Martin v Osborne [1936] HCA 23; (1936) 55 CLR 367 (5 June 1936)

HIGH COURT OF AUSTRALIA

Martin Informant, Appellant; and Osborne Defendant, Respondent.

H C of A

On appeal from the Supreme Court of Victoria.

5 June 1936

Latham C.J., Starke, Dixon, Evatt and McTiernan JJ.

Fullagar K.C. (with him Eggleston), for the appellant.

Sholl (with him Winneke), for the respondent.

Fullagar K.C., in reply.

The following written judgments were delivered:—

June 5

Latham C.J.

I agree with the judgment to be delivered by my brother Dixon.

Starke J.

The respondent was charged upon information under the Transport Regulation Act 1933 of Victoria that he was, on 1st November 1935, the driver of a commercial passenger vehicle which operated on a public highway without the vehicle being licensed in accordance with the Act. A commercial passenger vehicle means any motor car within the meaning of the Motor Car Acts which is used or intended to be used for carrying passengers for reward at separate and distinct fares for each passenger (Act, secs. 5, 45). The Act requires that the prosecutor shall first prove that the passengers carried upon such vehicle were carried for reward, but the onus is then upon the defendant of proving that the passengers carried upon such vehicle were not carried for reward at separate and distinct fares for each passenger (Act, sec. 47). The police magistrate before whom the information was heard convicted the respondent, but his decision was reversed in the Supreme Court. This court, however, granted special leave to appeal.

On Friday, 1st November 1935, the respondent drove a motor car from Melbourne to Ballarat, and several persons were thereby carried all or part of the way to Ballarat. But there was no evidence that these persons were carried for any reward. This evidence, as Gavan Duffy J. held in the Supreme Court, was obviously insufficient to establish that any of these persons had been carried for reward. The prosecution, however, also proved that the respondent on the two preceding days, namely 30th and 31st October, drove the same car between Ballarat and Melbourne and back again, and that several persons were thereby carried between Melbourne and Ballarat or Ballarat and Melbourne. It was further proved that he carried luggage in his car for some of the persons he drove. And the evidence also proved that the respondent commenced his journeys from Ballarat at a point in Sturt Street outside a tea-room or sweet-shop, or in Melbourne from the corner of Collins and Spencer Streets; and that the tea-room or sweet-shop had a notice posted upon it: "Osborne's Motor Service. Ballarat to Melbourne, 8-cylinder sedans leave here 10.30 daily, extra trips 1.30 Saturdays." But it was admitted that the respondent's brother had a licence for a six-cylinder Chrysler car and used it for the service described. Gavan Duffy J. held that all this evidence should have been rejected.

Evidence is confined to the fact in issue, and facts similar to but not part of the fact in issue are not, in general, admissible to prove its existence. On the other hand, where the fact in issue and other similar facts consist of a series of acts done in pursuance of some business operation or some system so as to form a continuous course of dealing or conduct, the similar facts are admissible in evidence.

In the present case, the act charged against the respondent related to a particular day, but the prosecution tendered evidence, which was regarded as inadmissible in the Supreme Court, for the purpose of showing that it was not an isolated act but one of a series of acts in the conduct of a transport service between Ballarat and Melbourne. In my opinion, the evidence was admissible for this purpose. It tended to establish the fact that the respondent was regularly engaged in driving motor vehicles between Ballarat and Melbourne for the carriage of passengers. But the force of the evidence depends upon its sufficiency to exclude every other hypothesis than the one under consideration, namely, whether the respondent was carrying the passengers for reward. The facts may be consistent with the commission of the offence charged, but are they inconsistent with any other rational conclusion? (Taylor on Evidence, 12th ed. (1931), pars. 63-69). The prosecution failed to call any of the passengers, and rested simply upon the circumstances already mentioned for proof of the charge. The evidence, such as it was, established the charge to the satisfaction of the police magistrate, but Gavan Duffy J. thought there was not sufficient evidence to justify the finding that a fare was charged on Friday, 1st November, the day on which the offence was charged. Apparently the learned judge considered that there were other rational hypotheses than the commission of the offence—such, for instance, as the voluntary carriage of the respondent's friends, or a joint adventure with the motor vehicle. It is a mere inference of fact, but the learned judge was entitled to use his knowledge of the world, and that attempts were being made in various directions to avoid the provisions of the Transport Regulation Act 1933, and that the respondent was acting openly and asserting to inspectors appointed pursuant to the Act that he was not taking fares. The conclusion reached by the learned judge is as likely, I think, to be right as the opposite one. At all events, I share it, and think the evidence is consistent with rational hypotheses other than the guilt of the respondent.

The appeal should be dismissed.

Dixon J.

The ingredients of the offence of which the defendant was convicted before the magistrate are to be gathered from sec. 45 of the Transport Regulation Act 1933 and from the definitions contained in sec. 5 of the expressions "commercial passenger vehicle" and "operate."

The charge was that he was the driver of a commercial passenger vehicle which operated on a public highway and was unlicensed. "Commercial passenger vehicle" is defined to mean any motor car used or intended to be used for carrying passengers for reward at separate and distinct fares for each passenger. "Operate" means carry passengers for reward at separate and distinct fares for each passenger. The essential elements of the offence, therefore, were that the defendant drove an unlicensed motor vehicle on a public highway carrying passengers for reward at separate and distinct fares. The burden of proving each of these matters, except one, lay as usual upon the informant. The exception is made by sec. 47 which provides, in effect, that, if the informant proves that the passengers were carried for reward, the burden should be on the defendant of proving that they were not so carried at separate and distinct fares.

In support of the prosecution ample proof was adduced that the defendant drove a motor vehicle on a public highway, that it carried passengers and that it was not licensed as a commercial passenger vehicle. The question in the case relates to the sufficiency and admissibility of the evidence offered to establish that the passengers were carried for reward. The charge was that he operated a car—a Hupmobile—between Ballarat and Melbourne on Friday, 1st November 1935. The proof given was circumstantial. The circumstances relied upon were briefly these. The defendant's brother, named Ernest Osborne, had a commercial passenger licence in respect of a six-cylinder Chrysler car. This car with a number of other cars was seen to stand in one of the principal streets in Ballarat outside a tea-room or sweet-shop which bore a sign to the effect that the eight-cylinder sedan cars of Osborne's Motor Service from Ballarat to Melbourne left there daily. Not far away was a garage. On the morning of Wednesday, 30th October 1935, the defendant and the Hupmobile car and other cars were outside the shop. Baggage was handed in and out of the shop and into various cars. The defendant, who took part in handling the baggage, placed some of it in the Hupmobile, in which four passengers sat. A lady with a suit case then alighted from a passing bus and the defendant took her suit case to the car, in which he gave her a seat. He then drove down to Melbourne where he dropped his passengers at various points, some of them taking rugs or suit cases. On the same afternoon, the defendant and the Hupmobile car were stationed at the corner of Spencer and Collins Streets in Melbourne. At intervals three passengers got into the car, which the defendant then drove off on the road to Ballarat. In Footscray the car picked up another passenger, whom the defendant helped in. The car was not followed further than Bacchus Marsh. On the following morning, Thursday, 31st October, the defendant and the Hupmobile were outside the shop in Ballarat. A passenger sat in the car. The defendant spoke to another man who drove the car away. It returned shortly with two more passengers and a quantity of luggage. After standing a little, it was driven off, but whether by the defendant or someone else did not appear, at any rate by direct evidence. Two hours later it was standing empty at the corner of Spencer and Collins Streets in Melbourne with the defendant beside it. An hour passed and it was still there, but with people in it. An Auburn car stood near it. The Hupmobile and its passengers were then driven off, but not by the defendant, who remained standing in the street. What he did then did not appear, but next morning, the date laid in the charge, he was outside the shop in Ballarat. The Auburn car was driven up to the shop carrying passengers, and, an hour or more later, it carried down to Melbourne a number of passengers, including the defendant, who rode next to the driver. In the afternoon of that day the Hupmobile and the Auburn cars were again standing at the corner of Spencer and Collins Streets and the defendant was there also. Two ladies and a child were seated in the Hupmobile. The Auburn drove off first. A little later the Hupmobile with the defendant in charge began its journey to Ballarat. On the outskirts of Bacchus March it stopped to pick up a youth. The car would not start again and the youth went for a mechanic, who remedied the defect. It carried the youth as a passenger to Ballarat, where he was dropped at the corner of a street. A lady and a child were dropped at another point. There the defendant spoke to the officers who were following him and said that they need not follow him further as the remaining lady was a friend of his wife's and he was taking her home. As it was raining hard, the officers took his word for it.

On the part of the defendant an objection was made to the evidence of what occurred on the two days preceding that laid in the charge. The Court of Petty Sessions overruled the objection, but, on review, the Supreme Court decided that the evidence ought not to have been admitted and that the admissible evidence did not authorize the inference that the defendant carried passengers for reward. I am unable to agree in the opinion that the evidence was inadmissible. If an issue is to be proved by circumstantial evidence, facts subsidiary to or connected with the main fact must be established from which the conclusion follows as a rational inference. In the inculpation of an accused person the evidentiary circumstances must bear no other reasonable explanation. This means that, according to the common course of human affairs, the degree of probability that the occurrence of the facts proved would be accompanied by the occurrence of the fact to be proved is so high that the contrary cannot reasonably be supposed. The circumstances which may be taken into account in this process of reasoning include all facts and matters which form constituent parts or ingredients of the transaction itself or explain or make intelligible the course of conduct pursued. The moral tendencies of persons, their proneness to acts or omissions of a particular description, their reputations and their associations are in general not matters which it is lawful to take into account, and evidence disclosing them, if not otherwise relevant, is rigidly excluded. But the class of acts and occurrences that may be considered includes circumstances whose relation to the fact in issue consists in the probability or increased probability, judged rationally upon common experience, that they would not be found unless the fact to be proved also existed. The application of this, as of any other general statement about relevancy is subject to the well-known specific rules of exclusion. For instance, the rule against hearsay and the warning implied in the descriptive phrase res inter alios acta lead to the exclusion of evidence not only of what a stranger to the cause has said but also of what he has done, if it is offered to prove his knowledge of some fact and thus the existence of that fact, notwithstanding that the fact itself be relevant and its proof by another medium be receivable. But it is at least true, I think, that the acts of a party are admissible against him whenever they form a component in a combination of circumstances which is unlikely to occur without the fact in issue also occurring. The repetition of acts or occurrences is often the very thing which makes it probable that they are accompanied by some further fact. The frequency with which a set of circumstances recurs or the regularity with which a course of conduct is pursued may exclude, as unreasonable, any other explanation or hypothesis than the truth of the fact to be proved. For example, the probability that the neglected condition of a barber's implements was the cause of his customer contracting barber's itch becomes much higher when it appears that about the same time two more of his habitual customers also contracted the disease (Hales v. Kerr[1]). If four close relatives of a woman, dwelling in her house and eating meals prepared by her, die of arsenical poisoning one after another within a few months, the inference that she wilfully administered the poison has more support than if one death only occurred in such circumstances (R. v. Geering[2]). The discovery of a number of dead bodies of infants buried in the ground at different premises lately occupied by a baby farmer greatly increases the probability of her having murdered an infant entrusted to her charge which has disappeared (R. v. Makin[3]; R. v. Knorr[4]). An inference from circumstances that on a specific occasion an act of adultery or of incest took place between a man and a woman may be uncertain until it appears that a previous sexual relationship existed between them, but the addition of that fact may remove doubt (R. v. Ball[5]; McConville v. Bayley[6]; R. v. Goldsworthy[7]). For a medical man or midwife frequently to procure abortion makes it unlikely that his or her proved association with a specific case of abortion was not criminal (R. v. Bond[8]; R. v. Graham[9]). The repetition by an accounting party of the same error or kind of error in calculating at excessive amounts the totals of his disbursements makes it probable that the overstatement was fraudulent (R. v. Richardson[10]; cf. R. v. Proud[11]; R. v. Garsed[12]; R. v. Hiddilston[13]; Hardgrave v. The King[14]; R. v. Finlayson[15]). In the same way repeated utterings of coins or notes in fact counterfeit or forged, and repeated obtainings of money by representations in fact untrue increase the probability that on a specific occasion a coin was uttered or a pretence made with guilty knowledge and intent (R. v. Whiley[16]; R. v. Forster[17]; R. v. Weeks[18]; R. v. Francis[19]).

In the present case the evidence to which the defendant objected, when combined with that describing the actual journey laid as the offence, shows that for three consecutive days the Hupmobile car behaved exactly as service cars do, and that for the greater part of that time it was under the control of the defendant. It appeared to be following a regular course of business in standing at convenient rendezvous for passengers, picking up and setting down passengers where they desired at either end of the journey, carrying their luggage and plying between large cities upon a set route. The place of rendezvous at Ballarat exhibited a sign notifying that a service was conducted from that point, although, it is true, by cars of a description to which the Hupmobile did not answer. Different passengers were carried on each journey and they possessed no common characteristic. In my opinion such evidence was admissible, because it tended to show that the defendant was operating the car regularly for the carriage of passengers between the two cities, and thus to make it improbable that the passengers were not carried for reward. When this evidence is admitted and taken into account, I think the finding of the magistrate is amply justified. The conclusion is well warranted that the defendant was using the car in aid of his brother's motor service and was carrying passengers for reward between Ballarat and Melbourne.

It was urged on his behalf that all the circumstances were consistent with his carrying persons only who had combined to form some transport club or other association designed to enable them to travel without paying separate and distinct fares. There are, in my opinion, two answers to this contention. The first is that the informant has the burden of proving only that passengers are carried for reward. Even if it be practicable so to organize an association as to avoid the payment of distinct fares for each journey, the hypothesis that the car operates without any reward at all is quite unreasonable and ought to be rejected from consideration. If some reward is to be inferred, it is for the defendant to disprove that it takes the form of separate and distinct fares. The second answer is that the hypothesis that the passengers did not pay separate and distinct fares for each journey itself appears unreasonable. It is sufficiently negatived by the circumstances of the case. The considerations which are opposed to it include the notice at the shop, the fact that it is a long journey between two large cities and not a short city and suburban service, the places where one or two of the passengers were picked up, the varied description of the passengers, and, finally, the difficulty of organizing a body of persons in such a way as to avoid the payment of separate fares. These matters combine to make it too improbable that such a plan was followed.

I think the appeal should be allowed and the order of the Supreme Court discharged. In lieu thereof the order nisi should be discharged with costs. Pursuant to his undertaking given on obtaining special leave to appeal, the appellant must pay the respondent's costs of the appeal to this court.

Evatt J.

The information upon which the respondent was convicted alleged that on Friday, November 1st 1935, between Ballarat and Melbourne, he was the driver of an unlicensed "commercial passenger vehicle" which operated on a public highway. By sec. 45 of the Transport Regulation Act 1933, the driver of a "commercial passenger vehicle" which (a) operates on any public highway, and (b) is not licensed as such under Part II. of the Act, becomes liable to a penalty.

By sec. 5 of the Act "commercial passenger vehicle" and "operate" are so defined as to show clearly that the offence charged against the respondent was that, on the day and at the place mentioned, he was driving a motor car in which there were being carried passengers for reward "at separate and distinct fares for each passenger."

Next, sec. 47 of the Act provides that, upon such a prosecution as the present, "the onus shall lie upon the defendant of proving that the passengers carried upon such vehicle were not carried for reward at separate and distinct fares for each passenger; but the defendant shall not be under any obligation to discharge such onus until the informant first discharges the onus of proving that the passengers carried upon such vehicle were carried for reward."

The respondent called no evidence and the questions which now arise are whether the magistrate was correct (1) in finding that the informant had proved that the passengers were being carried "for reward" and (2) in admitting evidence—strongly objected to by the defendant—as to the conduct of the defendant in relation to the same motor vehicle on the two days prior to Friday, 1st November 1935 in respect of which alone the respondent was charged and convicted.

As to the Friday itself, the evidence showed that the car with passengers in it was driven by the defendant along the road to Ballarat. It left Melbourne at about 4.40 p.m. and arrived at Ballarat at about 7.20 p.m. The manner of collecting, carrying and attending to passengers was such as would characterize the operation of a motor service being conducted for reward. But there was no direct evidence of any payment or promise to pay on the part of any passenger.

The effect of the evidence objected to and admitted may be summarized as follows:—(1) Wednesday, 30th October. In the morning the defendant handled the baggage of passengers at Ballarat. This took place both inside and outside a tea-shop. He placed the baggage of passengers in the same car, conducted them to the car and then proceeded to Melbourne. There the passengers alighted, the defendant assisting them. In the afternoon, the defendant again took the car to the place from which he departed for Ballarat on the Friday. At about 5 o'clock passengers arrived at this place and were then carried far along the road to Ballarat. The passengers were not those who had travelled down from Ballarat to Melbourne in the morning. (2) Thursday, 31st October. At about the same hour of the morning, the defendant was again busying himself in and about the tea-shop at Ballarat. He conducted a passenger to the same car, which he drove away. Soon afterwards, he returned to the tea-shop with two additional passengers and additional luggage. In the afternoon the defendant was again with the car, which was then standing at the position in Melbourne whence it was driven off towards Ballarat with passengers both on the Wednesday and on the Friday. (3) On the Friday morning a notice "Osborne's Services, Ballarat to Melbourne" was displayed in the window of the Ballarat tea-shop.

Let us first consider the case without reference to the authorities.

The two questions (a) whether the magistrate was in error in admitting evidence as to the events of the Wednesday and Thursday and as to the notice displayed, and (b) whether he was in error in finding that, on the Friday, the passengers carried by the defendant were so carried for reward, are closely related. There was no direct evidence of any payment by the passengers to the driver. But the course of operations on the Friday itself made it seem quite possible or even probable that the defendant was driving passengers in the car in the course of conducting an organized and systematic motor service between Ballarat and Melbourne. In order to test the validity of this prima facie inference as to organization and regular conduct of business, the previous course of events (say) on the Wednesday and Thursday would obviously be of the greatest relevance. The actual course of events on those two days made the prima facie inference of regular motor service business on the Friday not only likely but so highly probable as to be almost certain. This practical certainty was strengthened, not weakened, by the display of the motor service notice at Ballarat, assuming of course that the "Osborne" referred to was not the defendant.

It would be strange and irrational if the common law, to which this part of the law of evidence belongs, forbade the proof of facts which, regarded in their totality, made "so highly probable as to be almost certain" an inference as to the nature of the services rendered to the passengers on the Friday. For if the defendant was, on the Friday, driving the car in the ordinary course of conducting a passenger service between the two cities, the further inference is that he was doing so for reward. This is only another way of saying that most sane persons do not conduct businesses except for reward.

In the ordered processes of logical thought, in matters such as these, regard must be had to every hypothesis which can compete with the hypothesis suggested by experience and knowledge of human affairs. In the present case the latter hypothesis is that the defendant was pursuing his business for reward.

An illustration of the proper method of approach is provided by Morgan v. Babcock and Wilcox Ltd.[20]. There the managing director of a company was proved to have parted with a very large sum of money to be used for the purpose of bribery in the interests of the company. The precise issue was whether "the company" itself had provided the bribe, and the only hypothesis set up to compete with that of the prosecution was that the managing director acting personally had furnished the money from his own funds. But, in the absence of any positive evidence supporting this competing hypothesis, the court deemed it to be so highly improbable as not to be worthy of serious consideration. Accordingly the company was convicted.

In the present case the course of argument before us also laid bare the paucity both in number and quality of the hypotheses which might be advanced to explain the services rendered to the passengers by the defendant. I shall write down three, viz.: (1) his object was purely charitable and philanthropic, i.e., to carry on the passenger service out of pure goodness of heart; (2) the journeys proved were visits for some non-commercial purpose such as some family or sporting reunion with which the defendant was closely connected; (3) there was some form of joint ownership of the car on the part of the passengers, e.g., they were members of a transport club.

Merely to write them down demonstrates the improbability of each of these three hypotheses. As to (1), such a rare manner and form of public charity would probably be accompanied by great public notoriety, perhaps public alarm. Certainly human nature would be contradicted if the numerous beneficiaries could be restrained from explaining and vindicating the high and very rare motives of the defendant, not only to the world of Ballarat and Melbourne, but also to the police magistrate.

As to (2), the hypothesis of a family re-union might conceivably explain some of the facts as to Friday itself. But what a very curious form of festive reunion—so quick to meet, so soon to part! And why delay the joyous reunion by letting down the passengers, not at the place of reunion, but separately and accompanied by luggage?

As to (3), a transport club the members of which would be entitled to conveyance between Melbourne and Ballarat would certainly be an extraordinary form of club. Even if such a club had been promoted and organized it could hardly be expected to continue in operation if persons such as the defendant, rendering such valuable services as drivers and attendants, were not rewarded for their services.

In truth, the only rational explanation of the defendant's systematic driving of passengers between the cities is that he did so for some payment or reward. Here, as in Morgan's Case[21], the defendant called no evidence to support his defence. And it is impossible to say that the magistrate was not entitled to pay regard to that fact, especially as the complete information of his arrangement or understanding with all the passengers lay within the defendant's knowledge.

Considered from the point of view of strict logical reasoning, the two questions raised upon the appeal admit of only one conclusion, viz.: (1) that the evidence was rightly admitted, being not only strictly relevant to the question in issue but in its own nature so highly persuasive and probative as to be almost conclusive of the question in issue, and (2) that, when admitted, the evidence not only entitled but practically compelled the magistrate to yield to its great compelling force and convict, especially as all the hypotheses competing with the hypothesis of reward were not only fantastic in themselves but were quite unsupported by evidence.

The question remains whether any infringement of the rules of evidence was committed by the magistrate in his decision to admit the evidence. If so, the claims, both of logical reasoning and of common sense, will have to be dismissed and the conviction quashed. The matter is not only of general importance but of compelling interest.

Sir James Stephen, who endeavoured to explain the rules of evidence upon a rational basis, adopted the principle of relevancy to facts in issue as the criterion for allowing proof of any fact (Digest of the Law of Evidence, 5th ed. (1899), art. 2, p. 3). It was by way of application rather than qualification of this general principle that he regarded the power of the Court to exclude evidence of facts which, though relevant to a fact in issue, appeared to have too distant or remote a bearing upon the existence of such fact in issue.

In art. 10 Stephen, 5th ed. (1899), laid it down that proof could not be given of a fact which rendered the existence or non-existence of any fact in issue probable merely by reason "of its general resemblance thereto" (ibid., p. 15). He added: "You are not to draw inferences from one transaction to another which is not specifically connected with it merely because the two resemble each other. They must be linked together by the chain of cause and effect in some assignable way before you can draw your inference" (Digest, 5th ed., p. 171).

Here it may be suggested that Stephen's view may be expressed by the proposition that, as a general rule, no satisfactory or probable inference of fact B (e.g., that X killed Y) can be drawn from the knowledge or proof of fact A (that X killed Z). The incidents of the two homicides would have to be related together in such a way that some inference as to the fact in issue could be drawn with some degree of probability from the fact or facts sought to be proved.

Art. 12 of Stephen deals with the proof of "facts showing system." But his proposition is expressly limited to cases where a question arises whether an act "was accidental or intentional." In such cases, he says, the fact that the act in question formed "part of a series of similar occurrences in each of which the person doing the act was concerned" is deemed to be relevant.

Under this article Stephen mentions R. v. Gray[22] where, upon a charge that the accused had burnt down his house in order to obtain insurance moneys, evidence was admitted that, in two other houses of the accused, fires had occurred and insurance payments were obtained by him. Stephen, however, thought that the authority of Gray's Case[23] was weak, as the admission of the evidence of the other fires "may practically involve the trial of several distinct charges at once, as it would be hard to exclude evidence to show that the other fires were accidental" (Stephen, p. 20).

It is not easy to appreciate why this consequence should exclude the admissibility of the evidence in question. If unexplained by the accused it would properly have great and, perhaps, overwhelming force. The authority of Gray's Case[24] is now established by Makin v. Attorney-General for New South Wales[25], and in his charge to the jury in R. v. G. J. Smith[26] Scrutton J. treats the case as correct.

As has been noticed, art. 12 of Stephen is confined to cases where the question has arisen whether an act is accidental or intentional. He explains that in each case which is illustrative of art. 12 the evidence admitted went to prove "the true character of facts which, standing alone, might naturally have been accounted for on the supposition of accident—a supposition which was rebutted by the repetition of similar occurrences" (ibid., p. 172).

The leading case of R. v. Geering[27] is treated by Stephen as an application of art. 12. There, evidence was admitted against the accused that, after the administration of poison to the husband, her three sons had the same poison administered and that the meals of all four had been prepared by the prisoner.

In the application of the general principle of relevance, it is plain that the degree of resemblance and connection between the fact in issue and the fact sought to be adduced in evidence must be closely examined. Mere general resemblance is insufficient. It is not permissible to show that the accused or party is the kind of person who might be expected to do the kind of thing which is imputed to him. On the other hand, poisonings and fires, though often the result of accident, do not, in ordinary human experience, recur in the same family circle or in the case of the same occupier. Accordingly, evidence is allowed to prove the recurrence of such poisonings or such fires respectively without proof that the party concerned was more than "involved," in order to show the high degree of improbability attending the hypothesis that the poisoning or fire under particular scrutiny is an accident. As human experience negatives the likelihood of any repetition of disastrous accidents of a peculiar kind involving the same person, proof of such repetition is allowed to destroy or reduce the probability of the hypothesis of accident in the given case.

These illustrations are sufficient to show why Phipson insists upon the necessity of some "specific nexus" between the fact tendered in evidence and the main fact or transaction, and also why he explains that "the admissibility of similar facts as presumptive proof of the fact in issue is in truth mainly a question of degree, or of our knowledge and understanding of the causes of events, as to which, in many cases, the progress of science may change the law" (7th ed. (1930), p. 154).

The final reason why the question as to the relationship between the fact tendered in evidence and the issue required to be proved is one of "degree" is that the judgment or inference involved is not one of deductive logic. What is involved is the task of admeasuring the probability or improbability of the fact or event in issue, if we are given the fact or facts sought to be adduced in evidence. A similar scientific and rational inquiry is often required at another stage of the judicial process when the proper inference to be drawn from circumstantial evidence is in dispute. This is adverted to in Morgan v. Babcock and Wilcox Ltd.[28] where, in considering what conclusion should be drawn from certain circumstances, Knox C.J. and Dixon J. pointed out that the sufficiency of the circumstances "must inevitably be judged by considering whether general human experience would be contradicted, if the proved facts were unaccompanied by the fact sought to be proved"[29].

Up to this point our consideration of the applications of the general principles of the rules of evidence does not cover the precise question of admissibility raised in the present proceedings. That question is whether, on the Friday mentioned in the charge, the defendant operated the motor vehicle "for reward." There is no issue of accident or intention or mistake, so that the principles and cases discussed in Stephen's art. 12 are not of direct assistance. The question is whether, in order to show that the car was driven by the defendant "for reward" on the Friday, the prosecution was entitled to show that the defendant's operations on Friday were part of the general conduct of a motor service business, so that the inference of business reward on the Friday also might fairly be drawn. The difficulty confronting those arguing in favour of admitting this evidence is that, no actual reward on the Friday having been proved, the additional evidence falls short of directly proving an actual reward on the earlier occasions.

A consideration of certain leading cases may show that, under certain circumstances, it is permissible to prove a system, design or course of business extending beyond the particular act under investigation, which system, design or course of business gives a special significance to the particular act under investigation so as to enable a further and additional inference to be validly drawn as to whether the particular act was accompanied by another act.

In Makin's Case[30] the evidence tendered showed (1) that in addition to the body of the infant whose death alone was the subject of the relevant count in the indictment, the bodies of eleven other infants had been found in several houses occupied by the prisoners; (2) that a number of infants (other than the particular child) had been "adopted" by the Makins from the mothers; (3) that in such cases the Makins received small sums of money quite inadequate to meet the expense of upkeep for more than a short time; (4) that the infants were in good health when handed over; and (5) that, although only several months had elapsed, the infants had not since been heard of.

In the Supreme Court of New South Wales, it was argued that, in the absence of prima facie evidence of a homicide of the particular child whose death was charged against the Makins, evidence could not be admitted of the facts from which the other killings might be inferred. Windeyer J. said:—

There must certainly be some evidence connecting the prisoner with the death under investigation, and which he is accused of causing; but it is not necessary that such death should be first proved to have been caused by the prisoner. If that were so, in many cases the evidence as to the other deaths would become unnecessary, and in others antecedent proof of the killing in question impossible to give[31].

No one would suggest that a murder by poison of A in Sydney by B could be proved by showing that B, who had never been in Sydney, had murdered C and D by poison in Melbourne. But once show that B has been in attendance on A in Sydney and had the opportunity of giving A poison, evidence that C and D died in Melbourne by poison and with similar symptoms when attended by B, would at once become admissible on the trial of B for poisoning A[32].

This illustration of a poisoning case seems to show the tendency of the law to regard the poisoner as a person whose stamp or finger print, so to speak, tends to be sufficiently impressed upon all his dealings with a particular poison to justify the inference that he may be connected with a number of deaths by poisoning in all of which he is involved by circumstantial evidence. It must be observed that "to rebut accident" is often a euphemism for "to prove wilful murder by the accused." This problem arose in a very acute form in R. v. Armstrong[33]. In that case the Court of Criminal Appeal held that circumstantial evidence of an attempt by the prisoner to kill a rival solicitor by arsenic in the month of October 1921, a large quantity of arsenic having been found in the prisoner's possession when arrested in December 1921, was admissible in proof of the charge that the prisoner murdered his wife by the same poison in February 1921, seven months before the later attempt. The Court based its decision upon R. v. Geering[34]. It was said:—

The fact that he was subsequently found not merely in possession of but actually using for a similar deadly purpose the very kind of poison that caused the death of his wife was evidence from which the jury might infer that that poison was not in his possession at the earlier date for an innocent purpose, and such use of the same poison is more cogent than the mere fact of death from the same poison as in Geering's Case[35]. See Thompson v. The King[36] and the illustration there given[37].

The reasoning of this part of the judgment is that proof of the use of a specific poison for the purpose of injuring or killing A is evidence which can be used for the purpose of inferring that, from the time of its earlier possession by the prisoner, the same poison had quite probably been used for a similar felonious purpose. It is suggested that the major premise implicit in this reasoning seems to be that a person who uses a particular poison to kill or injure is likely to become so addicted to its use that it is reasonable to infer that either before or after the fact proved he used the same kind of poison for a similar purpose.

Yet the same judgment rejects the view that the evidence can be used to show the prisoner "was a man who was in the habit of committing and might be expected to commit this particular crime."

It is sometimes said that in Armstrong's Case[38] the admission of the evidence was justified upon the footing that the "possession" of arsenic by the prisoner was given a particular "significance" by the evidence as to the attempt to poison the rival solicitor. But the "possession" of arsenic can never obtain significance in itself. Arsenic in the hands of any person is a weapon that may be used for killing. In Armstrong's Case[39] the significant thing was not possession but the possessor. The tendency of the additional evidence was to show that arsenic, being in Armstrong's hands at the relevant time, was in the hands of a person who was prepared to use it when it suited him. This aspect seems to have been appreciated by Mr. Justice Avory who, in the argument on appeal, when the question of system was being discussed, asked "is there a system of poisoning everybody?"[40].

In Armstrong's Case[41], the Criminal Appeal Court was content to rely upon the decision in Geering's Case[42], and the only distinction between the two cases was that in Geering's Case there was evidence of the actual administration of food by the prisoner, whereas in Armstrong's[43] there was evidence that the prisoner had the opportunity of such administration.

But this distinction in fact was suggested by the defence as a vital one. For, it was argued, in the absence of evidence showing system or course of business why should accident or intent be discussed until a prima facie case is made? That there are some limits to the use of evidence of other poisonings which is tendered to rebut the defence of accident is suggested by a case mentioned by Hawkins J. in a discussion of the decision in Makin's Case[44]. He referred to a case of the murder by a woman of her daughter where it was attempted to prove that the prisoner also poisoned her mother shortly before. Hawkins J. said:—

I rejected the evidence, because none of the circumstances attendant upon the mother's death had the least connection with the death of the daughter. The poisons were different, the prisoner's action was different and therefore I thought the circumstances of the mother's death were not relevant evidence to prove the murder of the daughter[45].

Whilst difficulties arise in the use of evidence which, strictly speaking, is admitted solely upon the question of accident or intent, the question before us is of a somewhat different order. In Makin's Case[46] Windeyer J. was very concerned to show, not only by reference to Geering's Case[47] but by general principles of reasoning, that, when admitted, evidence of other poisonings is meant to establish not merely what he called "the abstract fact" that poisoning had taken place in the case under investigation, but also that it was a poisoning by the prisoner. He illustrated the position by reference to the case of the body of a person found cremated by an old camp fire, there being no trace whatever as to the mode of death and the accused being the last person seen at the camp fire with the missing person. Windeyer J. supposed that, on a charge of murder, the defence was that there had been a death from natural causes and a cremation by the accused because of the absence of any means of burial and to prevent the corpse being devoured by wild animals. He considered it certain that a similar disappearance of other men last seen travelling with the accused whose bodies had also been cremated would be relevant evidence on the charge of murder. The main principle of reasoning involved was that

a series of natural deaths of healthy men when camping out with a travelling companion, and their subsequent cremation, would be so extraordinary as to leave no doubt, when coupled with motive or other evidence, that A had been murdered (R. v. Makin[48]).

Windeyer J. discussed the case of R. v. Hall[49] where the Appeal Court in New Zealand thought that evidence was inadmissible of the fact that six months after the death by antimony laid to the accused's charge, the prisoner's wife had also exhibited symptoms of antimony. In each instance the prisoner had the opportunity of administering the poison. The ground of the New Zealand Court's decision was there was not sufficient evidence that the prisoner had administered antimony to Cain, the man with whose death he was charged. Before the hearing of the charge in respect of Cain, the accused had been convicted of administering poison to his wife with intent to murder. Windeyer J. rejected the theory that evidence of similar occurrences could only be admitted in relation to an issue of "accident or design." He said that

the whole object of such evidence is to establish the guilt of the prisoner, and if the nexus between the two events is such as to irresistibly lead the mind to the conclusion that the guilty agent in one case must be the guilty agent in the other, the admission of such evidence as leads to this conclusion must be right, irrespective of the existence of any abstract question of accident or design[50].

It may be observed that, in the illustration of the killings and cremations, the facts supposed by Windeyer J. would, in their totality, support the inference of a general scheme or design or course of business in which circumstances the use of such evidence might well not be restricted to anything short of inferring the particular homicide charged. In Hall's Case[51] the evidence of the other poisoning could hardly have been regarded as proving a system of poisoning, and, therefore, if the evidence was admissible at all, its use might properly be subject to restriction.

Windeyer J. referred to R. v. Gray[52] and pointed out that the evidence of the other fires was admitted not merely to negative accidental burning but to prove a deliberate burning by the prisoner for a similar motive of gain. In Gray's Case[53] Willes J., after consulting Martin B., had admitted evidence of the receipt of payment from different insurance offices. In this case moreover the series of fires tended to show a course of business which may distinguish the case from those like Armstrong[54] and Hall[55].

With Windeyer J.'s view that, in Makin's Case[56], evidence of other killings was not dependent upon prior prima facie proof of the particular homicide charged, Hawkins J. subsequently stated his full concurrence. "I dissent," he said,

from the suggestion that such evidence of prior transactions involving criminality for other offences can only be admitted in corroboration of a prima facie case which a judge would be justified in leaving to a jury if it stood alone. The admissibility of evidence in itself material and relevant to the inquiry can never be dependent upon whether it is used to corroborate evidence already given or is offered as an independent piece of evidence[57].

Here again it may be pointed out that the facts in Makin's Case[58] showed the existence of a general system. Further, it has also to be remembered, in dealing with the admissibility of evidence, that the test of admissibility of evidence is satisfied if the evidentiary fact points to the required conclusion as either the more plausible explanation out of those which are conceivable or at any rate a plausible one among those conceivable (Wigmore, 2nd ed. (1923), vol. 1, p. 243).

In Makin's Case[59], therefore, Windeyer J. was insistent upon showing (1) that the evidence of the other deaths and transactions proved was admissible even though there was no prima facie proof of the particular homicide charged against the accused; (2) that from the facts of such other deaths and transactions the jury could properly draw an inference not merely as to (a) absence of accident, and (b) presence of intent, but also (c) deliberate homicide by the accused.

In the result, the judgment of Windeyer J. was affirmed on appeal to the Privy Council. There it was held that evidence (1) that the other infants had been received on like representations to bring them up and educate them upon payment of a sum sufficient to support a child for only a short period, and (2) that the bodies of a number of infants had been found in the houses occupied by the prisoners, was all admissible. The judgment asserts the familiar principle that evidence of criminal acts of the accused outside those covered by the indictment cannot be used to show "that the accused is a person likely from his criminal conduct or character to have committed the offence for which he is being tried"[60].

But it also asserts that the general principle of admissibility is that of relevance, and that relevance may be shown if the commission of other crimes bears on the question whether the acts charged "were designed or accidental" or if a defence was thereby rebutted. As the Privy Council held that there was sufficient evidence aliunde to connect the Makins with the homicide of the particular infant, it became unnecessary for them to analyze the two propositions of Windeyer J. which I have summarized above.

In my opinion both propositions of Windeyer J. were rightly applied in Makin's Case[61] but, upon a further analysis, they turn out to be particular applications of a broader principle which may, under special conditions, be capable of adoption both in civil and criminal trials. That principle may perhaps be expressed as follows:—Where a question in issue is whether, on a particular occasion, the proved acts of a party were accompanied by the performance by such party of a further act, it is permissible to show that such party was, at or about the time in question, engaged in a special kind of business, line of conduct, or manner of living according to the exigencies of which the proved acts would ordinarily be accompanied by the performance of the further act in issue so as to constitute a typical instance of the business, line of conduct or manner of living in which the party was so engaged.

Provided that the business, line of conduct or manner of living is of such a character as to render it very highly improbable that on the occasion in question the performance by the party concerned of the further act in issue would not accompany his proved acts.

If this general principle is sound two conclusions follow:—

It will be observed that, if this manner of analysis is permissible, the facts of the present case are brought within the same principle. The receipt on Friday of some reward from the passengers is not sufficiently proved. But the course of business is proved and is revealed as that of a motor service business conducted in all respects in the same way as an ordinary business. Reward was essential to its proper conduct. Accordingly the proper inference was that the services performed on the Friday were rewarded in the ordinary course of business.

It is convenient to see whether the general principle suggested finds sufficient support in the decided cases. On the whole I think it does.

First of all, it is implicit in the reasoning of Windeyer J. His two propositions already analysed may be treated as applications of it.

Second, in R. v. Ball[64] Scrutton J. stated, in reference to Makin's Case[65], that the evidence as to the other deaths, &c., could not have been regarded as merely rebutting accident or proving intent. Scrutton J. said:

I do not think that that evidence was given to show intent, because the first thing to show was not intent, but that the prisoners had done the act at all, that they had actually killed the child; it was not till they killed the child that the question of the intent with which they did it arose, and I think that that evidence must have been given to enable the jury to draw the proper inference as to the sort of business or transaction that the prisoners were carrying on, of which the disappearance of this particular child was one incident.

The issue raised was therefore exactly that raised in Regina v. Hall[67], for there was not even evidence of the cause of death, much less that the prisoners caused it, and the argument for the defence followed exactly the line taken in that case. The evidence, however, was clearly relevant both to show whether the adoption was bona fide, and to show whether the death was accidental, and the Privy Council, on those grounds, held that it was admissible (Harvard Law Review, vol. 46, p. 974).

Mr. F. L. Stow in a valuable discussion of the principles has said:—

People may fancy they understand how such evidence can be admitted and yet not used in proof of the commission of the act charged; but in reality they understand nothing of the sort. In fact, it is now too late to deny that this kind of evidence may legitimately be used in proof of the act, and that Mr. Justice Wills is right when he recognizes, in his work on Circumstantial Evidence, that such evidence should not be excluded whenever it is relevant to make out any step in proof of the offence charged (Law Quarterly Review, vol. 38, at p. 67).

As a mere historical statement, Scrutton J.'s comment needs qualification for, at the trial of the Makins, the evidence of the other transactions was admitted by the Judge in order to show merely that the death of the particular child was to be attributed to something more than accident. But Scrutton J.'s comment on the facts is correct, and it supports the attempt to restate on broader grounds the principle upon which evidence of systematic conduct may properly be admitted.

In Ball's Case[68] the question was whether the Crown could give evidence as to prior acts evidencing sexual passion between the brother and sister charged with incest. The charge was not restricted to a specific day but to a "day between 1st and 14th July 1910 to the jurors unknown." The indictment as originally drawn confined the charge to 12th July, but, at the suggestion of Scrutton J., the date was amended. Scrutton J. admitted the evidence of acts evidencing sexual passion between the two accused although they took place at a date long prior to the time mentioned in the indictment. The decision of the Court of Criminal Appeal delivered by Darling J. was that Scrutton J. was in error, that the evidence was admitted "not to show the mens rea with which the act was committed, but to show the commission of the act itself" (R. v. Ball[69]), and that the prior authorities did not justify such admission. The House of Lords supported the decision of Scrutton J., regarding the evidence as admissible upon the issue of incest or no incest. Lord Loreburn said that "their passion for each other was as much evidence as was their presence together in bed of the fact that when there they had guilty relations with each other"[70].

It might, perhaps, have been said that in the special circumstances of the case the passion of the two accused for each other was far more cogent evidence in proof of the offence charged than their presence together in bed.

It is better to regard R. v. Ball[71] as an illustration of the general principle suggested above than as a special type of case illustrated by the practice in divorce causes. The Crown had to prove the commission of a specific act, not the intention with which the act was performed. The Crown was entitled to show that there existed a special manner of living according to which acts of the kind alleged frequently took place. That manner of life having been proved, an inference that the specific act took place at some time or other between the dates alleged might properly be drawn.

I now turn to another case in which Scrutton J. figured, the so-called "Brides in the Bath" case, R. v. G. J. Smith[72]. Smith was charged with the murder of a woman with whom he had gone through the form of marriage. She was found dead in her bath on 12th July 1912. At the trial evidence was admitted by Scrutton J. as to the deaths of two other women on dates subsequent to the death of the woman with whose death Smith was charged. One death took place in December 1913, the other in December 1914. Evidence was given that Smith had also gone through a form of marriage with these two other women and that they had both died in their bath in circumstances similar to those accompanying the death of the first woman. In his charge to the jury Scrutton J. said that there were no less than thirteen points of similarity between the three deaths[73]. Smith was the only person at hand on all three occasions.

It is to be noted that in Smith's Case[74] the particular fact of homicide was directly in issue. Evidence as to the other deaths and the circumstances surrounding them went to establish not merely intent of the accused in killing, but the killing itself. His real defence was not "By accident I caused her death" but "By accident she caused her own death." There are two important passages in the charge of Scrutton J. He told the jury to examine the question of motive and opportunity in the light of all the circumstances and added:—

It may be that even then they are not sure whether it is accident or design. And then comes in the purpose, and the only purpose for which you are allowed to consider the evidence as to the other deaths. If you find an accident which benefits a person and you find that the person has been sufficiently fortunate to have that accident happen to him a number of times, benefiting him each time, you draw a very strong, frequently an irresistible, inference that the occurrence of so many accidents benefiting him is such a coincidence that it cannot have happened unless it was design[75].

It is to be noted that even this ruling extends well beyond the tentative rule as stated in Stephen, and illustrated by the statement of the New Zealand Court in R. v. Hall[76] to the effect that

everything to be found, whether in the cases themselves or in the textwriters, establishes that evidence of other similar but unconnected acts of the prisoner is admissible only for the purpose of proving what is shortly called "guilty knowledge" ... It is always supposed that the doing of the act ... is first confessed; or that there is sufficient independent testimony on the subject to be laid before the jury[77]. (Harvard Law Review, vol. 46, p. 972.)

Further, Scrutton J. also stated to the jury in Smith's Case[78]:—

You may use the evidence as to the other deaths for this purpose—to see whether it helps you as to whether the death of Miss Mundy was accidental or designed. It is putting it in a different way, but you may use it for this purpose; if you think that the prisoner has a system of obtaining money from women by going through the form of marriage with them and then getting the money either by robbery or murder, you may use the evidence of the other deaths for that purpose[79].

In this direction, we have a further illustration of the principle inherent in those cases when other transactions may be used to prove the occurrence of some disputed part of the particular transaction in issue, as distinct from the mere intent or motive which characterized such transaction. In the case of Smith[80], not only would justice have miscarried, but every principle of rational scientific inquiry would have been denied, if the investigating tribunal had been denied the assistance of the additional evidence. For the result of extending the scope of the inquiry was to make it morally certain that Smith was engaged in a definite plan or system of fraud and murder in which the killing of the particular woman mentioned in the indictment was merely an incident. The legal position is realistically described by Stone in the following words:—

Evidence must be excluded which indicates that the prisoner is more likely than most men to have committed it, but evidence must be admitted which tends to show that no man but the prisoner, who is known to have done these things before, could have committed it. There is a point in the ascending scale of probability when it is so near to certainty, that it is absurd to shy at the admission of the prejudicial evidence (Harvard Law Review, vol. 46, pp. 983, 984).

Before the Court of Criminal Appeal in Smith's Case[81], counsel for the prosecution contended, in favour of the opinion of Windeyer J. in Makin's Case[82], that it was not necessary that a prima facie case as regards the particular act should be established before the evidence of the other transactions was admissible, and that the latter evidence was admissible both (1) to show design, and (2) to show that the person did the act charged. Inasmuch as the Court held that there was prima facie evidence to support a finding of homicide in respect of the woman mentioned in the indictment, it decided not to deal expressly with the contention of counsel. But the contention should have been considered and accepted, because Scrutton J.'s charge made it perfectly clear to the jury that they might properly infer the particular homicide charged from the proof of Smith's general modus operandi. The mere existence of prima facie evidence of such homicide might not have satisfied the jury of the fact of homicide without the assistance of the evidence of the other deaths, &c. Therefore it was necessary to determine whether Scrutton J.'s direction was correct, and it is to be assumed that its correctness was accepted by the Court. And if it is correct, so that the additional evidence tends to prove the fact of killing, why should the Court exclude this method of proving such fact and insist also upon prima facie proof of it by some other method?

The examination of the British cases justifies the acceptance of Wigmore's distinction between evidence to prove design and evidence to prove intent. Makin's Case[83] is to be regarded as establishing "a broad rule of admissibility where there is relevance, except where the only relevance is via disposition" (Julius Stone, Harvard Law Review, vol. 46, at p. 975). Relevance being the general principle, its application in the case of proof of intent is necessarily different from its application in proof of a design producing an act, the commission of which act is disputed.

According to Wigmore,

the peculiarity of Design is that the act is not assumed to be proved and the design is used evidentially to show its probable commission. It is obvious that something more definite and positive is here involved than in the case of Intent. In proving Intent, the act is conceded or assumed; what is sought is the state of mind that accompanied it. In proving Design, the act is still undetermined, and the proof is of a working plan, operating towards the future with such force as to render probable both the act and the accompanying state of mind. The Intent is a mere appendage of the act; the Design is a force producing the act as a result (Law of Evidence, 2nd. ed. (1923), vol. 1, p. 609).

From this distinction, it would appear to follow that evidence to show intent need not necessarily reveal any feature of common purpose or general scheme as a necessary requirement of admissibility. Thus Wigmore says:—

It is not here necessary to look for a general scheme or to discover a united system in all the acts; the attempt is merely to discover the intent accompanying the act in question; and the prior doing of other similar acts, whether clearly a part of a scheme or not, is useful as reducing the possibility that the act in question was done with innocent intent. The argument is based purely on the doctrine of chances, and it is the mere repetition of instances and not their system or scheme, that satisfies our logical demand. Yet, in order to satisfy this demand, it is at least necessary that prior acts should be similar. Since it is the improbability of a like result being repeated by mere chance that carries probative weight, the essence of this probative effect is the similarity of the instance (ibid., p. 615).

Accordingly, upon the issue of intent, a precise resemblance between the various instances is unnecessary. Upon such issue, therefore, the evidence in R. v. Armstrong[84] and R. v. Hall[85] was receivable.

Moreover, the strength of the inference does not rest exclusively on a given person's connection with the other transactions, for it is possible "to infer deliberate human intent without forming any conclusion as to the personality of the doer" (Wigmore, vol. 1, p. 616). Where, however, "the very doing of the act is still to be proved, one of the evidential facts receivable is the person's Design or Plan to do it" (ibid., p. 617). The object here, says Wigmore,

is not merely to negative an innocent intent at the time of the act charged but to prove a pre-existing design, system, plan or scheme, directed forwards to the doing of that act. In the former case (of Intent) the attempt is merely to negative the innocent state of mind at the time of the act charged; in the present case the effort is to establish a definite prior design or system, which included the doing of the act charged as a part of its consummation. In the former case, the result is to give a complexion to a conceded act, and ends with that; in the present case, the result is to show (by probability) a precedent design which in its turn is to evidence (by probability) the doing of the act designed. The added element, then, must be not merely a similarity in the results, but such a concurrence of common features that the various acts are naturally to be explained as caused by a general plan of which they are the individual manifestations (ibid., p. 617).

Wigmore states that the difference between requiring a sufficient similarity of other events in order to negative innocent intent and requiring sufficient features in common so as to prove design is in one sense a difference of degree, but the distinction is a real one. In the former case the mere prior occurrence of an act with the same doer and the same sort of act but not necessarily the same mode of acting nor the same sufferer may suffice (ibid., p. 619).

But where the very act has to be proved, and it can be proved only by inference from a plan or system, so much higher a degree of similarity is required as to constitute a "substantially new and distinct test." It follows that what he calls "anonymous acts" are available as evidence of intent but not to prove design, for the whole purpose of the evidence in the latter case "is to fix a design upon the defendant, as making it likely that he carried it out, and thus that it was he who did the act charged" (ibid., p. 619).

Wigmore also appears to consider that, as the object is to argue to the act charged from a design to do it, the acts proving design must be prior in time to the act charged (ibid., pp. 619, 620).

But it is difficult to appreciate why this must necessarily be so. In Smith's Case[86], for instance, the other transactions admitted in evidence were all subsequent to the particular homicide, and in Makin's Case[87] some of the other transactions were also subsequent to the particular homicide. It is true that, from the point of view of logical inference, it is easier to draw a conclusion as to the commission of a particular act if the general design is proved to have come into existence before the time when such commission is alleged. Consequently, it is often safer to rely upon prior acts than upon subsequent acts in order to prove the existence of a design at the time of the act charged against the defendant. But under certain circumstances there is no logical difficulty in inferring that the act charged was the first act in the general design. And in Smith's Case[88] the repetition of the act charged showed clearly that all three homicides were properly to be included in the general plan or design so as to make it legitimate to infer the existence of the plan and to reason from its existence to all its particular manifestations, including its first manifestation. A business must have some commencing point, and the performance of the earliest acts as well as of the latest acts may be inferred from the existence of the course of business, although care must be observed before drawing the inference.

I am therefore of opinion that the above analysis of the leading cases supports the validity of the principle of evidence suggested earlier in this opinion. Wigmore's general discussion strongly supports it, and the propositions advanced by Mr. F. L. Stow (Law Quarterly Review, vol. 38, at pp. 71, 72) are also in accord with it.

Mr. Stow's proposition is that, subject to the exclusion of evidence which merely shows a propensity, disposition or character which renders it likely that a person will act in a certain way, evidence is admissible to show, inter alia,

that any occurrence is one of a series of two or more similar occurrences, and that the facts of the occurrences comprised in the series indicate some system, plan, or design on the part of any person ... with a view to establishing—(i) that the occurrence or act was the act of, or was caused or committed by, the person concerned in such system, plan or design, or so implicated as aforesaid; (ii) that, being his act or caused or committed by him, it was done, caused or committed by him intentionally or with any particular intent (Law Quarterly Review, vol. 38, pp. 71, 72).

Mr. Stow's full proposition is only quoted so far as it bears upon the question of admitting evidence in proof of design. It is criticized by Mr. Julius Stone "for unnecessarily referring to intention" (Harvard Law Review, vol. 46, p. 977). But this criticism rather ignores the fact that the evidence of the other transactions, proving system or design, once it is admitted, tends to prove not only the commission of the particular act charged but the deliberation and intent accompanying it. The other ground of criticism has no relation to the proposition I have quoted.

In the present case, it follows that the decision to admit the evidence as to the defendant's earlier possession and manner of control of the car was correct. This evidence established that on each of three successive days a well-defined course of business activity was being pursued by the defendant. The inference that this business in all its manifestations was being pursued for reward is practically irresistible. The famous dictum of the American financier, "we are not in business for our health," is so close an approximation to universal truth that it would be absurd not to act upon it in this case, no evidence whatever having been called by the defendant.

The appeal should be allowed and the conviction restored.

McTiernan J.

The Transport Regulation Act 1933 of Victoria provides that a commercial passenger vehicle shall not operate on any public highway unless such vehicle is licensed in accordance with the Act (sec. 7). The licence is granted in respect of the vehicle. "Commercial passenger vehicle" is defined by sec. 5 to mean any motor car of the class described in the section, used or intended to be used for carrying passengers for reward at separate and distinct fares for each passenger, and "operate," to carry passengers on these conditions. Sec. 45 makes it an offence to drive an unlicensed commercial passenger vehicle operating on a public highway. The defendant who is the respondent was prosecuted under this section on the information of the appellant who is an inspector under the Transport Regulation Act which will be referred to as the Act. The substance of the information was that the defendant drove an unlicensed motor car which carried passengers for reward at separate and distinct fares for each passenger on a public highway. In a prosecution under sec. 45 of the Act, the onus of proving that the passengers in an unlicensed vehicle, alleged to have been driven in contravention of the section, were carried for reward rests upon the informant: if that onus is discharged, the onus of proving that the passengers were not carried at separate and distinct fares for each passenger lies upon the defendant (sec. 47).

The informant's evidence was that on Friday, 1st November the defendant drove a Hupmobile car, carrying three passengers, from Melbourne to Ballarat. The evidence further related to the circumstances in which the car left Melbourne. This car and an Auburn car had been standing near the corner of Spencer and Collins Streets, Melbourne, and the defendant was standing near them. The Auburn car was the first to leave, it being driven in the direction of Ballarat by Mrs. Osborne. The Hupmobile car left shortly afterwards and upon its arrival at Ballarat two of the passengers, a lady and a child, were set down in a side street, although according to the informant "it was raining hard at the time." It was deposed that the defendant said that the other passenger was a friend of his and would be driven home. It was proved that the car was not licensed, but no direct evidence was given of any payment or agreement to pay for the drive from Melbourne to Ballarat. The informant also gave evidence, which was unsuccessfully objected to by the defendant, of the operations of the Hupmobile car on the Wednesday and Thursday immediately preceding Friday, 1st November, the date of the alleged offence. The magistrate convicted the defendant. The conviction was set aside by the Supreme Court of Victoria on the ground that the evidence dealing with the events of Wednesday and Thursday was inadmissible and that the evidence relating to the movements of the car on Friday was insufficient to prove the guilt of the defendant.

The admissibility of the evidence to which the defendant objected depends upon its legal relevance to the fact in issue, namely, whether passengers were carried for reward in the Hupmobile car on Friday, 1st November. Without recapitulating the whole of the evidence, the following facts were proved. On Wednesday the defendant was seen in Ballarat outside a shop which bore the sign "Osborne's Services," and the Hupmobile car and other cars were near the shop. The defendant spoke to a number of people and handled luggage, some of which he put into the Hupmobile car. Two men and three women were seen to get into the car, and the defendant drove to Melbourne where the occupants of the car were set down in different places. In the afternoon the car with the defendant alongside was near the corner of Spencer and Collins Streets. Here a woman and two men got into the car, and the defendant drove off, proceeding to a suburb of Melbourne where another person joined the car. On Thursday morning the defendant had the car outside the same shop in Ballarat. A person came out of the shop and got into the car, and the defendant drove away. In the afternoon the car and the defendant were again near the corner of Spencer and Collins Streets where other people got into the car, and it was driven away. The defendant did not leave by the car. The Auburn was also standing there and on the next morning, Friday, the defendant and other persons were seen in this car which was then outside the shop in Ballarat.

The principles underlying the inadmissibility of evidence of a similar act on another occasion to prove the fact in issue are stated by Willes J. in Hollingham v. Head[89] in these terms:—"It is not easy in all cases to draw the line, and to define with accuracy where probability ceases and speculation begins: but we are bound to lay down the rule to the best of our ability. No doubt, the rule as to confining the evidence to that which is relevant and pertinent to the issue, is one of great importance, not only as regards the particular case, but also with reference to saving the time of the Court, and preventing the minds of the jury from being drawn away from the real point they have to decide." See also at pp. 392 and 393. But, as Willes J. explained in the course of that judgment, the rule against the competence of such evidence is not an absolute one. The numerous cases in which evidence of similar acts has been admitted provide instances of the legal relevancy of such acts. In the present case, there is that degree of probative force in the evidence of similar acts which qualifies it to be a basis for judicial inference concerning the existence of the fact in issue. The similar acts proved were links in the chain of proof that the passengers in the Hupmobile car on Friday, 1st November, were carried for reward. The evidence relating to these acts on Wednesday and Thursday was admissible to prove that the Hupmobile car was systematically driven as a service for the conveyance of passengers between Melbourne and Ballarat. The existence of this business and the fact that the car was operating on Friday in the ordinary way of business—facts which the evidence proved—provided a sound basis for the inference that the passengers were carried for reward on that occasion, and, in my opinion, established that fact beyond reasonable doubt. Any other hypothesis would not, in my opinion, be reasonably consistent with the evidence.

In my opinion the appeal should be allowed.

Appeal allowed. Order of the Supreme Court discharged. Order nisi discharged with costs.

Solicitor for the appellant, F. G. Menzies, Crown Solicitor for Victoria.

Solicitor for the respondent, D. Lazarus.