76 In considering the proximate cause of loss in the insurance context, the Court has regard to the reality, predominance and efficiency of a cause, rather than proximity in time: HIH Casualty & General Insurance Limited v Waterwell Shipping (at 608) per Sheller JA (Beazley and Stein JJA agreeing); see generally Lasermax Engineering Pty Limited v QBE Insurance (Australia) Limited  NSWCA 66 (at ff) per McColl JA (Ipp and Tobias JJA agreeing). The Court applies common sense standards in determining what is the proximate cause, approaching the question by reference to the understanding of “the man in the street, and not as either the scientist or the metaphysician, would understand it”: Lasermax (at  – ).
77 The respondent bore the onus of proving that E15 applied: Trickett v Queensland Insurance Co Ltd  AC 159 (at 164); see also Provincial Insurance (at 545) per Kirby P; State Government Insurance Commission v Lane (1997) 68 SASR 257 (at 260) per Debelle J (Cox and Millhouse JJ agreeing).
78 In my view the primary judge’s conclusion (at ) that the reference to “hail” in E15 meant damaged by hail in the sense that hail was the proximate cause of the damage was correct. Damage which was caused by hail, without, at least, elements of a storm coinciding, would be a meteorological anomaly. In other words, the Policy was looking at the situation where hail fell as part of a storm but, in the circumstances, was the proximate cause of the damage.
79 The primary judge concluded that hail was the proximate cause of the damage and, accordingly, sub silentio, held the respondent had discharged its burden of proof in this respect.
80 His Honour did not regard the Agreed Facts as determining the issue of proximate cause. As his Honour observed (at ), Mr Seton approbated and reprobated on this question, at times asserting the Agreed Facts were binding, and at others accepting the court should determine the issue on the basis of the evidence. Mr Seton did not challenge his Honour’s assessment of the conduct of the trial in this respect. I would also agree with his Honour “that the agreed facts [paragraphs xii and xiv] are imprecise in failing to identify precisely how damage was caused in the ways suggested” and that it was necessary to look at the evidence to determine the proximate cause issue: primary judgment (at ) – ).
81 His Honour found (at ) that the sequence of events required to attract the operation of E15, and, therefore, the limit of $100,000 that clause prescribed, involved “(1) ‘Damage to … Caravans (etc) … by hail’; (2) … the hail (penetrating) the entire thickness of the material damaged; (3) causing water to enter.” He concluded that sequence was established by the following evidence:
“91 In paragraph 6 of his affidavit sworn on 1 October 2003, Mr Caine says:
On 16/1/02 a severe hailstorm struck Chinderah and caused severe damage to our business including hail damage to our caravans and their annexes’.
In paragraph 14 of his affidavit sworn on 8 October 2004, he says:
‘… all of the vans were severely pitted on their external sheeting and some had the sheeting actually torn. Water penetrated into every caravan whether through the torn portion of the aluminium sheeting, the tiny pit holes that were caused by the hailstones, the seams of the sheeting that were evidently stretched by the excessive pitting or the hatches in the roofs. Damage was also caused to every caravan through the hail breaking the glass, splitting and opening seams and causing hairline fractures to the caravans’ linings. All the annexes and their attached Vans were similarly damaged and left non waterproof after the storm…’The photographs in evidence prove the same point.
92 Mr Caine gave evidence that some of the hailstones were ‘as big as house bricks’ (T44.12); he saw other hailstones of ‘all shapes and sizes’ (T44.41); the rain and hail travelled almost horizontally because of the ferocity of the storm (Para 9, affidavit of 8 October 2004 &T44.35); and the hailstorm was of such ferocity that hailstones came through one window of the residence, travelled across the room and reached the opposite wall (T43.32). The intensity was such as to tear away some of the sheeting on the caravan walls (T42.5); the hail dented the walls of the caravans, leaving little pin-holes in those walls (T42.10); and the windows of the vans were smashed (T42.24). The vans were ‘all penetrated to some degree’ (T42.); some four inches of rain fell in ten minutes (T44.45); and, in the result, ‘All the vans were left uninhabitable due to water ingress’ (Para 15, affidavit of 8 October 2004).
93 It is true to say that damage was caused to the caravans by storm including hail and torrential rain, as per paragraph 12 of the Agreed Facts. That is precisely what happened. But it is not inconsistent with hail being the proximate cause of the damage. Proximate cause is a concept which requires an assessment of the qualities of reality, predominance and efficiency in circumstances in which a number of factors contribute to the happening of the damage in question: HIH Casualty & General Insurance Ltd v Waterwell Shipping Inc (1998) 43 NSWLR 601 at 608.
94 In the present case, damage was occasioned to the caravans in a manner in which a storm took place, including hail and torrential rain; the hail penetrated the materials from which the caravans were constructed; the entire thickness of the walls and windows of the caravans was penetrated by the hail when holes of various sizes and descriptions were made, and the physical integrity of the structure was thereby breached, allowing water, in the form of rain and the hail itself, to enter the caravans. In that way, the hail was the active, efficient cause of the damage. It set the relevant train of events in motion, in the common sense way in which proximate cause must be ascertained. As earlier submitted, the undisputed facts afford a paradigm example of the mechanism of loss contemplated by the policy.
95 It is not correct to say that two concurrent causes, properly so-called, were involved. In that regard, the reasoning of the Court of Appeal in Petersen & Shadomill Pty Ltd v Union des Assurances de Paris IARD (1997) 9 ANZ Ins Cases ¶61-366 is instructive. In that case, the evidence disclosed two concurrent causes properly so called: flood water from drains (which was covered by the policy) and flood water not from drains (which was excluded). Priestley JA, with whom Mason P & Powell JA agreed, dismissed the insured’s appeal from the decision of Rolfe J in favour of the insurer, on the basis that, general flooding intermingled with flooding from drains did not come within the exception to the flood exclusion, therefore the exclusion did not apply. Secondly, the court held that the general flooding was the dominant cause because the torrent of rain was such that the general flooding would have happened whether the drains had been there or not.
96 In the present case, it is true that high winds were driving the hail but that is not a concurrent cause of the damage by hail. The presence of high winds is a necessary condition, without which the damage by hail would not have occurred, but that is not the same as a concurrent cause. The caravans did not blow away. If damage had been done by the wind blowing the caravans off their supports and tumbling them over, the proximate cause of the damage would be the wind. If the caravans were at the same time pounded by hailstones and penetrated in the manner required by E15, each of which damaged the caravans, two concurrent causes may be said to operate, wind and hail, but that is not this case. If all that happened was wind and rain, the damage in question would not have occurred. In that way, Petersen is distinguishable on its facts; the exception in the exclusion is invoked; and the limit applies. Indeed, if the plaintiffs are correct in their contention that the storm itself, including high winds and the torrential rain, were concurrent causes, properly so-called, Petersen should have been decided in the opposite way because the flood waters in question in that case were all occasioned by torrential rain.
97 It is no answer to contend that the opening words of E15 are confined to hail, as distinct from ‘wind, rainwater or hail’. The clause does not purport to exclude cover for wind or rainwater. If the caravans had blown away and broken themselves up as they tumbled over, the loss would be covered. If rainwater alone had caused the loss, by for example, entering an open door, that too would be covered. But that is not what happened. The mechanism of the loss was damage by hail and the clause is therefore engaged.
98 In the present case, there is an abundance of evidence that the effective, or dominant or operative cause of damage to the caravans, including the annexes and tropical roofs, was hail.
99 It is no answer that the hail fell during a storm which also involved torrential rain and high winds. Hailstorms almost always involve rain and wind. The policy singles out hail and specifically excludes cover for that risk, except in the circumstance in which the hail penetrates the entire thickness of the damaged matter, whereupon the $100,000 limit applies.”
82 This analysis of the evidence compellingly demonstrates, in my view, that it was the penetration of the hail into, or through, the various fabrics of which the caravans (including the tropical roofs and annexes) were constructed which was the proximate cause of the damage to the caravans. Indeed Mr Seton did not really challenge the proposition that hail was a proximate cause of the damage. Rather, he argued that it was but one of several proximate causes, the others being the wind and the rain. In my view Mr Seton did not establish error in the primary judge’s analysis of the evidence and his conclusion on proximate cause.
83 However, assuming Mr Seton is correct and the proximate cause of the damage was the concurrence of wind, rain and hail then, in my view, the principle the respondent invokes in its Notice of Contention that where there are two causes, one covered and the other excluded, the insurer is not liable, would arguably exclude the appellants from recovering: see Wayne Tank and Pump Co Ltd v Employers Liability Assurance Corp Ltd; HIH Casualty & General Insurance Ltd v Waterwell Shipping Incorp (at 610).
84 The Wayne Tank principle was recently considered by Allsop J (with whom Kiefel and Stone JJ agreed) in McCarthy v St Paul International Insurance Co Ltd  FCAFC 28; (2007) 157 FCR 402 (at  ff). His Honour explained (at ) that the principle comes into play where there are “two proximate causes which are concurrent and interdependent, in the sense that neither would have caused the loss without the other … the two causes can be seen as inseparable and so, in effect, as joint”. After analysing the cases considered in Wayne Tank, texts and later cases, his Honour concluded:
“109 In these cases, even though it could be posited that the damage may or would have occurred in any event by the cause that was not excluded, the fact is that the policy in each case was construed as excluding damage caused in a particular way. As a matter of fact the damage was caused in that way (whether or not there was another concurrent cause). Thus, recognising the limits of the cover agreed upon, the loss fell outside the terms of the policy. Wayne Tank has become the best known illustration of this result. But the result is not the consequence of the application of a principle other than that which truly underlay Wayne Tank – the ascertainment and application of the contractual intentions of the parties.” (emphasis added)
85 I have already set out the definition of “Damage” which would clearly extend to damage caused by wind and rainwater. It was not suggested that E14 was a relevant exclusion. This was, therefore, a Policy in which cover was provided for damage caused by wind and rainwater, but excluded (save in restricted circumstances) for damage caused by hail. Mr Seton submitted thatWayne Tank did not apply because there were probably three or four proximate causes. I do not understand the Wayne Tank principle to be constrained by the number of proximate causes. Rather, it applies wherever there is more than one concurrent and interdependent proximate cause, one or more of which is covered, while one (or possibly more) is excluded. In such cases the court construes the parties’ intention to have been that the loss will not be covered. In the present case, on Mr Seton’s argument, the forces of the storm, the wind, hail and rain were interdependent proximate causes. That attracts the operation of the Wayne Tank principle.
86 Thus, even if Mr Seton’s argument is correct and the concurrent and interdependent forces of the storm, the wind, rain and hail, were proximate causes of the damage, the appellants could not recover more than they have because cover for hail damage was prima facie excluded. Cover was provided if, as the primary judge found occurred in this case, the hail penetrated the entire thickness of the material damaged, causing water to enter, in which case the respondent’s liability was limited to $100,000: E15.87 It is unnecessary to reach a final conclusion on this point. There may be an argument as to whether Wayne Tank applies because E15 was a hybrid of an exclusion clause and one limiting liability. This was not argued and should be left for determination to an occasion where it is a live issue.