
question. Indeed, Sir John Donaldson attributes some significance to this by
referring to ‘quite separate sites in different parts of the country’. However,
this really ought to be irrelevant to any legal determination as to whether there
is more than one claim being made against the architect under the single
contract by the single claimant. If it were relevant, it would entertain the
possibility of more than one claim existing in a development built in phases
with different defects arising in each of the different phases, or perhaps even
defects arising in quite separate parts of the same development. Clearly, a
distinction of this nature, based upon location in which the defects arise, ought
to be disregarded.

It is also difficult to understand the materiality of the distinction which is

being made between the first and second limbs of the second example, namely
the fact that the defects might be the same in the latter, but different in the
former. The architect would have a general duty to carry out its services,
either expressly or impliedly, with reasonable skill and care. The services
may encompass design, supervision, checking, specification, and quite
possibly more. There may well be a number of separate breaches of that
general duty, such as a failure to specify the correct type of windows or
properly to supervise the laying of the foundations, or there may be a single
failure to design properly, which manifests itself in the same type of defect in
different parts of the works.

Whether there are separate breaches of the same duty or one breach, both of
which result in damage and a claim for compensation in respect of that
damage, should not really matter in determining whether there exists one
claim or more than one claim. Indeed, to focus on the type of breach which
has occurred seems to be contrary to the accepted position that a claim is not
be understood by reference to the cause of action which may have arisen. If in
the Haydon case, each theft or different method used to steal did not give rise
to a separate claim, why should different breaches of the general duty be any
different?11

The way in which a claim or claims for compensation may be presented in a

construction context was recognised by Stocker LJ in the Thorman case. He
said that where substantial building works are concerned, there may be a
variety of heads of damage and a variety of breaches of duty arising out of
different aspects of the general duty owed, for example a breach of the
specific duty to design and another breach of the specific duty to supervise.

The overall complaint might be set out in one set of proceedings with
particulars given of the specific aspects of the general duty which have been
breached and distinct losses pleaded arising from those specific breaches.

11 Haydon: note 8.

6
Where there are a variety of problems arising out of a project, these may well
be set out in a Scott Schedule, as was the position in the Thorman case itself.

However, Stocker LJ also recognised that the specific breaches of the general
duty which is owed might form the subject matter of separate actions. He
thought that the word ‘claim’ would embrace both the general claim
particularised as a series of separate breaches of the general duty, or each of a
series of separate and distinct claims. In other words, if the specific breaches
are presented as part of an overall claim for compensation, or as individual
claims, this would dictate whether they are to be regarded as one claim or
more than one claim.

These observations lead to a consideration of the second limb of Sir John

Donaldson’s third example, namely the manifestation of unrelated defects
seriatim, each giving rise to a separate complaint. He entertained the
possibility that these might be regarded as separate claims, or enlargements of
the original claim that the architect had been negligent. What the position
might actually be would depend on the facts.

However, it is interesting that the emergence of defects and the consequential

presentation of individual complaints over a period of time gave rise to a
temptation to regard the claims as separate. The material issue cannot be the
temporal manifestation of defects, simply because it may not be obvious to a
defendant when a defect emerged, and there cannot be any room for an
enquiry as to when a claimant should reasonably have been aware of defects at
any point in time. Instead, the only factor that may have a bearing on the issue
could be the manner in which the complaints are made known to the architect.
That echoes the approach which Stocker LJ was also inclined to follow as set
out above, when he expressed the view that complaints which are presented
separately as individual claims might each be regarded as a separate claim.

However, if the presentation of a complaint as a breach of a general duty

particularised by individual breaches of aspects of that overall duty, or as a
series of individual complaints, can dictate whether there is one claim or more
than one claim, that tends to suggest that a claimant can manipulate the
operation of a defendant’s professional indemnity policy. Depending upon the
way in which a defendant’s insurance programme has been put together and
the terms of the policy and excess layers in place, that may well be
commercially attractive to a claimant if it has concerns about the scope of the
defendant’s cover and the ability of the defendant to satisfy any judgment.

Instinctively, this seems incorrect and at odds with the views expressed in the
Haydon case that, although the way in which a claim is formulated is a good
starting point for determining whether there is one claim or more than one
claim, it cannot in all the circumstances be decisive.12

Of course, the Court of Appeal in the Thorman case did not say that the
formulation of the claim or claims was decisive, and therefore the underlying
facts were important, but there remains a powerful invitation to a claimant to

12 Haydon: note 8.

7
construct its complaints in a way which will satisfy its own commercial
agenda.

Thorman and the notification of circumstances

The Thorman decision also reverberates beyond the confines of the particular
issue which it was seeking to address. It has implications in construction
cases where defects emerge which do not immediately give rise to a claim
against an insured, but which are instead notified to professional indemnity
insurers as ‘circumstances’.

Most professional indemnity policies require an insured to give its insurers

notice not only of claims which are made in the policy period, but also of
circumstances of which it becomes aware which may give rise to a claim. The
precise notification requirements vary from one policy to the next, but the
approach remains the same, which is to allow the insured to benefit from the
cover being extended to claims which are made against the insured once the
policy has expired. If the circumstances are properly notified in the policy
period, the subsequent claim which is made against the insured, arising out of
the notified circumstances, is deemed to have been made in the policy period.

It is a feature of construction cases that all the defects in the works may not
immediately be apparent, and that what might at one point in time seem to be
confined to particular issues becomes a much greater problem when
investigations are carried out.

This was the position in Kajima v The Underwriter Insurance Company.13

Kajima had designed and built a block of flats in Leeds, which involved the
installation of a stack of pre-constructed pods and flat-pack construction. The
defendant’s insurers provided professional indemnity cover to Kajima for the
period from 20th May 2000 to 19th May 2002, and during that period Kajima
made a notification of circumstances which might be expected to produce a
claim against it. The circumstances notified were excessive movement and
settlement causing distortion of the adjoining roofing and balconies of the
flats. The notification also mentioned that investigations were underway into
the problems in order to identify the cause and what the implications might be.

As the investigations into the problems proceeded, various other defects came
to light, culminating in the discovery in 2005 that there were serious problems
relating to the stability of the building due to the underestimation of the wind
loadings. There was no means of economic repair and Kajima resolved the
claim which had then been made against it, but after the policy had expired, by
agreeing to purchase the building.

Akenhead J found that the insurers were not liable for the extensive further
defects which had been discovered after the investigations had been
completed. The circumstances which had been notified were confined to the

13 Kajima UK Engineering Ltd v Underwriter Insurance Co Ltd [2008] EWHC 83, [2008]
1 All ER (Comm) 855, 122 Con LR 123, [2008] Lloyd’s Rep IR 391, [2008] CILL 2567
(TCC).

8
particular matters set out in the notification, and although the fact of
investigations was referred to, these investigations were only in relation to the
settlement of the pods. It did not matter if the investigations happened to
reveal other defects which had no relationship to the circumstances notified.
Knowledge is important, in the sense that the insured must be aware of the
circumstances which it is notifying and it cannot guess that there is a defect of
which it is not then aware, albeit that it is possible to notify a ‘hornets’ nest’ or
a ‘can of worms’ if the facts support a notification of this character.

The Kajima decision highlights a further problem with the principles which
are set out in the Thorman case.14 If, as in the Kajima case, circumstances are
notified during the currency of the policy of the defects which have then been
identified, but investigations later reveal the existence of new defects which
had not previously been identified, what happens if the employer makes a
claim after the policy has expired?

If the claim which is made against the contractor is composite and

encompasses both the pre-existing defects which are the subject matter of the
notification and the further new defects which are later discovered, the first
limb of Sir John Donaldson’s third example in the Thorman case would
suggest that this is to be treated as one claim.

However, the insurers subscribing to the subsequent policy can legitimately

contend that the single claim which is made in their period of insurance arises
out of the circumstances notified in the previous policy period and therefore is
not covered by them. Equally, the insurers of the policy for the period in
which the circumstances were notified could contend that this single claim is
not covered because it encompasses a claim which does not arise from the
circumstances notified on the basis of the Kajima decision.

Clearly, it cannot be correct that the insured would be left with no cover at all,
and therefore the only solution would be to split the single claim over two
policy years. The first would provide an indemnity for liability in respect of
the defects which had been notified as circumstances in their policy year, and
liability for the defects which were subsequently discovered would be
indemnified by the insurers of the later policy year.

However, this would involve revisiting the Thorman case and the principles
which emerge from it by treating the first set of defects as one claim for policy
purposes and the second set as another. By the same token, to do so would
then emphasise the particular breaches, albeit of the general duty owed, and
stray into placing reliance upon the causes of action which had accrued, but of
course that was rejected by Devlin J in West Wake Price & Co v Ching, and
subsequent authorities.15

14 Kajima: note 13.

15 West Wake Price: note 6.

9
Revisiting Thorman
Unsurprisingly, the law which has developed around the interpretation of
professional indemnity policies has had to cater for the many different factual
situations which give rise to claims under the terms of such policies. The
basic structure of such policies, whether covering accountants, solicitors or
construction professionals, is fundamentally the same. However, claims
against professionals or design and build contractors in a construction context
have their own unique characteristics: there are often complaints about a
variety of matters in a single project, and defects can come to light over the
course of time, which may lead to claims being presented on a piecemeal basis
or, depending upon how a claimant decides to proceed, consolidated into one
claim.

The examples given in the Thorman case attempt to provide some guidelines,
but they have led to confusion and reveal inconsistencies when analysed. The
particular facts allowed the Thorman appeal to be resolved in favour of the
architects and the insurers who took over the insurance cover, and this was
enough to deal with the matter.16 There was, therefore, no need for Sir John
Donaldson to try to assess what the position might be in the various factual
scenarios set out in his examples.

Whether there is one claim or there are multiple claims must, in each case,
depend upon the particular facts. That much is clear, but the examples set out
in the decision provide an uneasy template by which to assess the extent to
which insurers and an insured are exposed under a professional indemnity
policy. The implications of the Kajima decision highlight the problems which
arise when trying to apply such a template.17

It is hard to escape the conclusion that whether there is one claim or more than
one claim for policy purposes is ultimately a matter of impression against the
background of the particular facts of an individual case.

Although previous decisions may be of interest, they should provide limited

guidance as to how that impression should be formed in each case. Indeed,
how a claim is formulated by the claimant should really make no difference at
all, even as a useful starting point, because it invites manipulation of the
contractual position between insurers and an insured.18

Furthermore, it should be noted that claims for indemnity under a contractor’s

design and build policy are quite often made under a mitigation of loss clause
in the policy. This allows a contractor to recover its loss and expense in
repairing defects before practical completion. In other words, the contractor
mitigates the claim which might otherwise have been made against it. In such

16 The architects would of course have had to bear another deductible had the second set of
insurers been on risk, and they also faced a denial of cover by these insurers on the
grounds of material non-disclosure.
17 Kajima: note 13.
18 In Australia & New Zealand Bank Ltd v Colonial & Eagle Wharves Ltd [1960] 2 Lloyd’s
Rep 241 (Comm), McNair J cautioned against placing reliance upon the form in which
the claim is put forward, albeit in the context of an all risks policy.

10
a situation, as Morison J observed in the Mabey & Johnson case, it is not
possible to see how the claim against the insured is formulated in any event.19

Dispensing with the Thorman guidelines may well be unsatisfactory to those

parties who are trying to calculate how a policy might operate at some point in
the future, but the price of uncertainty seems better than the cost of applying
some tests that do not really fit and which could lead to a result which is
commercially unacceptable to either party.

Michael Regan is a partner of Mayer Brown, solicitors, and head of the

construction and engineering group in their London office.

© Michael Regan and the Society of Construction Law 2015

The views expressed by the author in this paper are his alone, and do not necessarily
represent the views of the Society of Construction Law or the editors. Neither the
author, the Society, nor the editors can accept any liability in respect of any use to
which this paper or any information or views expressed in it may be put, whether
arising through negligence or otherwise.

19 Mabey & Johnson: note 10.

11
 ‘The object of the Society
is to promote the study and understanding of
construction law amongst all those involved
in the construction industry’

MEMBERSHIP/ADMINISTRATION ENQUIRIES
Jill Ward
The Cottage, Bullfurlong Lane,
Burbage, Leics LE10 2HQ
tel: 07730 474074
e-mail: [email protected]

website: www.scl.org.uk

12