England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM QUEEN'S BENCH DIVISION
COMMERCIAL COURT
Mr Justice Aikens

		Royal Courts of Justice Strand, London, WC2A 2LL Tuesday 31st July 2001

LORD JUSTICE RIX:

1. This appeal concerns film finance insurance, which I have sought to describe at the beginning of my judgment in HIH Casualty and General Insurance Limited v. New Hampshire [2001] 2 All ER (Comm) 39. In the present case, although the commercial structure of the transaction is somewhat different, the essence of the insurance remains the same, namely that the parties providing the syndicated finance for the production of a film, in this case represented by The Chase Manhattan Bank ("Chase"), stipulate for and receive the security of an insurance which is designed to pay, up to the maximum of the sum insured, if for any reason the revenues generated by the film within a certain period are insufficient to repay the loan finance plus associated expenses.
2. Because the assured is a lender rather than a film producer, it is distanced from an intimate knowledge of the proposed film-making and its marketing, and it is the producer and the entrepreneurial investors in the film who need to procure the insurance policy for the benefit of the assured, as a condition precedent of the lending transaction. In the present case the three policies with which these actions are concerned were placed by Heath North America & Special Risks Ltd and/or Heath Insurance Broking Limited ("Heaths"), who, although in form the agent of the assured, was in practice instructed by a Mr Graham Bradstreet through his company Premier Media Limited ("Premier") pursuant to an agreement with a film production company called Phoenix Pictures Inc ("Phoenix"). Heaths, together with a company called Screen Partners Limited ("SPL"), had developed the use of the form of insurance with which these actions are concerned, known as "time variable contingency" (or TVC) insurance, and had been prominent in placing it in the London market since 1992 and in the international market since 1995.
3. Since Chase's role was limited and the broking was done by Heaths, the policies contained a clause, the "Truth of Statement" clause, whose purpose may be said, in general, to distance Chase from responsibility for the placing of the insurance. The issues which arise on this appeal are all concerned with the extent to which that clause does or does not operate successfully to protect Chase from the complaints of misrepresentation and non-disclosure in the placing of the policies which the insurers have alleged against Heaths, as Chase's agent to insure. It is to be noted that no complaints are made against Chase personally. It is also to be noted, however, that the misrepresentations and non-disclosures alleged against Heaths are all said to have been made either fraudulently or negligently.
4. This appeal arises from a trial of preliminary issues. The trial took place before Aikens J, whose judgment is reported at [2001] 1 Lloyd's Rep 30. The facts were taken from the insurers' particulars of claim, as assumed facts. Thus when I speak of any fact in this judgment, it has only been assumed for the purpose of the preliminary issues. The misrepresentations and non-disclosures complained of are all denied, as are the allegations of negligence and fraud.
5. Since the Truth of Statement clause (also referred to simply as "the clause") is at the heart of this appeal, I shall set it out immediately. In due course, however, I will have to put it in its contractual context and also to say something further about the structure of the overall transaction in which the insurance policies containing this clause are themselves set.

The Truth of Statement clause

6. The Truth of Statement clause is found as the first of a series of clauses under the general heading of "CONDITIONS PRECEDENT". In his judgment below, Aikens J split the clause up into eight phrases, and on appeal the parties have been content to argue its construction by reference to those subdivisions. I shall therefore set it out in the same way, numbering the eight phrases, thus –

"1. It is a condition precedent to this Policy that: Elmwood Films Inc has truthfully completed Section 1 of the Questionnaire to the best of its knowledge. Any reference in the Questionnaire to the Revenue Participation Agreement is qualified by reference to the copy thereof attached to the Questionnaire or the Declaration. In completing the Questionnaire, Elmwood may rely on certificates of third parties to the extent such reliance is disclosed on the Questionnaire.

2. Provided that Elmwood Films Inc completes the sections of the Questionnaire required to be completed by it and delivers the same to the Lead Insurers…

3. …(it being acknowledged that any misstatement in any part of the Questionnaire (other than Section 1 thereof) shall not be the responsibility of the Insured or constitute a ground for avoidance of the Insurers' obligations under the Policy or the cancellation thereof).

4. In addition, the failure of Elmwood Films Inc to update Section 1 of the Questionnaire for a Film Production shall not be the responsibility of the Insured or constitute a ground for avoidance of the insurers' obligations under the Policy or the cancellation thereof.

5. Subject to the obligation of the Insured under "General Conditions – Due Diligence Clause" after acceptance by the Lead Insurers of the Declaration with regard to the Film Production, the Insured…

6. …[the Insured] will not have any duty or obligation to make any representation, warranty or disclosure of any nature, express or implied (such duty and obligation being expressly waived by the insurers), and …

7. …[the Insured]…shall have no liability of any nature to the insurers for any information provided by any other parties […including, but not limited to, Heath North America & Special Risks Ltd (other than Section 1 of the Questionnaire)]…

8. …and any such information provided by or nondisclosure by other parties including, but not limited to, Heath North America & Special Risks Ltd (other than Section 1 of the Questionnaire) shall not be a ground or grounds for avoidance of the insurers' obligations under the Policy or the cancellation thereof."

7. In the original version of the clause as set out in the policies the only break in the language came between phrase 1 and the rest of the clause respectively. Thus phrase 1 constituted a single paragraph, and all the other seven phrases constituted a single further paragraph. Moreover it is well to bear in mind that phrases 7 and 8 originally ran together without the interpolation of the language found in square brackets in phrase 7. That language is inserted into phrase 7 to indicate that it is common ground that the reference to Heaths and the exception relating to Section 1 of the Questionnaire apply equally to both phrases 7 and 8.
8. I shall explain below the references to Elmwood and to the Questionnaire.
9. It is necessary to have regard to the whole clause in order fully to understand the parties' submissions. However, the critical part of the clause is to be found in phrases 6, 7 and 8. There are to be found exclusions successively of "any duty or obligation to make any representation" (phrase 6), of "[any] liability of any nature to the insurers for any information" (phrase 7), and of any information or non-disclosure providing "a ground or grounds for avoidance…or cancellation" (phrase 8). Prima facie, phrase 6 excludes the duty of disclosure, phrase 7 excludes liability for what others including Heaths may say, and phrase 8 excludes the remedy of avoidance (or cancellation).
10. It may be observed that it is a complexity of the clause that "Insured" is defined elsewhere in the policy as meaning Chase, whether itself or as agent for the syndicate participants, and that the clause therefore may be said to distinguish between Chase on the one hand and third parties including Heaths on the other (phrases 7 and 8). This feature has led to the submission on the part of the insurers that phrase 6 only excludes Chase's duty of disclosure, and not Heaths' separate duty as the agent to insure.

A synopsis of the dispute

11. In a nutshell, and subject therefore to the danger of over-compression, the submissions of the parties with respect to these phrases are as follows. Chase submits: that phrase 6 excludes the duty to speak, not only of Chase but also of Heaths, on the basis that an agent to insure can have no duty to speak where his principal has none; that phrase 7 excludes the authority of Heaths to speak and thus makes it impossible for Chase to be liable for anything said by Heaths; and that phrase 8 excludes both the remedy of avoidance (for breach of the duty of utmost good faith found in insurance law) and the remedy of rescission (under the Misrepresentation Act 1967). Chase submits that the effect of such exclusions is to protect Chase absolutely against any misrepresentation or non-disclosure committed by Heaths, even if committed negligently or fraudulently. Chase submits that such a construction is exactly what the parties bargained for and intended and reflects the parties' understanding that (with the exception of Section 1 of the Questionnaire for which responsibility is accepted under phrases 1, 3, 7 and 8 of the clause) Chase is an outsider without knowledge of or responsibility for the film production which is at the core of the transaction.
12. On the other hand, the insurers submit that the exclusions are much more limited: that phrase 6 says nothing about excluding Heaths' duty of disclosure as agent to insure; that phrase 7 does not exclude Heaths' authority, only Chase's liability; and that phrase 8 does not exclude the right of rescission under the Misrepresentation Act. It also submits that the word "information" as found in clauses 7 and 8 is too weak a word to include misrepresentation. Above all, the insurers submit that nothing in these phrases protects Chase against the fraud or negligence of its agent, whether those are viewed as giving rise to remedies in damages or in avoidance or rescission. That submission is put as a matter of construction. Where fraud is concerned, however, it is also put as a rule of law.
13. In the court below Aikens J by and large acceded to the insurers' submissions. In particular he held that the clause did not protect Chase against the negligent or fraudulent misrepresentations or non-disclosures of Heaths. Nor did it protect Chase against claims under section 2(1) of the Misrepresentation Act. It only protected Chase against innocent misrepresentation or innocent non-disclosure: but then none such had been alleged.
14. In only three respects did he hold in favour of Chase's submissions. First, he agreed that Lord Morton's three rules in Canada Steamship Lines v. The King [1952] AC 192 did not apply directly to the duty of good faith found in insurance, because that was a unitary and absolute duty which did not depend on negligence or fraud but which could be broken even innocently. That, however, did not avail Chase, for he went on to apply the spirit of those rules by analogy in seeking out the presumed intention of the parties. Secondly, he held that it is possible as a matter of law to exclude liability for the fraud of an agent. That did not avail Chase either, for he went on to hold as a matter of construction that Heaths' alleged fraud was not excluded in any event. Thirdly, he held that in the absence at any rate of any special factors, none of which had been pleaded, no duty of care arises between insurer and assured to parallel the duty of good faith, and it therefore followed that the insurers' alternative pleas for damages for negligent misstatement (ie for breach of the Hedley Byrne duty of care) failed. However, in the overall scheme of his judgment that was of no consequence.
15. Thus it might be said that on points of principle Aikens J held in favour of Chase, but on decisive points of construction the victory went to the insurers.
16. The consequence is that Chase has appealed against the judge's rulings on construction, and the insurers have either cross-appealed or issued a respondent's notice against his rulings on principle.
17. The insurers also have a cross-appeal on a minor point which they say is a matter of mere omission on the part of the judge: having found that the clause does not protect against negligence or fraud, or against damages under section 2(1) of the Misrepresentation Act, he ought to have gone on to state expressly that the clause does not protect against rescission under section 1 of the Misrepresentation Act in the event of negligence or fraud.
18. There is one brand new point which has arisen on appeal, which was not argued before Aikens J. It is a point of construction. It is submitted that phrases 6, 7 and 8 give no protection whatsoever to Chase in respect of the period before the contract is made because, beginning with phrase 5, the whole of the rest of the clause is dealing only with the post-contractual period.
19. Having sought thus to introduce and summarise the dispute which arises on this appeal, I will now set out the facts and issues in greater detail.

The three actions, their parties and others

20. There are three separate but consolidated actions, each involving a separate policy. Between them they cover the financing of five films.
21. Two of the films were made the subject of individual policies. They are The Mirror Has Two Faces and The People vs Larry Flynt. The slip in respect of the former was signed by the insurers between June and September 1996: that slip was superseded by Policy No 614/NKA/118A. The slip in respect of the latter was signed by its insurers later in 1996: that slip was superseded by Policy No 614/NKA/119A. Chase is the claimant against the insurers in respect of those two policies, claiming a total of about US$16.5 million. Chase has brought a separate action in respect of each such policy.
22. An earlier slip had been broked about six months previously in respect of The Mirror Has Two Faces, but the lead company underwriter withdrew as did subscribing Lloyd's underwriters. Thus the slip which led on to the policy in respect of that film was a replacement slip.
23. The insurers, led by HIH Casualty and General Insurance Limited ("HIH"), are the claimants in the other of the three actions, which was the first to be commenced. The insurers' action concerns all five films, both the two already mentioned and the other three films – Amy Foster, U-Turn, and Apt Pupil. These three films are not the subject of separate policies, but of separate declarations made under a line slip facility, which, as is common ground, is a contract for insurance rather than a contract of insurance. However, when the declarations were made and accepted, these then became contracts of insurance. The line slip contained essentially the same wording as the individual policies. As and when a declaration was made and accepted that wording became incorporated into the contracts of insurance then concluded. The line slip was signed by the insurers in June to September 1996. The three declarations were made in September and October 1996 and February 1997 respectively. The declarations ultimately became the subject of Policy No 614/OL077A, which superseded the original line slip. In their action the insurers claim to be entitled to avoid or rescind the line slip, the declarations and the two individual policies because of the alleged misrepresentations and non-disclosures made by Heaths in the course of negotiations. The insurers also claim damages from both Chase and Heaths in deceit, negligent misstatement and under the Misrepresentation Act. Their amended particulars of claim are the assumed basis of the preliminary issues. There are as yet no defences in Chase's two actions.
24. I have already introduced Chase and Heaths as the assured and the agent to insure respectively. Each is mentioned in the policies. Thus "The name of the Assured" is given as "THE CHASE MANHATTAN BANK, itself and as Agent for the Participants under the Loan Agreement", and in the definition section the "INSURED" is defined as –

"The Chase Manhattan Bank (formerly known as Chemical Bank), itself and as Agent for the Participants from time to time…"

25. Heaths is expressly mentioned in the Truth of Statement clause (and also in the premium payment clause as payee). It is relevant to Chase's submission that phrase 7 of the clause is an exclusion of Heaths' authority to provide any information on behalf of Chase to point out that under the policies' "Miscellaneous" clauses there appears the following:

"Insurers understand and agree that Premier Media Limited, Graham Bradstreet, Tristar Pictures Inc and Phoenix Pictures Inc are not agents or representatives of the Insured."

26. It will be noted that there is no reference to Heaths in that clause. The insurers rely on that omission. Chase agrees that the omission is of course deliberate, since Heaths must be its agent at least for the making of the insurance contracts. It nevertheless submits that Heaths' authority to do more than just enter into the policies on its behalf is limited by the Truth of Statement clause.
27. Chase has been represented on this appeal by Lord Grabiner QC. Heaths has not been represented, as the issues on appeal do not directly concern it.
28. The insurers to the policies are listed in them. They are numerous, each for its respective percentage, which varies. The two lead insurers are XL Reinsurance Company Limited (with the largest single percentage at 48.6%) and HIH. XL Reinsurance has paid the claims and thus is not a party to these proceedings. (Two of the other insurers have since dropped out of the insurers' action.) The lead insurer in these proceedings has therefore been HIH, who heads the title to the insurers' action. However, shortly before the hearing of this appeal, HIH went into provisional liquidation and did not appear at the appeal. Its role was taken over by the other insurers. With one exception, they were represented by Mr Jeremy Cooke QC. The exception, who briefed separate counsel, was Axa Colonia Versicherung AG ("Axa"). It was represented by Mr Peter Gross QC, who was the originator of the new point that the second half of the Truth of Statement clause was dealing only with the post-contractual period.
29. I have already introduced Mr Bradstreet, Premier and Phoenix (see para 2 above) who were instrumental in obtaining the policies. They are all mentioned in the miscellaneous clause cited above as not being agents or representatives of Chase. The fourth person mentioned in that clause is Tristar Pictures Inc ("Tristar"). Tristar is a company in the Sony (ie Sony Pictures Entertainment Inc) group. Phoenix had entered into long term distribution arrangements with Sony and another company, Canal+, for the production by Phoenix of a library of films. Tristar was the Sony vehicle which owned and distributed the rights to The Mirror Has Two Faces, one of those films.
30. Tristar in turn sold a revenue participation to another Sony vehicle, Elmwood Films Inc ("Elmwood"). That sale was pursuant to a contract known as the Revenue Participation Agreement ("RPA"). Elmwood was the recipient and obligee of the Chase syndicate's finance. That finance was provided pursuant to a Reimbursement and Term Loan Agreement ("RTLA"). The finance enabled Elmwood, which was otherwise merely a special purpose vehicle, to buy its revenue participation from Tristar. Elmwood assigned its revenue participation to the Chase syndicate and the policy in question insured Chase against a shortfall of revenue with which to repay the loan. Thus the insurance was the sole security for any shortfall in the revenues.
31. It was common ground that the policy relating to The Mirror Has Two Faces would suffice as an exemplar for the line slip facility and all three policies in question. In the circumstances it is unnecessary to go into much detail about the other transactions. Suffice it to say that in the case of The People vs Larry Flynt different Sony vehicles were involved, but the structure of the transaction was otherwise entirely the same. In the case of the other three films, which were declared under the line slip facility, the structure was a little different. The obligee of the finance in their case was Phoenix itself. There the security of the insurance represented only 35% of the loan. There were three other pieces of security, involving payments by Sony (25%), by Canal+ (14%) and by Phoenix (26%). In every case, however, an insurance in the required form was a condition precedent of the lending, and the insurance itself, as will be seen below, referred to and/or annexed the relevant underlying agreements.
32. For a fuller summary of the structure of the arrangements, see paras 5/6 of the judgment of Aikens J.

The policy terms

33. I have already set out the critical Truth of Statement clause in para 6 above and will not do so again here. Express reference is made to the RPA and the RTLA in the Definitions clause and elsewhere in the policy. Other clauses of relevance are as follows:

[In the preamble]: "If the Assured shall make any claim knowing the same to be false or fraudulent, as regards amount or otherwise, this Policy shall become void, and all claim thereunder shall be forfeited."

[Under "General Conditions"]:

Due Diligence Clause

The Insured shall at all times do and concur in doing all things reasonably necessary after the date hereof to avoid or diminish a loss to insurers under this Policy. Insurers confirm that the due diligence obligation of the Insured shall mean that it shall at all times do and concur in doing all things reasonably necessary to avoid or diminish a loss to the insurers under this Policy in the same manner as similarly situated prudent uninsured banker…

[Under "General Exclusions"]: This Policy will not indemnify the incurring of Time Variable and/or Contingent Expenses directly or indirectly arising out of , contributed by or resulting from:

Fraud: Any fraud, misrepresentation, or concealment by the Insured, any employee of the Insured or the Distributor [defined as Tristar] or any agent of the Distributor.

[Under Claims Procedure clause]: …Confirm the facts in writing as soon as reasonably practicable with as much information as available.

[Under Miscellaneous clauses]: …The rule that a contract is to be construed against the party drafting the contract is hereby waived, and shall have no applicability in construing this Policy or the terms of this Policy.

Insurers confirm that the absence of Available Revenue in circumstances which would not otherwise entitle insurers to reject any claim will not give insurers any grounds to avoid this Policy or reject any claim…

The Insured is a large financial institution and for purposes of this condition precedent and other provisions of this Policy, matters known to the Insured shall only refer to matters known to or delivered to those officers of the Insured servicing the [RTLA].

Insurers understand and agree that Premier Media Limited, Graham Bradstreet, Tristar Pictures Inc and Phoenix Pictures Inc are not agents or representatives of the Insured.

Revenue Analysis:

Insurers acknowledge that they have been informed by the Insured that it has neither requested nor received nor intends to request any revenue projections from the Distributor for the Film Production or any other film production into which insurers may have crossing rights. In addition, insurers acknowledge that they have received no such projections from the insured, the Distributor or any of their respective agents except as set forth in the Questionnaire. Except as set forth in the Questionnaire any revenue analysis or financial projections for the Film Production at the time the Film Production is submitted for Declaration upon which insurers are relying were prepared for them by their agents or independent contractors and not by the Insured, the Distributor or their respective agents.

[Under the Declaration attached to the policy as Schedule 1]: …Any references in the Policy (including the Schedules thereto) to the [RPA] or the [RTLA] covered by this Declaration are qualified in their entirety by reference to such agreements copies of which have been made available to the Lead Insurers and are attached hereto.

[Under the Questionnaire attached to the policy as Schedule 2, and which is headed "Elmwood Films Inc"]:

Section I – General

The undersigned [Elmwood] represents and warrants to the insurers ("Insurers") under the time variable contingency insurance policy with respect to the film described below (the "Policy") that the information set forth in Section 1 of this Questionnaire is at the date hereof true and correct in all material respects…

Section II – TVC Insurance

The information set forth in Section II of this Questionnaire is provided without representation or warranty of any kind. All persons to which this information is provided acknowledge and agree that estimates and projections are inherently uncertain and that the undersigned shall have no liability of any nature for the accuracy of the estimates, projections and other information provided in Section II of this Questionnaire…The Insurers acknowledge that the estimates set out in this Section II are purely informational, that they are not relying on these estimates in issuing the Policy, and that the Insurers are relying on their own analysis of revenue estimates. The Insurers under the Policy acknowledge and agree that they will be and remain obligated to pay any claim under the Policy with respect to the Film Production and will not be excused from paying any claim if for any reason the actual revenues are less than the estimates set forth in this section II…

34. Section I of the Questionnaire set out information concerning the film such as the names of producer, director and principal cast, the attachment of the script, the amount of the final budget, the length of the film, the timing of the production, the distributor, and other insurances, such as completion guarantee insurance and errors and omissions insurance. Section II set out revenue estimates and other financial information.
35. It may be recalled that one of the points of construction that arises on phrase 7 of the Truth of Statement clause is as to the meaning of "information". It will be seen that that word occurs elsewhere in the policy.

The RTLA

36. The "Introductory Statement" of the RTLA, under which Elmwood is described as "Obligor" and Chase is described as "Issuing Bank", explains the structure of the transaction and contains the following:

"Concurrently herewith the Obligor is purchasing Contingency Expense Insurance (the "Insurance") that will insure that the participants will receive the amount necessary to repay the Reimbursement Obligation…"

37. The policy wording was annexed. The Reimbursement Obligation was subsequently described as "absolute, unconditional and irrevocable and shall be paid strictly in accordance with this Agreement…" Section 4.1, headed "Conditions precedent" provided, under sub-section (x), that the "Insurance" was among the documents that Chase must have received. Another condition precedent (sub-paragraph (xvi)) was that Chase should have completed its due diligence investigation to its own satisfaction. Section 6, headed "Events of Default", provided by sub-section (i) that a combination of the failure of the Insurance followed by the failure of Elmwood to repay the loan to Chase would be a default. The failure of the Insurance was expressed as follows:

"[if] the Insurance shall for any reason, not be or shall cease to be in full force and effect, or shall be declared null and void, or becomes unenforceable, or it shall be terminated, or disaffirmed by the underwriters thereunder…"

The allegations made by the insurers against Chase

38. In paragraph 48 of their amended particulars of claim the insurers set out their case as to misrepresentation. The misrepresentations relied on are collected in Schedule 2 to their pleading, which runs to 58 pages. It is said that before the conclusion of each of the contracts of or for insurance Heaths made the misrepresentations set out in Schedule 2 and did so "fraudulently either well knowing that the same were false or recklessly not caring whether the same were true or false or alternatively negligently in breach of a duty of care to the Claimants". The principal complaints are set out by Aikens J (in para 10 of his judgment) as follows:

"(1) that Heaths represented that the policies that were being broked for the [Phoenix] films were the same as the TVC policies that had been concluded in 1992 and subsequent years, whereas the later policy terms were materially different. The key differences pleaded were that under the later policy terms (i) it was agreed that revenues would not be available to repay the insured lender banks unless and until all advances and all distribution expenses (which were defined broadly and imprecisely) had been recouped; and (ii) that insurers had little or no control over the production in the event of a claim on the insurance.

"(2) Heaths represented that the previous TVC facility had been claims free, whereas that was not true.

"(3) Heaths represented that Lloyd's underwriters and the lead company insurer had withdrawn their lines from the slips for The Mirror Had Two Faces and the line slip because the risk was akin to financial guarantee which they either could not or did not wish to underwrite and also for "political reasons" between the lead company insurer and its head office; whereas the true reason was that Lloyd's withdrew because of non disclosures or misrepresentations by Heaths and the company lead had withdrawn because its head office had expressed fundamental concerns about the risks."

39. The insurers' complaints about non-disclosure are introduced in paragraphs 49/50 of the amended particulars of claim and are collected in Schedule 3 thereto, a mere 5 pages. The insurers again plead that such non-disclosure was fraudulent "inasmuch as it was effected deliberately with the intent of hiding such matters from the Claimants and inducing them to enter into the said contracts", alternatively was reckless or made negligently in breach of a duty of care. Aikens J describes (in para 11 of his judgment) the principal non-disclosures alleged as follows:

"(1) that there had been claims on the TVC facility in the 1992-3 year;

(2) that the Lloyd's market withdrew as a result of non-disclosures and misrepresentations;

(3) that the lead company insurer had withdrawn from the Phoenix facility because of fundamental technical concerns as to the Phoenix risk;

(4) that SPL was critical of the viability of the Phoenix programme and had communicated those concerns to both Chase and Heaths."

40. On the basis of these allegations the insurers claim the following relief in their action (see paras 51/59 of their amended particulars of claim): (1) avoidance or rescission; (2) damages for fraudulent misrepresentation, against Heaths, who is said to be liable "on its own account", and against Chase who is said to be liable as Heaths' principal; (3) damages for negligent misstatements made in breach of a duty of care owed at common law; (4) damages for misrepresentation under section 2(1) of the Misrepresentation Act; (5) damages for non-disclosure: it is not said on what basis, and one might presume that, reflecting the earlier pleadings in relation to the right to avoid or rescind, that the basis was in deceit and/or negligence; Heaths is again said to be liable on its own account and Chase as principal.
41. It will have been observed that the insurers' case is, or purports to be, made entirely (i) in fraud or negligence; (ii) in respect of matters which lie outside those dealt with in the Questionnaire or the Revenue Analysis clause; and (iii) against Heaths, such that Chase's susceptibility to avoidance or rescission or liability in damages arise out of principles of agency and/or vicarious liability.

The preliminary issues

42. These are stated as follows:

"On the true construction of the contracts of or for insurance pleaded in the Particulars of Claim No 1999 Folio 1413 [the insurers' action] and on the assumption that the facts and matters pleaded in those Particulars of Claim are true, are the Claimants (HIH) entitled:

(a) to avoid and/or rescind the contracts of or for insurance against the First Defendants (Chase); and/or

(b) to damages from the First Defendants (Chase) for misrepresentation or non-disclosure; and/or

(c) to damages from Heaths for non-disclosure?"

Issues (a) and (b) were described by Aikens J as preliminary issues one and two respectively. Issue (c) is not relevant to this appeal.

The judge's answers

43. The answers entered into the judge's order are as follows. He distinguished between contracts of and for insurance. As to the first issue (issue (a) above), he held that although the Truth of Statement clause would exclude the right to avoid the contracts of insurance for innocent misrepresentations or non-disclosures, none were pleaded; and that on the pleadings as they stand, if true, the insurers would have the right to avoid or rescind the contracts of or for insurance, since the right to avoid contracts of insurance for negligent or fraudulent misrepresentations or for negligent non-disclosure or deliberate concealment on the part of Heaths was not excluded, and neither was the right to rescind the contract for insurance (the line slip facility) excluded in the case of a misrepresentation of fact made intentionally, recklessly or negligently by Heaths or made without Heaths having a reasonable ground to believe that the fact was true.
44. As to preliminary issue two (issue (b) above), he held that insurers would have a right to claim damages based upon Heaths' intentional or reckless misrepresentations either in deceit or under section 2(1) of the Misrepresentation Act and whether in respect of contracts of or for insurance; but that there was no right under the contracts of insurance to claim damages either for breach of the duty of good faith or for negligent misstatement in breach of a duty of care; although there was a right to claim damages for negligent misstatement in relation to the line slip facility, for that was a contract for insurance.
45. It may be noted that there is nothing in his order, either one way or the other, in relation to damages for non-disclosure, even though that was included in the question under preliminary issue two.
46. As to preliminary issue three (issue (c) above), he held that there was no right to claim damages against Heaths for non-disclosure as presently pleaded. That was on the basis, as explained in his judgment at para 97, that it had been common ground before him that this third issue was intended to deal with a pure case of non-disclosure, ie was "not intended to deal with claims based on allegations of deceit or any kind of positive misrepresentation" including misrepresentations to be derived from non-disclosure (see, for instance Pryke v. Gibbs Hartley Cooper [1991] 1 Lloyd's Rep 603 at 615/6). In the circumstances, the questions raised under this third issue were whether the duty of good faith or a duty of care at common law could give rise to damages for non-disclosure. The judge held that they could not. There is no appeal from that decision, and Heaths is not involved in this court.
47. In the circumstances, it is not clear to me what, if any, the judge's conclusions were with regard to the claim to damages (against Chase as Heaths' principal) based on the allegations of fraudulent, as distinct from negligent, non-disclosure. That is nowhere dealt with expressly in his order, even though that claim was included as part of the question dealt with under preliminary issue two. In any event there has been no appeal from that omission. If, nevertheless, I consider his judgment with that query in mind, it appears that he treated the claim for damages for fraudulent non-disclosure as being inseparable from a claim based on damages for breach of the duty of good faith. As to that he held that damages are simply not available for breach of that duty (paras 82/84). When he next proceeded to consider whether damages for the tort of deceit had been excluded by the Truth of Statement clause (at paras 86/87), he seems to have confined his consideration to deceit by reason of misrepresentation alone and not also by reason of non-disclosure. It is not immediately obvious why he did so, given the state of the pleadings, unless it reflected the way in which the matter was argued. In the circumstances, there is no consideration in the judgment as to the possibility of damages in deceit for fraudulent non-disclosure against the background of a duty to speak based on the duty of good faith, and the absence of any such consideration is reflected both in his summaries at paras 95 and 117(2) and in his order. There is, as I have said, no appeal from that omission or treatment.

Some basic principles

48. Where fraud is concerned, and a contract has been induced by fraud, the victim of the fraud is entitled both to rescind the contract and to claim damages for the tort of deceit. Fraud is usually defined in terms of a fraudulent misrepresentation. The general rule is that mere non-disclosure does not constitute misrepresentation, and that in the absence of a duty to speak there can be no liability in fraud, however dishonest the silence. However, in certain circumstances a combination of silence together with a positive representation may itself create a misrepresentation. Such a situation may be called partial non-disclosure, and such cases may be explained as either instances of actual misrepresentation or as cases where a duty to speak arises because of matters already stated. An exception to the general rule relating to non-disclosure, however, arises in the insurance context, because of the duty of good faith, more strictly the duty of utmost good faith (uberrimae fidei). That creates what may be called a duty to speak, defined in the terms of sections 17/19 of the Marine Insurance Act 1906 (the "MIA 1906"). The question might arise, but as I have just explained, was not apparently argued before the judge below, whether a combination of the duty to speak and dishonest non-disclosure can create a cause of action in deceit.
49. There is no doubt, however, that the duty of good faith, standing alone, does not give rise to a cause of action in damages. The only remedy for its breach is the remedy of avoidance, the right to avoid. That is the effect of section 17 of the MIA 1906 ("A contract of marine insurance is a contract based upon the utmost good faith, and, if the utmost good faith be not observed by either party, the contract may be avoided by the other party.") It is also what was decided in Banque Keyser Ullmann SA v. Skandia (UK) Insurance Co Ltd [1990] 1 QB 665 at 773/781 ("Issue 5"), Banque Financière de la Cité SA (formerly Banque Keyser Ullmann SA) v. Westgate Insurance Co Ltd (formerly Hodge General & Mercantile Co Ltd) [1991] 2 AC 249. By itself, the duty of good faith gives rise to neither contractual nor tortious obligations, but exists sui generis as a matter of law.
50. Outside the tort of deceit and the duty of good faith, lies the area of negligence. In matters of misrepresentation, a remedy for negligence is a recent arrival, long post-dating the established remedies in deceit and avoidance to which may be added the right to rescind for an innocent misrepresentation. It was not until Hedley Byrne & Co Ltd v. Heller & Partners Ltd [1964] AC 465 that a duty of care not to make negligent misstatements was established. That was followed by the Misrepresentation Act 1967, which allowed damages under section 2(1) where a defendant could not prove that "he had reasonable ground to believe and did believe up to the time that the contract was made that the facts represented were true": a remedy for what has come to be known as negligent misrepresentation. There is also section 2(2) of that Act, which allows damages to be awarded in lieu of rescission, where that is an available remedy, even in the case of innocent misrepresentation ("a misrepresentation…made to him otherwise than fraudulently").
51. The question then arises as to what role these new remedies play in the context of insurance. Chase reserves the right hereafter to submit that the Misrepresentation Act does not apply to contracts of insurance at all. For the present, however, it accepts that at this level it is bound to recognise that there is scope for a claim to damages under section 2(1): see Banque Keyser at 789/790 (where the claim, however, failed) and Aikens J at para 90. It is plain, however, from Banque Keyser at 790E that mere non-disclosure, even when underpinned by the duty to speak imposed by the duty of good faith, does not amount to the "making" of a representation within the meaning of section 2(1). As for a claim to damages in lieu of rescission under section 2(2), Aikens J did not deal with this expressly, but in Highlands Insurance Co v. Continental Insurance Co [1987] 1 Lloyd's Rep 109 at 118 Steyn J expressed the view that where a contract of insurance has been validly avoided on the grounds of a material misrepresentation, it would be "difficult to conceive of circumstances in which it would be equitable within the meaning of s.2(2) to grant relief from such avoidance…". As for section 3 of the Act (which renders ineffective a term which would exclude or restrict liability for any pre-contractual misrepresentation or any remedy otherwise available by reason of such misrepresentation), Aikens J recorded (at paras 80 and 94) that the point was deliberately excluded from argument.
52. As for negligent misstatement and the Hedley Byrne duty of care, it was established in Banque Keyser that at any rate in a case of mere non-disclosure and in the absence of fraud (the circumstances there in issue), there was no remedy in negligence simply premised on the duty of good faith which existed between insurer and assured: see issue 8 at 790/805. Since I shall have to go further into the reasoning in Banque Keyser for the purpose of the insurers' submission that its rationale does not extend to a case of misrepresentation as distinct from non-disclosure but on the contrary runs in favour of promoting a remedy in negligence in the case of misrepresentation, I shall for the present simply state the conclusion there of Slade LJ (giving the judgment of this court) to be found at 801A/D:

"We do not think that the nature of the contract as one of utmost good faith can be used as a platform to establish a common law duty of care. Parliament has provided that in the case of marine insurance the consequence of a failure to disclose a material fact, and by inference the only consequence, is that the contract may be avoided. It is not suggested that the consequences in non-marine insurance should be different. In those circumstances it is not, we think, open to the court to assist the banks by providing a supplementary remedy in tort…What the banks cannot do in our judgment is to invoke the nature of the contract as one of good faith, with its limited contractual remedy, to bridge the gap so as to give them a cause of action in tort. The error in the submission made for the banks is that it ignores the nature of the special obligation imposed in contracts of the utmost good faith. The obligation does not, if we are right, create a duty to speak for breach of which the law attaches the consequences which flow from an ordinary breach of duty, whether statutory or otherwise. It is a rule of law which provides, and provides only, that certain stated consequences (namely the assured party's right to avoid the contract) will follow if utmost good faith be not observed."

53. Aikens J regarded this rationale as applying not only to negligent non-disclosures but also to negligent misrepresentations (see at para 88): "in each case there is no right at all". He held that special facts would be needed to create a duty of care as between an insurer and an assured and that none were pleaded. That view is put in issue on the insurers' cross-appeal, and I will consider it below.

Three points of principle

54. There are therefore three points of principle which are raised by the appeals in this court, which I propose to deal with in advance of arguments of construction. The first is the one I have just mentioned, whether the rationale of Banque Keyser applies to a claim in tort for negligent misstatement as well as to a claim in tort for negligent non-disclosure. The second is whether it is possible at all to exclude liability for an agent's fraud, or whether there is a rule of law against that possibility. The third is whether Aikens J was right to apply the Canada Steamship rules even if not directly at any rate by way of analogy in the case of what he called the unitary and absolute duty of good faith.

Contracts of or for insurance

55. Before turning to those three points of principle in turn, I need to say something about the distinction which is drawn throughout the judgment below between contracts of and for insurance. There is only one contract for insurance in this case, and that is the line slip facility. There is no appeal from Aikens J's decision that the line slip facility was not in itself a contract of insurance and therefore was not itself a contract of the utmost good faith.
56. The insurers' concern is that if they can attack the line slip facility itself, either by rescinding it or by claiming damages for being induced to enter into it, then they might be able to protect themselves against having entered into the three contracts of insurance which began life as declarations under it, even in circumstances where they were unable to vindicate any remedy in respect to those contracts of insurance themselves. Quite what remedies might in practice be open to the insurers in respect of the line slip facility on the alternative hypotheses that remedies were or were not available to them in respect of those three contracts of insurance were not explored in argument (see, for instance, Aikens J at para 91 of his judgment). Nevertheless, the argument has encompassed the possibility that the issues debated on appeal may produce different answers dependent on whether the contract in question was a contract of insurance or a contract for insurance.
57. It is not easy to restate compendiously the effect of the judge's judgment as to where the results reached in relation to each type of contract differ and where they are the same, but I think that it may be put as follows. (1) There is a right of avoidance in relation to contracts of insurance. (2) Where there is a fraudulent or negligent misrepresentation, there is a right of rescission in relation to a contract for insurance. (3) Where there is a fraudulent misrepresentation, damages are available in relation to contracts for insurance and, it is assumed without deciding (para 86 of the judgment below, citing Banque Keyser at 774A/B), in relation to contracts of insurance as well. (4) Damages under section 2(1) of the Misrepresentation Act are available in relation to both types of contract. (5) Damages for the tort of negligent misstatement are available in relation to contracts for insurance, but not (without special facts) in relation to contracts of insurance. (6) The Canada Steamship rules of construction apply directly only to contracts for insurance, not contracts of insurance. (7) Even so, similar results are achieved as a matter of construction of the Truth of Statement clause in both contexts, viz to the effect that they do not protect against avoidance or rescission, or any liability in damages, where there has been fraud or negligence. (8) In the circumstances, it may not matter that damages are available for negligent misstatement only in the case of the line slip facility, since in any event there is a right to avoid the contracts of insurance.
58. It is at first sight somewhat paradoxical that there should be a greater array of possible remedies (at any rate outside the case of pure non-disclosure) in relation to a contract for insurance than in the case of a contract of insurance: after all, the latter contract is the same as any contract save that it is also a contract of the utmost good faith. Moreover, it is in theory possible that the presence or absence of a possible remedy might affect the construction of an exclusion clause such as the Truth of Statement clause. Nevertheless, in my judgment it would be astonishing, or at any rate anomalous, if that clause were to have a different meaning or effect depending on whether it was the line slip facility or a contract of insurance made under it which was in question. After all, the Truth of Statement clause was included in the line slip facility for the very purpose that it could be incorporated in any contract of insurance made pursuant to it. The reference to the Questionnaire, the fact that it speaks to disclosure and non-disclosure and (passim) to avoidance, demonstrate that it was drafted for the purpose of and for inclusion in a contract of insurance. Therefore, prima facie it is the meaning and effect which it would have in the contract of insurance context that should guide its construction. I shall proceed provisionally on that basis, and reconsider the matter towards the end of this judgment.

The tort of negligent misstatement

59. The question raised by the first point of principle is this: Is the tortious remedy of damages for negligent misstatement available, subject to exclusion, in the contracts of insurance found in this case?
60. On behalf of the insurers Mr Cooke submits that it is, and that what this court said in Banque Keyser is to be limited to the case there under discussion, which was a case of mere non-disclosure. He submits that Slade LJ recognised the potential existence of a duty of care not to make negligent misstatements at various stages of the court's judgment. Thus Mr Cooke emphasised the following aspects of that case. What the claimant assured there complained about was the insurer's failure to disclose the broker's deceitful representation that full cover had been completed on an earlier tranche of the lending transaction. When the borrowers defaulted, the claimant's insurance became worthless because the insurer was able to rely on a fraud exclusion clause in the policies. It was a pure case of non-disclosure, for the court rejected any submission to the effect that the insurer's failure to disclose the broker's fraud amounted to a representation as to the honesty of the broker. In those circumstances Mr Cooke relied on the following passages:

"Now if the facts of the present case had been that Mr Dungate had carelessly represented to the banks that Mr Lee was an honest man, the Hedley Byrne principles would, in our judgment, plainly have applied, subject to the conditions we have mentioned, even though the misrepresentation was made in a pre-contractual setting" (at 793D).

61. The conditions mentioned were those of a voluntary assumption of responsibility by the defendant to the claimant and the reliance by the claimant on the defendant's misstatement (see at 792H). Thus Slade LJ said (at 794B):

"Accordingly, if Mr Dungate had carelessly misrepresented to the banks that Mr Lee was an honest man, the Hedley Byrne principles would, in our judgment, have enabled Kusa to succeed in its claim for negligence if, though only if, it could have proved a voluntary assumption of responsibility in respect of such representation of Mr Dungate coupled with reliance by the banks."

62. On the facts the court there was unable to find any voluntary assumption of responsibility either in the established business relationship between the parties, or in the relationship of the utmost good faith engendered by the insurance context; nor could it find any reliance. Mr Cooke, however, submitted that, in the case of a positive misrepresentation, the appropriate special relationship where responsibility is assumed could readily be inferred from the professional context of insurance broking and the doctrine of utmost good faith which accompanies it; and reliance can also be much more easily inferred. What destroyed the claim in Banque Keyser was simply that the duty of good faith could not by itself create a misrepresentation out of a duty to speak which existed for the separate purposes of the right of avoidance: see at 801C/D cited above at para 52.
63. This is in my judgment a formidable argument, as far as it goes. I accept that the relevant passages in the judgment of Slade LJ repeatedly emphasise the importance of the fact that the court there was dealing only with a failure to speak, not with a misrepresentation. For instance, to add to the citations emphasised by Mr Cooke, there is the passage at 797D/798G, where Slade LJ said –

"For better or worse, our law of tort draws a fundamental distinction between the legal effects of acts on the one hand and omissions on the other…The example given by Lord Keith [in Yuen Kun Yeu v. Attorney-General of Hong Kong [1988] AC 175, 192] is of a failure to prevent physical harm. But the same reluctance on the part of the courts to give a remedy in tort for pure omission applies – perhaps even more so – when the omission is a failure to prevent economic harm. As will appear below, a corresponding distinction is drawn by the law of contract which in general imposes no liability by virtue of a failure to speak, as opposed to a misrepresentation…The general principle that there is no obligation to speak within the context of negotiations for an ordinary commercial contract (though qualified by the well known special principles relating to contracts uberrimae fidei, fraud, undue influence, fiduciary duty, etc.) is one of the foundations of our law of contract, and must have been the basis of many decisions over the years."

See also further passages at 802E, 802H/803B and 804G. Moreover, I would also accept that the court's unwillingness there to find a special or sufficient relationship for the assumption of responsibility itself applies to a situation where what was being talked about was the assumption of a responsibility to speak and thus also to be careful in what was not said and not merely to be responsible for what was in fact said (see at 799G/H and 800F/801D). It may be said, therefore, that in essence the banks failed for want of a misrepresentation.

64. In the present case, however, there are 58 pages of misrepresentations pleaded in the insurers' Schedule 2 to their amended particulars of claim.
65. Thus far, therefore, Banque Keyser is an authority which can properly be relied on for dicta to the effect that in a case of a positive and negligent misstatement the position is different, and there may well be room to find the conditions necessary for Hedley Byrne liability in the negotiation of a contract of insurance.
66. However, there are three other strands in the Banque Keyser judgment which press in the opposite direction. The first is that, in an established relationship with which the law is perfectly familiar, that of insurer and insured, there is simply no need to find a duty of care which has not previously been held to exist. As Slade LJ said (at 801E):

"…we are not satisfied that justice and reasonableness imperatively require the finding of duty of care owed by Hodge to Kusa. This is not a case in which the denial to the banks of a cause of action in negligence will mean that the banks are left without any remedy at all…"

Slade LJ pointed out that in that case the broking firm would have been vicariously liable for its manager's deceit, just as in this case there are claims against Heaths. The reference to "justice and reasonableness" in Slade LJ's 1988 judgment predates the applications of the relevance of such considerations to the Hedley Byrne context in the +mportant cases of Smith v. Bush [1990] 1 AC 831 and Caparo Industries Plc v. Dickman [1990] 2 AC 605; but it must reflect a similar line of thought. In Caparo Lord Bridge summed up these considerations in his statement (at 618) that "the situation should be one in which the court considers it fair, just and reasonable that the law should impose a duty of a given scope on the one party for the benefit of the other".

67. The matter goes well beyond the availability of rights against other parties for in most cases of pre-contractual misrepresentation the right of the insurer to avoid the contract of insurance is all that is needed to protect him. If there is a problem at all, it lies in the highly unusual case where the breach of good faith is on the other side, as in Banque Keyser, and it is the assured who is seeking a remedy. It is of course well established that the duty of good faith is mutual and reciprocal, as Lord Mansfield himself recognised in the seminal case of Carter v. Boehm (1766) 3 Burr 1905 at 1909/1910 and as is now also recognised in section 17 of the MIA 1906. Where, then, the insurer is in breach, avoidance of the contract is perhaps unlikely to provide a satisfactory result for the assured, who would thus lose his insurance: but it is not possible to be confident about that, since examples of a breach by an insurer are so rare, and the usual example cited, for instance by Lord Mansfield and others, of the insurer taking a premium for a risk which he knows no longer exists, is one where avoidance and the recovery of the premium will entirely satisfy the assured. The fact remains that, save perhaps in a highly unusual, even unprecedented case, the duty of care adds nothing to the right of avoidance. Thus there is no need for a new remedy in the present case, where the current dispute is generated not by any lacuna in available remedies but by the insurers' concern to demonstrate that an exclusion clause has not covered all available heads of damage.
68. Secondly, however, there is the theme which runs throughout Banque Keyser which is that the duty of good faith which the law had developed especially for contracts of insurance provides a remedy only in avoidance and not in damages. It seems to me to follow that if the established remedy in this context grants no remedy in damages, even though it encompasses misrepresentation as well as non-disclosure, and even though the remedy of avoidance is available whether or not the misrepresentation is negligent, then it ought to require very special factors to make it just or reasonable to superimpose an additional remedy for the narrower case of a negligent misrepresentation made in breach of a different duty, a duty of care. This again ties in with the insistent query: What is the need?
69. The third strand is Banque Keyser's insistence on a proper and pleaded foundation for the assumption of a duty of care (or exceptionally at least a proper case for finding in law if not in fact a situation where a defendant should be treated as if he had assumed such a responsibility, see at 797B). The relationship of insurer and assured may create by law a duty of good faith, but that is not the same as finding in that relationship, either as a matter of fact or law, an assumption of a duty of care. If it did, Slade LJ could not have said (at 801A) – "We do not think that the nature of the contract as one of the utmost good faith can be used as a platform to establish a common law duty of care." If therefore the assumption of a duty of care is not to be found automatically in the relationship of good faith, where is it to be found? It is perhaps prudent to allow the possibility that it might arise on special facts, for instance where, exceptionally, an insurer had undertaken the task of advising the proposer about the suitability or scope of cover (see CLARKE, The Law of Insurance Contracts, at para 23-1B). In the ordinary case, however, this strand of the Banque Keyser judgment again teaches that the standard duty is that of good faith and the standard remedy that of avoidance.
70. The question might be raised whether a remedy in damages might be a more flexible instrument than the remedy of avoidance. It might be said that the remedy provided by the duty of good faith was developed at a time when damages were not available for negligent misstatement, and that the new remedy should fill a lacuna which the old law left unprovided for. In principle, flexibility in remedies is a matter to be encouraged. It is difficult to visualise, however, that such flexibility is either needed or useful where one is considering the position from the point of view of negligence which has caused an insurer to insure. Insurance is a matter of risk: an insurer is either satisfied to be on risk or not. If he discovers that he has been misled, he needs to make up his mind whether he is going to avoid his contract or not. He is put to his election. If, however, he has a remedy in damages as well, he could in theory waive his right to avoid without thereby waiving his right in damages. That would be a most unsatisfactory situation. From the point of view of an assured who has been caused by the insurer's negligence to enter a contract of insurance, the position is probably more complex, for if he had not been misled he might have been able to buy a better or more suitable policy in time: that is the example contemplated above of an assured for whom avoidance is not in itself a solution. It seems to me, however, that the case of such an assured is so rare, that it would be wrong to base any general decision upon it. It is sufficient to say that, in an exceptional case, it might be possible to find an assumption of responsibility proved. For the rest, most assureds have an insurance broker to look to in the case where an unsuitable insurance has been purchased.
71. There is this further consideration. As matters stand with the traditional remedy of avoidance, there is no need to consider whether a misrepresentation is negligent or not. Any material misrepresentation within the terms of section 20 of the MIA 1906 will suffice. A duty of care on the other hand would invite litigation about the boundaries of negligence. It is difficult to see to what advantage. This factor is connected with another, which is that it can often be difficult, especially against the background of the duty of good faith requiring disclosure, to distinguish between a misrepresentation and a non-disclosure: in Pan Atlantic Insurance Co Ltd v. Pine Top Insurance Co Ltd [1995] 1 AC 501 at 549G Lord Mustill said that "in practice the line between misrepresentation and non-disclosure is often imperceptible" (a citation for which I am grateful to McDonald Eggers & Foss, Good Faith and Insurance Contracts, 1998, at para 7.16).
72. In Banque Financière the House of Lords upheld the result in Banque Keyser on the different ground that the insurance was in any event doomed by reason of the borrower's fraud and the fraud exclusion clause. Therefore there could have been no loss caused by any breach of a duty of care in any event. In the circumstances their Lordships did not have to consider the reasoning in the court of appeal on issue 8. Lord Templeman merely confined himself to saying (at 275B) that –

"The person who makes a negligent mis-statement which is relied upon by a negotiating or contracting party may be liable in damages. In the present case there was no negligent mis-statement and the silence of Mr Dungate did not amount to an assertion that Mr Lee was trustworthy…"

There was no submission before this court that the speeches in the House of Lords throw any further light on the problem.

73. The parties were content to debate this issue almost exclusively by reference to Banque Keyser itself. In the circumstances, a wider range of authority was not invoked. Such a range of authority, particularly from cases decided in the decade after Banque Keyser, might well, however, have been looked to for the purpose of illustrating the importance in such a case as this of the three strands which I have derived from that judgment.
74. In sum, my view is that Aikens J was right to conclude that the rationale of Banque Keyser ultimately supports the conclusion that, at any rate in the usual case, there is no duty of care not to make negligent misstatements that is implicit in the relationship of insurer and assured or proposer. Whether a special case can be made out in any particular situation is a matter on which I would prefer to reserve my opinion. It is not suggested that a special case exists on the pleadings here.

Excluding liability for an agent's fraud

75. It is common ground that there is a rule of law founded in public policy which prevents the exclusion of liability for one's own fraud. But is it possible to exclude liability for an agent's fraud?
76. Mr Cooke accepts that it is possible to exclude liability for an agent's dishonesty in the performance of a contract, but submits that it is not possible to do so for an agent's dishonesty in the negotiation for a contract. For these purposes there is no difference between a principal and his agent. In either case, the principal seeks by his exclusion clause to keep the benefit of a contract which has only been obtained by fraud for which he is, exclusion apart, responsible. Lord Grabiner on the other hand submits that, whatever may be the difficulties either as a matter of business presentation or as a matter of clarity in drafting a clause which will be effective as a matter of construction, it can be done.
77. The issue to a large extent turns on the proper interpretation of three decisions in the House of Lords, on which Mr Cooke relies, namely S Pearson & Son Limited v. Lord Mayor of Dublin [1907] AC 351, Mair v. The Rio Grande Rubber Estates Limited 1913 SC (HL) 74, and Boyd & Forrest v. The Glasgow and South-Western Railway Company 1915 SC (HL) 20.
78. In his judgment below, Aikens J concluded (at para 35) that "in principle and on the balance of the case law" it is possible to exclude liability for the fraudulent misrepresentations of an agent.
79. Pearson concerned a contract for the construction of sewage works. Dublin's engineer put forward plans concerning a wall which were inaccurate: there was evidence that he knew the plans to be false. Dublin however was innocent of any complicity. The false information made extra work necessary. Pearson sued Dublin in deceit for the cost of the additional work. Dublin relied on a clause in the contract which provided that the city corporation did not hold itself responsible for the accuracy of the information provided and that no charges for extra work would be allowed in consequence of any such inaccuracy. The trial judge therefore withdrew the case from the jury. The House of Lords held that the matter should have gone to the jury, since Dublin could not rely on the clause.
80. The dispute in this court was whether the ground on which the House of Lords remitted the case to trial was a rule of law or a matter of construction or both.
81. Pearson is complicated by a matter on which the law has subsequently been clarified. It was argued that because the principal was innocent, therefore, at any rate in the light of the exclusion clause, there was no fraud or no liability for fraud. The argument was rejected. Lord Loreburn LC said (at 354) –

"The principal and the agent are one, and it does not signify which of them made the incriminated statement or which of them possessed the guilty knowledge"

and the Earl of Halsbury (at 357) said –

"If it was supposed to decide that the principals and agent could be so divided in responsibility that – like the schoolboy's game of "I did not take it, I have not got it" – the united principal and agent might commit fraud with impunity, it would be quite new to our jurisprudence"

and at 358/9 he expressly associated himself with Lord Loreburn's philosophy that –

"it matters not in respect of principal and agent (who represent but one person) which of them possesses the guilty knowledge or which makes the incriminating statement."

82. This language was subsequently misunderstood so as to allow a finding of fraud even where both principal and agent were innocent, as for instance where the agent was inaccurate but did not know he was, and the principal knew the true facts but had not authorised the agent's statement. That heresy was, however, exposed and eliminated in Armstrong v. Strain [1952] 1 KB 232. The true position is, I would have thought, that where the guilty agent is authorised by the innocent principal, the principal's liability is not personal but vicarious. It may be, however, that in 1907 the vicarious liability of a principal for the fraud of his agent committed within the scope of his authority was not free of the doubts and difficulties which only a few years later Lloyd v. Grace Smith & Co [1912] AC 716 finally eliminated. It is true that at 358 the Earl of Halsbury cited a passage from the judgment of Willes J in Barwick v. English Joint Stock Bank (1867) LR 2 Ex 259 at 265 which is in line with the doctrine settled in Lloyd v. Grace Smith; but the speeches in the latter case show that Barwick had itself been misunderstood, for instance in Ruben v. Great Fingall Consolidated [1906] AC 439 at 446, which was heard only the year before Pearson.
83. At any rate it is, I think, in the light of the rejected argument that the innocent principal can escape liability for his agent's fraud that some of the dicta in Pearson can best be understood.
84. Lord Loreburn gave the first speech. At 353/4 he said:

"Now it seems clear that no one can escape liability for his own fraudulent statements by inserting in a contract a clause that the other party shall not rely upon them. I will not say that a man himself innocent may not under any circumstances, however peculiar, guard himself by apt and express clauses from liability for the fraud of his own agents. It suffices to say that the clauses before us do not admit of such a construction. They contemplate honesty on both sides and protect only against honest mistakes."

85. It seems to me that Lord Loreburn there is contrasting the rule of law in the case of a person's own fraud with a rule of construction which is applicable to his agent's fraud. Ultimately he decides the case ("It suffices to say…") on the rule of construction. When he goes on to say that "The principal and the agent are one…", I do not think that he is reverting to a rule of law under which it does not matter whether the exclusion is of one's own fraud or of an agent's fraud: he is rather adverting to a different strand in the argument relating to what is now recognised as vicarious liability.
86. The Earl of Halsbury, by contrast, appears to have thought of the matter purely in terms of a rule of law, for he said (at 356):

"The action is based on the allegation of fraud, and no subtilty of language, no craft or machinery in the form of contract, can estop a person who complains that he has been defrauded from having that question of fact submitted to a jury."

87. Lord Ashbourne, by contrast again, appears to have looked on the matter as one of construction, for he reasoned (at 360) that:

"Such a clause might in some cases be part of a fraud, and might advance and disguise a fraud, and I cannot think that on the facts and circumstances of this case it can have such a wide and perilous application as was contended for. Such a clause may be appropriate and fairly apply to errors, inaccuracies, and mistakes, but not to cases like the present."

88. Mr Cooke and Mr Gross relied on the earlier part of that dictum as supporting a rule of law that puts any distinction between exclusion of one's own and an agent's fraud to one side. However I do not read it in that light. To the extent that it is expressive of a rule of law at all, it is raising the different and well known problem, which is for instance the subject matter of section 3 of the Misrepresentation Act, of whether it is possible by contract to exclude a misrepresentation, in particular a fraudulent one, which induces the contract in the first place. If a contract is impeached by fraud, can any exclusion save it? What Lord Ashbourne is saying is that for an exclusion to have any chance of effectiveness it must at least be clearly seen to apply to fraud as distinct from innocent errors.
89. Lord Macnaghten merely said that he agreed in the motion proposed (at 360). Lord Robertson (at 362) and Lord Collins (at 368) did the same.
90. Lord James of Hereford, like Lord Ashbourne, appears to have been influenced by the concern that fraud may impeach the contract itself, but, unlike Lord Ashbourne, he was therefore inclined to a rule of prohibition rather than one of construction. Thus at 362 he said:

"The protecting clause might be inserted fraudulently, with the purpose and hope that, notwithstanding its terms, no test would take place. When the fraud succeeds, surely those who designed the fraudulent protection cannot take advantage of it. Such a clause would be good protection against any mistake or miscalculation, but fraud vitiates every contract and every clause in it. As a general principle I incline to the view that an express term that a fraud shall not vitiate a contract would be bad in law, but it is unnecessary in this case to determine whether special circumstances may not create an exception to that rule."

91. Thus ultimately he did not determine the case on a rule of law. What then was the ground of his decision? It is not plain, but it can only be that the clause did not as a matter of construction cover fraud in any event, as distinct from "any mistake or miscalculation". I would add that his concern that an exclusion clause might itself be inserted fraudulently is if anything looking to a personally fraudulent rather than innocent principal.
92. Finally, Lord Atkinson, was also inclined (at 365) to the view that "such a contract" (ie one which purported to exclude liability for the frauds of both principal and agent, or a "contract to submit to a fraud") "would be illegal in point of law", but he likewise ultimately decided the case on a point of construction –

"…I do not think that the articles of the contract relied upon can, on their true construction, be held to have fraud, whether conscious or unconscious, within their purview or contemplation, or to apply at all to such a case of fraud as the present is alleged to be. They were, I think, intended to apply, and do apply, to inaccuracies, errors, and mistakes, or matters of that sort, but not to fraud, whether of principal or agent or of both combined."

93. He then went on to consider the different submission that, by the terms of the contract, the agents were not authorised to make any statements, so that the inaccurate plans were presented by the city corporation itself, which was innocent and therefore entitled to rely on its own exclusion of liability. He rejected that submission on the ground that the engineers were, by the express terms of the contract, constituted the city's agents to draw up and communicate the plans to the contractors. That is a point which seems to me to go to the ambit of the city's grant of authority and thus to the separate point of vicarious liability: as Lord Atkinson concluded (at 367): "I think, therefore, that the defendants are responsible for the fraud of their engineers in framing the plans, if fraud there were".
94. Thus of the five members of the House who gave a reasoned speech, only the Earl of Halsbury appears to have decided the case on a rule of law, the other four decided it ultimately on a rule of construction, two of those four (Lords James and Atkinson) were inclined without deciding to favour a rule of law, but the other two (Lords Loreburn and Ashbourne) did not speak to similar effect, and Lord Loreburn specifically contemplated as a possibility a different rule for the exclusion of one's own fraud from the rule for the exclusion of an agent's fraud. Moreover, in as much as there is any talk of an absolute rule of law, it is difficult to disentangle three separate strands: one is that it is never possible to exclude fraud; one is that there is a special problem where the exclusion is of fraud in the negotiation of a contract and particularly if it can be posited that the exclusion clause is itself part of the fraud; and one is that it is in any event impossible to distinguish between a principal and his agent because "the principal and the agent are one".
95. In my judgment, it can be concluded that Pearson is not a decision in favour of a rule of law that would prevent the exclusion of liability for the fraud of an agent, that the balance of views expressed are fairly evenly divided on the question at issue in this court, and that the views expressive of a rule of law are, in my respectful opinion, to a certain extent undermined both by the variety of strands of argument and by an uncertain appreciation that the liability of an innocent principal for the fraud of his agent is only a vicarious one.
96. It remains to be seen how these conclusions fare in the light of the two succeeding cases.
97. In Mair the claimant had bought shares in the defendant company on the basis of a prospectus which contained a clause to the effect that the directors expressly refrained from committing themselves to the truth of statements made in a report from one of their number, a Mr Littler, which was quoted in the prospectus. The claimant alleged that the report was fraudulent and claimed rescission. The report was a critical document, for it expatiated on the abundance of rubber and likely profitability of the venture. The issue was whether there was a case to go to trial. It was assumed that Mr Littler had been fraudulent, but that the company was innocent of his fraud. Nevertheless the House of Lords held that there was a case for trial, since the company had authorised and adopted Mr Littler's report. Lord Haldane LC (at 77) cited Lloyd v. Grace Smith as embodying the broad principle "which is not now disputable" that a corporation is liable for the fraud of its agent and continued –

"which applies at all events in the absence of some much more extended and precise limitation of responsibility than was expressed in the present case."

That, it seems to me, is a decision on a point of construction.

98. The other two members of the House, however, (there were only three), clearly went further than that. Lord Shaw of Dunfermline began by accepting that it was possible for directors to dissociate themselves from the statements of one of their number. Thus (at 81) he said –

"This would make a peculiar document; but it might be done. Yet the doing of such a thing would require to be in the most clear and unambiguous terms, so as to constitute a specific warning…"

He found, however, that in this case (ibid) –

"there was no such dissociation…On the contrary, the directors expressly based the prospectus upon his report, and they and the Company stood pledged to the position that they held it out to be true."

That again, it seems to me, is a decision on construction.

99. Lord Shaw went on, however, to say (at 82) that it would be contrary to good faith to hold a plaintiff to a bargain which had been induced by (even an innocent) misrepresentation:

"Fraud is not far away from – nay, indeed it must be that it accompanies – a case of any defendant holding a plaintiff to a bargain which has been induced by representations which were untrue; for it is contrary to good faith and it partakes of fraud to hold a person to a contract induced by an untruth for which you yourself stand responsible. It is elementary that a party cannot take advantage of a benefit derived from a contract sprung out of his own fraud, and I think it is equally sound that a party cannot take a benefit from a contract sprung out of a falsehood which he has placed before the other party as an inducing cause. And in applying this rule to a case of the inducing misrepresentation being made by one of a body of directors and used by his associates, the company, as a means of obtaining a contract or raising money, the principle at stake undergoes a legitimate and proper amplification.

"But this record on the part of the pursuer alleges that the statements made were not only false but fraudulent…"

100. I have said above that in this passage Lord Shaw is applying his doctrine of the requirements of good faith even to a case of innocent misrepresentation. While this is not entirely clear, it is I think the necessary result of his beginning with the analogy of fraud founded on seeking to keep the benefit of a contract induced by an untrue representation and of his ending with the remark that the allegation is not merely one of a false statement but a fraudulent one. In this passage, therefore, Lord Shaw is plainly affirming a far-reaching position. However, I do not regard it as being an alternative ground to that of construction, for it is not expressed as such, and it is difficult to see how it would be consistent with his earlier statement that a clause of dissociation could have been expressed, however difficult it was in practice to achieve. What, therefore, it seems to me that Lord Shaw is saying is that, where the directors had adopted the report of one of their number and where the attempt at dissociation had failed, they could not rely on their personal innocence to win the day. Lord Shaw does not, however, go so far as to say that, whatever exclusion might apply, it could not be effective to protect against an agent's fraud.
101. Finally, Lord Moulton stated that an exclusion for the misrepresentation of an agent (even a director) could be framed (at 83), but that the courts below had overlooked the fact that the allegation in the case was that the statements in question had been "fraudulent as well as false" (at 84). He continued –

"We have, therefore, an allegation of a fraudulent statement made by a director of the Company, used by the Company for the purpose of obtaining subscriptions, and effective in inducing the pursuer to take the shares. Now it is elementary law that no person can take advantage of the fraud of his agent. It would therefore be contrary to good conscience that this Company should hold the pursuer to [such] a bargain…"

That appears not to have been a decision on construction, but on a rule of law, albeit on the facts of that case. In the circumstances it is not easy to say what the true ratio of that case was.

102. The facts in Boyd & Forrest were in all material respects identical to those in Pearson, save that it was ultimately found that there was no misrepresentation at all. Again the employer was regarded as innocent, his agents were accused of fraud, but in the end, even on the hypothesis that there had been any misrepresentation, were found to be innocent. For good measure, reliance was negatived as well. The exclusion clause stated that the employer would not be liable in respect of any inaccuracy or error in the information or specification provided to the contractor. In the circumstances, with fraud so firmly rejected on so many grounds, Pearson was mentioned by only one member of the House, Lord Shaw of Dunfermline, who was also the only member to consider what the position might have been if fraud had been proved. He was clearly of the view that fraud could not be excluded as a matter of law, and cited Lord James from Pearson for the opinion that "fraud vitiates every contract and every clause in it", but did not expressly consider whether there was for these purposes any difference between the fraud of a principal and that of an agent. Otherwise, he cited from Lord Loreburn and Lord Ashbourne on the question of construction. His view as to the rule of law is put as follows (at 35/6) –

"For it is a sound principle, to which I do not hesitate to give adhesion, that the terms of a contract, however far they may extend in putting a burden of risks and speculation upon the contractor, cannot be founded upon as a protection against fraud of either party. It could not be so, in my opinion, even although the contract were expressed in such a particular. The law excludes from the range of agreement what is openly contrary to legal principle and to honest dealing."

As far as it goes, that statement of opinion is in the insurers' favour.

103. In this inconclusive state of the authorities, I have considered the issue as a matter of principle. It is common ground that the fraud or dishonesty of servants and agents in the performance of a contract can be excluded. It is also common ground that a principal can protect himself from his agent's fraud by making it clear that his agent has no authority to act in relation to a certain contract. Thus it was accepted by Mr Cooke and Mr Gross that if the Truth of Statement clause were to be construed as excluding Heaths' authority to speak on behalf of Chase, then its fraud could not affect Chase. It is also not disputed that, absent fraud, it is a matter of construction and not a rule of law whether a clause such as the Truth of Statement clause succeeds or not, for instance, in excluding the right of avoidance. Thus, one of the essential issues, when it comes to construction, is whether the Truth of Statement clause covers negligent and/or fraudulent misrepresentation or non-disclosure as well as innocent examples of breach of Chase's obligations. There is no submission that the clause does not at least cover innocent misrepresentation or non-disclosure. Thus Lord Shaw's views, if they were to be interpreted as standing separate from his earlier conclusion as to the ineffectiveness of the exclusion clause in Mair, were not adopted by the insurers.
104. In my judgment there is a real difference between an exception for one's own personal fraud and an exception for the fraud of an agent. It is plainly contrary to public policy to be able to enforce a contract obtained by one's own fraud or to avoid a liability for one's own fraud. Such self-justifying exceptions are analogous to a contract to commit a fraud or to an insurance against the consequences of one's own fraud or crime. Such contracts will not be permitted. It is another matter, however, to insure against liability for the consequences of one's servant's or agent's fraud or crime: see Lancashire County Council v. Municipal Mutual Insurance Ltd [1997] QB 897. I do not see why it is not possible, however difficult it may be in practice, for a principal to exculpate himself from his purely vicarious liability for fraud.
105. It is true that a contract induced by an agent's fraud raises a particularly difficult problem. Whereas the liability produced by the agent's tort of deceit may be referred to as vicarious, the insistence of retaining the benefit of a contract procured by the agent's fraud may be another matter. When in the context of such a contract the principal makes claim to his rights under the contract, he is not seeking to evade vicarious liability but directly to enforce his rights under it. Moreover, those rights may have only been obtained with the help of the agent's fraud, and in any event the exclusion clause may be said to be as susceptible to rescission as the rest of the contract. In such circumstances the survival of the contract may on any view look unattractive. Those problems, however, are no different in principle (even if they are in degree) in the case of fraud as distinct from innocent (or negligent) misrepresentation. It is the problem which section 3 of the Misrepresentation Act is designed to deal with, specifically addressing both liability and remedy. Outside that solution, however, which allows the court to put the claimant to proof that the term satisfies the Unfair Contract Terms Act requirement of reasonableness, the common law would (at any rate outside fraud) permit a properly worded exclusion clause to be effective: see the House of Lords cases cited above. In HIH v. New Hampshire this court decided that clause 8 was effective to exclude the remedy of avoidance (see at para 182; fraud was not there in issue). I do not think that reliance on an exception clause for the purpose of excluding a remedy of avoidance which arises because of an agent's fraud is in principle ultimately different from reliance on such a clause for the purpose of excluding a remedy of avoidance which arises because of the principal's own (non-fraudulent) misrepresentation. It is a matter of construction. Provided that the clause is sufficiently clear, and the contract partner properly warned of the extent of the exclusion sought, the survival of the contract reflects the autonomy of the parties. On that basis its survival is not unacceptably unattractive (or if it is, that may be a matter for section 3). Unattractive or not, public policy does not demand that the parties' autonomy be disregarded.
106. Of course, if the principal can be implicated in the agent's fraud in some way, then the fraud can become that of the principal himself. The dicta of Lord Shaw and Lord Moulton in Mair are possibly to be explained on that basis.
107. Having considered the authorities and pondered the issue as a matter of principle, I turn finally to ask what the textbooks say about this problem. The parties have helpfully included in the materials put before the court the relevant extracts from the following works. In CHITTY on Contracts, 28^th ed, 1999, at para 6-129 Pearson is cited for the proposition that a person could not contract out of liability for fraud inducing the making of a contract with him "at least where the fraud was his own". HALSBURY's Laws of England, 4^th ed reissue, 1998, Vol 9(1) at para 800 cites Pearson only for the proposition that "no exclusion clause can protect a party from the consequences of his own fraud". TREITEL, The Law of Contract, 10^th ed, 1999, at 222 cites Pearson for two propositions, neither of which supports the insurers' submission on a rule of law, viz that "Such a clause would not protect a party from liability for his own fraud and an exemption clause cannot by general words exclude liability for fraud". ANSON's Law of Contract, 27^th ed, 1998, notes that "an exemption clause can never exclude liability for personal fraud", again citing Pearson. GLOAG, The Law of Contract, 2^nd ed, 1929, was specifically referred to as containing the most extensive treatment of Mair, at 465: but I find nothing there inconsistent with my consideration of that case above. Moreover at 466 GLOAG considers Pearson from the point of view of what it has to say about construction and continues:

"But it has never been decided that an express contract that a principal is not to be responsible for his agent's fraud is illegal or ineffectual; and it is conceived that there are cases where it might be a reasonable precaution. When a contract has to be carried out by subordinates, it may often be impossible for the principal to exercise any proper control over them, while it may be easy for the other party to verify the accuracy of their statements."

It is only then at footnote 6, following the remark that "The question is considered in Pearson…", that GLOAG adds a reference to Dig ii 14, 27: "Illud nulla pactione effici potest, ne dolus praestetur" (That can not be effected by any contract, in case fraud is achieved).

108. In none of these citations do I find anything to support the insurers' submission; on the contrary. Mr Cooke in particular urged reliance on Spencer Bower, Turner and Handley, Actionable Misrepresentation, 4^th ed, 2000, at para 189, where the learned authors cite both Lord Loreburn and the Earl of Halsbury from Pearson but, to my mind, without assistance, for there is no discussion of the question as to whether the quotations are inconsistent with one another on the question of the possibility of excluding liability for an agent's fraud. The learned authors go on to discuss section 3 of the Misrepresentation Act, and suggest that it could never apply to fraudulent misrepresentations since in such a case there can be no "term which would exclude or restrict" liability or remedies. On that basis, their text is inconsistent with the possibility that there could be any protection against an agent's fraud, and supports the insurers' submission. But it is the only treatise to do so.
109. In conclusion, there is no binding authority on the issue. Such dicta as exist, all of it from judges of the highest distinction, are evenly balanced, but, in the light of the different strands of argument under consideration, of uncertain guidance. In principle, I do not see the same public policy compulsion to strike down the exclusion of an agent's fraud as in the case of one's own fraud. With one exception, the treatises which have been consulted appear to read the authorities in the same way. In my judgment there is and should be no rule of law making it impossible to exclude one's own liability or another's remedy for the fraud of one's agent. It may be different, however, if the principal is in some way implicated in the agent's fraud or if, a concern touched upon in the authorities, the exclusion is itself part of some fraudulent scheme.
110. Having said that, and having considered the argument in some detail in deference to the extensive submissions made by counsel on this issue, it remains the position that for fraud to be excluded requires the clearest possible wording. Parties to a contract assume honest dealing with one another. To warn a potential contract party of an agent's possible fraud would be a remarkable clause to find in any agreement. That is no doubt one reason why the point of principle has never previously had to be decided. There is no reference to the word "fraud" or its equivalent anywhere in the Truth of Statement clause. It remains to be seen where the argument on construction, which is considered below, ends up.

The Canada Steamship approach.

111. The general question of principle intended to be raised under this heading is whether the Canada Steamship approach to the construction of exclusion clauses is directly applicable or is to be applied by way of analogy in ascertaining the presumed intention of the parties.
112. I can be relatively brief about this question, for it was considered by this court in HIH v. New Hampshire by reference, inter alia, to the judgment of Aikens J in the present case. The clause in question in HIH v. New Hampshire, clause 8, was a different clause, but it raised the same issue, namely whether it covered and excluded negligent as well as innocent misrepresentation and non-disclosure. That issue was discussed at part (viii) and in particular paras 129/140 of that judgment.
113. In the circumstances there is no need to repeat that reasoning here. The essence of it, where the right to avoid is concerned, is that because the duty of good faith is unitary and absolute in the sense that breach of it does not depend on innocence or negligence, therefore it does not make sense to apply Lord Morton's structured rules for presuming that the parties intended, if the language of the clause permits it, to concern themselves only with a cause of action which does not depend on negligence.
114. Moreover, since breach of the duty of good faith does not give rise to a cause of action in damages at all, not even in a case of negligence, there is again no reason to distinguish between a negligent and innocent breach or to presume that the parties did not intend to cover a case of negligence.
115. As for the Hedley Byrne duty of care, even on the assumption that in a special case it could apply to a contract of insurance, liability depends on negligence: in the absence of negligence there can be no liability. Since there can be no damages for breach of the duty of good faith, any exclusion of liability to pay damages must have been intended to apply to a case of negligence – unless at any rate there can be a liability to pay damages for some other cause of action which does not depend on negligence.
116. In this connection, the position under the Misrepresentation Act must be considered. Aikens J was not asked to consider the significance of section 2(2), which raises the possibility of damages in lieu of rescission even in the case of an innocent misrepresentation. This sub-section was nevertheless relied on by Mr Cooke on this appeal. However, I respectfully agree with Steyn J in Highlands Insurance (see para 51 above) that "it is difficult to conceive" of circumstances in which there would be room for the application of section 2(2) in a case where an insurance contract had been validly avoided. If, on the other hand, there is no right to avoid, it is hard to think that there will be a right to rescind, and if there is no right to rescind there can be no remedy under section 2(2). In theory, however, it is possible that the exclusion of a right to avoid may have to be distinguished from a right to rescind, and I will have to consider that argument (at paras 171/177) below in the context of questions of construction.
117. The position with relation to the Misrepresentation Act is complicated by the fact that section 2(1) distinguishes between an innocent and a negligent misrepresentation, although not in those terms. It does so by providing a cause of action in damages where a contract has been induced by misrepresentation, but also providing a defence to a defendant who can prove that "he had reasonable ground to believe and did believe" that the representation was true. The section 2(1) cause of action can in my judgment be looked at in two ways: either it is viewed as a provision imposing a duty not to make negligent misrepresentations (with a reverse burden of proof, cf the similar way in which Diplock LJ analysed the duty of a banker under section 4 of the Cheques Act 1957 in Marfani & Co Ltd v. Midland Bank [1968] 1 WLR 956 at 972); or it is to be looked at as providing a cause of action for an innocent as well as a negligent misrepresentation subject to a no-negligence defence. If it is the former, then negligence is its essence and a case where a defendant has failed to meet the burden of proof is therefore to be viewed as one where that defendant is liable for a negligent misrepresentation. In practice, that is the effect of the subsection. In that case, the cause of action under section 2(1) is like the cause of action in Hedley Byrne in that its essence is negligence, there is only liability to pay if there is negligence albeit negligence is assumed in the absence of proof to the contrary, and an exclusion of liability to pay for misrepresentation ought naturally to cover negligent misrepresentation. If, however, it is the latter, then there is liability to pay whatever the nature of a misrepresentation albeit subject to a defence, and an exclusion of liability for misrepresentation should be viewed as excluding that cause of action, even without specific reference to a possible ground of defence.
118. In the light of HIH v. New Hampshire this is not the occasion on which to conduct a detailed review of the Canada Steamship line of cases, even though the submissions of Mr Cooke may have sought to tempt the court to do so. Lord Morton's third rule derives from Alderslade v. Hendon Laundry Ltd [1945] KB 189, a classic case relating to the construction of exemption clauses, but it may be noted arising in the sphere of a consumer contract. In Canada Steamship that test was applied by the Privy Council to a commercial contract, and in Smith v. South Wales Switchgear Co Ltd [1978] 1 WLR 165 the House of Lords in turn applied Lord Morton's tests to a construction contract. The Canada Steamship approach has been regularly invoked in the context of commercial and construction contracts, and often applied, as in The Raphael [1982] 2 Lloyd's Rep 42, E E Caledonia Ltd v. Orbit Valve Co Ltd [1994] 2 Lloyd's Rep 239, and (a recent unreported authority supplied to the court by the insurers following the hearing) Casson v. Ostley PJ Ltd [2001] EWCA CIV 1013. It has, of course, also been distinguished, as for instance, in Mark Rowlands Ltd v. Berni Inns Ltd [1985] 1 QB 211 (an authority of this court not apparently cited in Casson v. Ostley), Deepak Fertilisers and Petrochemicals Corporation v. ICI Chemicals & Polymers Ltd [1999] 1 Lloyd's Rep 387 at 397/8 and BHP Petroleum Ltd v. British Steel Plc [2000] 2 Lloyd's Rep 277 (a limitation rather than an exclusion clause). In particular, counsel debated a number of cases in which a controversial distinction has been drawn between what have been called allocation of risk clauses as contrasted with exclusion clauses: see for instance Morris v. Breaveglen Limited (unreported, 9 May 1997), where this court followed Rowlands v. Berni Inns.
119. One of these days it may perhaps be necessary to consider the strictness of the Canada Steamship approach in commercial contracts in the light of the development of the law in regard to exclusion clauses begun in Photo Production Ltd v. Securicor Transport Ltd [1980] AC 827 following the passing of the Unfair Contract Terms Act 1977. Despite that development, however, in Ailsa Craig Fishing Co Ltd v. Malvern Fishing Co Ltd [1983] 1 WLR 964, Lord Fraser of Tullybelton at 970 still contrasted the "very strict principles" demanded of exclusion clauses by Canada Steamship with a somewhat less rigorous attitude to be permitted in the case of limitation clauses. The strictness of those principles led Sedley LJ to say in Casson v. Ostley that whereas it may be axiomatic that the court is to deduce the parties' intention from the words they have used, "the intention itself is in most such cases a fiction". It nevertheless remains true that the parties' intentions are what any maxim of construction is seeking to find, and no such maxim should be permitted to force a result which does not do justice to those intentions as expressed in their language.
120. I have made these comments in deference to counsel's submissions. However, none of this line of cases, other than HIH v. New Hampshire, has considered the duty of good faith in the insurance context, a duty which it may be noted does not fit with the premise of Lord Morton's third rule, which was to consider whether "a possible head of damage other than that of negligence" exists.
121. With these principles in mind, I can now turn to the specific issues of construction which fall to be decided on the Truth of Statement clause.

Phrases 1 to 4 of the clause

122. No issue arises on phrases 1 to 4, but I pause to draw attention to the reliance that Lord Grabiner seeks to place on this first part of the clause. In effect he submits that it sets out in positive form the limits of what Chase accepts is its duties in respect of disclosure and representations to the insurers. Thus Chase accepts responsibility for section 1 of the Questionnaire and (because it is a proviso) for the completion (but not the contents) of the remaining sections of the Questionnaire, but not for anything else in relation to the Questionnaire, and not for the updating of section 1. Even section 1, for the truth of which Chase does accept responsibility (for it is a "condition precedent to this Policy"), is not something which Chase has to address itself (nor is it the responsibility of Heaths), but it is for Elmwood to complete "to the best of its knowledge". Phrase 3 makes it plain that any misstatement in any part of the Questionnaire other than section 1 "shall not be the responsibility of the insured or constitute a ground for avoidance…" Ultimately the limit of Chase's responsibility to the truth of section 1 is a matter returned to in phrases 7 and 8.
123. No misrepresentations or non-disclosures with respect to the Questionnaire, let alone to its section 1, are alleged against Chase, so it has not been necessary to test the limits of the clause's protection for Chase in that connection. I would assume, however, that, consistently with the insurers' submissions in relation to phrases 6/8, they would say that phrase 3 would not protect Chase against any negligent misstatement with respect to the parts of the Questionnaire for which it purports to exclude all responsibility.

Phrase 5 and the "post-contractual" theory.

124. I have said above that Mr Gross advanced a new point (for which he was given leave to serve a further respondent's notice) to the effect that the whole of the second half of the clause beginning with phrase 5 was dealing only with the post-contractual period and on that ground gave Chase no protection whatsoever in relation to pre-contractual negotiations.
125. His thesis went as follows. Until the recent decision of the court of appeal and House of Lords in Manifest Shipping & Co Ltd v. Uni-Polaris Insurance Co Ltd (The Star Sea) [1997] 1 Lloyd's Rep 360, [2001] 1 Lloyd's Rep 389 (see especially at 398, 405), the extent of the obligations which the duty of good faith continued to impose on the parties to a contract of insurance even after they had made their contract was uncertain: see The Litsion Pride [1985] 1 Lloyd's Rep 437 at 511/2. This was therefore a legitimate area for parties to resolve by agreement: that was the subject matter and function of the second half of the Truth of Statement clause.
126. Such a construction was supported, Mr Gross submitted, by the following considerations. Phrase 5 dealt expressly with the post-contractual period: it referred to the Due Diligence clause, which it is common ground only relates to performance of the contract, and to the period "after acceptance…of the Declaration". Therefore, prima facie the whole of what followed related to that period and there was no good reason to revert to the pre-contractual period which was the subject-matter of phrases 1 to 4. Thus the word "will" in phrase 6 was, appropriately, in the future tense.
127. This submission is a nice conceit, but in my judgment cannot be sustained. The language of the whole clause from beginning to end, with the single exception of the reference to the Due Diligence clause, is redolent of the period before the inception of contract. The second half of the clause is just too elaborate for any concern relating to the period post-contract. The reference to the Questionnaire again in phrases 7 and 8 ties the whole clause together and shows that it is not divided into pre- and post-contractual periods. The "will" of phrase 6, which becomes "shall" in phrases 7 and 8, is not a future tense but the mandatory language of contract: it is also found in phrases 3 and 4. The reference to Heaths, the agent to insure, in phrases 7 and 8 is also more compatible with the period before the inception of contract. The reference to the Due Diligence clause is a natural precaution, given the width of the wording which follows. I do not place any reliance on the positioning of commas, but I would point out, since some reference was made in argument to the comma in phrase 5, that on Mr Gross' construction, the comma should rather have come after "Due Diligence Clause" and should have been omitted where it is in fact found.

Phrase 6

"…the Insured will not have any duty or obligation to make any representation, warranty or disclosure of any nature, express or implied (such duty and obligation being expressly waived by the insurers)…"

128. Aikens J held that since there is no duty or obligation to make a representation or give a warranty, and since a party cannot, subject to any further exclusion, negate any representation or warranty that he chooses to make or give, phrase 6 is effective only in relation to the waiver of any duty of disclosure (para 63). That much is not in dispute on this appeal. The issue in relation to phrase 6 is as to the extent of that waiver: is it limited to Chase's duty of disclosure or does it also embrace Heaths' duty of disclosure as agent to insure? Aikens J held that it did not extend to Heaths' duty: he emphasised (at paras 64/67) the reference to the "Insured", which is defined as Chase (and its syndicate), the absence of any reference to Heaths or to any agent to insure, and the distinction which is made in the MIA 1906 between the duty of an assured (section 18) and the duty of his agent to insure (section 19). Since the insurers make no allegation of non-disclosure against Chase personally as distinct against Heaths, the distinction is important.
129. The MIA 1906 provides as follows:

"18. Disclosure by assured

(1) Subject to the provisions of this section, the assured must disclose to the insurer, before the contract is concluded, every material circumstance which is known to the assured, and the assured is deemed to know every circumstance which, in the ordinary course of business, ought to be known to him. If the assured fails to make such disclosure, the insurer may avoid the contract…

(3) In the absence of inquiry the following circumstances need not be disclosed, namely: -

…

(c) Any circumstance as to which information is waived by the insurer;

(d) Any circumstance which it is superfluous to disclose by reason of any express or implied warranty.

"19. Disclosure by agent effecting insurance

Subject to the provisions of the preceding section as to circumstances which need not be disclosed, where an insurance is effected for the assured by an agent, the agent must disclose to the insurer –

(a) Every material circumstance which is known to himself, and an agent to insure is deemed to know every circumstance which in the ordinary course of business ought to be known by, or to have been communicated, to him; and

(b) Every material circumstance which the assured is bound to disclose, unless it comes to his knowledge too late to communicate it to the agent."

130. Lord Grabiner submits that the duty to disclose of the agent to insure is dependent on the duty of the assured in that the former has no duty in respect of matters where the latter has none, and that this is emphasised both by the opening words of section 19 ("Subject to the provisions of the preceding section as to circumstances which need not be disclosed…") and by the words of section 19(b) ("Every material circumstance which the assured is bound to disclose…"). The argument is that in so far as the principal has no duty to disclose, then the agent to insure has no duty either; and if the assured has no duty to disclose at all, then the agent to insure can have none either.
131. Mr Cooke submits, on the other hand, that the judge was right to stress the independent duty of the agent to insure, and that that analysis is supported by the authorities to which the judge referred, viz SAIL v. Farex [1995] 1 LRLR 116 at 142, 150, and PCW Syndicates v. PCW Reinsurers [1996] 1 Lloyd's Rep 241 at 255, 257. The agent's duty is truly independent, not derivative. Moreover the reference to the "Insured" alone in phrase 6 was to be contrasted with the reference to other parties including Heaths in phrases 7 and 8. As for the reliance placed by Lord Grabiner on the wording of section 19, that only applied where specific matters were waived from disclosure. Where on the other hand there purports to be a total waiver of the duty of a party, the focus is not on the subject matter of disclosure but on the party owing the duty, and an exemption in favour of one is not to be construed as an exemption in favour of the other: see Aikens J at para 67.
132. It seems to me that the duty of the agent to insure can be described as both independent and derivative. It depends on the question being asked. If the question is: Is material information known to the agent, but not to the principal, matter to be disclosed? The answer is: yes, at any rate if it falls within section 19. From that point of view, since the premise of the question is the ignorance of the principal, one can speak of the agent's independent duty to disclose. If however the question is: Does the agent have any duty to disclose what the principal would have a duty to disclose, on the assumption of the principal's knowledge? The answer is: yes, provided, as is implicit in the question, the principal does have a duty to disclose such information; if, however, he does not, despite knowledge of the circumstance, then neither does the agent. Thus the answer to any particular enquiry cannot simply be found in labelling the agent's duty independent or derivative.
133. Thus I do not find anything in PCW Syndicates v. PCW Reinsurers, in the passages emphasised by Mr Cooke, of particular assistance. The question there was whether the separate knowledge of the agent as to his own fraud was to be imputed to the ignorant (and deceived) assured. Such passages naturally emphasise the independent aspect of the duty of the agent, because the argument was premised on the fact that, if the assured had known the circumstance in question, but had failed to disclose it, there would have been a good ground for avoidance. The issue in the case was as to the extent of the agent's independent duty in circumstances where the information, if it had been known to the assured, would have had to be disclosed. The question was not as to the extent of the agent's duty where his assured's duty was itself limited.
134. The most relevant passage in the authorities in my judgment is that in SAIL v. Farex at 156/157 where Saville LJ said this:

"The third argument relates to the separate obligation of the agents effecting an assurance on behalf of an assured, and again the parties were content to assume that s.19 of The Marine Insurance Act correctly sets out the non-marine as well as the marine position…

The argument here starts with the correct assertion that the duty on the agent is not confined to knowledge acquired from the assured but extends to knowledge otherwise acquired…

In my view this argument fails for the same reason as the argument based on s.18. Section 19 is made expressly subject to the provisions of s.18 "as to the circumstances which need to be disclosed". For reasons already given therefore, the alleged knowledge was no more a circumstance that needed to be disclosed under s.19 than it was under s.18, since each employs the same qualification on what should be disclosed. It was objected that this ran counter to the principle of utmost good faith between assured and insurer (cf s.16 of the Act) since it would mean that an agent could effect a valid insurance with knowledge that the insurer was ignorant of some matter of which he ought in the ordinary course to have been aware, for the very reason that the agent had deliberately refrained from informing the insurer of it. This, however, ignores the fact that the doctrine of utmost good faith attaches to the two parties to the contract, the assured and the insurer. It is the contract the assured makes that the insurer can avoid if there has been material non-disclosure by the assured or his agent. Why should it be a breach of good faith sufficient to deprive the assured of his contract if the agent fails to disclose something which , had the assured known of it, would not have had to have been disclosed by the latter?"

135. That rhetorical question can be applied to the present dispute. Why should it be a breach of good faith sufficient to deprive the assured of his contract if the agent fails to disclose something which, had the assured known of it, would not have had to have been disclosed by the assured because of the exclusion of any duty of disclosure owed by him to the insurer? It is true that in SAIL v. Farex the item of information in question was held (under the "second argument" at 156) not even to have been material, whereas in the present case the argument proceeds on the assumption that the circumstance in question is material but the duty to disclose it by the assured has been waived. In my judgment, the principle of the matter is the same.
136. It seems to me therefore that the present dispute is reduced to an implication, or at any rate an inference of construction, to be derived from the use of the word "Insured" in phrase 6 and the absence there of any reference to Heaths: by comparison to the broader language embracing third parties and Heaths by name in phrases 7 and 8. This in effect invokes the maxim that the expression of one excludes the other (expressio unius exclusio alterius). Mr Cooke also submits that it is precisely because the duty is not excluded under phrase 6 that liability and the remedy of avoidance have to be excluded under phrase 7. However, that argument proves too much, because in any event Chase's duty of disclosure is excluded under phrase 6 and yet its liability is also excluded under phrase 7 and the remedy of avoidance is excluded in its case too under phrase 8. Therefore one is back to the simple point that phrase 6 only mentions the "Insured".
137. In my judgment that point by itself is insufficient to bear the weight placed on it. At the end of the day it is because the circumstance in question is material that it has to be disclosed, whether by the assured or his agent. If, however, the principal assured is excused the duty of disclosure, because of waiver, the waiver negatives materiality and the waiver applies to the agent too. The argument to the contrary comes in effect to this: where both principal and agent know the circumstance in question, a total exclusion of duty on the part of the principal leaves the circumstance in question still material to be disclosed on the part of the agent. That is neither what the statute says, nor is it good faith. If, however, one assumes, again against the background of a total exclusion of the duty of disclosure on the part of the assured, that the assured is ignorant of what the agent knows, the case for upholding that exclusion in favour of the assured is stronger, not weaker.
138. In the circumstances it may not be profitable to speculate why there is no mention of Heaths in phrase 6. In any event, draftsmen do not anticipate every argument. I would hazard the guess however that it was reference to "any other parties" in phrases 7 and 8 which led the draftsman, out of an abundance of caution, to mention Heaths expressly there.
139. Aikens J (at para 68) considered what would have been the position if he had accepted that phrase 6 was apt to negate the duty of disclosure of the agent. He records the argument of Mr Cooke that even so, the exclusion would not assist if there had been a "fraudulent" non-disclosure by Heaths. There appears to have been no argument in relation to phrase 6 that the exclusion should be limited as a matter of construction to negation of an absolute duty, so as not to embrace negation of a duty not to be negligent (or fraudulent) in failing to disclose. Thus the distinction raised in relation to fraudulent non-disclosure appears to have been a reference to the rule of law submission. Aikens J had rejected that submission, but in any event he had also and separately held that where a duty of disclosure (of the assured, or his agent) had been excluded or waived altogether, then it would not matter if any non-disclosure had been deliberate or fraudulent. Thus at para 24 he had asked:

"Can the duty itself be limited? There are many cases where the courts have said that the insurer has "waived" the duty of disclosure of the assured, usually by limiting the scope of the questions that it asks in a proposal form. It can also waive the duty by the nature of the insurance itself. If the scope of disclosure can be limited by these means then I think it can be accepted that it is conceptually possible to draft a clause in a contract of insurance whereby the parties agree that the duty of disclosure of the assured (or his agent) is excluded, or waived, altogether. If that is so, then if it turns out that there was a deliberate non-disclosure by an assured or his agent, that cannot matter. If there is no duty, then adding the epithet "fraudulent" to the description of the deliberate non-disclosure of a material fact does not advance the argument. If there is no duty in the first place then the assured is entitled to keep quiet."

140. Thus, when it came to Mr Cooke's separate (contingent) argument in relation to fraudulent non-disclosure under phrase 6, Aikens J merely said (at para 68):

"I would not have accepted that argument, for the reasons that I have already given: if there is no duty to disclose at all, then there is no question of "concealment", let alone a deliberate or "fraudulent" concealment of material facts."

141. There is no cross-appeal or respondent's notice in respect of this point. If, therefore, the waiver of the duty of disclosure found in phrase 6 embraces the duty of disclosure of the agent to insure, then it matters not, just as in the case of Chase itself, whether a non-disclosure by Heaths might otherwise have been attacked as negligent, deliberate or even fraudulent. In effect it is accepted that where phrase 6 excludes or waives the duty of disclosure, it does so totally.
142. There is one further point that I should note in passing under this heading. Before Aikens J the insurers submitted that phrase 6 was wholly ineffective because the total exclusion of the duty of disclosure would be so inconsistent with the nature of the contract of insurance as to strike as its root and render the contract absurd. Aikens J rejected that submission at para 26 of his judgment. The insurers cross-appealed on this point, and addressed it in their skeleton argument, but did not press it in oral argument. Therefore, I need say nothing more about it, save that I agree that the point is a bad one.

Phrases 7 and 8

"the Insured…shall have no liability of any nature to the insurers for any information provided by any other parties…including, but not limited to, Heath North America & Special Risks Ltd (other than Section 1 of the Questionnaire)…"

"and any such information provided by or nondisclosure by other parties including, but not limited to, Heath North America & Special Risks Ltd (other than Section 1 of the Questionnaire) shall not be a ground or grounds for avoidance of the insurers' obligations under the Policy or the cancellation thereof."

143. Aikens J held that phrase 7 was of no assistance at all to Chase, since (1) it did not limit Heaths' authority to speak or act on Chase's behalf; (2) the word "information" in any event was too weak to protect Chase against anything that amounted to a misrepresentation; (3) even if the phrase could protect Chase against an innocent misrepresentation, it did not do so against a fraudulent or negligent misrepresentation; and (4) the exclusion of liability was not an exclusion against a remedy such as avoidance: see at paras 69/74 of his judgment.
144. Chase disputes each of these holdings on this appeal. Lord Grabiner's primary submission (point (1) above) is that phrase 7 excludes Heaths' authority to speak on Chase's behalf. If that submission were correct, then the insurers concede that it would not matter whether Heaths provided information negligently or even fraudulently. Lord Grabiner's submission is in effect that "no liability for…any information provided by…Heath" means or entails an exclusion of Heaths' authority. That, however, is not what is said expressly, and I do not think that meaning can be found in the words. The language of phrase 7 can be compared, as the judge himself did (at para 71), with the language of that part of the Miscellaneous clauses which expressly provides that the insurers understand and agree that PML, Mr Bradstreet, Tristar and Phoenix "are not agents or representatives of the Insured". Of course, Heaths could not be named there and was thus deliberately omitted from that list, for Heaths was always going to be the agent to insure who would present the insurance proposals. If, therefore, Chase wished to limit Heaths' authority to act as agent to insure other than in the most mechanistic of ways, it had in my judgment to do so by language which clearly expressed the limits of Heaths' authority. In this respect phrase 7 can be compared with the language Brightman J regarded as a successful limitation on an agent's authority to speak in Overbrooke Estastes Ltd v. Glencombe Properties Ltd [1974] 1 WLR 1335, at 1339H –

"The vendors do not make or give and neither the auctioneers nor any person in the employment of the auctioneers has any authority to make or give any representation or warranty in relation to these properties."

Thus on this point Chase's appeal fails.

145. I next refer to point (4) above, but only to defer discussion of it. Lord Grabiner submits that the phrase "no liability of any nature" excludes not only a liability in damages but also a liability to the right of avoidance. In the light of phrase 8, which on any view expressly deals with the exclusion of the remedy of avoidance ("shall not be a ground or grounds for avoidance"), it is not clear what the function of a second front in respect of the exclusion of such a remedy under phrase 7 was intended to achieve. For it might be said that if the remedy of avoidance is in any event excluded under phrase 8, it hardly seems to matter whether it is also intended to be covered under phrase 7. However, the argument in this respect has been complicated by a new submission which Mr Cooke has raised, apparently for the first time on this appeal, in relation to phrase 8, namely that the exclusion of the right of avoidance does not embrace the right of rescission. For that reason, I will postpone discussion of point (4) to a separate heading "Avoidance and rescission" (at paras 171/177) below.
146. The core of the argument under phrase 7, however, relates to points (2) and (3) above and to its subject matter "information". The phrase has no other subject-matter, since, as I have already remarked, non-disclosure appears only under phrase 8. If therefore "information" cannot apply to any misrepresentation at all, then phrase 7 loses all content and effect. That is in essence what Aikens J has held. That holding would also have the effect that phrase 8, dealing with the remedy of avoidance, would only apply to non-disclosure and not to misrepresentation, since phrase 8 begins by looking back to phrase 7 – "and any such information". If "information" cannot cover any misrepresentation, not even innocent misrepresentation, for the purposes of phrase 7, then it cannot do so under the guise of "any such information" for the purposes of phrase 8. That is what Aikens J said, in discussing phrase 8, at para 78 ("First, as I have already stated, in my view, the word "information" is not apt to cover misrepresentations"). The oddity is that in his recapitulation of his answer to preliminary issue one, the judge said that the clause "would protect Chase against avoidance in the case of an "innocent" misrepresentation" (para 81(3)) and it is that formulation which has been reproduced in the order (at declaration (b)(i): "the clause excludes the right of Insurers to avoid the contracts of insurance for innocent misrepresentations or non-disclosures by Heaths…"). There is no cross-appeal against that order.
147. In the circumstances, it might appear that the proper way to view the judge's reasoning is that, despite severe doubts about whether the word "information" would suffice at all to cover misrepresentations of any hue, he was in the end prepared to accept or assume that it would cover innocent misrepresentations (after all, none were pleaded) but not negligent or fraudulent ones. If that is the right explanation, then the word "information" plays some role, but not in itself a crucial role, in his overall construction of the phrase. That appears to be the way in which Mr Cooke has argued the matter on appeal.
148. In any event, I can see no reason why the word "information" cannot cover misrepresentation. Plainly, there can be no "liability" in any event unless the information is erroneous. It is true that the word "information" in phrases 7 and 8 can be contrasted with the word "misstatement" in phrase 3 and "representation" in phrase 6. Nevertheless it is an entirely general word and, in combination with its accompanying participle "provided", can plainly cover representations and thus misrepresentations. It is found in section 18(3)(c) of the MIA 1906 ("Any circumstance as to which information is waived by the insurer") and it is commonly used in insurance broking to describe matters of materiality which are disclosed to the insurer but are not intended to find their way into the contractual language of the policy. In a context where it is plain that liability is being excluded ("shall have no liability of any nature…for any information") and the remedy of avoidance for breach of the duty of good faith is also being excluded (in phrase 8), and where non-disclosure too is being put for that latter purpose in tandem with "any such information", it seems to me to be obvious that the word "information" is being used as a synonym for a false representation, in other words a misrepresentation. This may be somewhat mealy-mouthed language on the part of the draftsman of the contract, but I do not think that any other interpretation is feasible or business-like. This interpretation is supported by the reference at the end of phrase 7 to Section 1 of the Questionnaire, where the parties are reminded that the exclusion of liability for any information is not intended to cover Section 1 of the Questionnaire. Therefore Section 1 of the Questionnaire (and it follows the Questionnaire as a whole) is "information" (see also the citations from the Questionnaire at para 33 above). It is information for the truthfulness of which, but only in the case of Section 1, Chase accepts responsibility. Outside Section 1, however, Chase has no responsibility for "any misstatement" in the Questionnaire nor can any such misstatement constitute a ground for avoidance (phrase 3). Thus it is clear that "information" in phrases 7 and 8 is not contrasted with "misstatement" in phrase 3 but allied with it.
149. At the end of the day, the plain fact is that an assured is not liable or subject to avoidance for giving correct information, only for giving incorrect information. Just as he is not subject to avoidance for immaterial non-disclosure, only for material non-disclosure. Yet no one would submit that the exclusion of the remedy of avoidance would fail in the absence of qualifying "non-disclosure" with the epithet "material". To construe "information" as not including misrepresentation is to deprive phrase 7 of all content (and half of phrase 8). I do not think this should be done unless there is no alternative, which is plainly not the case here. In my judgment "information" in its context plainly stands for misinformation and misrepresentation.
150. The question then arises whether the exclusion for misrepresentation is intended to cover negligence and fraud as well as innocent inadvertence. As a matter of principle, this court is now bound by the decision in HIH v. New Hampshire, which I have discussed above. Ultimately, however, the question is one of construction on the wording of the clause in its context in this case. What then are the factors which bear on that question of construction here?
151. At para 73 of his judgment Aikens J linked that question very closely to the word "information". He held it was not clear enough to cover negligence or fraud. He referred to Steyn LJ's dictum in E E Caledonia Ltd v. Orbit Valve Co plc [1994] 1 WLR 1515 at 1522C that "…the ordinary meaning of the words in their contractual setting is the dominant factor". He said that the contractual setting was an insurance contract and in that setting the parties would not give to the word "information" a meaning that would embrace negligent or fraudulent misrepresentation.
152. For reasons that I have already given a few paragraphs above in relation to the word "information", I would not for myself read that word, particularly in the insurance setting with its unitary and absolute duty of good faith, in so limited a way. Nevertheless, although Aikens J was prepared to deal with phrase 7 relatively briefly in this way, in truth it is clear from his treatment of the similar question that arises under phrase 8 and the exclusion of the remedy of avoidance that his reasoning goes further and wider. In counsels' submissions in this court the wider argument was invoked in relation to both phrases 7 and 8, and it is in my judgment right to do so.
153. It is necessary therefore to see what further reasons the judge gave under phrase 8 for his similar conclusion that neither negligence nor fraud (nor "deliberate concealment" which the judge sometimes used as an equivalent for fraud) were within the exclusion there. He reasoned (see paras 77/78) that he would have expected the parties clearly to identify their intention to cover such risks. He contrasted the absence of such express language with the position under the General Exclusion clause where "fraud, misrepresentation or concealment" are named, adding "So when the parties wished to deal expressly with the question of fraud they did so". He again cited from Steyn LJ in E E Caledonia (at 1523), this time as to the reason why an express reference to negligence is not inserted – "one does not want to frighten off one of the parties". For these purposes he made no distinction between non-disclosure (para 77) and misrepresentation (para 78). Similarly, the language would have to be clearer to exclude the liability under section 2(1) of the Misrepresentation Act (paras 90, 92/3). As for rescission for misrepresentation at common law (as distinct from the specific remedy of avoidance pursuant to the insurance duty of good faith), that was in exactly the same position as avoidance: that is to say at para 80 the judge appears to be saying that the right to rescind has not been excluded at all, whereas it may be that the judge intended to say only that the right to rescind was not excluded in the case of negligent or fraudulent misrepresentation (see para 146 above and at para 81(5) of the judge's judgment).
154. Mr Cooke and Mr Gross supported the judge's reasoning. They relied on the insurance context, contrasting that with a guarantee. Chase and the other bankers may have been looking for an unconditional guarantee, but they made an insurance contract, with its duty of good faith. In any event, fraud was something apart. Pearson, Mair, and Boyd & Forrest all emphasised the difficulties of excluding fraud. It is so outside the contemplation of the parties, that for it to be excluded requires express language. Apart from the insurance duty of good faith, there are the separate torts of deceit and negligent misstatement and the statutory tort under section 2(1) of the Misrepresentation Act.
155. In my judgment the first thing to be said is that there is a real distinction to be made between negligence and fraud. When parties are considering the exclusion of liability, the question of negligence can never be all that far from their contemplation, even if the real possibility of some other kind of liability, not requiring negligence, may in the typical case squarely covered by the Canada Steamship guidelines lead to a reading down of even very wide language. Parties to a contract plainly look to performance rather than non-performance or mis-performance, but they also contemplate the latter. It seems to me, however, that fraud is a thing apart. Parties contract with one another in the expectation of honest dealing. That is a fortiori the position in the insurance context with its entrenched duty of utmost good faith. It is necessary to be particularly sceptical of the submission that fraud is excluded by merely general language. I will therefore deal with fraudulent and negligent misrepresentation separately.
156. As for negligence, I am not persuaded either by the reasons given by the judge or by the submissions on behalf of the insurers that the wording of phrases 7 and 8 should be construed as leaving it on one side. For these purposes phrases 7 and 8 have to be looked at together. I have already said that "information" has to be read as referring to misrepresentation. There is therefore a plain exclusion of liability for misrepresentation. That exclusion is stated in wide terms, viz "no liability of any nature…for any information provided". Similarly, there is a plain exclusion in phrase 8 of the right to avoid for any misrepresentation or non-disclosure. The judge relied, however, on the insurance context. But that is exactly the context in which the unitary and absolute duty of good faith is the central and special feature of the parties' mutual obligations. For the reasons given by the judge himself at an earlier stage of his judgment, and for the reasons given by this court in HIH v. New Hampshire, that context and that feature are reasons for being reluctant to distinguish between negligent and non-negligent instances of breach of that duty, rather than the reverse. As for liability in damages for misrepresentation, whether for breach of the common law duty of care (unavailable for separate reasons, see at paras 59ff above) or under section 2(1) of the Misrepresentation Act, there can be no liability without negligence, and therefore the exclusion of such liability must involve an exclusion of negligence. Thus the combination of the exclusion of the right of avoidance (a right which does not necessarily involve a correlative liability in damages) with the exclusion of a liability in damages for misrepresentation shows that both in phrases 7 and 8 the parties intended to cover negligent as well as non-negligent instances of breach. It follows, as Mr Cooke appreciated in the course of his submissions, that even if the Canada Steamship guidelines were applied as a statute in this case, the exclusion of negligence would fall within them.
157. There is then the feature of the specific exclusion of "fraud, misrepresentation or concealment" in the General Exclusion clause: that, however, may bear on the question of an exclusion for fraud (see below), but cannot assist on negligence (there plainly included within the general expression "misrepresentation"), nor do I think it was utilised by the judge for that purpose. As for Steyn LJ's dictum about frightening off one or other of the parties, see HIH v. New Hampshire at para 139.
158. Thus I see no reason to find in the specific language of this contract any need to come to a different conclusion from that reached, albeit on different wording, in HIH v. New Hampshire. If anything, the exclusion of the contra proferentem rule under the Miscellaneous clauses is of some additional assistance to Chase in this case.
159. I turn therefore to the question of fraud, which was not in issue in HIH v. New Hamsphire. In the present case there is no reference to fraud in the clause. I am very reluctant to find that the fraud of an agent, for whom a principal otherwise has responsibility, can be excluded without express reference to fraud or without language which is in every way the equivalent of such express reference. No case has been cited in which the fraud of an agent has been excluded by merely general language. (Cf Tullis v. Jacson [1892] 3 Ch 441, where the exclusion of fraud was that of an architect acting as a valuer and was in any event express.)
160. If, therefore, it was only a question of phrase 7's exclusion of any liability in misrepresentation that was in question, I would not hesitate to say that fraud was not excluded. I say that despite Lord Grabiner's submissions as to the overall structure of the transaction and of the Truth of Statement clause within it, which I bear well in mind. I am prepared to accept – at any rate for the sake of argument, for these proceedings are only at the stage of preliminary issues and no findings of fact have been made and the case proceeds on the assumption that what the insurers plead is true, not on any assumption in favour of Chase, and in any event his submissions are controversial – that what he submits is correct: namely that it was part of the commonly understood context of the transactions that Chase had no knowledge of its own that was material to the policies, desired to have a policy that was as close to being an unconditional guarantee as it was possible to achieve, and put forward a Truth of Statement clause that was understood as being designed to achieve such a policy. I am also prepared to assume, without deciding, that, so far as the Questionnaire itself is concerned, a combination of the Truth of Statement clause and the opening language of section II of the Questionnaire mean that even fraudulent misrepresentation outside Section I would not avail the insurers. (This point was not the subject matter of any real consideration.) Even so I would not be persuaded that the parties are to be understood as contemplating the exclusion of fraud in phrase 7 by reason of merely general language. I say that without any reliance on the specific reference to fraud in the General Exclusion clause: that clause is there excluding cover for losses caused by the assured's fraud. Even if fraud were not mentioned, such losses would not be covered. Similarly, there would be protection for the insurers against fraudulent claims even in the absence of the clause to that effect in the preamble to the policy. Similarly, the exclusion of the contra proferentem rule contained in the Miscellaneous clauses does not in my judgment assist Chase. The court's reluctance to find an exclusion of fraud in merely general language goes well beyond reliance on a maxim of contra proferentem. Fraud is a thing apart.
161. Why then the note of caution sounded at the beginning of the previous paragraph? It is because phrase 7 is tied in with phrase 8, and phrase 8 covers non-disclosure as well as misrepresentation. The question of the extent of the exclusion of the right of avoidance therefore has to be faced as well. Does it extend to fraudulent or deliberate concealment? It is at this point that the problem seems to me to become more complex.
162. I will seek to put the point in this way. Where misrepresentation is concerned, it is plain that the law now distinguishes between innocent, negligent and fraudulent misrepresentation. The first has long been recognised as giving a right of rescission, even though no right in damages (the remedy of rescission in the case of non-fraudulent misrepresentation apparently derives originally from equity, and firmly entered the common law only with the Judicature Acts: see CHITTY at para 6-101). The second is new to the second half of the twentieth century, but is currently well recognised, whether in the form of the common law breach of duty or the statutory remedy under section 2(1): it does provide a remedy in damages. The third is familiar as giving rise to the tort of deceit, where again damages are available but on a different basis from damages in negligence (save under section 2(1) where Royscot Trust v. Rogerson [1991] 2 QB 297, although subject to criticism, holds sway). So, it may always be necessary to ask whether a misrepresentation is innocent, negligent or fraudulent. That being the case, it is a perfectly understandable question to ask whether an exclusion of liability for misrepresentation excludes one, another or all aspects of misrepresentation.
163. Where non-disclosure is concerned, however, the law has not distinguished, as far as I am aware, certainly not in the same way, between innocent, negligent and fraudulent non-disclosure. In Banque Keyser Slade LJ emphasised that the court's consideration of the question of a remedy in damages for non-disclosure was given in the absence of any plea of fraud. It has not been suggested in this court, however, that non-disclosure of itself can give rise to any cause of action, other than to the remedy of avoidance in the context of the duty of good faith, even where the non-disclosure may be described as fraudulent. At para 47 above I have described the status of the judge's reasoning and order with respect to that very question, raised in preliminary issue two (ie issue (b) at para 42 above), of whether there was a cause of action in damages for inter alia fraudulent non-disclosure. In sum, the judge held that there was no remedy in damages for breach of the duty of good faith and said nothing in his judgment or his order to allow for any remedy in damages for fraudulent non-disclosure. There has been no appeal from that.
164. For the purposes of the current appeal and cross-appeal therefore I approach the matter on the basis that even where the non-disclosure is fraudulent there is no remedy in damages pursuant to the duty of good faith. Moreover, whether there is a remedy in the tort of deceit has simply not been the basis of any consideration, here or below (see para 48 above). It follows that for the purposes of the argument in this case no distinction arises in the matter of remedies whether a non-disclosure is innocent, negligent or even fraudulent: whichever it is, the only remedy contemplated is that of avoidance. The only relevant purpose, therefore, for distinguishing between an innocent, negligent or fraudulent non-disclosure is to test whether an exclusion of the right to avoid for non-disclosure operates equally for all three kinds of non-disclosure.
165. In this connection the next matter for consideration is what is meant by a fraudulent non-disclosure. The term has never been defined. Sometimes it is referred to as a deliberate concealment: but I am by no means sure that the two can be equated. A matter may be deliberately concealed in the honest but mistaken belief that it is not relevant or material or that enquiry of it has been waived. There may be nothing dishonest in that, but ex hypothesi the remedy of avoidance remains. In other circumstances, outside the insurance context a deliberate concealment might be described as dishonest, or at any rate extremely unattractive, and yet, in the absence of any duty to speak, cannot be additionally castigated and remedied as fraudulent. No authority has been cited for a definition of fraudulent non-disclosure: and the absence of such authority in my judgment is not a merely collateral matter but intimately connected with the current problem of asking whether it makes sense to distinguish between fraudulent and non-fraudulent non-disclosure. I am doubtful that it is. No case has been cited in which the distinction has been material to any remedy or absence of remedy under an insurance contract.
166. On the other hand a distinction is sometimes drawn by the policy itself. Thus in Arab Bank Plc v. Zurich Insurance Co [1999] 1 Lloyd's Rep 262 there was an "Innocent Non-disclosure" clause which excluded the insurer's right to avoid for non-disclosure or misrepresentation where the assured could establish that "such alleged non-disclosure, misrepresentation or untrue statement was innocent and free of any fraudulent conduct or intent to deceive". So there the clause itself distinguished on the basis of "fraudulent conduct or intent to deceive". It would seem that dishonesty rather than deliberateness was the test. In the absence of express language, however, I know of no authority which requires a distinction between the possible causes of non-disclosure.
167. In saying that I am conscious that in Carter v. Boehm itself Lord Mansfield does seem to have considered that there was a difference between that concealment which the duty of good faith prohibited and mere silence ("Aliud est celare; aliud tacere…"). As a result non-disclosure in the insurance context in the early years was referred to as "concealment", and the doctrine has sometimes been viewed and explained as constructive fraud. However, Lord Mansfield was seeking to propound a doctrine of good faith which would extend throughout the law of contract, and in that respect his view did not bear fruit. Where, however, in the insurance context it put down firm roots, it came to be seen as a doctrine which went much further than the antithesis of fraud, and, as it has come to be developed, "non-disclosure will in a substantial proportion of cases be the result of an innocent mistake" (Pan Atlantic Insurance Co Ltd v. Pine Top Insurance Co Ltd [1995] 1 AC 501 at 549D (per Lord Mustill).
168. In sum, I do not think that, in the absence of express language, any line is to be drawn between the various possible causes of or motives for non-disclosure. It is not in this way that the distinction is to be drawn. The question to my mind is whether a non-disclosure can support a claim in fraud, with its remedies in damages and/or rescission: either because on analysis it amounts or gives rise to a fraudulent misrepresentation or perchance for any other reason. Aikens J has held in effect, for the purposes of preliminary issue two, that only misrepresentation, and not non-disclosure, can give a remedy in damages for fraud, and there has been no appeal from that conclusion. In any event I would respectfully disagree with him about the need to distinguish between the causes of non-disclosure for the purposes of preliminary issue one. If a distinction is drawn by contract, then effect will have to be given to that distinction according to its terms (cf Arab Bank v. Zurich Insurance). In the absence of such contractual provision, however, I do not think that the law has so far equipped itself to mark out the fine lines which would have to be drawn between the various degrees of culpable non-disclosure.
169. If that is right, does it affect the attitude to be taken to misrepresentation? Having paused to consider the question, my conclusion is, No. Where misrepresentation is concerned, it is not merely a question of seeking to distinguish between various ways in which the duty of good faith can be broken. If a misrepresentation is fraudulent, the common law always gave a right of rescission and the tort of deceit remains with its remedy in damages. In my judgment, the exclusion of liability for breach of the duty of good faith is one thing; the exclusion of liability to pay damages for breach of any applicable common law duty of care or of the statutory duty under section 2(1) of the Misrepresentation Act is another thing; but apart from all of that, the exclusion of the common law remedies for fraud is yet something else. Fraud remains a thing apart. In my judgment, in the absence of express language covering fraud, those remedies for fraud have not been excluded. It follows that to the extent that fraud would have given a remedy under the common law it has not been excluded, either from phrase 7, or from phrase 8. In such a case, the insurers can rescind and/or claim damages in tort under the common law, even though any remedy pursuant to the duty of good faith may have been excluded: see Pan Atlantic at 543F/545B and section 91(2) of the MIA 1906.
170. It follows that on points (2) and (3) above in relation to phrase 7, and in relation to the judge's limitation of phrase 8 to an exclusion of innocent non-disclosure alone, Chase's appeal succeeds in large part. It fails only to the extent that anything in Schedules 2 or 3 can support a remedy at common law in fraud.

Avoidance and rescission

171. There is one point that remains under phrase 8, which I think is new to the appeal, and that turns on the possible distinction to be made between the right of avoidance for breach of the insurance duty of good faith and the right of rescission available at common law in the case of misrepresentation (but not non-disclosure). I say that the point is new to the appeal because, when Aikens J considered the right to rescind (at para 80), he merely said that it had not been excluded and that "The reason for this conclusion is exactly the same as before". By that he could only have meant to refer back to his holding that "information" did not embrace misrepresentation (point (2) above under phrase 7). He might have added that in any event there was no exclusion of a right to rescind for negligent or fraudulent misrepresentation. That is exactly what he did say in his summary at para 81(5). However, there was no suggestion that phrase 8's express reference to "avoidance" did not encompass rescission.
172. On this appeal, however, Mr Cooke has submitted exactly that. He points out that phrase 8 excludes two matters, viz "avoidance" and "cancellation", and submits that the former relates only to the insurance duty of good faith and the latter is a concept of termination under contractual provision. Neither, he says, amounts to rescission. Warming to this theme, he submitted also that the whole of the Truth of Statement clause was in truth premised on the contract of insurance context with its duty of good faith and was not concerned with other remedies in the wider sphere at all. This led him to submit, or at any rate to submit in the alternative to earlier submissions that he had made, that "liability" in phrase 7 was therefore concerned only with the duty of good faith, and thus was not concerned with a liability in damages. If that was so, it would of course go far to undermine some of his primary submissions about phrase 7, although it would I suppose promote a submission that phrase 7 could not be in any way concerned with liability for breach of a duty of care, or for negligent misrepresentation or deceit.
173. As it was, this submission relating to rescission had not been addressed to phrase 7 at all, but was introduced in relation to the line slip facility on the basis that as a contract for insurance it was not subject to avoidance, only to rescission: thus the point was made that the contract did not exclude any liability to rescission.
174. Phrase 8 does not at first sight look as though it is addressed to the line slip facility at all, since it refers expressly to avoidance of the insurers' obligations "under the Policy". However, I remind myself that the line slip facility became itself the subject of a "Policy", viz Policy No 614/OL077A. Thus Mr Cooke did not even rely in the context of his current submission on that aspect of phrase 8. It follows that phrase 8 applies as much to the line slip facility as it does to the contracts of insurance. It also follows that when phrase 8 speaks of "grounds of avoidance" it is to be presumed to be addressing the line slip facility as well as the contracts of insurance. In that context I do not think that "avoidance" is to be given a meaning which would not encompass rescission as well. In any event, I would not limit the term "avoidance" merely to the duty of good faith. Albeit that in the insurance context the MIA 1906 uses the verb "avoid", the language of avoidance and rescission is often found more or less interchangeably within and without that context to describe both the remedy for breach of the duty of good faith and the common law response to misrepresentation (see, eg, CLARKE at paras 23-1, 23-17C and Clough v. London and North Western Railway Company (1871) LR7Ex 26 at 35). This is perhaps partly due to the fact that the particular language of the MIA 1906 is, strictly speaking, applicable only to marine insurance, albeit many parts of it, such as sections 17/19, have come to be regarded as applicable, as a codification of the common law, to non-marine insurance as well. It must also be due to the fact that for the purposes of the MIA 1906 itself the remedy of rescission for misrepresentation had in practice been subsumed into the right to avoid (but only in practice: see, again, Lord Mustill in Pan Atlantic at 543/5). I would therefore make no distinction between the right to avoid and the right to rescind for the purposes of phrase 8, and Mr Cooke's new point fails.
175. If, however, that conclusion is wrong, I would revert to Lord Grabiner's submission in relation to point (4) under phrase 7. His submission was that the words of exclusion "no liability of any nature" were apt to exclude the remedy of avoidance, since on any common sense approach to construction Chase was liable, in the absence of an exclusion, to have the policies avoided for Heaths' misrepresentation. This is a short point on which it is possible to have two views, as Mr Cooke's alternative submissions referred to at para 172 above themselves demonstrate. My initial view was that the question hardly mattered, since the right to avoid is excluded in any event under phrase 7. On reflection, however, I have come to the conclusion that Lord Grabiner's submission is well founded, and also answers Mr Cooke's new point.
176. I would seek to put the matter in this way. Phrase 7's reference to "no liability" would primarily be understood to refer to a liability in damages. To the extent that phrase 7 is dealing with a potential liability in Hedley Byrne negligence or under section 2 of the Misrepresentation Act (whether in relation to the contracts of insurance or the line slip facility), that primary reference to a liability in damages is clearly apposite. If there could be no liability in damages, then there would be no need for phrase 7 at all (see phrase 8): and to the extent that phrase 7 bites on a potential liability in damages, it excludes it. The exclusion, however, is "no liability of any nature". One possible liability where misrepresentation is concerned, is a liability to a right to avoid or rescind. If the liability in damages is excluded, it would seem unbusinesslike to assume that a liability in avoidance or rescission remained. Indeed, the liability in damages might arise (in the case of an innocent misrepresentation and section 2(2) of the Misrepresentation Act) precisely because of a liability to rescission. I would therefore conclude that the wide language of "no liability of any nature" was intended to embrace any right of rescission which accompanied or could give rise to a right to damages. On that basis, it does not matter whether phrase 8 is dealing only with the strict insurance right of avoidance, and the express exclusion of that right under phrase 8 may even suggest that phrase 7 is dealing only with liabilities other than for breach of the duty of good faith.
177. Therefore, on one or other or both of these grounds, I would reject Mr Cooke's new point.

Contracts of or for insurance (revisited)

178. I refer to paras 55/58 above. For the reasons stated there and under the heading "Avoidance or rescission" above, I do not think that the Truth of Statement clause was intended to or does have any different effect in relation to any of the contracts in this case. Essentially, the clause protects Chase save to the extent that the insurers can prove a positive case of fraud such as would entitle them to rescind the relevant contract at common law for fraudulent misrepresentation and/or to recover damages in deceit.

Conclusion

179. In sum, therefore, the insurers' cross-appeal has failed, and Chase's appeal has succeeded in part. It has succeeded to this extent, namely that the insurers can neither resist a claim under the contracts of insurance nor cross-claim for damages merely on the basis of Heaths' negligence or non-disclosures. To do either they must prove a positive case of fraud such as would entitle them to rescind the relevant contract at common law for fraudulent misrepresentation and/or to recover damages in deceit.
180. In terms of the two preliminary issues which were involved in this appeal (see para 42 above):

(a) the insurers are entitled to avoid and/or to rescind the contracts of or for insurance against Chase provided that they prove a positive case of fraud as stated above;

(b) the insurers are entitled to damages from Chase only on the basis of a good claim in deceit, and for these purposes it has not been suggested that such a claim can be premised on the pleaded non-disclosures alone.

MR JUSTICE LLOYD:

181. I agree.

LORD JUSTICE ALDOUS:

182. I also agree.
ORDER: Appeal allowed in part and cross-appeal dismissed; Chase to receive 65 per cent of its costs here and below; permission to appeal to the House of Lords granted to all parties. PRIVATE  (Order not part of approved judgment)