Supreme Court Judgments

Supreme Court of Canada

North British & Mercantile Insurance Company v. Tourville, (1895) 25 SCR 177

Date: 1895-12-09

THE NORTH BRITISH & MERCAN TILE INSURANCE COMPANY (DEFENDANT)

Appellant;

And

LOUIS TOURVILLE AND OTHERS (PLAINTIFFS)

Respondents.

1895: Oct 3; 1895: Oct 4; 1895: Oct 5; 1895: Dec 9

PRESENT :—Sir Henry Strong C. J., and Taschereau, Gwynne, Sedge wick, King and Girouard JJ.

ON APPEAL FROM THE COURT OF QUEEN'S BENCH FOR LOWER CANADA (APPEAL SIDE).

Insurance against fire—Condition of policy—Fraudulent statement— Proof of fraud— Presumption —Assignment of policy—Fraud by assignor—Appeal—Questions of fact—Reversal on.

Where an insurance policy is to be forfeited if the claim is in any respect fraudulent it is not essential that the fraud should be directly proved; it is sufficient if a clear case is established by presumption, or inference, or by circumstantial evidence.

The assignee of the policy cannot recover on it if fraud is established against his assignor.

If a sufficiently clear case is made out the court will allow an appeal on mere questions of fact against the concurrent findings of two courts. Arpin v. The Queen (14 Can. S.C.R. 736); Schwersenski v. Vineberg (19 Can. S.C.R. 243); and City of Montreal v. Lemoine (23 Can. S.C.R. 390) distinguished.

APPEAL from a decision of the Court of Queen's Bench for Lower Canada (appeal side) affirming the judgment of the Superior Court in favour of the plaintiffs.

The action was upon a policy of insurance against fire issued by the defendant company to one Duval on a quantity of lumber in his yard on the river Nicolet and assigned by Duval to the plaintiffs. One of the conditions of the policy was that it should be forfeited if the claim was in any respect fraudulent and the defence of the company to the action was that

[Page 178]

Duval in his application for insurance had materially [exaggerated the quantity and value of the lumber and obtained thereby insurance above its value and that he had fraudulently exaggerated the amount of his loss. The contention of the plaintiffs on this appeal was that the fraud charged had not been directly proved, but had to be presumed from the evidence which was not sufficient, and also that Duval's fraud could not deprive them of the benefit of the policy. The courts below held that the charge of fraud had not been made out and gave judgment against the company.

The material facts of the case with the pleadings and substance of the evidence are set out in the judgment of the court.

Trenholme Q.C. and Lafleur for the appellants. Á clear case of fraud has been made out and the evidence can be dealt with by this court as well as the court of first instance. Bland v. Ross ([1]).

Duval's fraud forfeited the policy and the assignees are in no better position than he would be.

Beique Q.C. and Geoffrion Q.C. for the respondents. There is no direct evidence of fraud and it cannot be presumed If it could Duval's fraud cannot affect the innocent assignees.

Two courts have found that no fraud was committed and this court will not interfere with those findings. Arpin v. Tue Queen ([2]); Schwersenki v. Vineberg ([3]); City of Montreal v. Lemoine ([4]).

The judgment of the court was delivered by :

TASCHEREAU J. By this action instituted in March 1884, the respondents, as assignees of one Evariste

[Page 179]

Duval, claim from the company appellant the sum of $5,000, being the amount of an insurance policy issued' on the 7th September 1883, by the appellant to the said Duval, concurrently with other policies in various other companies, amounting altogether to $17,000, on a quantity of lumber then piled in a yard on the river Nicolet, which lumber was but two weeks afterwards destroyed by fire.

The appellants pleaded in answer :

1 That the policy was obtained by the false and fraudulent representations of Duval that the lumler insured was worth $30,000 whereas at no time during the existence of the policy was it worth half that sum :

2. That Duval, in the application, materially exaggerated the quantity and value of the lumber mentioned therein, and thereby obtained from the appellants and other companies, represented by the same agent simultaneous insurances to the amount of $17,000 over and above $12,000 prior insurance —thus making $29,000 of insurance in all whereas the lumber thus insured was worth not more than $11,500; the whole contrary to one of the conditions of the policy which was to be null in such an event:

3 That the insurance was forfeited in accordance with a clause in the policy, because Duval falsely and fraudulently exaggerated the amount of the loss in his claim, by putting it at $36,515.68, whereas it did not exceed $11,500.

After a protracted and voluminous enquête the Superior Court gave judgment for the amount claimed. This judgment was confirmed by the majority of the Court of Queen's Bench, Hall J., in a dissenting opinion, holding that though the charge of fraud had not been made out, yet the lumber destroyed was proved to have been worth not more than $15,482.

The company now appeals from that judgment.

[Page 180]

The controversy here as in the courts below, bears exclusively on questions of fact.

We are of opinion that the appellants have fully made out their case.

It is in order, before reviewing succinctly the salient parts of the evidence adduced on both sides to consider a proposition of law strenuously relied upon by the respondents. Conceding, on this argument at least, that if the appellants' contentions as to overvaluation and over-insurance by Duval prevail a clear case of fraud has been made out against him, they pressed upon us the uncontrovertible maxim that fraud is not to be presumed, odiosa et inhonesta non sunt in lege prœsumenda, and argued there from that as the appellants' proof of overvaluation rests entirely upon presumptions and inferences of facts their defence must fail. The respondents would thus seem to contend indirectly at least, that the courts cannot find fraud unless it be directly proved. But, for obvious reasons, this proposition is untenable.

There would be very little protection against fraud if such was the law. Those who intend to defraud do all in their power to conceal their intent.

Their acts could not defraud if they were not clothed with the garb of honesty. Á maxim of the criminal law, based on the same principle is that the guilt of the accused is never to be presumed. But that does not mean that a criminal shall not be convicted if he has not taken a witness for his crime.

It is likewise, as a general rule, only by presumptions and circumstantial or inferential evidence that dishonesty can be proved.

As Coquille said, a long time ago :

Selon les règles do droit, la fraude ne peut être prouvée que par conjectures, parce que ceux qui veulent frauder travaillent de tout leur* pouvoir pour la couvrir.

[Page 181]

Or, as says Dumoulin :

Elle ne serait pas fraude, si elle n'était occulte. Ce sont donc les circonstances qu'il faut principalement considérer, fraus consistit in circumstantiis. '.

It is useless to insist further on this point.

Another legal proposition put forward by the respondents at the hearing is just as untenable. They argued that even if Duval's fraud has been established, they nevertheless are entitled to recover against the company, because, as they contend, they cannot be held answerable for his fraud. This is a startling pro position. They, as assignees, would have a right of action though their assignor had none. They would have been subrogated to a claim vitiated by fraud, but would yet claim the right to pocket the benefit of that fraud. What a protection to frauds on the insurance companies would such a doctrine carry if it were to prevail.

I will now briefly review the facts of the case.

They, in limine, are of a nature to throw discredit on the respondents' claim. Duval, when he took this insurance in his own name, did so, he has to admit, in direct violation of a contract he had with the respondents, by which he had covenanted that all insurances on this lumber would be taken in their name as security for their advances. And he not only concealed this from the agent but concealed it also from the respondents till after the fire. Nay, more, during two days after the fire that one of the respondents was down at Nicolet discussing with him the loss and the claim against the insurance companies, he, Duval, never said a word of these additional insurances he had so taken on the 7th of September. It is only later, and then not from him at all, but from the companies, that the respondents heard of these new insurances.

[Page 182]

Now this suppressio veri, though perhaps not alone directly affecting the result here, as it may be that Duval was not bound to disclose it, yet cannot but at the very outset of the case, under the circumstances, tell unfavourably against him. And it may be doubtful whether if he had revealed the fact that he was so acting in fraud of an express agreement with his creditors the agent would have taken the risks at all.

Another feature of the case which, at its inception, cannot but strike one's attention, is the enormous addition made by Duval to the insurance previously carried by the respondents on this lumber. The latter, though they had over $25,000 at stake, and usually kept this lumber pretty fully covered, had insured for $ 12,000 only, and Duval was aware of it. He, however, on the 1st of September, not only doubles that amount but takes additional insurances to the amount of $17,000, thus, behind the respondents' back, increasing the insurance from $12,000 to $29,000. The reason he gave to the agent for this large increase was the accumulation of sawn lumber in his yard, caused by the Whitehall Company not taking delivery as agreed. Now, it was then not over two working weeks since this Whitehall Company had ceased their shipments. And so it would have been in that short space of time, if we believe him that the insurable value of the lumber in this yard would have increased from $120000 to $29,000. The thing is incredible on its face. But we have, moreover, direct evidence by Kelly the agent of this Whitehall Company, from a statement he personally prepared for his principals three days only before the fire that the whole quantity of sawn lumber in the yard sold to them, but not yet delivered, amounted to only 545,000 feet, of the value of $5,523.75. So that Duval's additional insurance for $17,000 was over three times more than the

[Page 183]

value of the lumber upon which he then himself, instilled it.

The controversy, I ought to have remarked before, turns principally on the amount of lumber that the logs must have produced during the season of 1883, the respondents contending that the fire destroyed 3,820,348 feet, as sworn to by Duval in his proof of loss, whilst the appellants say that there cannot have been in the yard then more than 1621162 feet. As to the value of the lumber, and the quantity of logs that came down to the mill, there is no dispute.

The plan resorted to by Duval and the respondents to establish the quantity of lumber burned is this: to take, in the first place, the amount of sawn lumber carried over from the season of 1882 as per inventory of December of that year viz. 844828 feet 2nd the number of logs made in the winter of 1882.83, and a few scattered logs picked up or bought from others; then deduct from the total the lumber sold before the fire the lumber saved from the fire, and that produced from the logs un sawn at the time of the fire, and the difference should, as they contend, represent the quantity burned, which by that method they would make out to have been 3,820,348 feet, of the value of $36515.88.

The respondents' case rests, it is rightly remarked by the Court of Appeal, almost entirely on the oral evidence of one Marchand, Duval's culler, and on four specifications professing to be four original reports made by him to Duval of the logs cut in the shanties in the months of December January February and March of the winter in question. He says those are the original statements made each month by Albert Duval brother and clerk of his employer, from his, Marchand's, dictation and reading from his culler's book which he brought down from the shanties; that

[Page 184]

after A. Duval had completed the statement it was again checked over to insure correctness; that he Mar chand, then signed the statement, and went back to the shanties for another month's operations. It is a singular fact that a copy of these so-called specifications was never sent to the respondents, though Duval, by his contract with them had bound himself to do so. The respondents never saw them till after the fire. And one cannot but be struck with the similarity in the appearance, as exhibited to us in manuscript, of the paper, the writing, the ruling, which is by hand and consists of double lines of red and blue pencilling, which would lead one who had to do with documentary evidence to say at once that they were all prepared at the same time. They profess to contain an inventory of the different kinds of logs, their length, and contents in board measure. But Marchand's original culler's book, from which all these figures were read out, has disappeared and that disappearance has taken place only after the insurance companies' contestation of this claim. Now Marchand's statements, it is amply proved by the best possible evidence that an insurance company can almost ever bring in such a case, cannot be accurate and no credence can be attached to his testimony. According to his calculations the cut of logs produced on an average during that season :—

Pine	159.	ft.	per log
Spruce	87	"	"
Hemlock	121	"	""
Bass.	132	"	"
Ash	109	"	"

 

His logs, however, were of the same quality and size as those cut by George Ball and, McCaffrey, two respectable mill owners on the same river. Yet for the same year, Ball's pine logs gave only 70 feet, and McCaffrey's

[Page 185]

89 whilst Duval claims 159 feet for his. In spruce, McCaffrey's logs only produced. 53 and a half feet Ball's logs produced 57 whilst according to Duval's theories his produced 87 feet. In hemlock, McCaffrey and Ball got 90 feet per log, whilst Duval claims that he got 121. In bass, Ball got 80 feet per log, but Duval claims to have had 132. In ash Ball got 80 feet per log but Duval claims he got 109 feet. On an average, upon the whole of the operations, Ball & McCaffrey got 78 feet per log, but he, Duval, claims to have got 116. So that according to Marchand if his statements were correct Duval would have got, out of the same quantity, quality and kind of logs, over 2,000,000 feet more than his neighbours in the same business on the same river in the same year, and made over $20,000 more than they did. Or, to put it in another form, if Duval and Marchand are to be believed, they got out of 59,000 logs as many feet in quantity and as much in dollars as any other mill owner on the same river got The same year, or ever got any year before or after the fire out of 90,000 logs of the same kind and size. Or, Duval would have made, according to the calculations of one Welch, an expert examined in the case, a profit, in 1883, of 57 and a half per cent. And yet his neighbours were doing a flourishing business and he was a bankrupt.

If a comparison is made with the result of 1882, the year preceding the fire, taking Duval's own figures, his 59,000 logs gave him, in 1883, 2,300,000 feet more than the same number would have given him in 1882. And the average, upon the whole of his operations would be 116 feet per log for the year of the fire, though only 78 feet for the preceding year. An explanation of how he could in 1883 get 38 feet more per log than his neighbours, whilst in 1882 he got only the same number

[Page 186]

as they did, has not been attempted. The reason is plain. Logs have not such a power of expansion.

If we apply the same test to the years succeeding the fire, as far as proved in the case, the result is the same, over 2,000,000 feet more for the same number of logs in 1883.

Every such test that can be applied reveals the extraordinary coincidence that the over-valuation by Duval is over two million feet. This harmony in the results tells heavily against the respondents.

Duval would have us believe that his 59,000 logs of 1883 were all of 11 inches and over. But that is incredible. It is in evidence that of the whole cut of 1885 for the same mill, from the same limits one third and, of the whole cut of 1887 more than one half, were under eleven inches. McCaffrey's and Ball's logs for 1883. also comprised a large number under eleven inches It is moreover in evidence that instead of the logs of 1883 being cut on the 11 inch limit and being unusually large, as Duval and Marchand swear, the foremen who cut the logs and the men who handled them were ordered to cut them of nine inches and over and that they did cut them that size and even down to eight inches. And the evidence is all one way by the men who made and handled and saw the logs, that they were logs of the same size and description as were made in all other years on the same river from 1882 to 1887 inclusive for that mill and for all the other mills on the Nicolet; all the witnesses say they were the ordinary logs of the river Nicolet.

Not a single reason has been given, or attempted to be given, to explain why in 1883 alone a different kind and size of logs should have been made or their production so enormously increased and a result attained so much larger than that of every other year and every other mill on the same river.,

[Page 187]

Tourville himself, one of the respondents, has to admit that it is the same description of lumber that is sawn from year to year in the locality.

There is another piece of evidence, the result of which also carries great weight against the respondents. in fact, in every form in which an outside check can possibly be availed of by the appellants, as ' well remarked by Mr. Justice Hall in the Court of Appeal, the case presents the clearest evidence of uniform and systematic exaggeration of such an extent, and under such circumstances, as to he absolutely in compatible with good faith.

It is in evidence that all the lumber sawn at the mill up to the 14th of August was piled and loaded under contract at 40 cents per 1000 feet, for which Duval paid $605.64. Now $605644 at 40 cents per 1,000 feet gives 1,514,100 feet, or say in round numbers 1,600,000 as the total output up to the 14th of August, two weeks before the application for insurance and five weeks before the fire. Now as he claims that the fire destroyed 3,820,348 feet, and that he sold 2,232,279 feet before the fire all sawed during that season except 844828 feet it follows that he claims that he sawed 5207 999 feet before the fire. And if 1,600,000 feet only were sawn up to the 14th of August, it follows he sawed the balance of 3600,000 in the five weeks from the 14th of August to the 21st of September, whilst it took him eight weeks after the fire from the 21st of September to the 17th of November, running under pressure, to saw 1,427,351 feet, in that same mill, after it had been put in a better condition.

Or to put it in another way, his mill during 30 days would have cut 120,000 feet a day. And yet the respondents have to admit in their factum that from 35, 000 to 40,000 a day was the utmost that it could ever give. And here again this evidence establishes over 2,000,000 feet as Duval's over-valuation.

[Page 188]

The same result is obtained by a comparison of the cost of sawing. Taking Duval's own figures again, he would have to be able to saw 2,000,000 more feet before the fire for the same wages that it would have cost him after the fire when the mill had been repaired. How it happened could not, of course, be explained. l Then, by asserting as he does that he sawed 5207, 000 feet before the fire, he claims that he sawed before the fire for $1.50 per 1,000 feet, the same lumber that cost him $2.50 per 1.000 feet after the fire in a better mill.

Again, it cost him in wages to run that mill 48 days after the fire $3,555.51 or $74 a day, against $7.862.84 for a pretended 144 days before the fire or $54 perday. At the same rate of $74 per day he must have run only 106.days before the fire, and at 30,000 feet per day cut Only 3,180,000 feet before the fire and not 5207000 as claimed:

The respondents attempted to support their estimates by proving the capacity of the mill and the number of days it was in operation during that season. But far from succeeding in doing so their evidence on this point turns out to be more favourable to the appellants' contentions than to theirs

According to one Chabot's evidence upon which they mainly rely on this part of their case, the mill would have cut 75,000 logs. Now Duval himself cannot claim more than 59,000 the boomage account is there to check him. So that Chabot evidently proves too much; his exaggerations result from his own figures. .Moreover, according to his own estimates, the cut gave in 1883 only 80 feet per log, whilst Duval claims 116. So that, on the controversy as to the average output, the respondents' principal witness entirely supports the appellants' contentions. That which makes against

[Page 189]

his point who swears may be believed, although that which makes for it is disbelieved.

The respondents' evidence as to the number of piles in the yard is also unreliable. Assuming the number '. claimed by Duval to be proved we still are without satisfactory evidence of the quantity contained in each pile. We have on this point nothing but opinions of a vague and unreliable nature, proved withal to be untenable by the various tests I have alluded to. The same may be said of the evidence as to the number of logs sawn after the fire.

As to the evidence of the two Duvals the remarks that I made as to Marchand's evidence fully apply. Their figures are based on Marchand's statements and they, like him, swear to what is conclusively proved to have been physical impossibilities. The number of witnesses who swear to such things cannot have any weight. Non numerantur, sed ponderantur.

Such are the principal features of the evidence in the case.

If as it has been well remarked ([5]) the force and effect of circumstantial evidence depend upon its incompatibility with, and incapability of, explanation or solution upon any other supposition than that of the truth of the fact which it is adduced to prove, the appellants' case is as clearly made out as a case of this nature can ever possibly be.

The evidentiary facts, the facts they rely upon, are unmistakably proved. Their absolute incompatibility with the respondents' theories is also patent. There is no room for any other solution, if these facts are true but that Duval grossly and wilfully exaggerated the quantity of his lumber, both on the 1st of September on his application for insurance and in his statement of loss after the fire It is an utter impossibility that

[Page 190]

the calculations resulting from respondents' own evidence could be correct and that Duval had the quantity of lumber he claims to have had and upon the incorrectness of these calculations there is no room for controversy. The logic of figures is irrefutable.

Such a number of cogent circumstances so closely connected with each other, each separately tending to the same mathematical result and rationally consistent with but one solution circumstances which it is impossible to conceive to have been fraudulently or designedly brought together, and as to which there is no room whatever for the hypotheses of confederacy or error, irresistibly lead to the conviction that the fact of over valuation by Duval to which they all unequivocally depose is true. The united force of so many coincidences carries of itself the conclusion to which its various elements converge. Such an array of facts and figures cannot possibly mislead. It amounts to demonstration, carrying with it absolute certitude, which no oral evidence can weaken.

The disappearance, unsatisfactorily explained, of the culler's pass books, and of all the papers which might have thrown any light upon the controverted facts, is a feature of the case that I should have alluded to previously. The rule omnia prœsumuntur contra spoliatorem is one based on common sense and reason. If these papers had supported the claim they would have been scrupulously taken care of and their non-production justifies us, in law, to come to the conclusion that they would, if forthcoming be adverse to the respondents' contentions. Mill owners it is proved by Rutherford Welch and Ward, always preserve these hooks. And when is it that they have disappeared d Only when a contestation by the insurance companies was dreaded They were in existence when an arbitration about this same fire mentioned in the record took place but were

[Page 191]

not produced before the arbitrators though called for. The ignorance or loose business habits of Duval are invoked as an excuse for their non-production, but ii ne faut pas prendre l'ignorance pour l'innocence, ni la rusticité ou la rudesse pour la vertu.

The appellants have made out the clear case that is required to justify us, nay to oblige us, on an appeal even upon questions of fact, not to adopt the conclusions of the court below. If the case had been tried by a jury a verdict for the respondents would undoubtedly have been set aside, as being against the weight of evidence and a new trial ordered. But, as we are here judges of the facts of the case, as the courts below were, our judgment must be to dismiss the action.

Further there are abundant reasons why this case should not be held to fall under the general rule that, upon such an appeal against the concurrent findings of two courts we should not interfere.

First it was not tried by a jury. 2nd. The judge who determined it in the first instance did not hear the witnesses but gave his judgment upon written depositions. 3rd. The Court of Appeal expressed great doubts in adopting the findings of the judge of first instance. 4th. The judgment of the Court of Appeal was not unanimous, Mr. Justice Hall finding it proved that Duval had over-insured for more than one-half the quantity and value of the lumber. 5th. By the considérants of the judgment of the Superior Court it does not appear that the non-production by the respondents of the written documents bearing on the controversy was taken into consideration. 6th. The Court of Appeal appears to have given weight to a piece of evidence of undoubted illegality the award upon a certain arbitration about this fire to which the appellants were not parties.

[Page 192]

On all these grounds the case is distinguishable from Gray v. Turnbull ([6]); North German Co. v. Elder ([7]); Allen v. The Quebec Warehouse Co. ([8]) * Council of Brisbane v. Martin ([9]); and that class of decisions which we have ourselves given effect to in this court in various in stances inter alia, Arpin v. The Queen ([10]); City of Montreal v. Le Moine ([11]); Schwersenski v. Vineberg ([12]); and from which we do not intend here to deviate.

The case falls under the exceptions foreseen in all the decisions wherein the general rule was followed, and the following have their full application; indeed, they enlarge the duties of a Court of Appeal further than is required to justify the allowance of this appeal l

The judicial committee is not bound by the decision of the court be low upon a question of evidence, although in general it will follow it ([13]).

The parties are entitled to have the decision of the Court of Appeal on questions of fact as on questions of law, and the court cannot excuse itself from the task of weighing conflicting evidence and drawing its own inferences and conclusions, though it should always bear in mind that it has not heard nor seen the witnesses for which due allowance should be made As a rule a court of appeal will be disinclined to interfere, when the judge hearing the witnesses has come to his decision upon the credibility of witnesses as evidenced by their de meanour but otherwise in cases where it depends upon the drawing of inferences from the facts in evidence ([14]).

And in Bigsby v. Dickinson ([15]) it was held that :

Although the Court of Appeal, when called on to review the conclusion of a judge of first instance, after hearing witnesses viva voce, will give great weight to the consideration that the demeanour and manner of the witnesses are material elements in judging of the credibility of the witnesses, yet it will in a proper case act upon its own view of the conflicting evidence." "Of course," said James L.J. in. that same case, "if we are to accept as final the decision of the court of first instance in every case where there is a conflict of evidence our labours would be very much lightened, but then, that would be doing.

[Page 193]

away with the right of appeal in all cases of nuisance, for there never is one brought into court in which there is not contradictory evidence."

And Bramwell L.J. said :

The legislature has contemplated and made provision for our reversing a judgment of a Vice Chancellor where the burden of proof has been held by him not to have been sustained by the plaintiff and where he has had the living witnesses and we have not If we were to be deterred by such considerations as these which have been presented to us from reversing a decision from which we dissent it would have been better to say at once that in such cases there shall be no appeal.

And in Jones v. Hough ([16]) Bram well L.J. said :

First, I desire to say a word as to our jurisdiction. If, upon the materials before the learned judge, he has, in giving judgment come to an erroneous conclusion upon certain questions of fact and we see that the conclusions are erroneous we must come to a different conclusion, and act upon the conclusion that we come to, and not accept his finding. I have not the slightest doubt that such is our power and duty. A great difference exists between a finding by the judge and a finding by the jury. Where the jury find the facts, the court cannot be substituted for them, because the parties have agreed that the facts shall be decided by a jury; but where the judge finds the facts, there the Court of Appeal has the same jurisdiction that he has, and can find the facts whichever way they like. I have no doubt, therefore that it is our jurisdiction, our power and our duty; and if, upon these materials. judgment ought to be given in any particular way different from that in which Lindley J. has given it, we ought to give that judgment.

The cases of Thurburn v. Steward ([17]), and Symington v. Symington ([18]), though they have but a limited application, yet may be referred to on the point. Also what our present Chief Justice said on the subject in Phœnix Ins. Go. v. Magee ([19]); and the case of Russell v. Lefran cois ([20]), where this court reversed the concurrent findings of the two courts below upon a question of fact, and the Privy Council refused leave to appeal. True it is, that there the credibility of any of the witnesses was not directly questioned; but here, even upon that

[Page 194]

point we are in the same position as the two courts below were, their conclusions having been exclusively reached, as ours have to be, upon the mere reading of written depositions.

In Aitken v. Mc Meckan ([21]) the Privy Council, and in The Queen v. Chesley ([22]) this court, also reversed on a question of fact.

We have here, according to the express terms of the statute, to give the judgment which, in our opinion, the Court of Appeal should have given, and that court should have exercised their power to reverse the decision of the Superior Court. The law would be absurd indeed if on the one hand it gave an appeal on questions of fact, whilst on the other hand such an appeal could never be allowed. It is on the assumption that there may be error in the judgment, although two courts have concurred therein, that the right of appeal is given in such a case, even on questions of fact.

The judges of the Appellate Court are as capable in such a case,

says Lord Kingsdown in Bland v. Ross ([23]),

(and indeed are presumed to be more capable), of forming an opinion for themselves as to the proof of facts and as to the inferences to be drawn from them.

In Chard v. Meyers ([24]) Strong V.C., now Chief Justice of this court said upon the same point :

I concede that when there is a balance of evidence causing the de termination of a question of fact to be dependent altogether on the credit to be given to particular witnesses, it is almost impossible for the court, on such an appeal as this, to overrule the decision of the master in whose presence the witnesses have been examined. But if there is as I find here a balance of direct testimony, and the circum stances point strongly to one conclusion, and against the other, I know: of no reason why the court may not review the evidence, and reverse the master's finding.

And the learned judge reversed the master's finding

[Page 195]

discrediting a witness upon whose evidence the master had determined the case.

Add in Morrison v. Robinson (1) the same learned judge held that the rule that where the decision of a question of fact depends altogether upon the credit to be given to the dire it testimony of conflicting wit nesses, the court, as a rule, will adopt the finding of the master, who has had the advantage of hearing the witnesses, applies only where the evidence being directly contradictory there are no circumstances pointing to the probability of one statement rather than of the other

We do not fail to take into consideration I need hardly say, that the fact of the two provincial courts having come to the same conclusion enhances the gravity of our duties, and imposes upon us more than might perhaps be required under other circumstances the strict obligation not to allow the appeal without being thoroughly convinced that there is error in the judgment. But, at the same time, we would unquestionably be forgetful of our duties if we did not form an independent opinion of the evidence and give the benefit of it to the appellants if they are entitled to it. Over-insurance must be put a stop to as much as it is in the powers of the courts to do it. Therein lies one of the greatest sources of fraud in connection with the insurance business. If the assured is not in part a co-assurer with the company, that is to say, if the par ties to the contract have not a common interest in the preservation of the property insured one of the most efficient safeguards against fraud and crime is removed. Any such contract where the assured might expect to make a profit by the destruction of the property in sured is, in law, tainted with immorality. And to re quire from a company, when called upon to pay a loss

(1) 19 Gr. 480. 131/2

[Page 196]

over which hangs any suspicion, a stronger proof than the appellants have made in this case to defeat a fraudulent claim, would be virtually to leave the assurer at the mercy of the assured, a result which obviously, in the public interest, even more than in the companies' interest, should by all possible means be averted. Interest reipubiicae ne Maleficia remaneant impunita.

Appeal allowed. Action dismissed. Costs in the three courts against respondents, distraits to their attorneys.

Appeal allowed with costs.

Solicitors for the appellants : Dunlop, Lyman & Macpherson.

Solicitors for the respondents : Beïque, Lafontaine, Turgeon & Robertson.