Galaxy Homes Pty Ltd v The National Mutual Life Association of Australasia Ltd (No 2) - [2013] SASCFC 66

SUPREME COURT OF SOUTH AUSTRALIA
(Full Court)
GALAXY HOMES PTY LTD v THE NATIONAL MUTUAL LIFE ASSOCIATION OF AUSTRALASIA LTD (NO 2)
[2013] SASCFC 66
Judgment of The Full Court
(The Honourable Justice Anderson, The Honourable Justice Peek and The Honourable Justice Stanley)
5 July 2013
INSURANCE - LIFE INSURANCE - THE POLICY - CONSTRUCTION
INTERPRETATION - GENERAL RULES OF CONSTRUCTION OF INSTRUMENTS - GENERAL MATTERS
The appellant company Galaxy Homes Pty Ltd appealed to the Full Court against a decision which dismissed its claim on an insurance policy issued by National Mutual Life Association of Australasia Limited. The life insurance policy was taken out by Galaxy on the life of its managing director. The policy included a terminal illness benefit in which the insured can request payment of the life benefit immediately where the insured has a "terminal illness" as defined. The insured was diagnosed with advanced metastatic melanoma soon after the policy was cancelled, and made a claim under the terminal illness benefit. This claim was rejected by the insurer. The Full Court found that the judge correctly analysed the wording of the policy condition, and that the insured's illness did not meet the standard imposed by the terminal illness policy. The Court deferred consideration of the matters raised in the notice of contention. The appellant company is seeking special leave from the High Court in respect of that decision.
The Full Court is now asked to determine the matters raised in the respondent's notice of contention. The issues raised in the notice of contention are whether the cover under the policy is triggered merely by the suffering of an illness during the currency of the policy, whether facts which became known after the policy period are admissible and whether the insurer is protected by an exclusion clause for a pre-existing illness.
Held:
(i) On grounds 1-7 of the notice of contention, the Court finds in favour of the appellant. The appellant was still covered by the policy at the time the claim was made because a claimable event occurred during the currency of the policy.
(ii) On grounds 14-17, the Court finds in favour of the appellant. The judge was correct in determining that the admissible evidence to be relied upon in determining the prognostic character of the illness could only be that which was known at the time the illness was said to have occurred.
(iii) On grounds 20, 21 and 22, the Court finds in favour of the respondent. The judge erred in his construction of clause E.1 and his finding the appellant did not have a pre-existing illness so as to be excluded under clause E.1. The appellant did have a pre-existing illness so as to be excluded, and further, s 47 of the Insurance Contracts Act would not operate to preclude the exclusion in clause E.1 from applying.
Insurance Contracts Act 1984 (Cth), referred to.
Tower Australia Ltd v Farkas (2005) 64 NSWLR 253; Johnson v American Home Assurance (1998) 192 CLR 266; McMahon's Tavern v Suncorp Metway (2004) 89 SASR 125; Asteron Life Ltd v Zeiderman (2004) 59 NSWLR 585, applied.
Willis v Commonwealth (1946) 73 CLR 105; McArthur v Mercantile Mutual Life Insurance Company Ltd [2002] 2 Qd R 197, distinguished.

GALAXY HOMES PTY LTD v THE NATIONAL MUTUAL LIFE ASSOCIATION OF AUSTRALASIA LTD (NO 2)
[2013] SASCFC 66
Full Court: Anderson, Peek and Stanley JJ
THE COURT
Introduction

On 3 May 2013 the Court delivered its judgment in this matter.[1]
[1] Galaxy Homes Pty Ltd v The National Mutual Life Association of Australasia Ltd [2013] SASCFC 34.
The Court took the course of delivering judgment and publishing reasons on the grounds of appeal but deferring issues raised in the notice of contention. This course was taken to enable Mr Eden as much time as possible to consider his appeal rights and financial position. The Court has now been informed that an application for special leave to appeal has been filed in the High Court.
The parties have requested that the Court in these circumstances deal with the issues raised in the notice of contention (“the notice”). The reasons which now follow relate to the issues raised in the notice.
The notice dated 23 January 2013 raises 22 points of contention. Some have now been abandoned. Those matters argued before the Court can be reduced to five main heads. We will now set out the grounds in the notice under their respective generic description.
Grounds 1 to 5: The Occurrence of Terminal Illness which triggers Cover
The issue in relation to these grounds of contention is adequately summarised in the written submissions of the respondent as follows:
The issue raised by these grounds is whether cover under the insuring clause of the policy is triggered merely by the suffering of an illness by the life insured during a policy period or rather both the suffering of an illness and the formation by the insurer of the opinion that the illness will result in death within 12 months during the policy period.
The judge, in his first reasons delivered on 20 August 2012 - Galaxy Homes Pty Ltd v The National Mutual Life Association of Australasia Ltd[2] (first reasons), summarised the principal contentions of both parties. The arguments on appeal followed essentially the same course.
[2] [2012] SASC 141.
The judge dealt with the respondent’s grounds of contention 1-5 in his first reasons as follows:[3]
[3] [2012] SASC 141 at [20]-[25].
[20]At one level, the insurer’s contention is relatively straightforward. NLMA maintains that the policy cover is occurrence or event based, such that the relevant occurrence which would serve to trigger an entitlement to be paid the Terminal illness benefit, must occur within the policy period, in this case, during the period 23 February 2011 to 17 November 2011. The insured does not dispute this. However, NMLA maintains that what must occur within the policy period is a terminal illness as defined in the policy. According to the definition a terminal illness will be established only if the illness is one “which in [the insured’s] opinion, will result in the death of the person insured within 12 months, regardless of any treatment that might be undertaken”. According to NMLA this event will occur only when such an opinion is arrived at.
[21]It is accepted by the insured that for it to succeed it is necessary that the insurer forms the requisite opinion about the illness said to have been experienced by Mr Eden. However, the insurer maintains that the whole of the definition of “terminal illness”, including the reaching of this opinion, must be satisfied at a time prior to the expiration of the policy period. In the present case, at no time was this a possible outcome. Even on the assumption that at some time during the policy period Mr Eden experienced the metastatic melanoma, it was simply impossible for the insured to have satisfied the definition of “terminal illness” during the policy period given that, Mr Eden first attended St Andrews Hospital to be assessed by Mr Walsh on 27 December 2011, was first diagnosed with metastatic melanoma on 28 January 2012, the medical report and medical certificate of Mr Walsh could not become available until after 17 November 2011 and ultimately, a claim was not made by the insured for the payment of the terminal illness benefit until 27 April 2012. In these circumstances it was only possible for the insurer to embark upon the exercise of forming the requisite opinion some five months or more after the expiration of the policy. For this reason NMLA refused at the outset to enter upon the opinion forming exercise.
[22]In arguing this way, the insurer, as is commonly done, has imported the definition of “terminal illness” into the insuring clause, so that the insuring clause would read:
If the person insured has an illness which, in [the insurer’s] opinion, will result in the death of the person insured within 12 months regardless of any treatment that might be undertaken, you may ask us to pay the life benefit immediately …
According to the insurer a plain or literal reading of this requires that the whole of the event including the formation of the insurer’s opinion must occur within the policy period.
[23]The insured, in essence, submitted that such an approach cannot have been the objective intention of the parties to be gleaned from the whole of the policy and that the words of the insuring clause, when read in conjunction with the words of the definition, are capable of giving rise to an alternative construction that is more reasonable and more in accordance with the likely objective intention of the parties. The insured maintains that whilst the definition of “terminal illness” must, ultimately, be satisfied it is only the experiencing of the illness itself which needs to occur within the policy period. The policy is an occurrence or event type policy but the occurrence or the event is simply the experiencing of the illness, albeit the illness must in due course be shown to be of a particular type, that is, one that satisfies the definition of “terminal illness”. However, the process necessarily to be embarked upon in order to determine whether or not the subject illness satisfies the terms of the definition is not, of itself, part of the occurrence or event and does not, itself, have to take place during the policy period.
[24]Each of the competing constructions might be seen to present some difficulty insofar as consistency with other provisions in the policy is concerned and with respect to potentially or arguably anomalous outcomes should other hypothetical fact situations fall to be considered.
[25]In any event, both asserted constructions call for the definition of “terminal illness” to be applied according to its terms. This might present its own difficulties of application once all the facts were to be ascertained, giving rise to such questions as, for example: when is the insurer to undertake the opinion forming exercise? As at what date is the insurer to form its opinion? What evidence is the insurer to rely upon or, more precisely, is there a cut-off date for the provision of evidence relevant to the opinion? From what date is the insurer required to have formed an opinion as to whether or not the life insured had less than 12 months to live? Given my conclusions with respect to the preliminary points and, as a consequence, the manner by which the parties’ dispute should be resolved hereafter, these matters, to the extent they may arise can and should be left to another day.
[Footnotes omitted]
In summary, the appellant contends that it is sufficient to trigger cover under the policy for the illness to have occurred in the policy period, but it is not necessary that there be a diagnosis or the insurer’s opinion formed in the policy period.
It is not suggested by the respondent that the judge has misunderstood or failed to deal with its principal submissions in his reasons, but simply that he has reached the wrong conclusion.
Clause D1 of the insurance policy states in D.1.2 under the heading “Terminal Illness Benefit”:
If the person insured has a terminal illness, you may ask us to pay the life benefit immediately. This amount is paid once only as a lump sum.
Terminal illness, as discussed in the Court’s earlier reasons, is defined in the policy as–
Any illness which, in our opinion, will result in the death of the person insured within 12 months, regardless of any treatment that might be undertaken. Our decision will be based on medical evidence provided to us by the person insured’s doctor, and any other medical evidence that we may require.
The respondent in its written submission points to several factors in aid of its suggested construction. It contends that the actual terminal illness benefit is an accelerated payment of the life benefit. It points out that Mr Eden did not die during the currency of the policy. No trauma insurance cover was taken out for which there may have been a benefit in the event of cancer and that therefore the insured is seeking an early payment of the life benefit to which he would never have been entitled.
Both parties agree that the policy cover is triggered by an occurrence or in other words is event-based. The question is – what is the relevant occurrence?
In his first reasons the judge dealt with the interpretation of this part of the policy. It is clear that the judge considered both constructions as argued by the parties before him. He held finally that the words “claimable event” in the definition of terminal illness should be construed in favour of the argument presented by the appellant.
His Honour concluded as follows:[4]
[68]The Terminal illness benefit insuring clause is satisfied where an insured establishes that the life insured “has” an illness during the policy period which illness satisfies the definition of terminal illness, that is, any illness which in [the insurer’s] opinion will result in the death of the life insured within 12 months regardless of any treatment that might be undertaken.
[69]The word “has” should be given its ordinary meaning which, in this context, is to experience, or to suffer, or to be subjected to, a specified state.
[70]Ultimately, whether a person “has an illness” at a particular time or during a particular period of time will be a matter for expert medical evidence. In this case any evidence given by Mr Eden as to symptoms, if any, experienced during the policy period, cannot of itself be determinative but may, if accepted, be relevant to any medical opinion (diagnosis) reached as to whether or not, during the period of cover, he “[had] metastatic melanoma”. It also may be relevant to the formation or not of the opinion necessary to the characterisation of any such metastatic melanoma as a terminal illness.
[71]In order to satisfy the insuring clause (and definition) it is sufficient that the life insured “has”, during the policy period, an illness, ultimately found to be a terminal illness. It is not necessary for any formal diagnosis to have been arrived at, or any medical opinion to have been obtained, or any claim to have been made, or for the insurer’s opinion to have been reached, during the policy period itself.
[Footnote omitted]
[4] [2012] SASC 141 at [68]-[71].
The judge sets out his analysis of the competing constructions at [26]-[71]. The judge did not accept the literal interpretation urged on him by the respondent. He found that the objective intention of the parties in entering the contract required a reading of the whole policy and that meant a construction that was more reasonable than the literal interpretation.
It appears that the main factors influencing the judge are, first, the claims procedure provided for in clause F of the policy, and second, the construction of the heading in the policy “When will we pay you?” under the children’s trauma option section. That analysis enabled the judge to conclude that the construction argued by the insured is more natural and obvious and more reasonably reflects the objective intention of the parties.
The judge analysed the wording of clause F, which states:
You must notify us that you wish to make a claim as soon as possible after the person insured suffers, or is diagnosed with, a claimable event.
The judge reasoned that if the insurer’s construction of the words “claimable event” in the definition of terminal illness was applied in clause F, a significant unregulated process would take place prior to a claim being made under the policy. The judge stated that a number of preliminary steps would need to be taken, including notification by the insured to the insurer of its belief a terminal illness was present, then the insurer considering and forming an opinion that the illness is a “terminal illness”. This would clearly involve investigation by the insurer, including medical diagnosis and evidence being provided by the insured. It would only be following the formation of the opinion that the steps provided for in clause F could be commenced. This would include the requirements that the insured make it known that it wishes to make a claim before the necessary claim forms are provided, and the forms returned “within six months, or as soon as is reasonable possible, after the person insured suffers, or is diagnosed with a claimable event”. The judge noted the insurer can ask for further information at any time.
The practical effect of this construction is that these later steps, which would ordinarily be required by the insurer to assist it to come to the opinion as to whether a terminal illness was present, would instead be taken after the insurer had already formed the opinion.
The judge considered the wording of the children’s trauma option section “When will we pay you”, supported his construction. Clause D.1.4.3 of the policy states under that heading:
Subject to any qualifying period, we will pay you the amount as soon as possible after we are satisfied that a death, terminal illness or trauma event has occurred. The date that death, terminal illness or trauma event occurs is the date medical evidence, acceptable to us shows it to have occurred.
The judge reasoned that for the purpose of this provision, the insurer’s definition of “terminal illness” could not apply given the second sentence of the provision. It is expressly stated that the terminal illness occurs on the date medical evidence shows the illness to have occurred. The judge found this date would plainly be one different from any date on which the insurer’s opinion is reached, suggesting that whilst the forming of an opinion is needed prior to recovery under the policy, the date the opinion is formed is not significant to the entitlement to make the claim itself.
The judge also looked to the practical consequences of adopting the insured’s construction. The judge reasoned that, adopting the insurer’s construction, an insured taking out a life benefit cover for a 12 month period would receive much less than 12 months cover in regards to the terminal illness benefit, given the time it would take to satisfy the necessary steps involved prior to the claim “occurring”. While the judge noted this would not normally present a problem if the policy was renewed from year to year, he stated it should still be able to operate in a reasonable and businesslike manner even if only for one policy term or period as in this matter. Further, such a problem would also arise if it was the last year of the plan.
In other words the payment of a premium for 12 months cover might in reality only amount to, say, 6 months cover because of all the steps required to be taken whilst the policy was on foot if the insurer’s proposed construction is preferred.
The judge emphasised the importance of consistency in the use of terminology throughout a commercial document. The meaning of “claimable event” under this provision and clause F should therefore apply to the definition of “terminal illness” under clause D.1.2. The judge therefore decided that to define a claim as argued by the insurer would result in a clumsy, in fact unworkable, operation of the provisions of the policy.
The respondent submits that the judge erred in his reasoning in a number of ways. The respondent submits the fundamental and defining characteristic of the phrase “terminal illness” is that it is an opinion, and requires prediction. It submits the opinion is the insured event and it is contrary to the nature of the policy to conclude this event need not occur during the policy period.

The respondent further submits that the judge’s construction of the policy is in error because it contrasts with the prospective nature of the wording in the definition, as the phrase “will result” means the insurer must look forward at the time of forming its opinion. It is put to the Court that, on the judge’s construction, the insurer must instead look back to the policy period to form an opinion about what “will” occur, which in its submission is not practicable.
Finally the respondent submits the judge’s interpretation of the definition of “terminal illness” requires a modification to the wording of Clause F, importing substantially different language so as to accommodate his conclusions.
We consider that the reasoning of the judge in the paragraphs we have set out is correct. In our view the commercial consistency required in interpreting the policy is clearly against the construction submitted by the respondent.
While both constructions submitted by appellant and respondent were open to the judge, we consider he was correct to find there were good reasons why the occurrence of the illness itself should be seen as the insured event.
We consider that there is no difficulty in an opinion being formed outside the currency of the policy but related to the occurrence of the illness during the currency of the policy. Further, we find this approach is consistent with that followed in Tower Australia Ltd v Farkas[5] in which an opinion as to the prognosis was to be formed as at the earlier diagnosis date. This approach allows the policy to work in circumstances where the entitlement to claim occurs within a single policy period, or where the policy has come to an end.
[5] (2005) 64 NSWLR 253 at [5].
In regard to the respondent’s submission that the judge’s interpretation requires a modification to the wording of clause F, we consider that the judge was merely explaining his construction, and the preferred reading of the clause. He was not attempting to reconstruct the clause through his own wording, but analysing how it should be interpreted to enable a consistent commercial result on a reading of the whole policy.
In our view the respondent fails in its grounds of contention numbered 1 to 5.
Grounds 6 and 7: Ending the Plan
The policy provides in clause H under the heading “Ending the Plan”. H.1:
H.1You can end the Plan at any time. If you cancel it in the first 14 days from the date your plan document is first received, you get your money back. If you cancel it after the first 14 days, you do not get your money back.
The Plan ends on the date we receive your notice requesting the Plan termination.
When the Plan ends, you can no longer make a claim under the Plan and we do not have to pay you any benefits. We will not refund you any premiums, except if the Plan ends within the 14 day cooling off period.
H.2 “When the Plan will End”:
The plan automatically ends as soon as one of the following happens:
·    When we have paid the benefit in full, or
·    The person insured dies, or
·    When the Plan to which this option is attached ends (if applicable), or
·    On the expiry date.
The judge dealt with this aspect in his first reasons:[6]
[6] [2012] SASC 141 at [62]-[65].
[62]The insurer has also placed reliance on the third paragraph under “H.1 Ending the plan” which provides as follows.
When the plan ends, you can no longer make a claim under the plan and we do not have to pay you any benefits.
In its written outline the insurer submitted that the provision precludes the insured from bringing its claim after the plan (in this case, the single policy period from 24 February 2011 to 17 November 2011) had ended. Read literally, this paragraph appears to be to the effect that the insured’s claim to be paid the Terminal illness benefit, presented on 27 April 2012 a little more than five months after the plan ended, was made too late. Such a literal reading does not provide a reasonable and businesslike construction when considered in the context of the policy as a whole. In my view, this provision is not to be construed in this way.
[63]The paragraph in question also provides that “when the plan ends … we do not have to pay you any benefits”. Read literally this would apply to the situation of a valid claim made and accepted but, as yet, unpaid during the policy period. If the policy were not renewed, the insurer could point to the H.1 provision and refuse to pay. In addition, the policy also provides as follows, under the heading “H.2 When the plan will end”.
The plan automatically ends as soon as one of the following happens:
·…, or
·The person insured dies, or
·…, or
·…
Read and applied literally to this circumstance, the third paragraph under H.1 would preclude any claim being made for the Life benefit consequent on death of the life insured and would relieve the insurer of any obligation to pay the Life benefit.
[64]Neither of these possible outcomes is appealing and they serve to suggest that something other than a strictly literal construction was intended by the parties. I accept the insured’s submission, to the effect, that the third paragraph under H.1 is an attempted plain English statement to the effect that after the plan has ended the cover provided by the insurance ceases. Save for any accrued claim, that is, based on an event which satisfies the requirements of the relevant insuring clause and having occurred whilst the plan was on foot, coverage under the plan will have ended.
[65]If I am wrong, with respect to this construction of the H.1 provision, the policy would become, by implication, one requiring both the occurrence and the claim or notification to take place within the policy period. I have already, in effect, rejected a construction of the Terminal illness benefit insurance clause which seems to characterise the policy in this way. To read the third paragraph under H.1 literally in this way would run directly counter to what I have found to be the proper construction of the insuring clause. An alternative approach would be to see this provision as giving rise to an implied condition precedent to a right of recovery for a valid claim. In other words, the provision has no bearing on the content of the insured event itself but gives rise to an obligation to lodge the claim before the plan ends. I would reject such a construction as well. The provision is not clearly structured as a condition precedent to recovery nor located where one would expect such a condition to be. It would operate, in this respect, quite inconsistently with the provisions under “F How to make a claim” which contemplates a period of at least six months between the occurrence of the claimable event and the completion and return of the claim forms, without any hint under F that the claim forms must, in any event, be lodged before ending the plan.
[Footnotes omitted]
The respondent contends that regardless of the answer to the matters raised in the first five grounds of the notice of contention, clause H operates in the circumstances of this matter so that no claim can be made for the terminal illness benefit after 17 November 2012, the date on which the plan was terminated.
The respondent argues that the judge should have taken clause H into account in interpreting the policy in regard to the occurrence of terminal illness which triggers cover.
In our view the result would have been the same. If clause H was to be read literally, as submitted by the respondent, it would create practical problems as adverted to by the judge.
Clearly the cover provided under the plan must cease at some stage. However we consider that the clear intention of the parties in entering into the policy was to protect accrued rights, that is, provided the insured event occurs within the currency of the policy, the termination of the plan must protect the rights acquired by the insured. The construction for which the respondent contends would, in our view, convert an occurrence or event-based policy into a claims-made policy which was not the intention of the parties.
It is our view that the entitlement to make a claim for an event which occurs during the currency of the policy is not extinguished by the termination of the plan. The wording of clause H does not allow for the construction urged by the respondent because it only deals with the “making of the claim” as distinct from the entitlement to make a claim. Further, as noted by the judge, when read literally the clause provides “when the plan ends … we do not have to pay you any benefits”. That this would apply in the circumstances where a claim has been accepted but unpaid during the policy term demonstrates the construction the respondent has put forth is untenable.
We consider that the judge has correctly interpreted clause H in the context of the whole policy. We agree with his reasons, set out earlier, for reaching his conclusion.
The construction urged by the respondent is not consistent with the interpretation this Court has placed on the first aspect of the notice of contention. The interpretation we have reached has regard to looking to the policy as a whole. By consistent reasoning, grounds 6 and 7 of the notice of contention also fail.
Grounds 14 -17: Admissible Facts
The judge dealt with the respondent’s grounds of contention 14 -17 in his “main reasons” at paragraphs [66] to [71].[7]
[7] [2012] SASC 235. At paragraph [66] of the main reasons, both the date of “17 December 2011” (in the second line) and “17 November 2012” (in the sixth to seventh lines) should read “17 November 2011”.
The judge noted the respondent’s reliance upon the statement by Dixon J (as he then was) in Willis v Commonwealth[8] as to facts being preferred to prophecies, and correctly distinguished the present case from the type of situation there being addressed:[9]
[67]The notion that a court will not speculate where it knows is well accepted, but usually in the context of its taking into account future contingencies when assessing damages for future economic loss. The question before me in this case is not in that form. The court is not being asked what is the probability of a particular event occurring in the future. It is being asked whether, as at 17 November 2011, Mr Eden’s condition attracted a particular prognosis. (Emphasis added)
[8] (1946) 73 CLR 105 at 116.
[9] [2012] SASC 235.
The respondent also relied on appeal upon the decision of the Queensland Court of Appeal in McArthur v Mercantile Mutual Life Insurance Company Ltd[10] (McArthur). We consider that that case is to be distinguished on the basis adopted in the later decision of the New South Wales Court of Appeal in Tower Australia Ltd v Farkas[11] (Farkas) where Mason P (with whom Beazley JA and Tobias JA concurred) stated:
[19]The decision is, however, distinguishable. In McArthur, the insurer’s error in forming the requisite opinion meant that the court had to proceed to decide, as a question of fact, whether the contractual state of affairs existed. …
[20]As the respondent points out in his written submissions, the evidence in question in the present case must relate to the insured’s prognosis as at the date of diagnosis/occurrence of the relevant disease, 1 May 2002. But it is only the prognosis as at that date that governs entitlements under the policy. Facts relevant to that matter may be tendered, whether or not they were in the minds of any doctor or the insured at that time. But evidence that the insurer did or did not survive for 12 months casts no relevant light on that prognostic matter. For the Court to have regard to facts which could not be known at the contractually agreed date for assessment would effectively deny the bargain struck between the parties.
[21]The insurer does not suggest that the mere fact that the insured survived for more than 12 months negated his entitlement. Yet to pay regard to later events, except those reflecting on the prognosis that was or ought to have been formed on 1 May 2002, would be to depart from the relevant bargain which was necessarily forward-looking in a prognostic sense.
[10] [2002] 2 Qd R 197.
[11] (2005) 64 NSWLR 253.
As was the situation in Farkas, the present case involved the question of a prognosis at a particular time rather than, as was the case in McArthur, a question of whether, as a question of fact, a state of affairs then existed.[12]
[12] We consider, as did the Court in Farkas (2005) 64 NSWLR 253, that it is sufficient to proceed on the basis of this distinction. We note that the appellant in the present case submits (and the respondent disputes) that it was not necessary to rely upon that distinction because the Court in McArthur [2002] 2 Qd R 197 was only concerned with medical reports which were later formulated but whose contents were restricted to facts known or knowable at the relevant time as distinct from facts actually occurring after that time – and thus in fact conformed to the distinction between “knowable facts” and “facts which could not be known” applied in Farkas (2005) 64 NSWLR 253 at 257 [20] and by the judge in the present case. It is not necessary to resolve this dispute.
We agree with both the approach of the Court in Farkas[13] and with the judge’s ultimate acceptance of the following submission by the insured in the present case:[14]
[68]The insured submits that the admissible evidence available to be relied on by the medical experts, is as to facts (including inferences) probative of Mr Eden’s condition as at 17 November 2011, that were either known or knowable as at that time, once appropriate enquiries or investigations were to be undertaken. What is not admissible, submitted the insured, are matters of fact that were not knowable as at 17 November 2011 because these matters of fact had not come into existence. The most obvious example of the latter is the fact of Mr Eden’s continued survival as at September or November 2012.
[13] (2005) 64 NSWLR 253 at 257 [20].
[14] [2012] SASC 235.
We would simply add the following observations. Say the facts here had been somewhat different and were that as at 17 May 2012 (Mr Eden then being alive), the medical evidence was that Mr Eden’s prognosis as at 17 November 2011 had been that it was likely, but far from certain, that he would die from the terminal illness within 12 months from 17 November 2011. If Mr Eden had subsequently died on 17 October 2012, he would have died one month within that prognosticated 12 month period but such a time of death could neither confirm nor cast doubt upon the accuracy of the prognosis because it was entirely consistent with the terms of the prognosis itself.
Similarly, if Mr Eden had subsequently died on 17 December 2012, he would have died one month outside that prognosticated 12 month period but again such a time of death could neither confirm nor cast doubt upon the accuracy of the prognosis because again it was entirely consistent with the terms of the prognosis, namely that while it was likely, it was far from certain that he would die from the terminal illness within 12 months from 17 November 2011.
If the above propositions are clear, and we consider that they are, then the fact that Mr Eden was alive as at say September 2012[15] can neither confirm nor cast doubt upon the accuracy of the prognosis because survival at that time was entirely consistent with the terms of the prognosis.
[15] Dr Fox, an expert witness called by the respondent, relied upon clinical observations of Mr Eden as at that date, particularly the fact of his then survival.
We conclude that His Honour was correct in his conclusion on the facts of the present case that it was not appropriate to have regard to facts not in existence as at 17 November 2011.
We would simply add a caveat. One may imagine a case where a prognosis, say “the insured will certainly die within a month”, may be apparently so glaringly inconsistent with the facts as appear at a later trial (say, that the insured is still alive four years after the date of the prognosis) that such disparity may call attention to the possibility that the process by which the original prognosis had been formulated was faulty.
However, in such a situation it will still be a matter of expert evidence being called rather than simply relying on the fact of survival at a particular time; the significance of such a disparity will lie in its impact on the weight of the expert evidence in the case. To give an obvious example: say the insurer’s expert in the postulated case gives evidence based on an examination of the materials available at the time of the prognosis (and without reference to the fact of survival) that the prognosis by the insured’s expert (“the insured will certainly die within a month”) was an unsatisfactory and flawed prognosis; and say the insured’s expert maintains that his prognosis was correct at the time that it was given (but gives no convincing explanation why the insured is still alive four years after he had said that the insured would certainly be dead). In such circumstances, the judge, when deciding as to which expert to accept, may, depending on all of the circumstances, be entitled to take into account the disparity between the prognosis that “the insured will certainly die within a month” and the unexplained survival of the insured for four years.
The above discussion really just serves to confirm the close relationship between the various aspects of the policy (and the respective submissions in relation to them) in the present case. The judge here determined that the policy required a prognosis of near certainty that the appellant would die within 12 months and that the evidence and prognoses offered by the appellant (putting aside that proffered by the respondent) simply did not go anywhere near as far as that.
Obviously, if that be accepted (as we have accepted it in our previous judgment), there is no need to consider the matter of survival at all. However, if one does look at the matter of survival in the present circumstances, it does not give rise to the type of apparent inconsistency discussed above at all. The prognosis by the appellant’s experts in the present case was not that “the insured will certainly die within 12 months” but rather that “there is a particular percentage chance that the insured may die within 12 months”. The percentages chosen of themselves necessarily recognised the real, and not insubstantial, chance that the insured would not die within 12 months. Accordingly, survival was not so inconsistent with the prognosis so as to call into question the very methodology of the prognosis.
For all of these reasons, we would reject the respondent’s grounds of contention 14-17.
Grounds 20 and 21: Pre-existing illness exclusion (“the clause E.1 argument”).
The issue raised by these grounds is whether or not clause E.1 of the policy excludes cover for the illness suffered by the life insured.
The clause provides:
E. When we will not pay
E.1If the person insured had a medical condition, injury or sickness before the plan began.
We will not pay a benefit for a medical condition, injury or sickness that occurred before the commencement date unless you or the person insured told us in writing about the medical condition, injury or sickness, and we agreed to accept it, when you or he or she applied for the plan …
For the purposes of this clause, the person insured had a medical condition, injury or sickness if:
A medical practitioner or other health professional gave the person insured, or recommended that he or she receive advice, care or treatment …
At trial, there was evidence that in 1994 Mr Eden was diagnosed as suffering from a cutaneous malignant melanoma which was surgically excised at that time. Mr Eden gave evidence that in 1994 he was advised that as a result of this malignant melanoma he had a 40 per cent chance of dying within five years.

Fortunately, seventeen years later when the appellant came to take out the insurance cover, Mr Eden had not suffered any further symptoms, of which he was aware, related to this malignant melanoma.
The judge found that Mr Eden, in 2011, when the appellant took out the policy, was suffering from the same medical condition with which he had been diagnosed in 1994, namely, malignant melanoma, but at a significantly more advanced stage. The judge found that neither the appellant nor Mr Eden told the respondent in writing about that condition before the commencement date. Neither had the respondent agreed to accept it. The judge held, however, that the respondent was not able to rely on the exclusion clause because the medical condition suffered by Mr Eden in 1994 would not have entitled the appellant to have received the terminal illness benefit. As His Honour said:[16]
To use the language of clause E.1, it was not a medical condition for which the insurer is being asked to pay the benefit.
[16] [2012] SASC 235 at [96].
In other words, while Mr Eden in 2011 was suffering from the same disease with which he was diagnosed in 1994, namely, malignant melanoma, albeit, at a significantly more advanced stage, the malignant melanoma from which he suffered in 1994 was not the “medical condition, injury or sickness” for which the appellant was asking the respondent to pay the terminal illness benefit.
The judge reasoned:[17]
Clause E.1 will only have work to do or come into operation where otherwise the terminal illness benefit is found payable. The terminal illness benefit will be payable only with respect to an illness which has a particular prognostic character…
In the event that the insured were entitled to recover the Terminal illness benefit under the policy, it would not be because Mr Eden had suffered, during the term of the policy, the medical condition or sickness known as malignant melanoma but only if found that he suffered, during the term of the policy, from a medical condition or sickness (in this case, metastatic melanoma) that was properly characterised as an illness “which… will result in the death of [Mr Eden] within 12 months, regardless of any treatment…”.
[17] [2012] SASC 235 at [92] – [93].
The judge concluded that the evidence did not support a finding that Mr Eden suffered from metastatic melanoma stage IV with the relevant prognostic character prior to the commencement of the policy.
The respondent submits that this construction of the policy introduces a significant element to the exclusion clause that is neither warranted by the language of the clause, necessary or logical.
It submits that clause E.1 itself defines “medical condition” for the purpose of that clause. That definition does not refer to, let alone turn on, the prognostic character of the medical condition. It submits that once the evidence establishes that the medical condition for which the claim is made is the same medical condition as occurred before the commencement of the policy, the exclusion applies. To construe the clause in the way the judge did means that a metastatic melanoma likely to cause death within 18 months would not be captured by the exclusion clause. This is neither consistent with the language of clause E.1, nor reasonable.
It submits that as the medical evidence established that the medical condition suffered by Mr Eden in 1994 was the same medical condition that the claim for the terminal illness benefit was being made, namely, malignant melanoma, the exclusion clause was enlivened.
The appellant submits that the judge’s construction should be accepted because the reference in clause E.1 to not paying “a benefit for a medical condition … that occurred before the commencement date unless you … told us in writing about the medical condition”[18] makes plain that the pre-existing condition had to be one for which the benefit is being claimed. As the benefit is claimed for a terminal illness which will result in the life insured’s death within 12 months, this was not the medical condition from which the life insured was suffering in 1994. In any event, the appellant submits that, to the extent that there is any ambiguity in the language of the exclusion clause, that ambiguity had to be construed against the insurer, resulting in the adoption of the construction most favourable to the insured.
[18] Emphasis added.
In Johnson v American Home Assurance[19] Kirby J explained the approach to the construction of an insurance policy, as follows:[20]
An insurance policy is a species of commercial contract. It must be interpreted so as to give the words used their ordinary meaning. The primary duty of a court is to discern from the language, structure and apparent purpose of the document what it means. Subject to any special statutory rules governing the approach to interpretation and any interpretative rules lawfully contained in the policy itself, a court should give the words used their ordinary operation. But it should be an operation which takes into account the commercial and social purposes of an insurance policy. Wherever possible, an absurd or manifestly unjust result will be avoided upon the hypothesis that such would not have been intended by the parties.
[19] (1998) 192 CLR 266.
[20] (1998) 192 CLR 266 at 272.
Adopting that approach, we consider that the construction afforded clause E.1 by the judge is not correct.
Both the language and the purpose of the clause are inconsistent with the approach adopted by the judge. The purpose of clause E.1 is to ensure that the insurer is not exposed to the risk of paying out on a pre-existing condition, injury or sickness, of which it is unaware. While the terminal illness benefit will be payable only with respect to a condition, injury or sickness which has a particular prognostic character, there is no reason to suppose that in accepting the insured risk, the insurer was only concerned to be informed of pre-existing illnesses which fell within the particular prognostic character that would give rise to an entitlement to payment of the terminal illness benefit. Any medical condition, once contracted or suffered, will change. In some cases, its effect on the sufferer’s life expectancy will change. In the context of a life insurance policy, it is uncommercial to construe such a policy in a way which would deprive a medical condition of its character as a pre-existing medical condition, where it was suffered by the life insured prior to the insurance contract being made, on the basis that, at that time, the medical condition would not result in the death of the insured person within 12 months. As the respondent contends, the construction favoured by the judge leads to the conclusion that if the life insured had been diagnosed with metastatic melanoma with the prognosis that his death would result within 18 months, it would not have enlivened the obligation to disclose that illness as a pre-existing medical condition for the purposes of clause E.1. The purpose of the clause relieves the insurer of the obligation to pay out the terminal illness benefit where either the insured or the life insured has failed to disclose a pre-existing medical condition, or the insurer has refused to accept the risk of that condition once disclosed, if the insured subsequently claims a benefit for that condition. A medical condition does not cease to be a medical condition because its prognosis changes. That would deprive the provision of its commercial efficacy. It would not afford the insurer the protection for which it has contracted in relation to the assumption of risk with respect to pre-existing conditions.
That purposive analysis is not undermined by textual considerations. The judge considered that malignant melanoma was not “a” medical condition for which the insurer was being asked to pay the benefit, on the contrary, the medical conditions for which the insurer was being asked to pay the benefit was metastatic melanoma stage IV, which would result in the death of the life insured within 12 months. In our view, this overlooks the proper construction of the language of clause E.1. It relieves the insurer from paying a benefit for a medical condition that occurred before the commencement date, unless the insurer was told in writing about the medical condition and the insurer agreed to accept it.[21] The judge was correct in concluding that the text of the clause meant that the medical condition in respect of which the claim to be paid a benefit was made, had to be the same medical condition that occurred before the commencement date. The terms of the policy presuppose that the insurer might be prepared to accept the risk of insuring the proposed life insured. In our view, that is inconceivable if the insured was required to disclose to the insurer only those conditions which by definition would have required the insurer to pay the benefit i.e. a medical condition which would result in the death of the life insured within 12 months. Accordingly, we consider that the proper textual construction is to afford the words “medical condition, injury or sickness” their ordinary meaning without importing into that phrase the words “that will result in the death of the life insured within 12 months”.
[21] Emphasis added.
Finally, we reject the appellant’s submission that the judge’s construction should be adopted in reliance upon the contra proferentem rule. We reach this view on two bases. First, on the basis that we do not accept that the clause is ambiguous. Second, on the basis that the contra proferentem rule is not applied merely because the language of a provision is ambiguous. It is only to be applied where competing constructions are strongly supported by argumentation and where dictionaries and logic alone cannot readily carry the day for either construction.[22]
[22] Johnson v American Home Assurance Co (1998) 192 CLR 266 at 275; McMahon’s Tavern v Suncorp Metway (2004) 89 SASR 125 at 128 [7].
For the reasons explained above, not only do we not consider clause E.1 to be ambiguous, we consider that application of the ordinary principles of construction is sufficient to ascertain the correct meaning of the clause.
Ground 22: s 47 of the Insurance Contracts Act 1984 (SA)
The issue raised by this ground is whether or not s 47 of the Insurance Contracts Act 1984 (Cth) (“the Act”) prevents the insurer from relying on clause E.1 of the policy.
Given the conclusion we have reached as to the operation of clause E.1, the argument in relation to the application of s 47 becomes important.
Section 47 provides:
Pre‑existing sickness or disability
(1) This section applies where a claim under a contract of insurance is made in respect of a loss that occurred as a result, in whole or in part, of a sickness or disability to which a person was subject or had at any time been subject.
(2)Where, at the time when the contract was entered into, the insured was not aware of, and a reasonable person in the circumstances could not be expected to have been aware of, the sickness or disability, the insurer may not rely on a provision included in the contract that has the effect of limiting or excluding the insurer’s liability under the contract by reference to a sickness or disability to which the insured was subject at a time before the contract was entered into.
Section 47 is concerned with clauses in contracts which exclude or limit liability for sickness or disability of the insured existing before the contract was made. An insurer is not able to rely on such a clause where the insured was not aware of, and a reasonable person in the circumstances could not be expected to have been aware of, the sickness or disability.
The purpose of s 47 is to prevent avoidance of the Act’s regulation of misrepresentation and non-disclosure. As the judge observed:[23]
The intention behind s47 … is that an insurer should not be able to side-step or get around the carefully crafted statutory regime in part IV of the IC Act which serves to identify material misrepresentations and material non-disclosures and the full extent of remedies available to the insurer, by taking advantage of matters of form so as to re-characterise such conduct as either warranties by an insured or as falling within exclusions from cover.
[23] [2012] SASC 235 [123].
The judge found that by reason of the operation of s 47 the respondent would not have been entitled to rely upon clause E.1 to exclude liability under the insurance contract if he had reached a different conclusion as to its application. The judge reasoned that given the malignant melanoma had been removed in 1994, at which time Mr Eden had been given a 40 per cent chance of surviving five years, and he had not only survived that time, but seventeen years without any further recurrence or symptoms of the illness, he was not aware, nor could a reasonable person have been expected to have been aware at the time the contract was entered into, that he was still suffering from malignant melanoma.
The respondent submits that the judge misdirected himself by focussing upon the question of whether or not Mr Eden, or a reasonable person in the circumstances of Mr Eden, was, at the time the contract was entered into, aware that he was then suffering from the sickness. It submits that s 47 is not directed to the contemporaneous suffering of the sickness. It is directed to the awareness, at the time the contract was entered into, of the existence of a sickness suffered by the insured either at that time or at any time in the past. The respondent places particular emphasis on the phrase that concludes sub-section 47(1) “or had at any time been subject”.
The respondent submits that the judge found that Mr Eden was aware he had been subject to malignant melanoma in 1994. The evidence clearly established that the awareness was current at the time of entry into the contract because Mr Eden stated that he informed the insurance broker about it in detail.[24] Accordingly, the insurer submits, the judge erred in concluding that s 47 would disentitle the respondent from relying on clause E.1 to exclude a liability to pay pursuant to the policy.
[24] Peter Eden evidence given 13 September 2012 at T37-38.
The appellant submits that the respondent’s argument is misconceived because it overlooks the differing work to be performed by sub-sections 47(1) and 47(2), and the different enquiries to be made in respect of each.
It submits the issue under s 47(1) is whether s 47 applies, or is engaged, at all by the claim being made by the appellant. In that context, the issue is whether the appellant’s claim was based upon a sickness to which Mr Eden “was subject or had at any time been subject”. It submits the judge expressly addressed this issue and accepted that the claim did relate to such an illness, namely, malignant melanoma. In other words, the judge gave effect to the reference to “at any time” in s 47(1). However, it submits, the issue under s 47(2) is the separate issue of whether “at the time when the contract was entered into” the insured was not aware, and could not reasonably have been expected to have been aware, of “the sickness” existing at that time. If the insured was not so aware, then s 47(2) goes on to provide that the insurer cannot rely upon an exclusionary provision in the contract.
Accordingly, it submits, the separate inquiry under s 47(2) is clearly directed to the insured’s state of mind at the time of entry into the contract, and not some earlier time. In this context, particularly bearing in mind the purpose of s 47, it submits that the awareness must mean, as the judge found, an awareness that the life insured was suffering from the condition, as opposed to awareness of some historical state of affairs of no apparent continuing relevance. Therefore, it submits, as the judge found, there was no reason to think that some 17 years after his earlier diagnosis of malignant melanoma, Mr Eden was aware, or reasonably could have been aware, of the malignant melanomas from which he was suffering at that time.
In our view, s 47 does not prevent the respondent from relying upon the exclusion in clause E.1.
In our view, the respondent’s submission must be accepted.
We accept the submission of the appellant that sub-section 47(1) prescribes the circumstances whereby the provision applies. The operative part of s 47 is sub-section (2). It directs attention to the time when the insurance contract was made. It provides that where, at that time, the insured was not aware of, and a reasonable person in the circumstances could not be expected to have been aware of, “the sickness or disability” the insurer may not rely on a provision included in the contract that has effect of limiting or excluding its liability under the contract by reference to “a sickness or disability” to which the insured was subject at a time before the contract was entered into[25]. Consideration of the operation of s 47(2) begs the question: what is “the sickness or disability” upon which the provision operates? In our view, the answer is the “sickness or disability to which [the insured] person was subject or had at any time been subject” where a claim is made in respect of a loss that has occurred as a result of that sickness or disability. The “sickness or disability” in s 47(2) must be the same “sickness or disability” in s 47(1). The flaw in the appellant’s argument is that it involves reading into s 47(2) the words “existing at that time” after the words “sickness or disability”. It is not necessary to do so for the provision to operate. This construction is reinforced by reference to the words we have emphasised above, with which sub-section (2) concludes, namely: “by reference to a sickness or disability to which the insured was subject at a time before the contract was entered into”. In this phrase, the words “at a time before the contract was entered into” must be understood to refer to an unlimited temporal period prior to the contract being made. It is not confined to the time immediately preceding entry into the contract. Accordingly, the operation of s 47 precludes an insurer from relying on a provision in the contract limiting or excluding the insurer’s liability in respect of a claim for a sickness or disability to which the insured was subject at any time before the contract was entered into, where the insured did not know or could not reasonably have known of the condition. This construction is consonant with the underlying purpose of s 47.
[25] Emphasis added.
In Asteron Life Ltd v Zeiderman[26] Spigelman CJ considered the terms of ss 46 and 47 of the Insurance Contracts Act. He said:[27]
Section 46 and s 47 must be construed in the context of the Act as a whole and, in particular, in the context of the substantial amendment of the common law effected by Pt IV of the Act with respect to the consequences of nondisclosure. That Part, the provisions of which it is unnecessary to set out, has the effect that the insurer bears the risk of unknown defects, relevantly, about the health of the person insured at the date of the contract within certain limits that are not material here. Shortly and simply put, there is no non-disclosure, of a character entitling the insurer to avoid meeting his obligations under the contract, in the case of a failure to disclose a sickness or a disability of which the insured was not, and could not reasonably have been, aware.
In the absence of s 46 and s 47, it would have been possible for an insurer to circumvent the impact of Pt IV, by limiting or excluding liability for losses resulting from defects that existed at the time the contract was entered into, whether the insurer was aware of them or not. As the Australian Law Reform Commission recognised in its report, on which this Act is based (Australian Law Reform Commission Report, Insurance Contracts, No 20 (Canberra, AGPS, 1982)), such a contractual provision would have the same effect as a provision requiring the insured to disclose material facts, whether he knew of them or not. The Commission said (at 112):
“… Some absolute warranties of existing fact might be rephrased as exclusions from cover. An example is the common exclusion of pre-existing illness contained in a personal accident policy. This applies to any pre-existing illness, even if the insured was not, and could not reasonably have been, aware of it. Exclusions of that type are as objectionable as analogous warranties. Where exclusion is based on the state or condition of the subject matter of the insurance, the insurer should not be able to rely on that exclusion if the insured proves that, at the time the contract was entered into, he did not know, and a reasonable man in his circumstances would not have known, of the existence of the relevant state or condition.”
[26] (2004) 59 NSWLR 585.
[27] (2004) 59 NSWLR 585 at 589-590 [13] – [14].

While the Chief Justice was in dissent in the result of that appeal, there is nothing in the reasons of the majority to suggest that they took issue with his analysis of the provision.
The mischief to which s 47 (like s 46) is directed is the insertion into an insurance contract of a provision limiting or excluding liability for losses resulting from a pre-existing sickness or disability, whether the insured knew of it or not. Identification of the mischief does not lead to the necessity of reading into s 47 words limiting its operation only to pre-existing conditions, of which the life insured was aware or ought reasonably to be aware, which he or she was suffering at the time the contract was made. For the reasons set out above, the language of the section and its purpose contraindicate such a construction.
It follows that the protection afforded by s 47 to an insured does not apply to a claim for sickness or disability, where the insured knew or a reasonable person in the circumstances would have been expected to know, at the time the contract was made, that the insured was at that time, or had at any time been, subject to that sickness or disability.
On the findings of the judge, Mr Eden knew, at the time the appellant entered the contract, that he had suffered from a malignant melanoma. Accordingly, s 47 does not operate to preclude the exclusion in clause E.1 from applying.