The loss in this case was clearly covered by the terms of the policy. Those terms unquestionably embraced a loss by fire arising from the use of the basement of the premises as « cabinet-maker’s shop, which included the ordinary use of fire for varnishing and the melting of glue. The policy also, by implication at least, gives the assured the right to occupy and use the basement as it was used at the time when the insurance was made; for it contains an express provision that if the premises shall be occupied in any way so as to render the risk more hazardous than at the time of insuring, the policy shall be void. And the attempt now is to prove, by parol, that the assured, at the time this contract of insurance was made, agreed that he would thereafter occupy this basement room in such a manner as to render the risk less hazardous than it was at the date of the policy. The question then arises whether this supposed agreement, which, if actually made and if there has been no misunderstanding between the parties as to its nature and extent, was of itself a part of the contract of insurance and should have been insert*334ed in the written policy, as a condition or warranty, can be converted into what the defendant’s counsel calls a promissory representation ; and thus avoid the policy on the ground that the assured has not performed his part of the agreement, when no such agreement is either expressed or implied in the written policy which was executed by the agents of the insurance company.
Marshall, who I admit is a writer of very considerable authority on the law of insurance, does indeed speak of two different kinds of representation, one of which he calls an affirmative and the other a promissory representation. But I have not been able to find any case in which a count has adopted this distinction. And the only other writer on the law of insurance, who appears to. have considered a representation as a contract between the parties, is Ellis. He says “ a representation in insurance is in the nature of a collateral contract.” (Ellis’ Law of Fire and Life Ins. 29.) I have examined Millar, Weskett, Annesley, Hughes, Evans, Park, Beaumont, Phillips, Emerigon, Blaney, Quenault, Grun & Joliat, Vincens, Lafond, Persil, Merlin, Pardessus, Boulay Paty, and the works of some other English and foreign writers on the subject of marine, fire and life insurances ; and so far as they say any thing on the subject, I find them to concur in saying that misrepresentation, in reference to insurance contracts, is a false affirmation as to some fact, material to the risk; which affirmation is made by the assured, or his agent, either from a mistake as to the fact represented, or with a design to deceive the insurer.
Annesley says, if there be a misrepresentation, it will avoid the policy, as a fraud ; but not as a part of the agreement, as in the case of a warranty. And if the representation is false in any material point, even through mistake, it will avoid the policy ; because the underwriter has computed the risk upon circumstances which did not exist. (Ann. on Ins. 124.) Blaney says, it is necessary that the contracting parties should have equal knowledge, or ignorance, of every material fact or cir*335cumstance which may or can affect the insurance. And if on either side there is any misrepresentation, allegatio falsi, or supyressio veri, which would in any degree affect the amount of the premium or the terms of the engagement, the contract will he deemed fraudulent and absolutely void.- (Blan. on Life Assurance, 59.) Evans states the difference between a representation and a warranty to be, that the one induces an error in regard to the subject of the contract, and. the other is a stipulation of the contract itself. And he divides represen tations into but two classes—those which are intentionally false, and misrepresentations through mistake. (Evans' Law of Ins. 58, 64.) Hughes speaks of a representation as the assertion of a material fact which the insured knows to be false, or which he makes in an unqualified manner without knowing whether it is true or not. (Hughes’ Law of Ins. 345.) Phillips, an American writer, whose treatise on the law of insurance stands deservedly high, says, a representation is a material fact stated before completing the contract; and a misrepresentation is the statement of such a fact which turns out not to be true. (1 Phil. on Ins. 90.) And Mr. Justice Park, lately one of the English judges, a recent edition of whose valuable work on marine insurances and insurances on lives, and against fire, has been published by Barrister Hildyard, places misrepresentations under the head of frauds in policies. He divides them into two classes—representations intentionally false, and the misstatement of a material fact by mistake. And he defines a representation to be a state of the case; not a part of the written instrument, but collateral to it and entirely independent of it. He also says, if there be a misrepresentation, it will avoid the policy as a fraud, but not as a part of the agreement. (1 Park on Ins. 8 Lond. ed. 404, 433; see also Quenault Des Assur. Terrestres, 289, No. 374, 375; Persil Traite Des Assur. Terr. 297, No. 210, 211; Grun & Joliat Des Assur. Terr. 260, No. 208; and 2 Boulay Paty Cours De Droit Commercial Maritime, 87, tit. 10, § 14.) Chancellor Kent also, in his brief notice of contracts of insurance, *336speaks of two kinds of misrepresentations only ; those which are intentional and avoid the contract for actual fraud on the part of the assured or his agents ; and those which arise from mistake or oversight, which do not affect the policy unless they are untrue in substance and are material to the risk. (3 Kent’s Corn. 283.) It is hardly possible to suppose that if there was such a term known to the law of insurance as a promissory representation, rendering the contract void for the non-performance of a stipulation in the nature of a collateral executory agreement, which the parties did not think proper to make a part of the written contract, it would have'been passed over in silence by all the writers I have referred to.
Nor do I find any such thing as a promissory representation mentioned in the decisions of the courts. On the contrary, Lord Mansfield, who may be called the father of the present system of commercial law in England, clearly repudiates the idea of a representation being promissory. For in the case of Bize v. Fletcher, referred to in Douglass, but more fully stated by Mr. Justice Park, (1 Park on Ins. 441,) he told the jury that there was an essential difference between a warranty and a representation. That a warranty was a part of the contract, and a risk described in the policy was part of the contract; but there could be no warranty by any collateral representation. He said the ground on which a representation affects a policy is fraud ; the representation must be fraudulent; that is, it must be false and material in respect to the risk to be run. All risks are governed by the nature of them, and the premium is governed by the risk. Where a representation accompanies an instrument, it says, “ I will have this understood as my present intention, but I will have it in my power to vary it. ” His language in this case is certainly inconsistent with the supposition that a declaration as to any future event was in any way promissory, or any thing more than a declaration of a present intention ; which intention the assured would have the right to alter, according to circumstances, if the underwriter did not think proper to make it a part of the contract *337of insurance by inserting it in the written policy as a condition precedent to his liability.
It must be recollected too, that in the case referred to, Lord Mansfield was speaking in reference to a written representation, absolute in its terms, but relating to events which were to occur after the making of the policy. For the vessel, at the time of the insurance, was in port, and was insured upon a trading voyage from L’Orient to the Isles of France, and any port or places in the East Indies, China, Persia, or elsewhere beyond the Cape of Good Hope, and until her return to her port of discharge in France 5 and the written representation was as follows; Intends to sail in September or October next. Is to go to Madeira, the Isles of France, Pondicherry, China, the Isles of France, and L’Orient.” But she did not in fact sail until the 6th of December. And instead of proceeding to China, from Pondicherry, she went from thence to Bengal, and passed the winter there. She then returned to Pondicherry and took in her homeward cargo, and was on her way to her port of discharge, in France, when she was captured by a privateer. If the written representation therefore could be considered as promissory, or in the language of Ellis, ££ in the nature of a collateral contract,” the promise was not complied with either as to the time of sailing or as to the course of the voyage.
The effect of the representation is necessarily different where it assumes the form of an absolute affirmation of an existing fact or of a past event; unless from the very nature of the representation it is impossible for the underwriter to understand it in any other way than as a matter of opinion merely. Thus in the case of Macdowell v. Fraser, (1 Doug. Rep. 260,) where the broker represented, without explanation, that the vessel was seen safe in the Delaware on the 11th of December, when in fact she was lost two days before that time, the misrepresentation was .held to avoid the policy. Ho evidence was given, in that case, to show how the false representation occurred. But upon, the motion for a new trial it was stated that the error *338arose from a mistake of the master of another vessel, in reference to the time when the lost ship left New-York; which led the assured to make a mistaken computation. And the court very plainly intimate that if all the facts in the case had been communicated to the underwriter, to enable him to make his own calculation of the probable safety of the vessel on the given day, and he had then chosen to insure without requiring a warranty as to that matter, the result might have been different. So in the case of Curell v. The Mississippi Marine & Fire Insurance Company, (9 Louis. Rep. 163,) the representation was that the ship had sailed on a particular day, when in fact she had sailed twelve days before; so as to make her out of time at the date of the policy. And the court Very properly held that such a misrepresentation avoided the policy. The case of Denniston v. Lillie, upon an appeal from Scotland, (3 Bligh's Rep. 202,) is of a more doubtful character; and it is evident Lord Eldon was not perfectly satisfied with his own decision in that case, as he attempted to invite a re-argument. Not that the representation there related to something that was to take place after the making of the policy—for the vessel had been captured more than a month before that time—but because the representation was not in fact made at the time when the insurance was effected. For the underwriters were only shown a letter, written a long time before the vessel sailed, stating that she would sail on a particular day. His lordship stated the question to be, whether it was a misrepresentation of an expectation, or a statement as to a past fact which was material to the risk. He came to the conclusion that it was the latter; in which conclusion, with due deference, I think he was clearly wrong. For when the letter of Duff & Co., which the underwriter must have seen was dated on the 2d of April, stated that the vessel would sail on the 1st of May, he could not have supposed they intended to do any thing more than to express an opinion or an expectation that the vessel would sail at that time And if he considered' the day of sailing material to the risk, he should have inserted it in the policy as a part of .the *339agreement, instead of relying upon it as the statement of a fact. The case would have been entirely different if the person who applied for the insurance had stated, without explanation, that the vessel had not sailed previous to the first of May, when she had in fact sailed eight days before. For it was possible for the agent to have heard that she was at New-Providence as late as the first of May, by a vessel coming direct from that island with a fair wind. In the case of Baxter v. The New-England Insurance Company, (3 Mass. Rep. 96,) the insurers were informed that the master»of the James stated, that when he sailed, the Robert was not to sail till four days afterwards, when in fact she had sailed four days previous to the James. And no evidence appears to have been introduced in that case to show that there was any truth in this statement that the master of the James was the author of the misrepresentation. The statement that she had not sailed at the particular date was, therefore, a misrepresentation of a past fact, material to the risk as the proof showed, and rendered the policy void. But in the case of Rice v. The New England Marine Insurance Company, (4 Pick. Rep. 439,) upon an insurance on the same risk, proof being introduced that the master of the James did in fact give the information, as stated in the letter upon which the defendants relied as a misrepresentation, the supreme court of Massachusetts decided that the insurers were not discharged, And, what is more material to the case now under consideration, they held that the letter of the master of the Robert, stating that he should leave Kingston on the 12th of August, was not a representation of a fact, but of- an expectation merely ; and therefore not a misrepresentation which would avoid the policy. So in the more recent case of Bryant v„ The Ocean Insurance Company, (22 Pick. Rep. 200s) the present judges of the same court decided that a statement made at the time of effecting the insurance, as to the nature of the cargo which was to be thereafter shipped on board the vessel, if not fraudulently made, was not a representation which would avoid the policy *340if he afterwards changed his mind and took on board a different cargo. And in the case of Allegre's adm'rs v. The Maryland Insurance Company, (2 Gill & John Rep. 136,) where a written application for insurance stated, among other things, “ said brig will sail from La Plata in the course of this month,” the court of appeals in Maryland held that this was not a technical representation of the sailing of the. vessel, but a statement of the opinion or belief of the applicant that she would sail at that time; and therefore did not avoid the policy, although she did not in fact leave till more than a month afterwards.
These cases therefore showr that a statement as to a future fact or event which is in its very nature contingent, and which the insurer knows the party could not have intended to state as a known fact, but as an intention or expectation merely, if honestly made and not with an intent to deceive, is not a collateral contract or a promissory representation which the assured is bound to see performed to render his policy valid. But if the underwriter considers the statement material to the risk, and is unwilling to insure at the contemplated premium without binding the assured to the performance of it as a condition precedent to his liability, he should make it a part of the contract stated in the policy.
Where the assured acts in good faith without any intent to deceive, and without concealing or misstating any fact within his knowledge which it is essential to the underwriters to know, to enable .them to judge of the propriety of assuming the risk and the amount of premium and other conditions of the policy, common justice requires that the party who pays the premium should be informed, by the terms of the written agreement, what is the real contract between him and the underwriters; and it should not be left to the uncertain recollection of any one to prove a different agreement from that which is contained in the written policy. For it frequently happens that where negotiations are carried on between parties, and they suppose they understand one another as to the terms of the bargain, they find, when they come to reduce their agreement to writing, *341that they do not understand it alike. It is for this reason that parol proof is not admissible to vary or alter the terms or the legal meaning of a written contract, by showing what either party said while the negotiation was going on. Fraud, misrepresentation and deceit are necessary exceptions to this general rule ; but there is no good reason why any thing which is in fact a part of the contract between the parties, should form an exception to the rule in an insurance case,
The case now under consideration, I am inclined to think, shows the importance of adhering rigidly to this rule in insurance cases as well as others. For although I have no reason to suppose the president and secretary of the company have not stated the supposed agreement in relation to the use of the fire-place, exactly as they understood it, I have great doubts whether the plaintiff understood that he was to be precluded by that agreement from using the fire-place to heat his glue-pot and warm his varnish ; or that he was to remove his cooking apparatus from the basement room the instant the policy was signed, without giving him a reasonable time to put up his stove for cooking in another part of the house. It must be recollected that the conversation took place in dog-days, when a stove was not wanted to warm his shop ; but when his family were using the fire-place in that room for family purposes. He therefore most probably spoke in reference to that use of the fire-place, when he said he would abandon the fire-place and use his stove. And as the president and secretary do not themselves agree in respect to the words he used, it is possible that both have misapprehended what he did in fact mean to say on the subjectj or he may have inadvertently used language which did not properly express what he intended to agree to on the subject. That he understood he was to abandon the use of the fire-place for cooking, is very probable. For it appears the family only cooked there until the next day, when he had probably gotten his stove up in another part of the house, or had made some other provision for the necessary fire for family purposes. And if he thus discontinued cooking in the fire*342place in good faith, immediately after he had obtained his insurance, it is hardly probable that he would have used the fireplace for the temporary purpose of varnishing, if he had understood that his agreement with the officers of the insurance company extended so far as to embrace such a use. By the terms of his policy, the basement was privileged as a cabinet-maker’s shop, which of course included the necessary use of fire for gluing- and varnishing.
In Whitney v. Mayer, (13 Mass. Rep. 172,) the supreme court of Massachusetts decided that the underwriter could not set up a parol agreement between the parties, which was not inserted in the policy, to defeat the insurance ; but that if the underwriter intended to avail himself of it, he should have made it a part of the written contract. A similar decision was made by Lord Tenterden in Flinn v. Tobin, (1 Mood. & Malk. Rep. 369.) And in this case, no one who reads the testimony can for a moment doubt that a promise to abandon the fire-place and use a stove, was an agreement, and not a representation of a fact. I think the referees erred, therefore, in receiving parol evidence of such an agreement to defeat the policy j and that their report should have been set aside and a venire de novo awarded.
The judgment of the court below is therefore erroneous, and should be reversed.
Bockee, Senator.There is no rule better settled than that “ parol evidence shall not be admitted to contradict, add to or vary the terms of a written instrument.” Whether the admission of the testimony of Garfield and Starbuck was a violation of this rule, is the only point to be enquired into. If the evidence was properly admissible, or if the stipulation or agreement given in evidence had been contained in the policy, the case would present such a violation of contract on the part of the plaintiff in error as wbuld probably bar him from a recovery. It is not denied that a fraudulent representation, material to the risk, might be proved by parol, and would avoid the poli*343cy. Fraud is an element that vitiates all contracts. A representation is of some matter extrinsic the contract, and generally, if not always, relates to the present state and condition of the subject insured. The proof shews that the plaintiff said 11 he would abandon his fire-place in the basement altogether $ he would not use it himself or suffer any other person to use it for any purpose whatever, but would use a stove which he had.” It is contended by the defendant’s counsel that this is a prommissory representation, fraudulently made, material to the risk, and that the non-fulfilment of it precludes the plaintiff from recovering on the policy. How such a prommissory representation, relating to a thing which the party is to do in futuro, is to be distinguished from a contract or agreement, I am unable to comprehend. It is a representation in no other sense than every contract, promise or agreement is a representation that the party will do or refrain from doing a particular thing. By whatever name it is called, it is neither more nor less than an engagement that fire should not be used in the basement fireplace. If this had been contained in the policy, it would have been a warranty, binding upon the plaintiff. Being out of the policy, it is no more than conversation between the parties, inadmissible as evidence of their rights under this written contract. If it was material to the risk, it was material that it should be inserted in the policy, or at least that the evidence of it should be in writing. It was essentially a part of the contract—a stipulation by the plaintiff adding to and varying the terms of the policy. It can hardly be said that a contract is both written and verbal. To admit parol evidence in this case, would be breaking down a salutary and established rule ; and there would be no longer any security in written contracts. In the case of Bize v. Fletcher, (Douglas, 284,) cited by the court below, the question was not on the admissibility of parol evidence of a prommissory representation, but whether a writing annexed to the policy was to be considered a part of the contract, and therefore a warranty to be strictly performed; or whether it was to be taken as a representation, collateral to and out of the *344policy. So in the case of Edwards v. Footner, (1 Camp. 530,) where a ship was to sail with a certain number of guns and men, the question was whether a written memorandum not attached to the policy was to be considered a warranty, or a representation. The precise question before this court is not whether a prommissory representation may exist, but whether parol evidence of such representation can be given, where, if the representation was included in the policy, it would be a warranty, and where the effect of the evidence will be to add to and vary the written contract between the parties. On this question, perhaps no decision can be found in the books exactly in point,.for the reason that the rule applicable to the case is one of the most ancient landmarks of the law of evidence and has never been questioned or disturbed. The policy of insurance was the highest evidence of the contract between these parties, and parol evidence to shew the contract variant from what appears by its terms in writing, would be productive of dangerous consequences and was improperly received. The decision of the court below ought to be reversed.
The President, and Senators Dickinson and Hopkins delivered oral opinions in favor of reversing the judgment of the court below.
All the members of the court, nineteen being present, concurring in this result, the judgment was unanimously reversed.