Johnson v American Home Assurance Co

They shall bear thee up in their hands, lest thou dash thy foot against a stone. 1. Psalms, 91:12.

5. Unfortunately, angels did not intervene to protect the appellant's foot. But he had an insurance policy. This case concerns his attempt to obtain earthly rescue from the insurer. The question is whether the courts below misconstrued the policy. It provided coverage, in certain circumstances, for "permanent total loss of use of [a foot]". The insurer says that if, as a practical matter, use of the foot for mobility could be restored by the use of an insert in the shoe of the insured (an "orthotic"), the injury would necessarily fall outside the cover. The insured contends that the sole "loss of use" to be judged is that of the foot, as a foot. In issue is which construction of the policy is correct. Upon the resolution of that question depends the insured's entitlement to pursue his claim.

6. In the Supreme Court of Western Australia, the primary judge [2] upheld the insurer's construction. He dismissed the insured's claim. The Full Court [3] unanimously confirmed that decision. Now, by special leave, the insured appeals to this Court.

1. Steytler J. See Johnson v American Home Assurance Co (unreported; Supreme Court of WA; 16 November 1994). 2. Johnson v American Home Assurance Co (unreported; Supreme Court of WA (Full Court); 15 August 1996 (Kennedy, Rowland and Anderson JJ)).

Mr Andrew Johnson (the insured), was injured on 16 February 1991 when he was camping in the south-west of Western Australia. He took hold of a rope which was tied to a branch of a tree adjacent to a river. The rope gave way as he swung his body towards the water. He fell heavily on to the river bank. As a result, he suffered very severe fractures to the bones of the right foot. His orthopaedic surgeon, Mr Frederick Easton, described the case as "one of the most severe injuries to a hind foot" that he had seen "short of the foot being traumatically amputated".

The insured was admitted to hospital where he remained for seven weeks. The question of surgical amputation immediately arose. Mr Easton appears to have favoured that course as a means of removing the challenge which the profoundly damaged limb would present to the mobility of the insured and for the reduction of pain. Had the insured promptly submitted to the surgeon's recommendation and undergone the major surgery of amputation of the foot there is no doubt that he would have been entitled to recover as claimed.

The primary judge accepted that, consequent upon the injury, the appellant was severely restricted in the use to which he was able to put his foot. Even at the time of the trial, nearly four years after the injury, the surgeons were still debating whether, ultimately, amputation of the foot, or possibly of the leg below the knee, was "inevitable" or merely "probable". The primary judge accepted that, on the probabilities, at some future time, the appellant would undergo amputation of the foot from a point above his ankle.

There was a factual dispute between the parties concerning the mobility of the insured when he used the orthotic. A video film was shown at the trial in which the insured was described as walking without the use of crutches or a stick, without any trace of a limp, squatting, walking backwards and walking on rough and smooth ground. This evidence would ultimately be critical to a factual inquiry into the precise extent of the loss of use of the foot as a foot. The insured accepted that this would be so. There was also evidence that the insured retained passive movement of the foot through a range of 10 degrees, and movement with considerable pain through a range of 25 degrees. He admitted that, as a frequent traveller by air, both within and outside Australia, he had boarded aircraft without a stick or crutches. He was able to get about for various distances without such aids, for periods of time depending on whether he was having "a good day". However, according to the insured, this measure of mobility was entirely the result of the use of the insert in the right shoe and the use of modified thongs around the house. He said that he had also been advised "to wear supportive lace-up boots" to aid him in walking.

The precise extent of the insured's mobility has not been authoritatively resolved because of the opinion which the primary judge took concerning the meaning of the policy, confirmed by the Full Court. He rejected the insured's contention that the focus of attention had to be on the "use" of the foot "as a foot", unaided by orthotics. He accepted that regard could be had to the use of such orthotics. Once that was done, the use of the foot demonstrated by the evidence, and indeed accepted by the insured, necessarily put the case outside the meaning of the policy, as so construed.

The policy in question is an Individual Injury/Sickness Insurance Policy issued by American Home Assurance Co (the insurer). The policy is written in "plain English". According to material placed before the Court in support of the application for special leave to appeal, the terms used in it are common to personal injury insurance policies offered throughout Australia. The case law suggests, as does the international nature of the insurance market and the origins of this insurer, that the language of the policy also appears in like policies offered overseas.

By the opening words of the policy, the insurer, in consideration of the payment of the premium, and subject to the terms of the policy, promised to pay the insured the compensation specified in the Schedule "if during the Policy Period any of the Events specified in The Schedule shall happen to the Insured Person". The "events" are defined, relevantly, in Section A (Capital Benefits) as:

7. Permanent Total Loss of use of one limb 100% [of the compensation].

16. Permanent Total Loss of use of toes of either foot

In the Schedule of Selected Benefits, it is provided that the compensation is a capital sum of $500,000 to be payable in respect of events 1 to 19.

Three of the definitions in the policy must be noted:

2. Injury means an injury which occurs fortuitously to You whilst this Policy is in force and which results solely and directly and independently of any other cause in any of the Events specified in The Schedule within one year of the date of occurrence of such Injury.

5. Permanent means lasting twelve calendar months and at the expiry of that period being beyond hope of improvement.

8. Limb means a hand at or above the wrist or a foot at or above the ankle.

Provision is made, under event number 19, for the payment of a sum for "Permanent Partial Disability not otherwise provided for under Events 5 to 18 inclusive". The amount payable in such circumstances is described as "Such percentage of the Capital Sum Insured as We in Our absolute discretion shall determine and being in Our opinion not inconsistent with the Compensation provided under Events 9 to 18 inclusive". Whether this provision applies in the circumstances of the insured's loss, need not be considered. It was not the matter in dispute between the parties. That dispute arose from the insured's contention that his "foot", taken in isolation, was so damaged that he has suffered "permanent total loss of use of one limb", being the foot at or above the ankle. The insurer contested that interpretation. So far, it has done so with success.

As a result of the video film, taken when the insured was unaware, the primary judge concluded that the insured had "yielded to the temptation of exaggerating the effects of his injury" [4] . Nevertheless, this was immaterial to the construction of the policy. According to that construction it was permissible (as the insurer had argued) to measure the "use" which the insured had of his damaged limb, bringing into account the aid provided by the orthotics. Once "so aided", it was clear that the insured was able to make use of his foot and similarly to put his ankle to use. The primary judge concluded [5] :

[T]he fact that the use of the foot might often lead to pain does not, I think, detract from the fact that the foot is being put to "use" for the purposes of the policy. I do not consider that, even allowing for the fact of pain, and for the use of the inserts, the use to which the plaintiff has put, and continues to put, his foot could be said to be so "unreasonable" or "impractical" as not to amount to a "use" at all. 1. Per Steytler J, at 11. 2. Per Steytler J, at 11-12.

The Full Court noted the "central thrust" of the insured's argument that the loss to be measured was the "practical and effective use of his foot "as a foot" " [6] and that the use of orthotics was immaterial. However, their Honours were unconvinced [7] :

With the benefit of those aids, he has the practical use of his foot which enables him, and at all material times enabled him, to walk, to stand, to bear weight, to walk backwards and to squat. The function of the foot is much more than the "peg" which the appellant described. To ignore the uses referred to is to ignore the requirement in the policy that the loss should be "total" before the insured is entitled to payment for the loss of use of a limb. We cannot accept that the loss of use was "total" within the meaning of the policy while the appellant was able to achieve all that he could by the simple expedient of using an insert to support his injured foot [T]he restriction in the use by the appellant of his foot was both mechanical and due to pain. The use of the inserts reduced the pain by supporting the appellant's foot. In one sense, this use is little different from the use of analgesics. 1. Full Court judgment at 13. 2. Full Court judgment at 14.

The insured contended that, because of legal misdirection, neither the primary judge nor the Full Court had addressed the correct question when determining his entitlement to compensation under the policy. Instead of measuring the permanency and quantification of his loss of use of the foot, they had measured those factors as altered by the supplement of artificial inserts. The insured claimed that this constituted an error of law and that he was entitled to have the liability of the insurer assessed by regard only to the condition of his foot as such.

I accept the following principles as assisting in the approach which is to be taken to discovering the meaning of the disputed words of the insurance policy:

An insurance policy is a species of commercial contract. It must be interpreted so as to give the words used their ordinary meaning [8] . 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 [9] 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 [10] . Wherever possible, an absurd or manifestly unjust result will be avoided upon the hypothesis that such would not have been intended by the parties [11] . Because the primary search is for the ordinary and fair meaning to be attributed to the words used [12] , no court is authorised, under the guise of construction, to make a new contract for the parties which is at odds with the terms of the contract to which they have agreed [13] . In this respect, the primary rule for the construction of insurance contracts is no different from that which governs other written instruments. Maxims and rules of construction, developed by courts as tools to aid in the task of elucidation, are subordinate to the primary duty of construction. This is always to search for meaning of the words used. If, in those words, there is only one meaning, a court may not reject it simply because it regards the result as unfair or otherwise undesirable [14] .

Insurance is substantially a national, even an international, market. Many insurers, including the insurer in this case, are associated with international corporations offering the same or similar policies in a number of jurisdictions. Particular words and phrases in policies are often the subject of elucidation by courts in different jurisdictions. Although competition, legislation and public policy encourage variations in the writing of policies, so that they address the particular needs of the particular insured, insurance is commonly offered in standard form policies which have a national or international provenance. Courts recognise this fact and the consequence that risks may be assessed, and re-insurance procured, on the footing that settled interpretations of commonly used language will not be disturbed without good reason [15] . There are many illustrations of cases where a phrase, recurring in insurance documents, is taken by the courts to have a "well settled" meaning conformably with the commercial purposes of the document [16] . If a settled meaning is demonstrated, given the nature of the insurance market and its function, courts will hesitate before substituting a meaning which is at odds with that which is settled by past decisions.

The foregoing principles, combined with the practical approach which is ordinarily adopted to the construction of words in insurance documents, are relevant to the phrase "loss of use" in sickness and accident policies, such as the one in question here. That phrase could connote, especially where qualified by adjectives such as "total" and "permanent", a complete and entire loss of any use of the limb in question, equivalent to severance. However, over a very long period of time, and in many jurisdictions, the phrase has been construed to mean loss of use as a matter of practical utility. It is not necessary for the insured, in effect, to demonstrate complete severance of the limb in question, whether the phrase used is "permanent and total loss of use" or "loss of the entire" limb [17] . In this respect, a meaning has commonly been given to the phrase which takes into account "the purpose to which, in its normal condition, [the limb] is susceptible to application, in the absence of more specific definition in the policy" [18] . This test of practicalities has long been accepted, not only in Australia and England [19] but also in the United States of America [20] . It was not contested by the insurer in this case. It was accepted by all of the judges who have passed on the present policy. It demonstrates at once the willingness of courts to pay regard to shared jurisprudence in this area [21] and the practical approach adopted to the construction of contracts of this kind.

Notwithstanding the primary rule, there has long been recognised that a liberal approach should be adopted in giving meaning to words in the special field of insurance contracts and insurance documentation [22] . In part, this approach to problems of construction may have derived from an attempt by courts to balance the scales ordinarily unequal as between the insurer and the insured, at least in sickness and accident policies. In part, it may have arisen from points taken by insurers which seemed unmeritorious in particular cases [23] . In part, it doubtless derived from a recognition of the standard forms used in most insurance documentation and the control which insurers generally have over such forms, at least in the case of consumer insurance. Recognition of these characteristics of insurance contracts has led to two principles, often invoked. The first is that a fair and reasonable construction should be adopted which would take into account the variety of persons entering an insurance contract and the entitlement of such persons to know the bargain which they have secured [24] . The second may be traced to a bundle of maxims collected in Co Litt 36a [25] one of which is the maxim verba chartarum fortius accipiuntur contra proferentem. Insurance policies, containing words that are intractably ambiguous [26] , will be construed so as strongly to favour the resolution of the ambiguity in a way advantageous to the insured. This maxim, like many others commonly expressed in earlier times, is now not so much in favour. Courts today accept the duty to endeavour to find the meaning, even of ambiguous expressions, from the language and logic of the document rather than by resort to maxims and other rules of thumb. Even a century ago, the utility of the maxim contra proferentem was doubted in Taylor v Corporation of St Helens [27] . So also in decisions given in the early years of this Court [28] . Nevertheless, even then the principle was accepted as applicable "as the last resort in construction" so far as it required that "a document written by A is to be read as outsiders would read it, and not as A would read it" [29] . More recently, it has been accepted that the contra proferentem principle may still be useful where each of the competing constructions is strongly supported by argumentation and where dictionaries and logic alone cannot readily carry the day for either party [30] . Then, it is not unreasonable for an insured to contend that, if the insurer proffers a document which is ambiguous, it and not the insured should bear the consequences of the ambiguity because the insurer is usually in the superior position to add a word or a clause clarifying the promise of insurance which it is offering.

The giving of meaning to words of ordinary usage in a policy of insurance, as of another written document, is generally a question for the tribunal of fact; in earlier times ordinarily a jury [31] . In many of the reported cases on the disputed words of insurance contracts, the insurer has (as in this case) effectively demurred to the construction which the insured placed upon the language of the policy. Commonly, the insurer has sought to have the judge instruct the jury that one meaning only is open upon the words used or, alternatively, to take the matter from the jury on the basis that the policy does not, in law, respond to the uncontested facts [32] . In the present case, where the tribunal of fact was the primary judge sitting without a jury, he was obliged to direct himself as to the application of the words of the policy to the facts as found. The meaning of the words of the policy, read in their context, was one for the tribunal of fact, just as it was in former times a jury question - not a question for the judge. Yet if the judge misdirected a jury on the meaning to which the words were susceptible and, equally, where a judge sitting alone misdirected himself or herself as to that meaning, appellate relief is available. The object of such relief is to ensure that the correct question is addressed. Even on the appellant's case in this appeal, it would be open to the judge finally determining his claim, to decide that "total" and "permanent" "loss of use" of the foot was not established in the evidence. However, the purpose of the appeal was the insured's effort to have the tribunal of fact address itself to the condition of the foot alone, and not, as the judges below thought, to that limb supplemented by artificial aids.

1. Australian Casualty Co Ltd v Federico (1986) 160 CLR 513 at 527. 2. See, eg, Marine Insurance Act Cth, Sch 2. 3. Legal & General Insurance v Eather (1986) 6 NSWLR 390 at 394, 405. 4. cf Albion Insurance Co Ltd v Body Corporate Strata Plan No 4303 [1983] 2 VR 339; Sweetwater Progressive Mutual Life & Accident Association v Allison (1929) 22 SW (2d) 1107 at 1111. 5. cf Sneck v Travellers' Insurance Co (1895) 34 NYS 545 at 547. 6. John Hancock Mutual Life Insurance Co v Schroder (1938) 180 So 327 at 330. 7. cf John Hancock Mutual Life Insurance Co v Schroder (1938) 180 So 327 at 330. 8. cf Legal & General Insurance v Eather (1986) 6 NSWLR 390 at 394. 9. Legal & General Insurance v Eather was such a case. There, the meaning of the phrase "all reasonable precautions to avoid or minimise injury, loss or damage" was taken as determined by earlier decisions such as Kodak (Australasia) Pty Ltd v Retail Traders Mutual Indemnity Insurance Association (1942) 42 SR (NSW) 231; Fraser v B N Furman (Productions) Ltd, Miller Smith & Partners (A Firm) Third Party [1967] 1 WLR 898; [1967] 3 All ER 57. 10. Sneck v Travellers' Insurance Co (1895) 34 NYS 545; MacGillivray & Parkington on Insurance Law, 8th ed (1988), pp 816-817; Lord v American Mutual Accident Association (1894) 61 NW 293. 11. Life & Casualty Insurance Co of Tennessee v Peacock (1929) 124 So 229 at 231. 12. MacGillivray & Parkington on Insurance Law, 8th ed (1988), pp 816-817. 13. cf John Hancock Mutual Life Insurance Co v Schroder (1938) 180 So 327 at 330; Citizens' Mutual Life Association v Kennedy (1933) 57 SW (2d) 265 at 267; Life & Casualty Insurance Co of Tennessee v Peacock (1929) 124 So 229 at 231. 14. This Court did so in Australian Casualty Co Ltd v Federico (1986) 160 CLR 513 at 531. 15. Australian Casualty Co Ltd v Federico (1986) 160 CLR 513 at 520-521, per Gibbs CJ, citing Sutton, Insurance Law in Australia and New Zealand (1980), p 294, par 8.45; Halsbury's Laws of England, 4th ed, vol 25, par 594 n 1; cf Mount Albert City Council v New Zealand Municipalities Co-operative Insurance Co Ltd [1983] NZLR 190 at 193. 16. An instance is Bradbury v London Guarantee & Accident Co Ltd (1927) 40 CLR 127 where a widow answered a question in a proposal "Have you or husband ever had a fire?" in the negative. The Court held that "husband" there meant a husband then living, a construction denied for the insurer. 17. Anderson v Fitzgerald (1853) 4 HLC 484 at 510, cited Barton J in Western Australian Bank v Royal Insurance Co (1908) 5 CLR 533 at 559; cf Condogianis v Guardian Assurance Co Ltd (1921) 29 CLR 341. 18. See Western Australian Bank v Royal Insurance Co (1908) 5 CLR 533 at 574. 19. Western Australian Bank v Royal Insurance Co (1908) 5 CLR 533 at 559. 20. (1877) 6 Ch D 264 at 270-271, per Jessel MR. 21. See, eg, Western Australian Bank v Royal Insurance Co (1908) 5 CLR 533 at 574, per Higgins J. 22. Western Australian Bank (1908) 5 CLR 533 at 574. 23. Australian Casualty Co Ltd v Federico (1986) 160 CLR 513 at 520; see also Legal & General Insurance v Eather (1986) 6 NSWLR 390 at 394. 24. cf Life & Casualty Insurance Co of Tennessee v Peacock (1929) 124 So 229 at 231. 25. cf Life & Casualty Insurance Co of Tennessee v Peacock (1929) 124 So 229 at 231; Sneck v Travellers' Insurance Co (1895) 34 NYS 545 at 549; Theorell v Supreme Court of Honor (1904) 115 Ill App 313 at 318; John Hancock Mutual Life Insurance Co v Schroder (1938) 180 So 327 at 331; Lord v American Mutual Accident Association (1894) 61 NW 293.

The problem presented by this appeal is not, in my view, an easy one. To some extent, the resolution depends upon the reader's impression, not so simply reduced to rationalisation and justification. So that I do justice to the arguments which persuaded the judges below, let me acknowledge those of the insurer's contentions which need to be taken into account. By doing this, I can be sure that my mind is focused on the contest before reaching the resolution which I prefer.

First, the insurer argued that the ordinary and natural meaning of the words "total loss of use of a foot" drew attention to the practical condition of the insured's foot following an injury. It required an extreme disability both in terms of the effect of the injury on the limb and the duration of that effect. Even when the loss was judged by having regard to the "practical purposes", the precondition for entitlement to recovery remained an extreme loss of use. The insurer argued that, on no account, as a matter of practicality, could this be said of the insured's residual use here. It was therefore outside the terms of the policy.

Secondly, the insurer pointed to the need, in terms of the policy, for the insured to demonstrate that the total loss was "permanent". In the definition of that phrase in the policy reference is made to the condition being "beyond hope of improvement". Although not spelt out, "improvement" could be autogenous but also by the use of medication and artificial devices. Given the medical and technological setting in which the policy was written and accepted by the insured, the practical availability of such methods of "improvement" must have been within the contemplation of the parties when the particular definition of "permanent" was adopted. If, by autogenous healing or the provision of artificial aids, "improvement" could be secured which robbed the effects of the injury, as a matter of practicality, of the description of "total loss", the precondition to liability of the insurer did not exist [33] .

Thirdly, the insurer emphasised the minor nature of the adjustment which the insured had adopted here. It required no additional surgery - simply the use of special inserts for the right shoe and variation of the thong used about the house. The insurer called in aid a series of New Zealand cases where the question was whether an insured had suffered total loss of sight. In Buchanan v Brosnan [34] , Frazer J remarked:

Nobody would think of saying that a man had totally lost his sight if, with the aid of glasses, he could see well enough for all practical purposes. 1. [1929] GLR 40.

The approach in that case was followed in Boyes v Smyth [35] and, later, in Long v Graham [36] . In the lastmentioned case it was held that the loss of vision resulting from a cataract, consequential upon trauma, did not amount to "permanent total loss of the sight" of an eye. Because an operation would restore some vision, it could not be said that the loss was "permanent" or "total". By analogy, the insurer submitted that no one would think of saying that a person had totally lost the use of a foot if that person could walk well enough on the foot for practical purposes by the simple expedient of inserting an orthotic in the shoe or using an adapted thong. Similarly, if the loss or use of a limb could be restored by corrective measures, whether by medication, surgery or by use of an orthotic, it could not be said that the loss of use was "total". Nor was it "permanent", as defined.

I acknowledge the force of these arguments. The policy is, in my view, ambiguous. For a number of reasons, I prefer the construction advanced for the insured. I consider that the courts below misdirected themselves in the approach which they adopted. My reasons are as follows:

1. The relevant promise in the policy addresses attention to a part of the human body. It refers to "one limb" and elsewhere to "either foot" and "one foot". The definition of "limb" reinforces this impression by affording an anatomical description. The definition "means foot at or above the ankle". Thus, the promise on the face of the policy appears concerned with the defined portion of the human body, as such. It makes no reference to external objects or to taking prostheses or orthotics into account in determining the "use" which is in question. That "use" is "of" the limb itself. It is not "of" the limb as supplemented by an external object. Accordingly, on a literal construction of the policy, the focus of attention is, and is only, upon the relevant anatomical member. In this respect, the provision concerning a "limb" or a "foot" may be distinguished from the provision in respect of injury to the eye or eyes. There, differential entitlements are provided in the policy for "loss of sight" (which attracts 100 per cent of the compensation) and "loss of the lens" (which, in the case of one eye, attracts only 50 per cent of the designated compensation). The primary provision may also be contrasted with the entitlement for "loss of hearing". Where particular attributes of the body organ are intended to give rise to entitlement (such as sight of the eyes or hearing of the ears) this is expressly stated. There is no express reference to mobility, as such, in the case of a "limb" or "foot". In such case, the drafter has directed the attention of the reader of the policy to the anatomy itself and nothing more.

2. The purpose of the insurer in writing the policy in "plain English" was apparently to facilitate the understanding of its terms by the variety of people who could be expected to subscribe to such a policy. The list of injuries set forth in the Schedule would quite readily lead an insured to a belief that the purpose of the policy was to cover, relevantly, injuries to specified body parts in respect of which compensation is provided. The policy contains definitions which elaborate or qualify the meaning of several of the words used. However, there is no definition which elaborates or qualifies the meaning of "total loss of use". It would have been relatively easy for the insurer to have included such a definition, making it clear, on the face of the contract, that, in calculating "loss of use" regard would be had to prostheses, orthotics and other external aids or to surgery perhaps even of a radical kind. It was within the power of the insurer to so provide. In the absence of such provision, the inference is that the "use" to be considered is that of the limb itself, unaided, and not of the limb supplemented by an object external to the anatomy.

3. This view of the meaning and purpose of the Schedule of injuries is reinforced if regard is had to the universal character of the words used and of the fact that such words have commonly been found in sickness and accident insurance policies for a very long time. It must be assumed that such words, whether of this insurer's policy or of similar policies written elsewhere, were intended to operate in the variety of environments including those beyond Australia. There would be many places in the world where particular aids, external to the anatomy, would not be available. Prostheses of various kinds and even seemingly simple orthotics, and radical or complex surgery (such as micro-surgery) might not be accessible to the insured person, including within a year of the date of the occurrence of the injury. Availability would depend upon where that person was injured or normally resided. The words of the policy would remain the same. It cannot be supposed that those words would have different consequences depending upon the degree of sophistication or technological facilities that happened to exist in the location of the particular insured. The policy is concerned, relevantly, with an anatomical injury not the availability of palliatives external to the anatomy. This view of the policy is not inconsistent with the definition of "permanent" as meaning "lasting twelve calendar months and at the expiry of that period being beyond hope of improvement". The "hope of improvement" referred to, in the context of a policy written by reference to particular parts of the anatomy, is autogenous improvement of the kind commonly occurring in the response of the human body to injury, perhaps aided by universally available medication or surgery. If "improvement" were intended to refer to improvement by exceptional surgical intervention or use of external objects (prostheses, orthotics, crutches, walking sticks etc) it might reasonably have been expected that this would have been made clear on the face of the policy. Yet upon such matters, the policy is silent. It has thereby left only the anatomical part specified as the reference point by which "loss of use" is to be measured.

4. This approach to the language of the policy appears to be consistent with several decisions in the United States of America where analogous questions have been considered by the courts in insurance cases and beyond. Thus in Lord v American Mutual Accident Association [37] , the insured recovered upon a policy expressed in terms of "loss of one hand" where the hand was so injured as to be useless for its usual purposes. No one suggested that the possible insertion of a hook would disqualify the insured from recovery for the anatomical damage, that being the purpose of the policy. In Life & Casualty Insurance Co of Tennessee v Peacock [38] , the policy afforded cover for loss of both feet by severance. One foot of the insured and the toes of the other were severed. She could not walk barefoot without falling over. However, if she wore a shoe with a spring or brace insert, she was restored to some locomotive functions. The Supreme Court of Alabama held that it was open to the jury to find that the policy responded to the loss. The Court determined that "the word "loss" should be construed to mean the destruction of the usefulness of that member, or the entire member, for the purpose to which, in its normal condition, it is susceptible to application" [39] . Another decision on facts similar to the present case is John Hancock Mutual Life Insurance Co v Schroder [40] . It is also a decision of the Supreme Court of Alabama. There, the policy provided indemnity for "entire and irrecoverable loss of use of both feet". The insured was seriously injured. He was able to walk only with the aid of two canes. Standing and walking was painful. He could not balance without support. Alike with this case, the insured's physician testified that he would be "better off if both feet were amputated" [41] . The Court held that it was open to the jury to find that there had been "entire loss of use" of both feet. It rejected the submission of the insurer that the trial judge should have instructed the jury that, if the insured was able to walk with the aid of canes, he had not suffered the "entire loss of use" of both feet. This is, in effect, the construction of the policy for which the insurer has argued here. Despite the acceptance by the insurer, correctly in my view, that "permanent total loss of use" does not necessitate severance, the insurer is effectively arguing for that result. So long as the limb in question is not severed, it would necessarily have some "use", if only by passive manipulation or for a measure of balance or as adjunct to external objects or for aesthetic purposes. Once it is conceded that, without an express provision in the policy, severance is not required, the kind of "loss of use" referred to in the policy must mean loss of unassisted use of the limb in question, ie without supplementation by external objects [42] .

Outside the context of ambiguous insurance documents, analogous questions have arisen in the United States of America. Thus, in Gilday v Mecosta County [43] the issue was whether the word "disability" in the Americans with Disabilities Act [44] was to be construed by looking at the situation of the claimant when impairment was controlled by medication or treatment or without regard to such mitigating measures. Whilst the United States Court of Appeals for the 6th Circuit allowed the appeal, the Court divided on the issue of whether a disability (in that case diabetes mellitus) was to be determined as a disability in its uncontrolled state [45] . The case illustrates the divergencies of approach that can arise and the importance of resolving them in a way that will uphold the object of the protection offered.

1. Unreported; United States Court of Appeals (6th Circuit); 2 September 1997. 2. 42 USC s 12102(2)(A). 3. Moore J held that "disability" was to be determined as a disability in its uncontrolled state. Kennedy J held that it was not. Guy J found that the disputed questions of fact in the case made it unnecessary for him to decide the point.

5. The foregoing conclusion is not necessarily inconsistent with the approach of the New Zealand courts where, in cases involving claims for loss of sight or loss of useful vision, regard has been had to the corrective effect of the use of spectacles [46] . It is unnecessary to determine, in this appeal, whether those cases correctly state the law for this country. However, where the policy refers to a particular sensory capacity (such as "vision", "sight" or "hearing") attention is arguably shifted away from the body organ as such. Thus a "limb" or "foot", where used in the policy, addresses attention to the organ itself, not to a function, even a major function, of the organ, such as mobility in the case of a foot. Feet and the lower limbs have utility quite apart from their importance for mobility. "Vision" and "sight", where used in a policy, define the entirety of the capacity which is to be measured.

6. Whilst the contrary construction of the policy is certainly open, the foregoing reasons demonstrate, I hope, that the construction for which the insured argues is also available and, as I would think, is more persuasive. However, if this case is at the cusp and if each construction is arguable - each unable to expel the other by logic and reason alone - the case is one for the contra proferentem rule. The insurer made a promise. It said nothing in its policy about qualifying that promise in respect of loss of use of the lower limb by reference to the availability and use of objects external to that limb. It could have done so. By offering the policy in the terms it chose, it must be taken to have accepted the construction for which the insured contends because that construction is reasonably available in the language used.

1. (1894) 61 NW 293. 2. (1929) 124 So 229. 3. Life & Casualty Insurance (1929) 124 So 229 at 231, referring to Sneck v Travellers' Ins Co (1895) 34 NYS 545. 4. (1938) 180 So 327. 5. John Hancock Mutual Life (1938) 180 So 327 at 330. 6. Citizens Mutual Life Association v Kennedy (1933) 57 SW (2d) 265; Theorell v Supreme Court of Honor (1904) 115 Ill App 313; Curwood v R F Williams Pty Ltd [1942] VLR 61. 7. Unreported; United States Court of Appeals (6th Circuit); 2 September 1997. 8. 42 USC s 12102(2)(A). 9. Moore J held that "disability" was to be determined as a disability in its uncontrolled state. Kennedy J held that it was not. Guy J found that the disputed questions of fact in the case made it unnecessary for him to decide the point. 10. Buchanan v Brosnan [1929] GLR 40; Boyes v Smyth [1933] NZLR 1427; Long v Graham [1967] NZLR 1030; cf Re Lloyd & Commercial Union Assurance Co of Canada (1984) 45 OR (2d) 78.

Because it is common ground that success for the insured still leaves it open to the insurer to contest his claim by reference to the factual evidence addressed to the use of the insured's foot unaided by external adjustments, the matter must, on the view I take, be returned to the Supreme Court. I acknowledge that this would be an unfortunate result. It would require the continuation of litigation already long protracted. However, the insured is entitled to have his claim determined without misdirection as to the policy's meaning. This he has so far been denied.

The appeal should be allowed. The orders of the Full Court of the Supreme Court of Western Australia should be set aside. In lieu thereof, it should be ordered that the appeal to that Court be allowed and the orders of Steytler J set aside. The matter should be remitted to the Supreme Court of Western Australia for the determination of the appellant's claim under the policy. The costs at first instance should abide the outcome of that determination. The appellant's costs in the Full Court and in this Court should be paid by the respondent.

On 16 February 1991, the appellant was camping with others at a place known as Sue's Bridge Crossing on the Blackwood River in the south west of Western Australia. A length of rope was tied to a branch of a tree near the river and the appellant took hold of the rope and swung out over the water. The rope gave way and the appellant fell to the river bank. He suffered severe injury to his right foot.

In May 1990, the appellant's wife had taken out an individual injury and sickness insurance policy with the respondent insuring the appellant against certain events. One of those events was

The sum payable by the insurer on the happening of that event was 100 per cent of the Capital Sum Insured under the policy, namely, $500,000.

The appellant made a claim on the policy but the respondent denied liability, contending that there had not been "permanent total loss of use of one limb". The appellant brought an action in the Supreme Court of Western Australia claiming the amount payable under the policy. The trial judge held that the appellant had not proved that he had suffered permanent total loss of use of one limb within the meaning of the policy and accordingly dismissed the claim. An appeal to the Full Court of the Supreme Court of Western Australia was also dismissed. The appellant now appeals to this Court pursuant to special leave.

It was not disputed by the respondent that the appellant had sustained an injury as defined by the policy. It disputed whether there had been permanent total loss of use of one limb. "Permanent" was defined by the policy as meaning "lasting twelve calendar months and at the expiry of that period being beyond hope of improvement". "Limb" was defined as meaning "a hand at or above the wrist or a foot at or above the ankle". No other word, or combination of words, used in the expression "permanent total loss of use of one limb" was defined in the policy.

The injury suffered by the appellant was very serious. The trial judge found that the orthopaedic surgeon who attended the appellant immediately after the accident had considered amputating the leg somewhere below the knee but that was not done. Several operations were performed to deal with the fact that the appellant had suffered a severe open fracture dislocation involving several bones of the right foot. In the course of those operations, various pieces of bone were removed. The trial judge said that he had little doubt that the appellant was, at the time of trial, "severely restricted in the use to which he is able to put his foot" and that the appellant "will, at some future time, undergo an amputation such that he will entirely lose his foot from a point above the ankle". Nothing, however, was said to turn on this prognosis. The appellant's case at trial, and on appeal, was that he had suffered a permanent total loss of use of the right foot and the case was conducted on the basis that any such future operation on his leg was not relevant.

The trial judge said that it was accepted by counsel for the appellant at trial that the appellant "was never in the position that he was unable, at all, to move his foot, or his ankle, leaving aside those times when his foot was immobilised in hospital or in plaster" and that it was not suggested that it was "anticipated, at any relevant time, that he would, amputation aside, be entirely unable to move his foot or his ankle."

Some months after the accident, the appellant became able to stand and walk on both legs and to do so without crutches or sticks. At first, hardly surprisingly, he could do so for only very short periods. But those periods have grown longer as time has gone on. Both standing and walking have been accompanied by pain, usually very great pain. His foot now has an equinus deformity and the big toe joint is markedly stiff. We were told in the course of argument that much of the bone in the heel of the foot has been removed.

Because of the equinus deformity the appellant uses an orthotic or insert in his right shoe or, if he would otherwise have gone barefoot, he uses what was described as a "thickened thong".

The appellant contended that the question whether he had totally lost the use of his right foot must be determined without regard to the fact that, by using orthotics, he is able, for limited periods, to perform functions such as walking, standing or squatting which, so the argument went, "can only be performed if the foot is there, or is replaced by a prosthesis". It was said that the Full Court, and the trial judge, had applied wrong principle in considering whether there had been a total loss of use of the limb by having regard to the use that could be made of it with the aid of an orthotic.

To state the question as whether loss of use is to be judged with or without the orthotic distracts attention from the question presented by the terms of the policy. The insurer is bound to pay under the policy if there has been permanent total loss of use of the foot at or above the ankle. It was accepted by the parties that the "use" which is to be considered is practical use, not some theoretical use, and that what is relevant is use of the foot for the purposes to which the foot would ordinarily be put. Thus, central to the question whether the insured has suffered a permanent total loss of use of one foot is the factual question of what use, if any, he now can make of the foot. That question of fact is not decided by considering only whether the insured has the assistance of some artificial device in performing functions of the kind he would ordinarily perform with his feet. A person may wear a boot to kick a football but nevertheless that person uses the foot. A person may wear skis to travel on snow but uses his or her feet to do so. A person may wear climbing shoes to scale a rock wall but still uses his or her feet to do so. Some of those activities may be very difficult, if not impossible, without the device attached to, or enclosing, the feet, but the fact that a device is used does not determine whether the person is using his or her feet.

Nor is the question to be resolved by considering only whether the insured is able to perform functions of a kind usually performed by use of the particular limb or part of a limb. To take an obvious example, a person who has had the right foot amputated may, with the aid of crutches or a prosthesis, walk, but that person has permanently lost the use of the right foot. The fact that the person may walk does not mean that he or she has not lost the use of that foot. But at the other extreme, a person may find it more comfortable to walk with an orthotic inserted in the right shoe. That person has obviously not lost use of the right foot.

The question of what is permanent total loss of use of the foot is, then, not to be decided only by reference to whether some aid is used any more than it is to be decided by reference only to whether the person can perform functions for which the foot is usually used.

Reference was made to a number of decisions of courts in New Zealand and the United States [47] . Those decisions provide little or no guidance to the resolution of this matter. They concerned different expressions in different insurance policies (or in statutory contexts) and arose out of different factual circumstances. No principle can be deduced from them beyond the most basic - that the extent of the coverage provided by an insurance policy is determined by the words of the policy.

1. Buchanan v Brosnan [1929] GLR 40; Boyes v Smyth [1933] NZLR 1427; Long v Graham [1967] NZLR 1030; Sneck v Travellers' Insurance Co (1895) 34 NYS 545; Life & Casualty Insurance Co v Peacock (1929) 124 So 229 (Ala); John Hancock Mutual Life Insurance Co v Schroder (1938) 180 So 327 (Ala).

41. Here, it was found that with an orthotic the appellant could walk, squat and stand up. Most importantly, however, it was found that each of those actions was performed by the appellant using his foot and ankle, albeit with the aid of the orthotic or thong. The trial judge found that the appellant was able to turn his foot by means of his ankle, was able to use the foot for the purpose of bearing his weight while standing, for the purpose of going up and down steps, for the purpose of squatting and for the purpose of moving backwards. That being so, it cannot be said that he had totally lost use of that foot.

42. The policy of insurance gave cover only against the happening of specified events. The event which was alleged to have happened has not been demonstrated. It follows that there is no claim on the policy and the appeal should be dismissed.