When I subscribed to the view set forth in our original opinion, I did so because I had yielded to persuasion to the effect that there is ambiguity in the policy in that it is not expressly set forth therein that the so-called "Hazards not Covered" clause modifies the principal coverage of the policy which is against "Sprinkler Leakage."
I concede that except for this "Hazards not Covered" clause, the loss sustained would have been within the coverage of the policy which insures against "all direct loss and damage by sprinkler leakage * * *", but I attach much significance to the words which immediately follow the word "Leakage", — "except as hereinafter provided", and I think that these words have direct reference to the "Hazards not Covered" clause, and indicate an intention that by that clause the principal coverage is to be modified and limited.
The policy is one against loss resulting directly from sprinkler leakage, but in the "Hazards not Covered" clause it contains a warning to the effect that it affords no protection if the sprinkler leakage is caused by one of the events set forth in that clause.
It was argued, and I at first allowed myself to be persuaded that the "Hazards not Covered" clause was not intended as a modification of the principal coverage but was inserted merely as a warning that should loss occur, not from sprinkler leakage, but from one of the contingencies listed in the clause, then no protection should be afforded by the policy; that the framers of the policy should be held to have adopted the strange plan of setting forth not only the one thing against which the policy afforded protection, but also the many other dangers, from which loss might result but which have no connection whatever with sprinkler leakage. In other words, it provides that "the insurers protect you against sprinkler leakage only and we do not protect you against fire and do not protect you against windstorm" nor against certain other occurrences. How much more reasonable to interpret the "Hazards not Covered" clause as saying that the insurers protect against sprinkler leakage unless the sprinkler system is caused to leak by windstorm, etc.
In Industrial Paper Cordage Company v. Aetna Insurance Co., R.I., 14 A.2d 657, 130 A.L.R. 703, is found a case which I cannot distinguish, and in which there was presented the identical argument which is made here. There the court quoted at length and with approval from the opinion in Luckett-Wake Tobacco Co. v. Globe Rutgers Fire Insurance Co., C.C., 171 F. 147, 148, in which there had been likewise presented this exact contention, except that there the policy was one against fire. In it there was a clause providing that. "This company shall not be liable for loss caused directly or indirectly by invasion, insurrection, riot, civil war, or commotion, etc."
A riot had resulted in a fire, and it was sought to hold the company liable under the policy. It was argued on behalf of plaintiff that the clause which we have just quoted afforded no necessary modification or limitation of the principal coverage against fire because in it there was no express statement to that effect. In other words, it was argued that if it had been intended that that clause should limit the coverage against fire, it should have contained the words "by fire" immediately after the word "loss". Similarly here it is argued that because the words "sprinkler leakage" are not contained in the "Hazards not Covered" clause, that clause does not modify the principal coverage which is against sprinkler leakage. In the Luckett-Wake case the court rejected that argument in the following words: "* * * The view seems to be wholly unsound and unmaintainable. The only loss insured against or which is covered by the policies is `loss by fire,' and we do not doubt that the exception in the policies of loss caused directly or indirectly by `riot' must include those from fires which are the work of rioters. The excepting clause necessarily relates back to a `loss by fire,' as that phrase is previously used in the policy; otherwise, the excepting clause is meaningless as referring to a loss not covered by the insurance. * * * If the loss was not by fire, it was not insured against at all, and the excepting clause was useless. If the loss was by fire, it was insured against, unless the fire bringing about the loss was caused directly or indirectly by riot. If the latter, the loss comes within the excepting clause; but in the former the loss was not insured against at all, so that in either event the defense is good."
I am thoroughly in accord with that view and I cannot see that there is any necessity in including in the "Hazards not Covered" clause in a policy which insures only against sprinkler leakage, a statement to the effect that that clause has application to sprinkler leakage. *Page 392
In the original opinion it is recognized that the policy intended to afford protection against one hazard, and one hazard only — sprinkler leakage. In that opinion we noted that the "Hazards" not covered is plural and we said that "The use of the plural is confusing since there is only one hazard covered by the policy * * *." 1 So. 2d 830, 831.
If there is only one hazard which is covered by the policy, and that one is sprinkler leakage, why go further and say to the insured that the policy does not protect against a great number of other possible occurrences? Among the possible occurrences listed is "loss or damage caused directly or indirectly by seepage or leakage of water thru building walls, foundations, sidewalks, or sidewalk lights (unless caused by `sprinkler leakage')", and in that opinion we mention that as "remarkable" saying: "* * * it is equivalent to saying that the company is not liable for seepage or leakage of water through building walls caused by sprinkler leakage unless caused by sprinkler leakage. `* * * or by condensation or deposits on the "automatic sprinkler system", or by floods, inundations, or backing up of sewers or drains, or by the influx of tide water or water from any source other than the "automatic sprinkler system."'".
I see nothing at all remarkable about that provision. It very plainly provides that if any seepage through walls, etc., should occur without having resulted from sprinkler leakage, and should, after seeping through, cause other sprinklers to leak, there is no coverage, but that if as a result of sprinkler leakage there is seepage, and that seepage sets into action other sprinklers, then there is coverage.
I can easily conceive that a boiler might leak, and that as a result boiling water might seep through the wall or floor and melt sprinkler heads on the other side. The policy merely says that if such should occur, there is no protection afforded by the sprinkler system, but it very properly adds that if such seepage is itself caused by sprinkler leakage, then there is coverage for the results of that seepage even if it sets into action other sprinkler heads.
We reached the conclusion which prompted our original opinion, because we found it difficult to understand how sprinkler leakage might be caused by some of those potential occurrences set forth in the clause. We reasoned that the clause could not mean that there would be no liability for loss resulting from sprinkler leakage if caused by any of those listed contingencies because we could not understand how some of those contingencies might cause a sprinkler system to leak.
I now consider that reasoning unsound, or, as such reasoning is characterized in the law journals, "spurious". In the first place, we know nothing of the insurance business. We have had no experience with sprinkler leakage, and we do not know what possible occurrences might set into action a sprinkler system, and therefore we, I think, are not justified in rejecting, as ambiguous, language adopted by experts, after years of experience just because we may not understand how such occurrences may be caused. And when I more carefully analyze the various causes which are set forth in the "Hazards not Covered" clause, I find it not at all difficult to understand how any one of them might cause a sprinkler system to leak, and all that that clause provides is that if, as a result of one of those causes the sprinkler system is caused to leak, then the legal cause of the resulting damage is not the leakage in sprinkler system but is the occurrence which caused that leakage.
We persuaded ourselves that seepage could not cause a sprinkler system to leak, and that therefore the policy could not mean that there should be no coverage for loss by leakage of the sprinkler system if that leakage had been caused by seepage. Why not? What do we know about the material used in sprinkler heads? How do we know that water, hot or cold, may not dissolve sprinkler heads or the soft metal which holds the valves closed? How do we know that rain water, if it should seep through a roof or a wall, and fall upon such metal, might not cause it to dissolve, and set the sprinkler heads in action?
We also persuaded ourselves that an airplane on the ground could not cause damage to a sprinkler system, and that therefore the policy could not mean that there would be no protection against loss caused by sprinkler leakage, if that leakage had been caused by an airplane on the ground. Again I ask, how do we know this? Why couldn't an airplane on the ground strike a pipe or why couldn't its vibration set into action a sprinkler head? The same may be said of all of the other listed possible occurrences set forth in the "Hazards not Covered" clause. *Page 393
I notice in that clause the word "fire", the word "lightning" and also "cyclone", "tornado", "windstorm". Any one of these and any one of the others set forth might cause a sprinkler system to leak.
Let us consider for a moment the reason for including the word "fire". If this clause should be rejected as ambiguous, then might not a sprinkler leakage policy be held to afford protection against that part of the damage caused by the water from the sprinkler system even though the cause which set the sprinkler system into action might have been fire? How much more reasonable to say just what I think the policy plainly says, that sprinkler protection is not afforded by the policy if that sprinkler system is set into action by fire.
It should be remembered that a sprinkler policy is written at a very low rate and is not intended as a protection against the various dangers which beset any water system in any building. It is offered as a protection only against the peculiar hazards resulting from the installation in a sprinkler system of the soft metal heads and other devices which permit water automatically to escape when necessity arises.
Almost all buildings contain water pipes and mains of various kinds and sprinkler policies are not intended to insure against the dangers of leakage from water pipes or water mains, or even from sprinkler systems if caused by any of those fortuitous occurrences which may break any water system or water pipe, and therefore the framers of sprinkler policies recognize this fact, and expressly exempt from coverage any of the occurrences which might cause any other water system to break or to leak. As a matter of fact, the loss caused to the plaintiff in the case at bar resulted not from a leak in the sprinkler system proper, that is to say, not in that part which was in the premises occupied by the plaintiff. No sprinkler head gave way and no pipe joint in that part of the building leaked. It was a feed pipe on the upper floor, which was attached to the roof, which was broken when the building was unroofed by the windstorm.
I have examined carefully all of the cases cited in the original opinion. Some of them do squarely hold that such a clause is ambiguous. See Hanover Fire Insurance Company v. Newman's, Inc., 5 Cir., 108 F.2d 561, 563. I cannot see it so and I am in accord with those cases which hold that there is nothing ambiguous about it. Plaintiff relies largely on two cases decided by the United States Court of Appeal for the Fifth Circuit, to-wit, Hanover Fire Insurance Company v. Newman's, Inc., supra, and Maryland Casualty Company v. Razook, 24 F.2d 160, 161. The opinion in the Hanover Fire Insurance Company case makes no attempt to discuss the question, it merely announces the result. The Razook case appears to me to be difficult of distinguishment though the policy there came much nearer to being ambiguous than does the policy which is before us. In that case the policy insured against direct loss resulting from "rain or snow driven or admitted through broken or open windows and skylights — except as herein provided * * *."
The loss occurred during the great Miami storm in September, 1926. The tornado broke the windows in the building and rain was then admitted through the broken windows. It seems to me that there was ambiguity in that policy but I cannot find it in the one before us.
The loss here was caused by windstorm, as I think the record plainly shows. Therefore the "Hazards not Covered" clause should be held to exempt the insurer from liability.
I respectfully dissent. To hold otherwise is, I think, to rewrite the policy. In Luckett-Wake Tobacco Co. v. Globe Rutgers Fire Insurance Co., supra, the court used language of which I heartily approve: "* * * We know of no reason, or principle of justice, or rule of law which puts insurance contracts, plainly and unambiguously expressed, outside of the rules of proper construction, or exempts either party to them from the obligatory force of their stipulations. * * * True, insurance companies prepare the contracts, and where their terms are ambiguously expressed, and in certain other well-established instances, the courts lean against them in efforts to ascertain their meaning; but such rules have no application here, where there does not appear to be the slightest trouble in ascertaining the meaning of the terms and stipulations of the policies.'"
I therefore respectfully dissent. *Page 394