OPINION OP THE COURT BY
FKEAR, C.J.These cases were argued together. The facte, except as to the form of the policies, are substantially the same as in the case of Yee Wo Chan & Co. v. The Transatlantic Fire Ins. Co., ante p. 160. The first of these cases is a submission to this court on an agreed statement of facts and is based on a policy for $3000 upon a building at the comer of River and Pauahi streets, several blocks from where the fire originated. The second comes up on plaintiffs’ exceptions from the Circuit Court, Eirst Circuit, and is an action on a policy for $5000 upon the same merchandise as tliat covered in the other Yee Wo Chan & Co. case just mentioned. These cases differ from that in that the defendants rely not only on the defense of “civil commotion” as an excepted risk, which the decision in that case settles against them, but also on the exception of loss caused “by order of any civil authority,” which the policy in that case did not contain. The paragraph in which these words occur is the same in each of the policies now in question and is as follows:
“This company shall not be liable for loss caused directly or indirectly by invasion, insurrection, riot, civil war or commotion, or military or usurped power, or hy order of any civil authority; or by theft; or by neglect of the insured to use all reasonable means to save and preserve the property at and after a fire or when the property is endangered by fire in neighboring premises; or (unless fire ensues, and, in that event, for the dam*166age by fire only) by explosion of any kind, or lightning; but liability for direct damage by lightning may be assumed by specific agreement hereon.”
It is contended, particularly by counsel for the plaintiff in the first of these cases, that the words “directly or indirectly” do not apply to the words “by order of any civil authority,” but that they apply to the preceding words only. We cannot agree with him. It is true, as argued, that the enumerated risks ending with “usurped power” may be grouped together as implying violence or lawlessness to a greater degree or as being more uncertain as to the extent of their operation than the risk next named, that is, “order of any civil authority,” and it is argued that this indicates a break in thought in the enumeration of these two classes of risks and affords ground for belief that the words “directly” and especially “indirectly” were intended to apply to the first class only; and yet all these risks, including the last, may be classed together as possessing certain common attributes. They are, for instance, all of a more or less public or political nature. It is true also that at first impression the repetition of the words “or by” in the last clause preceding the first semicolon would seem to indicate a break in the thought; and yet “or” would naturally be repeated before the last of a series. It is here repeated even before the next to the last. And it as well as the word “by” is implied where not expressed before every one of the enumerated risks. The insertion or repetition or omission of these words in any particular instance is largely a matter of taste. The punctuation certainly favors the view that all these risks are classified together, for they are all separated from each other by commas at most, and are as a whole separated from what follows by a semicolon. But a controlling consideration is that the words “directly” and “indirectly” are adverbs modifying the word “caused.” I his is time also of the adverbial phrases “by invasion,” &c. These also modify “caused.” The words “directly or indirectly” do not modify “by invasion,” &c., nor was there any intention to distinguish between a direct and an indirect invasion, or direct and indirect riot, or direct and indirect order, &c. That would be absurd. The direct or indirect consequences of these are what *167was in mind. The “or” before “by order” connects “by invasion,” &c., with “by order,” not “caused directly or indirectly” with “caused” implied before “by order.” Whether the words “directly or indirectly” are continued by implication after the first semicolon or not, we need not decide, but the fact that they modify “caused” which certainly is implied throughout, as well as the fact that direct and indirect damage are distinguished in the last two clauses of the paragraph, would seem to point strongly that way.
, Whether the words “direct” and “indirect” are synonymous with the words “proximate” and “remote” which are so familiar in the law of cases of this class, we need not decide. Nor need we decide whether it would make any difference in this particular case if the word “indirectly” was not implied with the word “caused” before the words “by order.” For, if the order of the Board of Health was the cause of the fire in this instance, was it not the direct and proximate cause? There certainly was no independent efficient intervening cause between the fire originally caused by that order and the fire which consumed the insured property. But holding, as we do, that “indirectly” is implied before the words “by order,” at least makes a portion of this case clearer than it might otherwise be. For instance, in connection with the argument that the word “indirectly” does not apply to the words “by order,” it is urged that the word “order” implies intention and that therefore the excepted risk was meant to apply only to eases where there was a direct intention to burn the property covered by the policy and that, since the civil authorities in this instance not only intended to bum only particular buildings not including those now in question but also took every precaution and did everything they could to prevent the fire from spreading to other buildings, the buildings in question cannot be said to have been destroyed by “order” of the authorities. But when it is considered that the exception covers loss caused indirectly as well as that caused directly by order of any civil authority, whatever force this argument might otherwise have, disappears, “Loss caused * * * indirectly” by order of civil authority is not necessarily merely commensurate with *168the order. It may include loss caused by the order though not intended., We may add that the exception in the policy is not “wbat is done by or in pursuance of” an order, but “loss caused by,” that is, “resulting from” an order. The losses in question resulted from, that is, were caused by, the order, indirectly, if not directly.
It is further contended, especially by counsel for the plaintiffs in the second case, that to exempt the insurer, the order must be lawful and that the- Board of Health could not lawfully order the burning of buildings. The language of the policy is “loss caused by-order of any civil authority.” Nothing is said about the order being lawful, -and it could not have been the intention of the insurer in inserting this clause in the policy to make its exemption from liability depend on the solution of nice questions as to the lawfulness of the orders of the civil authorities. We do not mean to say that the insurer would be exempt in every case in which a person or body possessing civil authority might set fire to a building. In most cases if a person should cause another’s building to be burned, he might be taken to have acted in his private capacity, and not as a civil authority, even though he possessed civil authority of some kind and purported to act in his official capacity. But it is sufficient when, as here, the civil authority, the Board of Health, has the power under the statutes to destroy buildings when it is necessary to do this for the public health, and in the particular case purports to act in its official capacity and does act in good faith and within the apparent scope of its powers. In such ease the insured cannot under a policy of the kind now in question raise an inquiry as to whether the exercise of the lawful power was justified by the circumstances of the particular case. This is an action on contract against -one who did not cause the loss and must be governed by the terms of the contract. What questions the insured might properly raise in an action of tort against the alleged wrongdoer, though a civil authority, who has set his property on fire is another matter.
- One question remains to be considered. Was the order of the Board of Health the cause of the loss from a legal standpoint? *169Either the order or the plague was the cause. There was no independent efficient intervening cause after the order. Should the plague be considered the cause, and the order only an incident to that? Or should the order be considered the cause and the stamping out of the plague only a motive to that? Or should the plague be considered a remote cause, and the order an efficient intervening or proximate cause? Of course, if the order was the cause the defendants are not liable, while if the plague was the cause and the order was not an efficient intervening cause the defendants are liable. It is' often difficult to determine whether a cause is proximate, remote or efficient and intervening. Much law has been evolved on this subject. The proximate cause is not necessarily the nearest in time or place. It is the predominating, operative, efficient cause, — which sets the others in motion. If there are two’ or more independent efficient causes, the loss is usually attributed to the nearest. No case has come to our attention in which the phrase “by order of any civil authority” has been the subject of judicial construction. In the cases before us, should the order be considered the cause in point of law and the stamping out of the plague the motive or object, or should the plague be considered the cause and the order the incident or instrument, as if the order was for the destruction of the buildings in some other way as by pulling them down?
In Insurance Co. v. Boon, 95 U. S. 117, the insured property was destroyed by a fire that spread from another building which had been set on fire by order of the Commander of the United States forces to prevent the rebel forces from getting the military stores therein, the order having been given when it became apparent that the loyal forces could not hold out against the rebel attack then in progress. The policy did not except loss caused by order of any authority, whether civil or military, but did except loss caused by invasion or usurped power. The question was whether the invasion or usurped power of the rebels or the order of the Commander of the Union forces was the proximate cause of the loss. The Circuit Court held that although the attack was the reason for the order yet the order was the cause, an independent intervening cause, and that therefore the insurer was *170liable, inasmuch as that cause was not excepted in the policy, (40 Conn. 575; 3 Fed. Cas. No. 1639). But the Supreme Court reversed that decision and held that the invasion or usurped power was the efficient cause and that the order was only “an incident, a necessary incident and consequence of the hostile rebel attack on the town, — a military necessity caused by the attack. It was one of a continuous chain of events brought into being by the usurped military power, — events so linked together as to form one continuous whole.” But in Portsmouth Ins. Co. v. Reynolds, 32 Gratt, (Va.) 613, referred in 13 Am. & Eng. Encyc. of Law, 2nd Ed., 131, where tire insured buildings were destroyed by fire which spread from other buildings set on fire by order of the United States authorities under an apprehension of an attack by the Confederate forces, the court held that the insurers were liable, that the threatened attack was not the cause in a legal sense, or, if a cause, only a remoto cause of the loss, and that the act of the government authority was an independent intervening efficient cause. In this case there was no attack in progress nor such an urgent necessity as in the other. And in Harris v. The York Mut. Ins. Co., 50 Pa. St. 341, where the loss excepted was that caused “by fire occasioned by mobs and riot,” and where the insured buildings were destroyed by the spread of fire from a bridge set on fire by the Commander of the United States forces to prevent the advance of a rebel army, the court held the insurer liable, both because the force in question was not merely a mob or body of rioters and because the fire was occasioned proximately by the order of the military authorities, and only remotely by the invading army.
Thus there seems to be some difference of opinion as to whether the invasion or usurped power, where loss occasioned thereby is excepted, or the order given by the opposing commander in consequence of such invasion or usurped power, where loss occasioned by such order is not excepted, is the proximate cause of the loss in a legal sense. If the decisions of the United States Circuit Court and the Supreme Courts of Pennsylvania and Virginia should be followed, the order of the Board of Health should be considered the proximate cause and the insurers would not be *171Hable. But if the decision of the Supreme Court of the Hnited States is in point and is followed, the plague is the proximate cause and the insurers are liable. It seems to us that these cases are distinguishable from that decided by the Supreme Court of the Hnited States. In that case as in these cases the insurance was against fire, but in that the expected risk was invasion or usurped power, while in this it is the “order of any civil authority.” Would not the court in that ease have arrived at a different conclusion if the exception had been the “order of any military authority” and not “invasion or usurped power?” The question is not merely what the proximate cause is from an abstract or scientific point of view or what it might be in some other form of action. It is largely a question of the intention of the parties as determined by .the language of the contract and its subject-matter. In that case the court, in order to carry the cause back of the order to the usurped power, laid considerable stress upon the intention of the parties. It said, among other things:
“Policies of insurance, like other contracts, must receive a reasonable interpretation consonant with the apparent object and plain intent of the parties. This is entirely consistent with the rule that ambiguities should be construed most strongly against the underwriters, and most favorably to the assured. * * *
“Invasion involved, of necessity, resistance by the constituted authorities of the government, and the employment of its military force. Destruction of property by fire was quite as likely to be caused by resistance to the usurping miHtary power as by the direct action of that power itself. This must have been foreseen and considered when the insurance was effected. It is difficult, therefore, to believe .that the parties intended to confine the stipulated exemption within the limits to which the assured would now confine it. * * *
“It cannot be said that was not anticipated which military necessity recognized. And the insurers and the assured must have looked for such action by the Federal forces as a probable and reasonable consequence of an overpowering attack upon the city by an invading rebellious force. Having excepted from, the risk undertaken responsibility for such an attack, they excepted with it responsibility for the consequences reasonably to be anticipated from it.”
So here, assuming that we must go back for the cause at least *172as far as tbe order, as shown by all the cases cited above, could it be reasonably contended that the parties did not intend that “loss caused directly or indirectly by order of any civil authority” should cover eases like the present? When loss is caused by order of a civil authority, the officer or official body, unless it acts so far beyond the scope of its authority, that it cannot be regarded as acting as a civil authority at all, necessarily acts in consequence of some cause or reason. It may destroy property to prevent the spread of an epidemic or to prevent an invasion of an enemy or to accomplish other purposes. Whether, if the civil authority should' order the burning of property to prevent the spread of fire, the case would come within the exception, we need not say. In such ease the fire, the spread of which is sought to be prevented, is the very risk insured against. In this case, plague is not the risk insured against. If the exemption was not intended to cover cases like the present, what was it intended to cover? The determination of the proximate cause in the legal sense depends to some extent upon what the main risk insured against is and what the exception is. Eor instance, if loss by fire is the main loss insured against and loss by explosion is the exception, and if this is taken to include loss caused by fire occasioned by explosion as well as that caused by the shock of the explosion, so that the exception is in effect “loss caused directly or indirectly by explosion,” then in case a fire is caused by explosion and spreads to other buildings, the explosion is regarded as the proximate cause and the insurer is not liable. But if loss caused by explosion is not excepted, the loss is regarded as caused by the fire proximately and by the explosion only remotely and the insured is liable. St. John v. Am. Mut. Fire and Marine Ins. Co., 11 N. Y. 516. In neither case would the court go back of the explosion to see what caused that. And yet if invasion were an excepted risk and a shot from the invading army caused an explosion which in turn caused the fire, the loss would be regarded as caused proximately by the invasion and not by either the fire or the explosion and the insurer would not be liable. So here where loss by fire is insured ag'ainst and “loss caused directly or indirectly by order of any civil authority” is excepted, the or*173der and not the fire should be regarded as the proximate cause; and if loss caused directly or indirectly by plague were excepted and not loss caused by order of any civil authority, the plague and not the order might be regarded as the cause, within the meaning of the contract. But since loss by plague is neither insured against nor excepted, the plague cannot be regarded as the cause of the loss of property destroyed by fire ordered by civil authority, though in consequence of the plague. We may add also that here as in the Virginia case there was not the same pressing necessity for' the destruction of the property either in point of time or as to the method of destroying it as there was in the case of Insurance Co. v. Boon. Nor was there the same recognized duty to destroy’it at all. In cases of that kind there was a well recognized military necessity and duty to destroy property of that kind under such circumstances, so that in making the contract such losses could fairly be considered as intended to come within the scope of the exception. Bur there is no well known necessity or duty or practice of burning buildings in case of plague or other infectious or contagious diseases. On the whole we are of opinion that within the meaning of these policies the loss must be regarded as caused by the order of the Board of Health and not by the bubonic plague. Whether the Board was justified in issuing the order is a question not before us.
J. T. De Bolt for the plaintiff in the first case. W. R. Gastle and P. L. Weaver for the defendant. P. Neumann and W. A. Whiting for the plaintiffs in the second case. A. G. M. Robertson and L. A. Thurston for the defendant.Judgment for the defendant in the first of these cases. Exceptions overruled in the second.