This is a suit at common law on a marine insurance policy attaching on the steamtug Carbonero, indemnifying against liability to her tows. One or more barges in her tow were lost. The tug was libeled. The ultimate decision was in favor of the tug, and this suit was brought to make good the expenses of the litiga-*45tioti. The Circuit Court decided in favor of the underwriters, whereupon this writ of erroi- was sued out. The only question we need consider is whether, inasmuch as ihe result showed that the tug was not at fault, any claim can arise on the policy for reimbursement of the ex-peases of the litigation or any part thereof. As the plaintiff in the Circuit Court is the plaintiff in error, and the defendant in that court is the defendant in error, we will refer to them as plaintiff and defendant respectively.
We desire to remove at the outset from our consideration claims made by the plaintiff to die following effect: He says that, when the tug was libeled, it would have been sold unless stipulated for, and he ivas obliged to secure a surety on the stipulation at considerable expense. Also, the libel was originally brought in the District Court for the District of Massachusetts, and then appealed to the Circuit Court of Appeals, both courts deciding in favor of the tug; but the Circuit Court of Appeals found that the tug was not free from negligence, though it also found that it was not proven that the loss of the tows was iii consequence of that negligence. Under those circumstances, the Circuit Court of Appeals refused costs to the tug. Therefore the plaintiff says that the defense was unsuccessful to some extent. All these propositions are easily disposed of, because whatever may have occurred with reference to these details were but the incidents of litigation, and go the same way as the major elements thereof.
Of course, this policy, like all marine policies, was built up historically on the ancient forms; and so, instead of being drawn strictly as a policy of indemnity, it was undoubtedly adapted from the well-known policies on hull, cargo, or freight. Under such circumstances, some expressions will be found to be inconsistent with the main purpose of the policy, and not easily reconciled with other strong and clear language which it contains which must guide us. These facts are so far from embarrassing the court in its decision that they are easily disposed of by being referred to the historical nature of the policy sued on, by virtue of which old material always remains into which the new' is inserted, although the new ordinarily controls the old. There are some expressions which raise a suggestion that it was intended that always, whenever a claim of liability was made, the suit was to be defended at the cost of the underwriters. Many policies of indemnity expressly provide to that effect, especially the usual policies which insure casualties in manufacturing establishments, and other casualties appertaining to various buildings, and liabilities from employers to the employed. What there is in this policy, however, is carefully so ■expressed as to leave it entirely to the option of the underwriters whether or not they would elect to assume the defense of any litigation which might arise. With these explanations, practically all we need say is that, on their face, the terms of the policy are dear that there is no liability on the part of the underwriters when there is no liability on the part of the tug or its owner. The various portions thereof expressing what we refer to were carefully pointed out in the opinion of the learned judge of the Circuit Court, and need not be rehearsed by us.
*46Neither does the plaintiff show us any marine insurance usages, or any decisions of the courts, which justify us in relieving him from the express terms of the policy. On the other hand, the only decisions of the court in point are in favor of the defendant. The last case of authority cited by either party is a decision of the Court of Appeal. Cunard Steamship Company v. Marten (1901) 2 K. B. 611. The late leading case may be said to be Xenos v. Fox (1868) L. R. 3 C. P. 630, 4 C. P. 665, also finally decided by the Court of Appeal. We will refer to these again, merely adding here that their true relation to the law of marine insurance can be best understood, as cases can ordinarily be best understood, by examining them as they appear in their proper setting in .Arnold on Marine Insurance, (7th Eng. Ed., 1901), beginning with section 862 and ending with section 876.
The question we have before us relates to marine insurance, which, although it requires in many respects a broad and liberal treatment, is also in some respects technical, so that attempted analogies to other departments of the law may aid but little, and even not at all, in solving the issue which presents itself. As illustrating this proposition, we refer especially to what is said in Arnold on Insurance in the sections we have cited, and we might, if necessary, take time in referring to other writers on that subject with the same result. The marine insurance doctrine of contributions to “particular averages” and “particular charges,” with which averages or charges the claim now presented must classify itself, if it can be ■ recognized at all, might well have been laid upon a broad foundation, so as to be governed by the same rules which apply to contributions to general averages, or to the adjustment of marine salvages; but they never have been. One has always been distinguished from the other in usages of marine insurance by a broad line. “Particular averages” and “particular charges” must sometimes be contributed to by the underwriters, although bringing their liability in excess of the face of the policy, while not necessarily ' so with reference in general averages and salvages. Therefore even the rules applicable to general averages and salvages cannot be availed of, and so much less can the cases and principles brought to our attention by the plaintiff, but drawn from other departments of the law. For example, he relies on 1 Brandt’s Suretyship and Guaranty (3d Ed., 1905) § 238, as establishing a proposition that where a surety is sued on account of an alleged liability for his principal, and defeats the action, he may recover from the principal the expenses involved in defending the suit. Unfortunately the citation he makes does not sustain him, although probably his proposition is sound to a very considerable extent. This. arises because, within certain lines, the surety is regarded as the agent of the principal, and entitled to such protection as an ordinary agent is entitled to; and also because equity, which for the most part is looked to to protect the surety, has liberal rules and effective remedies with regard to the entire topic of suretyship. On the other hand, the question before us being peculiar to marine insurance, it happens, as was said by Lord Justice Lindley, in Johnston v. Salvage Association, 19 Q. B. D. 458, 460 *47(1887), that contracts like that before us are not contracts of indemnity in any proper sense of the term. What the learned lord justice was considering particularly was the sue and labor clause, which we will presently take up; but his observation applied to this class of policies at large. He said:
“Suc-fi a contract is not a contract of indemnity in any proper sense. It is a contract to pay the assured expenses which he might incur, but not to indemnify him against any claims made by other people against him.”
He then distinguished the broader rules of the equity courts to which we have already referred. All that he said is in line with the observations we have made-, that analogies governing this case cannot safely be found in other departments of the law.
The plaintiff, however, relies more especially on the sue and labor clause contained in this policy. That clause is quoted by him as follows:
‘•It shall be lawful and necessary for the insured, his, her or their agents, factors, servants and assigns, to sue, labor, travel for and make all reasonable efforts in and about the defense, safeguard and recovery of such vessels, crafts and cargoes, or any part thereof, without prejudice to this insurance, and the acts of the insured or this company or their agents, in recovering, saving and preserving the property in case of disaster shall not be a waiver or an acceptance of an abandonment, or as affirming or denying any liability under this policy, but such acts shall be considered as done for the benefit of all concerned, and without prejudice to the rights of either party.”
It is first of all to be noticed that this is not the true ancient sue and labor clause. It omits that portion by which the underwriters agree to contribute. It retains only that which the rigid ancient law thought necessary in order to preserve to the insured his right to abandon. Mr. Phillips seemed to think that possibly the underwriters might be held to contribute under the fundamental rules governing marine insurance, or as an implication from the license to sue and labor. Phillips on Insurance (5th Ed., 1867) par. 1742. We do not find that his suggestion is sustained. In fact, the sue and labor clause is so ancient that it is impossible to go back of it. Owen’s Marine Insurance Notes and Clauses (3d Ed., 1890) speaks of it as of great antiquity, and always embodied in the policy. Gow’s Insurance (1895) 120, places it as far back as the London policy of 1613, and merely adds that it seems to be indigenous to England. It certainly is so ancient that no trace of the English law of contribution goes back of it, so that it is not possible to sustain Mr. Phillips’ suggestion. However, this is not important, because the fact that the present policy made a departure from the ordinary policy in this respect indicates a purpose that the underwriters should not contribute. In this particular the case is less favorable to the insured than Cunard Steamship Company v. Marten (1903) 2 K. B. 511, 512, already referred to, where the sue and labor clause contained the usual agreement to contribute. Moreover, a very common running-down clause contained a stipulation for the payment of costs, meaning the expenses of litigation, in resisting claims of the vessel collided with, whether successful or unsuccessful. Owen, ubi supra, 96. The omission of any such provision in the policy before us is a warning that it ought not to be lightly inserted. Therefore, *48on. these grounds alone, we might properly reject the plaintiff’s proposition in reference to' this clause.
However, the authorities are all against him. We will not rely on Cunard Steamship Company v. Marten, just referred to, because that decision was put on narrow grounds. Lord Justice Roemer, at page 515, does, indeed, restate what we have already said with reference to the historical construction to be given insurance policies, with regard to the fact that language intended to accomplish a peculiar purpos'e is inserted in the printed parts of the customary form notwithstanding the printed parts may have little or no application. to the precise risk insured. But he let the case turn on the fact that the sue and labor clause ordinarily does, and then did, refer to “the said goods and merchandise and ship,” and indicated that it could have no relation, and is inapplicable, to an indemnifying policy. In the present case, Cunard Steamship Company v. Marten, if literally accepted, might compel us to hold that, if the tug had been liable for stranding the tows in question, the tug could not recover from the underwriters any expenditure made in relieving the tows from their stranded position. This we would be reluctant to do. It is sufficient for us that, we determine that the sue and labor clause has relation only to the subject-matter of insurance, which in the present case was the liability of the tug to the stranded tows, and nothing else. The cause of the legal expenses involved here arose, not out of the fact that the policy attached, but out of the fact that somebody claimed that it attached when in fact it did not.
Not only does the positive language of the policy lead to the conclusion we have stated, as explained by the learned judge of the Circuit Court, but we repeat that the authorities are decisive, and this even when the complete sue and labor clause is present. They are well summed up in the sections of Arnold on Insurance to which we have referred, and are very crisply stated in Tyser’s Marine Insurance Losses (1894) p. 51, as follows:
“The underwriter is not liable under this clause [meaning the sue and labor clause], unless he would be liable for the loss to avert which the labor or expense is incurred.” ' v
The earliest case to which we may refer is Biays v. Chesapeake Insurance Company, 7 Cranch, 415, 419, 3 L. Ed. 389 (1813). That was a suit on a memorandum policy limiting liability to a total loss. The court held that the sue and labor clause did not apply, unless, perhaps, in case where the services might have prevented an actual total loss. The principle involved was announced as follows:
“If this clause [meaning the sue and labor clause] be construed with reference to what is most evidently its subject-matter — that is, a loss within the policy — and in Connection with other parts of the instrument, it seems impossible to misunderstand it, or that it should receive so extensive an application as the plaintiff is desirous of giving to it. The parties certainly meant to apply it only to the case of those losses or injuries for which the assurers, if they had happened, would have been responsible.”
These observations were recognized as representing the law by Mr. Phillips in his work to which we have referred at section 1777.
*49We can pass over more than half a century to Xenos v. Fox, L. R. 4 C. P. 665, 667 (1869), already referred to. Here the sue and labor clause was in a policy which had a running-down clause. Xenos v. Fox has been many times relied on, both by the courts and the text-writers. Chief Justice Cockburn there said that the sue and labor clause applied to a loss or misfortune happening to the thing insured; and it was held that the underwriters were not liable for the expenses of a suit brought against the owners of the vessel in whose behalf the policy issued, which suit was unsuccessful. We might well have disposed of this case on the authority of the decisions last cited, namely, Biays v. Chesapeake Insurance Company and Xenos v. Fox, which have stood unquestioned; but the propositions submitted to us were of sufficient interest and importance to justify the consideration which we have given them.
The judgment of the Circuit Court is affirmed, and the defendant in error recovers its costs of appeal.