Peacock v R [1911] HCA 66; (1911) 13 CLR 619 (8 December 1911)

HIGH COURT OF AUSTRALIA

Samuel Peacock Appellant; and The King Respondent.

H C of A

On appeal from the Supreme Court of Victoria.

8 December 1911

Griffith C.J., Barton, and O'Connor JJ.

Duffy K.C., Bryant and Cussen, for the appellant.

Woinarski K.C., and Dethridge, for the Crown.

Duffy K.C., in reply,

December 8

Griffith C.J.

This is an appeal from a decision of the Full Bench of the Supreme Court of Victoria, dismissing an appeal by way of special case on a conviction for murder. The appellant is a medical practitioner, over seventy years of age, who conducted in Melbourne a private hospital which was used only by women. The deceased was a single woman named Mary Davies, about twenty seven years old, who was pregnant. The case made by the Crown was that in pursuance of an arrangement previously made with the appellant, by or with the assistance or connivance of a young man named Poke, who was responsible for her condition, she entered the appellant's hospital on 9th or 10th August for the purpose of having an operation performed with a view to bringing about abortion; that the appellant performed an operation which was followed by a miscarriage; that septicæmia ensued, and death; and that the appellant made away with her body, of which no trace has been discovered. The Crown contend that under these circumstances the jury could infer that the appellant had performed the illegal operation which resulted in her death. Evidence given by Poke, who was assumed to be an accomplice, was relied upon as an admission by the appellant that he had performed such an operation. The case was tried before the learned Chief Justice, who did not give to the jury the usual direction or caution that they should not convict upon the uncorroborated testimony of an accomplice. If the case is established without the evidence of the accomplice, of course such a direction is not necessary. The prisoner made a statement in the dock, as he was allowed to do by the law of Victoria. The jury found him guilty. The learned Chief Justice then stated a case, raising three points. First, was the evidence sufficient to warrant the jury in finding the fact of death. (I have said that the body was not discovered). Secondly, was the evidence sufficient to connect the prisoner with the death; and, thirdly, was the omission to give the jury the usual direction as to accepting the evidence of the accomplice fatal to the conviction. Subsequently, a further case was stated, raising a question of the propriety of a direction which the learned Chief Justice gave to the jury as to the effect to be given to the prisoner's statement in the dock, to which I will afterwards refer. I will deal with these points seriatim.

The first is whether there was sufficient evidence of the death of Mary Davies. The evidence in nearly all branches of the case was purely circumstantial. The rules as to circumstantial evidence are nowhere better stated than in a book, somewhat old it is true, but by an undoubted authority (Starkie on Evidence, 3rd ed., published in 1842). I quote from page 574. Speaking of circumstantial evidence, he says:—"Fourthly, it is essential that the circumstances should, to a moral certainty, actually exclude every hypothesis but the one proposed to be proved: hence results the rule in criminal cases that the coincidence of circumstances tending to indicate guilt, however strong and numerous they may be, avails nothing unless the corpus delicti, the fact that the crime has been actually perpetrated, be first established. So long as the least doubt exists as to the act there can be no certainty as to the criminal agent: hence upon a charge of homicide it is an established rule that the accused shall not be convicted unless the death be first distinctly proved, either by direct evidence of the fact or by inspection of the body; a rule warranted by melancholy experience of the conviction and execution of supposed offenders charged with the murder of persons who survived their alleged murderers, as in the case of the uncle already alluded to, cited by Sir Edward Coke and Lord Hale. So Lord Hale recommends that no prisoner shall be convicted of larceny in stealing the goods of a person unknown unless the fact of the robbery be previously proved. The same principle requires that upon a charge of homicide, even when the body has been found, and although indications of a violent death be manifest, that it shall still be satisfactorily proved that death was neither occasioned by natural causes, by accident, nor by the act of the deceased himself. In considering the probability of the latter supposition it is to be recollected that it is by no means improbable that a person bent on self-destruction would use precautions to protect his memory from ignominy, and his property from the forfeiture consequent on a verdict of felo de se. The force of circumstantial evidence being exclusive in its nature, and the mere coincidence of the hypotheses with the circumstance being in the abstract insufficient, unless they exclude every other supposition, it is essential to inquire with the most scrupulous attention what other hypotheses there may be which may agree wholly or partially with the facts in evidence. Those which agree even partially with the circumstances are not unworthy of examination, because they lead to a more minute examination of those facts with which at first they might appear to be inconsistent; and it is possible that upon a more minute investigation of these facts their authenticity may be rendered final, or may be even altogether disproved. In criminal cases the statement made by the accused is in this point of view of the most essential importance. Such is the complexity of human affairs, so infinite the combinations of circumstances, that a true hypothesis which is capable of explaining and reconciling all the apparently conflicting circumstances may escape the acutest penetration; but the prisoner so far as he alone is concerned can always afford a clue to them; and although he be unable to support his statement by evidence, his account of the transaction is for this purpose always most material and import tant. The effect may be on the one hand to suggest a view of the case which consists with the innocence of the accused, and which might otherwise have escaped observation. On the other hand its effect may be to narrow the question to the consideration whether that statement be or be not excluded and falsified by the evidence."

That passage is applicable to circumstantial evidence in general both as to proof of death, and of the fact that the prisoner caused the death. The question may be raised whether what I have read is to be regarded as a rule of law, or as a rule to be applied in the administration of justice, as indicating the duty of the Judge. It may mean that if the Judge considers that the evidence is insufficient he should direct an acquittal. But it is not material to consider that question. With regard to the well-known caution given by Sir Matthew Hale, I think it is now settled law that the fact of death may be proved by circumstantial evidence as well as any other part of the case. But another difficulty arises on that point, in this way: The rules of evidence are the same in criminal as in civil law, and the rules of logic and common sense as to what inference may be drawn from facts are the same whether the case is civil or criminal. In civil cases where the evidence is nicely balanced, the recognized practice is to leave it to the jury to say which hypothesis they accept, where there are two equally, or nearly equally, probable hypotheses. But that is certainly not the practice in criminal cases. It is the practice of Judges, whether they are bound to give such a direction or not, to tell the jury that, if there is any reasonable hypothesis consistent with the innocence of the prisoner, it is their duty to acquit. That I mention now, because it comes in appropriately after reading the passage I have read from Starkie, although it has more application to the second branch of the case than to that regarding proof of death.

As to the proof of death, the evidence may be briefly stated: Mary Davies went to the prisoner's house on either 9th or 10th August as a Mrs. Nelson, a married woman. She was there visited several times by Poke, who represented himself to be her husband. He went there on Thursday, 10th, Sunday, 13th, and Tuesday, 15th. Upon all these occasions she was in bed, but appeared in good health. The fair inference to be drawn from this was that something had happened to her which required her to remain in bed. During the night of Tuesday the 15th she had a fall. She fell down three or four stairs and cut her face and hand. On Thursday the 17th Poke again visited the house, and the appellant told him that Mrs. Nelson was worse, that she had absorption of poison, and had contracted puerperal fever. Poke saw her, and said she seemed drowsy and pale. On the Friday evening Poke again went to the house and saw the appellant, and he gives an account of a conversation which he says took place between himself and the appellant on that night.

"Poke said—I was present on Friday night about 8 p.m. in his waiting room. Miss Elliott admitted me." (She was the appellant's housekeeper). "I said—How are things? Prisoner said—Mrs. Nelson is very ill. In case anything happens what are you going to do? I said—I don't know."

Up to that point the Judge's notes are in the form of direct narration, but there they suddenly change to indirect narration. "Prisoner said the best thing I could do is to say she came here to try and bring about a miscarriage, and I had to attend to her at once." Then the notes return to the form of direct narration. Under these circumstances it is impossible to say what really was said—whether prisoner said—"The best thing I can do," or "the best thing you can do" is to say, &c. Perhaps we may take it that what was said is—"The best thing we can do is to say," &c. Say what? According to the statement on the Judge's notes, to say, "We agreed to bring about a miscarriage. I had to do it, and she died." That is to say that persons in peril of being found out in committing murder are to say that the best thing to be said is—"We did it." That is preposterous, and no jury would be justified in accepting it. Another suggestion was—"We must tell some lie or other. Let us tell that lie, that we killed her." That is altogether too absurd. The third suggestion is that the Judge misunderstood what Poke said, and that what was really said was—"The best thing we can do is to say she came here having tried to bring about a miscarriage." Mr. Duffy suggested that we should look at Poke's evidence in the depositions in the Police Court, and that we should find that this was the actual fact. From any point of view it is an extremely unsatisfactory statement. When that statement was made, whatever it was, Poke said—"We are not married." Up to that time the prisoner undoubtedly believed, according to the evidence, that they were married. He went on to say that the prisoner said—"If you are not married you had better leave it to me. Do not come to this house again. It might draw suspicion. Meet me to-morrow night." I shall have occasion to refer to that conversation again. On Sunday, the 20th, Poke again visited the house, and was told that Mary Davies was slightly better. On the 21st he went again, and prisoner said—"Things are as bad as they can possibly be." On Tuesday, the 22nd, he went again, and the prisoner said—"She is dead." Then, Poke says, they discussed the question of the disposal of the body. The doctor described the circumstances of her death and asked where her people lived, and stated that if her people did not inquire for her for a fortnight he could have her secretly buried. He said he would put her clothes in a bag and take them to his farm at Carrum, as he could easily burn them there, and Poke would have to keep his mouth shut. He told Poke to meet him again on the following Monday night. Poke went to meet the prisoner, but did not then see him, but he met him on Tuesday, the 29th, at his house in the evening, and he says that he had this conversation with him:—"I said, What have you done with the body? and the prisoner said—It is buried. I said—Where? Prisoner said—It is better known to myself. I said—What of the clothes? Prisoner said—They are burnt. I only got rid of the last yesterday. I said—I would like some of the jewellery as a keepsake; and prisoner said—The jewellery has gone with them. It is the first time I have had anything like this for years. I said—I think the case is all carelessness. Prisoner said—Oh, no! it was not. On the same evening the police came to the prisoner's house. He refused to answer any questions and denied everything. Poke was then brought in and confronted with him, and the conversation I have just read—which is said to have taken place between the prisoner and Poke—was stated to him. The prisoner merely refused to say anything. Search was then made by the police, and three pieces of jewellery identified as the property of Mary Davies were found in one of his pockets. There was also found in the scullery a neck wrap or fur boa which had been cut into several pieces, and which was identified as hers. It was proved also by other evidence that on Sunday, the 27th, the prisoner was at his farm at Carrum and lit a large fire in the scrub. Another witness saw him carrying a bag there on that day. On 2nd September the ashes of the fire were examined, and in them were found several metal bodies of buttons, other buttons, such as are commonly worn on women's clothing, including suspender buttons, safety pins, hooks and eyes and burnt clothes, boot eyelet holes and a boot heel."

None of these were identified as the property of Mary Davies, but they were common parts of a woman's wearing apparel and the apparel she wore when she went to prisoner's hospital had on it articles of that sort. There was also found on the 5th September in a small hut on the farm—the only building there—a fragment of an old tooth-plate which corresponded in some particulars with a tooth-plate which Mary Davies wore. There was also found outside the hut lying on the ground a fragment of a false tooth and a plate of the same material as the bit found inside the house. A dentist who attended her in 1905 said it was not the plate she had been wearing then. However, the woman has never since been heard of. On these facts I think it was open to the jury to find that she was dead, and that the appellant disposed of her body. And as to Poke's evidence in that respect—as far as it bears on this part of the case—there was corroboration, I think, in the fact that the statement that the prisoner was said to have made was repeated by the police in his presence and that all he said was "I will say nothing." Moreover, the circumstances deposed to by Poke as having been communicated to him by the prisoner were entirely consistent with the prisoner's doings on the previous day and with the discovery at Carrum and the discovery of the jewellery and neck-wrap in the house.

I think, therefore, that there was sufficient evidence of the death of Mary Davies.

The next question is an entirely different one, namely, what was the cause of death. Was the prisoner responsible for her death? The case was treated throughout, as I understand it, practically on the footing that, if the prisoner disposed of the body, the jury could find him guilty. But that argument requires to be carefully considered. I again read from the passage I read before from Starkie, p. 576:—"The same principle requires that upon a charge of homicide even when the body has been found and although indications of a violent death be manifest, that it shall still be fully and satisfactorily proved that death was neither occasioned by natural causes, by accident nor by the act of the deceased himself."

The learned Chief Justice in effect told the jury that if they found that the appellant disposed of the body because he knew that, if it was not disposed of, it might be discovered on inquiry that he had done something that would not stand the light of day, they might draw the inference that he had porformed an illegal operation, and that she died in consequence of it; and this, irrespective of the alleged confession to Poke. On that direction other questions arise, with which I will deal later on. I have referred to the rules generally followed in cases of homicide. In the case of R. v. Hodge[1] Alderson B. directed the jury "That the case was made up of circumstances entirely; and that, before they could find the prisoner guilty they must be satisfied, not only that those circumstances were consistent with his having committed the act, but they must also be satisfied that the facts were such as to be inconsistent with any other rational conclusion than that the prisoner was the guilty person." He then pointed out to them the proneness of the human mind to look for and often slightly distort the facts, in order to establish such a proposition—forgetting that a single circumstance which is inconsistent with such a conclusion is of more importance than all the rest, inasmuch as it destroys the hypothesis of guilt. It was a very strong case, but on that summing up the jury acquitted the prisoner.

The rule is sometimes stated that the circumstances must be such as to be inconsistent with any reasonable hypothesis other than the guilt of the accused. I inquire, then, Was this the only reasonable hypothesis or the only rational conclusion? and ask, first, What was the physical cause of death? before coming to the question of who caused it. In ordinary cases of death by violence, it may be caused by suicide; or by the act of another person or of several persons, any one of whom may have done it. In the present case we know what the cause of death was, if we know anything about the case at all. It was the result of a miscarriage. The next question is: How may that have been brought about? We all know that it may be the result of accident or of purpose. If of accident, there is no guilt anywhere. If of purpose, it may be the act of the woman herself, or of some other person. Here it becomes very important to consider the version that the prisoner gave in the statement he made from the dock. He said:—"About the second week in August, a tall dark woman giving the name of Mrs. Nelson called to see me professionally at my house. She complained of feeling very unwell, suffering from abdominal pains and a slight discharge of blood, and desired me to make an examination of her. I did so and found that what she said was correct. The mouth of the womb was soft and open and she was threatened with a miscarriage. I informed her of her condition and explained that there was a risk of dangerous consequences if she was not careful and advised her to take a rest and a course of treatment in my hospital for some little time. She consented to this. We discussed the terms and arranged that she should pay me £5 5s. per week in advance. From her symptoms I thought that it might become necessary in case the miscarriage resulted to perform an operation known as curetting on her, that is, to clean out the womb, and I told her to inform her husband of that possibility and to get his written consent to my performing it should the necessity arise. She left and a day or two later, came back bringing with her the document purporting to be the consent of her husband. I glanced at it and threw it in my desk in the surgery."

That is his statement. It is admitted—and it is to a certain extent supported by other evidence—that she went to prisoner's hospital as Mrs. Nelson, a married woman, and that her ostensible husband visited her. She had left her home, and had been for some days in a country town called Traralgon. What she did there is not accounted for. However, she returned on 7th October by a train which came from Traralgon, and it was said by another witness named Lack that on arrival in Melbourne she looked "drawn in the face and a bit red about the eyes." As for the statement of the prisoner that he asked for authority from the husband, a document was found in his desk purporting to be signed by C. Nelson—which was proved to be in the hand writing of the woman herself, consenting to her being treated as she desired. The prisoner had evidently asked for that document and it could only have been asked for for the purpose of being used as evidence in the event of some difficulty afterwards arising. All that is quite consistent with innocence. So far there are two hypotheses, neither improbable, that miscarriage was caused either of purpose or by accident. I have already said that she had an accident two days before she became suddenly ill. So far, the evidence is consistent with the prisoner's statement that she was threatened with a miscarriage, that she came to him for treatment, that the miscarriage occurred, and that she died. The learned Chief Justice told the jury, in effect, that if they believed that the prisoner made away with the body they could disregard his statement altogether. Now what is the real weight to be given to the fact of making away with the body? On this point I will read a passage from Best on Evidence (3rd ed.), p. 518:—"Undoubtedly suppression or fabrication of evidence by a party accused of a crime, is always a circumstance, frequently a most powerful one, to prove his guilt. Too many instances have occurred of innocent persons alarmed at a body of evidence against them which although false or inconclusive they feel themselves unable to refute, having recourse to the suppression or destruction of criminative and even to the fabrication of exculpatory testimony." These instances are given of the danger of drawing such an inference, of which one is the not uncommon case of a false alibi. It is contended, however, and the learned Chief Justice seemed to think, that everything may be presumed against a man who makes away with a dead body. In my opinion the fair inference to be drawn is that the object in making away with a body is to prevent the discovery of some fact that would be apparent, if it were not made away with, and would be likely to imperil the life or liberty of the man who does it. In this relation it is important to consider the facts of the particular case. What fact could have been ascertained by discovery of the body? The only fact that could have been ascertained was that the miscarriage had been brought about by some violence. In the nature of things nothing more could have been discovered. In some cases the mere making away might be sufficient, as in the Queensland cases which were referred to where the murderer burned the body. But in other cases it may not be sufficient. In cases of infanticide an inference of murder was often sought to be drawn from the concealment of the body. In England the legislature once passed an Act providing that, if a mother concealed the body, she could be convicted of murder, unless she could prove by some independent witness that the child had lived. That is not the law now. Juries were reluctant to draw such an inference; then the law was altered, and concealment of birth was made a substantive offence. It seems to me that suppression of evidence ought not to put a prisoner in any worse position than if the evidence had not been suppressed. He may not benefit by the concealment. But why should he be in a worse position than if the true facts had been discovered? What then could have been discovered in this case? Merely that the miscarriage was not accidental. That is all. It leaves the problem as to whether it was brought about before she came to the hospital or not as it was before. Otherwise you are in reality convicting a man not of killing a woman, but of concealing a body. The argument is, he concealed the body, therefore he is guilty of murder. It is very important again to remember that the idea of concealing the body did not occur to anyone until the prisoner discovered that Mary Davies was not married. When it was disclosed that she was not married an entirely different complexion might well seem to him to be put on the matter. He found himself involved in a case which would probably be presented as one of abortion of a young woman brought to his house for that purpose. His case from that point of view might well seem to him to be practically hopeless. Would anyone believe the story that she came as a married woman? That idea might, indeed, impress a man who was not of unusual fortitude of mind. In this connection, I will read another passage from Best, p. 543:—"This rule rests upon principles which have their foundation in the deepest equity and soundest policy. When the crime is separable from the person of the criminal many sources of error are introduced which do not exist in the opposite case. 1. A given event the origin of which is unascertained may be the result of almost innumerable causes having their source either in accident or the agency of different persons; 2. The danger of the inference of his guilt from the event being aggravated by the imprudence of the accused or even his criminal agency in other matters; 3. In witnesses and tribunals the love of the marvellous and the desire to detect great crimes committed secretly."

Applying to these facts the principle stated in R. v. Hodge[2], that the facts must be such as are inconsistent with any other rational conclusion than the guilt of the accused, it seems to me that they are not inconsistent with another rational conclusion.

If the question arose upon the same facts—say—in an action on a life policy, and the jury found that the act of violence causing death had occurred before the girl entered the hospital, the verdict could not, in my judgment, be set aside as one which reasonable men could not have found. Nor, perhaps, could the contrary verdict have been impeached. If that is so, there was a reasonable hypothesis. But the rule in capital cases has never been stated that the jury may make a mere balance of probabilities and accept that theory which they think the more probable. Unless this rule is now to be laid down for the first time I think that, apart from the alleged confession made to Poke, the evidence was insufficient; and so, indeed, it was assumed in the Supreme Court.

Then does that evidence carry the case any further? I have already commented upon it. It is almost impossible to believe that it is a correct statement of what actually happened. If it is, it is the evidence of an accomplice. There is one other statement of Poke's which I omitted to read. Poke says that the prisoner said—"You will have to keep your mouth closed or it will be bad for both of us." This was when the prisoner had been told that Mary Davies was not a married woman. This, as it seems to me, only means that Poke was involved in the same risk as himself. At best, all this evidence relied upon as a confession is ambiguous, but I will assume that it is sufficient if believed. Is there any evidence in corroboration of it? If in a narrative of facts part of a story is corroborated, that may be sufficient to give credence to the whole. But the corroboration must be of the evidence as to the fact to be proved, not of the general credibility of the witness. For instance, in the case of sexual crimes, suppose that a woman alleges that she has been ravished, and that there is independent evidence that such an assault has been committed, that is no corroboration of her evidence that a particular person assaulted her. It used to be thought to be sufficient corroboration to show that a witness told the truth as to some part of his story. But that is not the rule as now accepted.

I am unable to find any corroboration of Poke's story that any such conversation as he alleges took place. The only corroboration that can be suggested is that he told the truth as to another part of the case, i.e., as the prisoner's admission of having secretly disposed of the body under circumstances which of themselves did not point to anyone in particular as having committed the offence. If the evidence points to one of two persons, A. and B., having committed an offence, and C., an accomplice, says that A. confessed that he committed it, the fact that the other circumstances point to either A. or B. having committed the offence is no corroboration of A.'s having confessed. It leaves the matter where it was. There is, therefore, no corroboration that I can find of that part of Poke's story. It is not sufficient to show that another part of his story was corroborated as to another part of the case.

If this is the correct view, the usual caution ought to have been given by the learned Chief Justice. And so he himself thought. For he says—"I did not say that, although I intended to, in order to make assurance doubly sure." I refrain from comment. The question then arises: What are the consequences of the omission to give that caution? It will be convenient to defer consideration of that point until after dealing with the fourth question submitted. By the Victorian Evidence Act, sec. 52, an accused person may make a statement of facts without oath in lieu of or in addition to any evidence on his behalf. A later Act provides that the person accused shall be entitled to give evidence, but that no comment shall be made upon his failing to do so. The direction which the learned Chief Justice gave to the jury in this case I take from his own lips in delivering judgment on the appeal, although it is not exactly in the same form in which it appears in the special case.

Now as to his statement: The prisoner is at liberty to make a statement which is not on oath to a jury if he thinks fit and the jury may treat it as evidence and may act upon it if it is not in conflict with any other evidence in the case, but where it is in conflict with any other evidence the jury should disregard it and act upon the sworn evidence. He says "I never did anything to this girl except what was lawful and proper. I did not illegally act upon her at all." There is no definite and distinct evidence that he did. It is suggested to you that he did or he would never have observed all the secrecy he has done from beginning to end and finally disposed of the body by burying it. It is said that kind of action is wholly inconsistent with what he did and is consistent only with the fact that he acted illegally upon her and brought about her death. That seems to me to be the whole position.

That is, in effect, that if they believed that the prisoner made away with the body they could disregard his statement altogether, and the learned Chief Justice still thinks that that was good law. He says so, and thinks that he was supported by the case of Mack v. Murray[3], cited in an edition of the Statutes, but unfortunately wrongly quoted, in which the learned Judges properly held that when the evidence for the prosecution taken in connection with the statement of the accused was consistent with his innocence, justices were justified in dismissing the charge The learned Chief Justice, however, instructed the jury that when the prisoner's statement is inconsistent with the sworn testimony it must be disregarded altogether. However, he, on being asked to do so, recalled the jury, and gave them a further direction, saying:—"Learned counsel is of opinion that you may misunderstand what I put to you as to the difference between the sworn statement and the prisoner's statement. I have told you that whenever the prisoner's statement does not come in conflict with the sworn evidence you are at liberty to believe it, but where it comes in conflict with the sworn evidence you are not to accept it. Counsel thinks you might understand that, that you would not even accept the statement although you did not believe the sworn evidence which it contradicted. It is only when it is in conflict with sworn evidence which you can accept as true, that you must not accept it."

The proper direction to be given, it seems to me, is this: that the jury should take the prisoner's statement as primâ facie a possible version of the facts and consider it with the sworn evidence, giving it such weight as it appears to be entitled to in comparison with the facts clearly established by evidence. Instead of that the jury were advised that if they connected the accused with the concealment of the body they might infer that the appellant killed the deceased woman, and that if they drew that inference they might disregard his statement altogether. That was manifestly a wrong direction, and the conviction cannot stand. Indeed, all the other Judges of the Supreme Court thought such a direction would be wrong. àBeckett J. could not believe that such a direction had been given. Hood J. thought that it was immaterial, apparently, as I understand him, on the ground that, even if it had not been given, the jury would have found the prisoner guilty.

As the conviction cannot stand, the next question is whether there should be a new trial. According to the law of Victoria, the Court may grant a new trial. This power has been exercised by the Victorian Courts even in a capital case. Assuming the power to exist, I think that it should be used with great caution. I do not think it was intended that a new trial should be granted as of course in every case where there has been an irregularity. I think the proper rule is that where there was evidence to go to the jury and the error was of such a nature that, if it had not been committed, the verdict would probably have been the same, a new trial may be granted. On the other hand, if, on the whole case, it is reasonably probable that, but for the error complained of, the verdict would or might have been different, a new trial should not be granted.

The failure of the learned Chief Justice to advise or direct the jury that they should not convict upon the uncorroborated evidence of an accomplice, raises a very interesting question. The old rule in England was that although the Judge ought to give the direction, it was a matter of practice and not of law, and in the case of R. v. Stubbs[4], the Court held that a conviction could not be quashed on the ground that the presiding Judge had directed the jury that corroboration of an accomplice as to each of several prisoners was not necessary. That was the only point reserved. The point of the Judge having failed to caution was not taken; Parke B., however, intimated that it was a matter for the consideration of the Secretary of State. In more modern times all the books lay it down that it is considered in England to be a rule of practice now so generally followed as almost to have the force of law. And it is recognised in England now as a rule failure to comply with which must be followed by the quashing of the conviction. In the case of R. v. Tate[5], Lord Alverstone C.J. quoted a passage from Russell on Crimes, 6th ed., vol. iii., p. 646, where it is stated: "It may be observed that the practice in question has obtained so much sanction from legal authority, that it deserves all the reverence of law and a deviation from it in any particular case would be justly considered of questionable propriety." Lord Alverstone had previously quoted from a decision of Cave J. in In re Meunier[6], in which that learned Judge said: "No doubt, it is the practice to warn the jury that they ought not to convict unless they think that the evidence of the accomplice is corroborated; but I know of no power to withdraw the case from the jury for want of corroborative evidence, and I know of no power to set aside a verdict of guilty on that ground." Lord Alverstone added[7]: "I think he ought to have added, Assuming that the jury was cautioned in accordance with the ordinary practice." In R. v. Warren[8], Channell J. stated the law to the same effect.

That must be taken to be now the recognized rule in England, and if the Judge neglects to give the caution the conviction will be quashed. Whether that is to be regarded as a new rule established under the Criminal Appeal Act, or whether it is to be regarded merely as a more modern statement of the old law introduced into Australia, is an extremely interesting question which it may some day be necessary to decide formally. In the present case, where the conviction must be set aside, it is only material in considering whether a new trial should be granted. If the direction had been given, I think it must be assumed that the jury would have done what they ought to have done. I think that the Judge should have told the jury that Poke's evidence as to the alleged confession was not corroborated, and advised them not to act upon it. He should further have advised them that the rest of the evidence was consistent with the prisoner's statement that the deceased was already in danger of miscarriage when she entered the hospital, and that under these circumstances it would be dangerous for them to convict, and we must assume the jury would have acted upon that advice.

There is another matter to which also I think I should refer, in considering the manner in which this discretion should be exercised. Objection was made to the wrongful admission of evidence which had been tendered by the Crown to prove that the prisoner regularly carried on the practice of abortion in his hospital. That evidence was given and admitted, after objection, on a Friday morning. Late on Saturday the Judge said he would strike it out of his notes, but this was not said until it had been before the jury for two whole days and a night. I express no opinion on the abstract question as to whether evidence of this kind may be admissible in some cases. There is no doubt that the evidence in this case was not admissible. It appeared to be a mere statement of opinion by a witness as to a matter on which the police had instructed her. After the jury had finally retired, the Chief Justice recalled them and formally withdrew the evidence. What he said is thus stated in the record:—"That amounts to no evidence on which you are to act. You are to put quite out of your minds the fact that she said or might have said anything which would lead to the supposition that something different might have been said. You are to forget it altogether and not use it in your deliberations at all against the prisoner."

Judges are liable to make mistakes like other people, and it may be technically sufficient, as a matter of law, formally to withdraw evidence from a jury in that way. But that must depend, I think, on circumstances. I am not dealing with a dry point of law, but with the manner in which our discretion should be exercised. But, having regard to the nature of this evidence, that it had been tendered to prove that the prisoner was a person engaged habitually in the practice of abortion, and to the fact that it put an entirely different complexion on the whole case, I think that something more was needed as a matter of common fair play than a mere formal and frigid withdrawal of the evidence in such a manner. What, in my opinion, the learned Judge should have done, was to tell the jury to regard the case as if no such evidence had been given, and to treat the prisoner as a person against whom nothing was known except this charge. When I say "should have done," I mean that that is what I think a Judge who had present to his mind the traditions of English law would have done. That is a matter to be considered in determining whether there should be a new trial.

Apart from that, I think that, if he had given proper directions in regard to the corroboration of Poke's evidence as to the alleged confession, the verdict would or might have been different, unless indeed the jury were unable to shake off the influence of the inadmissible evidence or the horror of the secret disposal of the body, "the love of the marvellous and the desire to detect great crimes, committed secretly," and other influences which might unconsciously have affected them. We must, however, assume that they would have been able to do so, and would have found a verdict of not guilty. Under these circumstances I think that a new trial should not be granted.

Barton J.

I will first consider the law in relation to the direction given to the jury by the learned Chief Justice of Victoria upon the statement of the accused at the trial. That statement was made in exercise of the right given or formulated with respect to persons accused of indictable offences and defended by counsel, by sec. 38 of the Crimes Act 1891 (No. 1231). The Evidence Act 1890, sec. 52, had made it lawful for any person charged with any indictable offence on summary conviction, whether defended or not to "make a statement of facts (without oath) in lieu of or in addition to any evidence on his behalf." In former times the practice had varied somewhat, and statutory intervention became expedient to secure the statement as a right, to define its nature and limits, and to secure some consequent right to the Crown. Sec. 34 of the Crimes Act 1891 which allows the accused to be called as a witness if he consents, concludes with the proviso that (if he does not testify) "no comment shall be made upon the fact that any such person has not given evidence in his own behalf." That prohibition must be carefully borne in mind in charging juries where the prisoner has not given evidence, but has made a statement. The material provisions of the Crimes Act 1890 (No. 1079), relating to the reservation of questions of law and consequent statement of a case for the determination of the Judges of the Supreme Court, are secs. 481 and 482. I do not propose to read them again. It will have been observed that, besides giving the Full Court the ordinary powers of affirming, amending, reversing or avoiding the judgment at the trial, and a power to order an entry on the record that "the party convicted ought not to have been convicted," they include a power "to direct a venire de novo or new trial to be had, or to make such other order as justice may require." The exercise of this power is, of course, in the discretion of the Court, but the discretion must be exercised judicially.

Returning to sec. 38 of the Act No. 1231, I proceed to consider the nature of the statement which is authorized by that Act as a matter of right. First, it is to be "a statement of facts." Next, it is to be "in lieu of or in addition to any evidence on his behalf." Thirdly, the section expressly gives the counsel for the Crown the right to reply whenever any such statement is made, a right which apparently did not previously exist in Victoria except when some evidence was given by or on behalf of the accused. "A statement of facts" as the phrase is used in the section, means, I think, a statement consisting of assertion of the matters which the accused wishes to be accepted as facts. They may or may not be facts in the signification of truths, but they are to be facts as distinguished from arguments; that is, the statement is allowed in addition to any arguments which the accused or his advocate may put forward. Now, this statement is to be in lieu of or in addition to any "evidence" on his behalf—which term is obviously used in the sense of evidence which, so far as it is oral, is under the usual sanction of an oath, or of an affirmation where the law permits that sanction. If it is in lieu of such evidence, it is put in the place of it for some purpose or other. I cannot imagine any purpose for which it can take the place of sworn evidence other than the ordinary purpose with which evidence is tendered—that is, the obtaining of credence for the thing alleged. Every such statement, whether in lieu of or in addition to any sworn evidence, is of necessity an endeavour to gain credence for the version put forward by the accused in opposition to the case for the prosecution; whether in contradiction of statements of the opposing witnesses, or by way of explanation, so as to detract from the force which their statements would or might otherwise have. Like evidence in its usual meaning, it is included among "all the legal means, exclusive of mere argument, which tend to prove or disprove any matter of fact the truth of which is submitted to judicial investigation"—Taylor on Evidence, p. 1. In allowing a statement in addition to any ordinary evidence on behalf of the accused, the legislature must have considered it capable of adding to the weight of such evidence, if the statement is believed. Otherwise the grant would have been illusory. It is intended, then, whether given in place or in support of sworn evidence for the accused, to benefit him to the extent to which it gains credence from the jury as against the Crown evidence. If it fails to gain credence, the accused has taken the risk of that.

In either aspect, the statement which takes the place of or is added to sworn testimony by or for the prisoner, must be conceded to have some evidentiary status, though he who makes it is not called a witness, nor does his version go by the name of evidence. It must be either evidentiary or argumentative, and the term "statement of facts" is no name for argument. Although the Court or the Crown may not comment on the fact that the accused has not given sworn evidence on his own behalf, it is a fact which the jury cannot dismiss from their minds. And they cannot help observing that the test of cross-examination has not been applied to the statement—a test which so enhances the value of sworn testimony that has lived through it. These considerations, however, only relate to the weight of the assertions of the accused. They do not touch his right to have all the weight that they are worth assigned to them. Not only does the construction I have suggested seem to be the intention of the legislature, but that body would appear to have acted in accordance with such an intention in giving a new right of reply where a statement has been made by the accused without the addition of any sworn evidence. This privilege would seem to have been given to enable the Crown to point out any respect in which the statement is defective, or unworthy of credence in opposition to the facts deposed to on behalf of the prosecution. If that is so, surely the whole of it must be left to the jury, coupled with the comments of course, but side by side with all, the other evidentiary material in the case.

It follows then from the the terms of the section that the accused is entitled to have his statement considered by the jury, not only where it gives an explanation consistent with the assertions of fact sworn to for the prosecution, but where it flatly contradicts such assertions by its own. It is the right of the accused in such circumstances that the jury be asked to decide how much credence they will give it where the two sides are in absolute conflict. They may conclude that it is not worthy of any weight at all. On the other hand, from observation of the utterances and demeanour of the opposing witnesses, they may conclude, unsworn though it is, that it outweighs their testimony, or at any rate that it raises a reasonable doubt—and in either of the two latter views they may and ought to acquit. Such being the statutory right of the accused, the question is whether the direction of the learned Chief Justice assured it to him.

First, the accused was entitled to have his statement considered side by side with those of the Crown witnesses, which could not be effectively done unless the jury first possessed themselves of the Crown case as narrated by these witnesses, then compared it with the statement, and finally decided whether they believed the Crown's version or that of the accused, or whether they felt a reasonable doubt between the two. Now, his Honor did not adopt any such process. The further case stated informs us, to put it shortly, that his Honor told the jury in the first place to direct their attention to the statement, because, if they thought that it was true, they ought without more to acquit the accused. How could they decide whether it was true or false without comparing it with the evidence for the Crown? Then his Honor told them that, if they did not believe the statement (after this inverted process which was likely, if the jury succeeded in following it, to subvert the purpose of the Act), they should proceed to consider the several sets of facts which the Crown propounded as indicating the guilt of the accused, and also the evidence of Poke as to his alleged conversations with the prisoner. Thus they were, in effect, told that they need not consider the Crown case unless they first decided not to believe the statement. The duty of putting the Crown evidence and the statement side by side, and arriving at their conclusion after close comparison, was not suggested in that connection. The process actually suggested might, however, have been, by inference, unduly favourable to the accused, but for what followed it. His Honor, after examining all the evidence—meaning, we may take it, the evidence exclusive of the prisoner's statement—reverted to the statement, which he pointed out was not on oath, but which the prisoner was by law at liberty to make. They might treat it, he said, as evidence, and might act upon it if it were not in conflict with any other evidence which was sworn, but where it was in conflict with such other evidence they ought to disregard it and act upon the sworn evidence. The jurymen having retired, Mr. Bryant, for the accused, asked for a further direction to them, fearing that, if they did not fully rely on some of the sworn evidence for the Crown, they might still suppose they were bound to act on it merely because it was sworn, rather than on the prisoner's statement. Mr. Bryant's fears were fully justified by the terms of the direction, which indeed authorized the jury to omit that very duty on the performance of which, as I have pointed out, the Statute entitles the accused to insist—the duty, namely, of comparing and contrasting the statement with the sworn evidence for the prosecution (of course, with the right of preferring the statement), not only where it gives an explanation consistent with the case for the Crown, but where it is in absolute contradiction of the statements of the Crown witnesses. However His Honor very rightly recalled the jury, and I take from the report of his further direction the following statement, which we may use, and which was used in the Supreme Court. "Learned counsel is of opinion that you may misunderstand what I put to you as the difference between the sworn statements and the prisoner's statement. I have told you that wherever the prisoner's statement does not come in conflict with the sworn evidence you are at liberty to believe it, but where it comes in conflict with the sworn evidence you are not to accept it. Counsel thinks that you might misunderstand that; that you would not even accept the statement although you did not believe the sworn evidence which it contradicted. It is only when it is in conflict with sworn evidence which you can accept as true that you must not accept it." In the further case stated the last sentence is contained more fully. It reads thus—"I then told them that the sworn evidence which they should prefer to the prisoner's statement was so much of that evidence as they could fully rely on, and that they might certainly prefer the prisoner's statement to any evidence in the case of which they had doubt or on which they could not fully rely, notwithstanding that the latter was sworn and that the prisoner's statement was not."

It is true that this addendum somewhat modifies the original direction. But the fact remains that the jury had been told that where the statement was in conflict with any other evidence which was sworn, they ought to disregard it and act upon the sworn evidence. I regard that as a serious misdirection. It was in no sense withdrawn, but on the recall of the jury it was repeated with an explanation. With regret, I cannot see how one can say that this explanation removed the impression which the minds of the jury must have received from the error. But even if that could be said, the explanatory direction is itself objected to, and with reason. The jury were left to infer that, before they could prefer the statement to any portion of the sworn evidence, they must first have a doubt of that portion of the sworn evidence, or feel that they could not fully rely on it, while they could first come to the conclusion that they could fully rely on this or that portion of the sworn evidence, and having done so, prefer it to the statement of the accused, or in other words, discard the statement. The fault of the direction seems to me to lie in this, that in each of the several parts of it to which I have referred, the most important requirement of sec. 38 is left out of account. That requirement—that right of the accused—is that, whether the statement explains or contradicts, the jury are to be so guided that it may be placed in juxtaposition, contrasted and weighed with the portions of the Crown evidence to which it relates. This is merely to say that the case for the Crown and the case for the defence, of which the statement is a part, or it may be the whole, are, after both are clearly understood, to be simultaneously weighed, for so to weigh them is the right method to find out which of them makes the other kick the beam. Of course, the accused has the additional advantage that a reasonable doubt must be resolved in his favour.

For these reasons, I am of the opinion that the jury were misdirected with regard to the statement. Further, I am of opinion that the misdirection is of such a nature that it cannot be said that it was not likely to lead to a substantial miscarriage of justice. Therefore the conviction must be quashed. Whether there should be a new trial or not depends on the opinion of the Court on other points.

On the question whether there was evidence sufficient to enable the jury to conclude, if they chose, that Mary Margaret Davies was dead, and that the accused had caused her death, these being separable facts, it is to be observed that the evidence to establish these factors of the charge is largely circumstantial. I agree with a great deal that has been said as to the necessity of great caution being exercised before convicting on that class of evidence. Speaking of one division of it, with which we are nearly concerned in this case, Best, in his valuable treatise on Evidence, which is a work of great authority, says at p. 518, sec. 417, of the 3rd edition—"Undoubtedly the suppression or fabrication of evidence by a person accused of a crime is always a circumstance, frequently a most powerful one, to prove his guilt; but many instances have occurred of innocent persons, alarmed at a body of evidence against them, which, although false or inconclusive, they feel themselves unable to refute, having recourse to the suppression or destruction of criminative, and even to the fabrication of exculpatory testimony." For this, amongst other reasons, the author says of the maxim "omnia præsumuntur contra spoliatorem, Whatever weight may be legitimately attached to this presumption in civil cases, great care must be taken in criminal ones, where life or liberty are at stake, not to attribute to spoliation (such for instance as the suppression or destruction of evidence) or similar acts any force to which they are not entitled. I read this passage as impressing the necessity as to this class of evidence—and indeed the necessity exists as to all presumptive evidence—of care in its admission and consideration, and the avoidance by the jury of any inference from it beyond that which is reasonably warranted by the suppression or destruction itself, if traced to the accused. But I do not read it as suggesting that in any case of presumptive evidence, where a reasonable inference, being open, is distinctly adverse to the accused, the danger of attributing too much force to the inference will warrant a course by which the jury would be prevented from weighing the fact at all. When any case is left to them they will, of course, be told that if they believe the facts it is for them to say how much weight is to be given to them. If in a case, wholly or partly one of presumptive evidence, an adverse inference obviously beyond reason is suggested, they will be told not to accept it. If an inference or hypothesis is open which is consistent with the facts, but not adverse at all, or less adverse than that offered by the prosecution, they will be told to consider it, and reminded that they may adopt it, if they think it within reason. But the circumstances being before the jury, they must all be weighed in conjunction, and the question which of two or more varying inferences or hypotheses that are open is the most reasonable one, is for the jury, not for the Court."

If, however, a circumstance or a body of circumstances will bear no more than what is called a "light" or "rash" presumption, it does not tend to the proof of the matter in issue, and is entitled to no weight at all. Any such circumstance will not have been admitted in evidence unless, though apparently unimportant by itself, it is one of several facts which when taken in conjunction are material as being capable of a particular adverse inference. Whether the fact, or that body of facts which is called the "case" is capable of bearing a particular inference, is for the Court, and unless it is so capable, the Court's duty is to withhold it from the jury, as a single fact or as a case. But when the case is undoubtedly capable of the inference of guilt, albeit some other inference or theory be possible, it is for the jury, properly directed, and for them alone, to say not merely whether it carries a strong probability of guilt, but whether the inference exists actually and clearly, and so completely overcomes all other inferences or hypotheses, as to leave no reasonable doubt of guilt in their minds. The presumption in favour of innocence would in that event have been overcome by a stronger presumption raised against it by the evidence. To quote Best once more—"No person is to be required to explain or contradict until enough has been proved to warrant a reasonable and just conclusion against him in the absence of explanation or contradiction." If after such explanation or contradiction the case as a whole convinces the jury that the conclusion is not only reasonable but just, it is their duty to act upon that conviction. Much that I have said on this point may appear so elementary as not to be worth stating. But in a case demanding very close application, such as the present, it is so necessary that one should keep these considerations constantly in mind that one is justified in setting them forth.

The Chief Justice has dealt so fully with the evidence that I do not propose to state it at length. Undoubtedly, on the evidence as it stands, the young woman passed as Mrs. Nelson, and when Poke visited her in the house of the accused, which has been called his hospital, he asked for Mrs. Nelson, and gave his own name as Nelson. There is nothing to cast doubt on the prisoner's statement that he took them, as probably others did, for man and wife. There is evidence that when Miss Davies became an inmate of the house referred to, she was about five months advanced in pregnancy. The accused says that on examination he found that she was threatened with a miscarriage. She was treated by him up to the time of her disappearance, whether that was caused by bringing on abortion and so unhappily causing her death, or by death from innocent causes, or by her recovery and departure. Poke visited the house from time to time up to the 22nd of August, when the Crown alleges that she died, and he was generally admitted to see her until shortly before that date. The evidencc called for the Crown to prove the death and that the accused caused it, probably does not establish a strong case, if indeed it establishes any, without the testimony of Poke, and on the case put forward by the Crown he must be considered an accomplice. The results of that position, so far as the direction to the jury is concerned, I will deal with presently. For the purpose of considering the present question I will treat it as if the jury were entitled to act upon the evidence, as indeed I think they were. Poke, then, has several conversations with the accused, and by that means on Thursday, the 17th August, he learns a fact which is proved also from other sources, that on the 15th she had had a fall, and he sees her and finds that she has cut or scratched her nose and her hand. She was then drowsy and pale. There is no evidence, nor is there any statement by the accused, that the fall had anything to do with her subsequent illness, puerperal fever. Poke never saw her afterwards up to the 16th October, when he gave evidence at the trial. [His Honor then quoted from Poke's evidence of his conversations with the accused up to the 29th of August.] This evidence was offered in proof of death and in proof that the prisoner secretly disposed of the body and of her belongings. [His Honor then quoted from the evidence of the visit of the accused to Carrum, of his having been seen tending a fire there, and of the subsequent discovery in the ashes of partially burnt articles corresponding with articles that Mary Davies had worn, or had with her, when she left her home.] These circumstances, together with the statements of the accused to Poke, were of course evidence, if believed, that the girl was dead, and that he had burned her belongings and secretly buried the body, thereby carrying out intentions which he had previously expressed to Poke, and which he afterwards admitted to the same person that he had fulfilled.

Did the evidence then warrant the jury in inferring, as they obviously did, that the accused caused the death of Mary Davies? It is clear to my mind that it was quite open to them, if they thought it right, to draw that inference. As we all know, the ordinary course, where death occurs under circumstances of innocence, is for the practitioner to report it and to give a certificate of its cause. It is common ground that the accused administered some treatment. If there was no illegal operation, the appearance of the body would not indicate anything tending to criminate. The jury were, of course, not bound to infer guilt from the disappearance of the body; nor, on the other hand, were they bound, as seems to be thought, to draw the suggested inference that these facts and admitted intentions and doings were accounted for by the discovery by the accused that the patient was not a married woman, and his fear that the death of an unmarried woman in his house would place him, albeit an innocent man, in a very dangerous position. It might be that the accused feared a stronger force of public opinion against him if it became known that the body, which must be discovered unless he disposed of it, was that of an unmarried, and not as he had supposed when he undertook to treat her, a married woman. When women, married or single, offer themselves to be treated, not for the acceleration but for the prevention of abortion, there is always in the nature of the case some danger that under perfectly innocent circumstances the miscarriage may ensue and that puerperal fever and death may result. On the other hand, if the intention be to bring about abortion, a felonious act, the physical danger to the patient is the same whether she be spinster or wife. But whether the patient were married or single, and whether the treatment were innocent or criminal, the danger to the accused was that of a death in his hospital. It was entirely for the jury to consider whether the fear that public opinion would be more angry in the case of a spinster than in that of a wife, was sufficient to account in case of death for a resolve to destroy the traces of her having died there, if the intention of her treatment were innocent. It might be, and no doubt it was, urged on them that this is an instance, like those adduced by the author already quoted, "of innocent persons, alarmed at a body of evidence against them, which, though false or inconclusive, they feel themselves unable to refute, having recourse to the suppression or destruction of criminative ... testimony." If they found themselves unable to accept this hypothesis, and came to the conclusion that the inference suggested by the Crown was the only reasonable one, the conviction, whether they discharged their function wisely or not, cannot be disturbed on the ground that there is not sufficient evidence of the cause of death.

I pass to the question whether the conviction of the accused ought to stand in view of the fact that the learned Chief Justice of Victoria did not warn the jury that they ought not to convict on the uncorroborated evidence of an accomplice. In the case of R. v. Wilson[9], decided as lately as the 7th February this year, Lord Alverstone C.J., delivering the judgment of the Court of Criminal Appeal, said—"It must not be supposed that corroboration is required amounting to independent evidence implicating the accused ... In addition to Attwood's Case21 Leach C.C., 464., I would refer to the words of Maule J., in R. v. Mullins3Cox C.C., 526, at p. 531.—Confirmation does not mean that there should be independent evidence of what the accomplice relates, or his testimony would be unnecessary; and to the words of Cave J., in In re Meunier4(1894) 2 Q.B., 415, at p. 418.—I know of no power to withdraw the case from the jury for want of corroborative evidence, and I know of no power to set aside a verdict of guilty on that ground." Previously, in R. v. Tate[13], His Lordship, after expressing assent to the proposition that "there is no definite rule of law that a prisoner cannot be convicted on the uncorroborated evidence of an accomplice," had quoted the same passage, and remarked—"I think he ought to have added assuming that the jury was cautioned in accordance with the ordinary practice." In R. v. Tate[14] the Judge had not directed the jury, "in accordance with the settled practice," and he had led them to suppose that if they believed the accomplice they might properly convict, although his evidence was without corroboration. Under these circumstances the Court was of opinion that there had been a miscarriage of justice, and on that ground the conviction was quashed. This was apparently an exercise of the power conferred in the previous year by the Criminal Appeal Act, sec. 4 (1). The Court added[15]—"We should not, however, have taken this view, notwithstanding the Judge's departure from the practice, if we thought there was in fact substantial corroboration upon the evidence. But in our opinion there was no such corroboration."

On a careful consideration of these two cases, and of all the others cited to us, I think the following propositions are warranted by the authorities—(1) That where the evidence of the accomplice is not substantially corroborated, the duty of the Judge to warn the jury against acting upon it has not yet become a positive rule of law, although it is a matter of settled practice; (2) that in England, if there is an absence of substantial corroboration, and the Judge has failed to warn the jury according to practice, the Court will treat the conviction as a "miscarriage of justice" within the meaning of the Criminal Appeal Act, sec. 4 (1), and set it aside; (3) that in England or in Victoria, if substantial corroboration exists, the Court will not interfere with the conviction, though the Judge has departed from the ordinary practice by omitting to warn the jury; (4) that the corroboration will be deemed sufficient when it is substantial and is upon a material part of the case, and it need not amount to independent evidence implicating the accused.

It is not, I fear, settled that a conviction on the uncorroborated evidence of an accomplice, and in the absence of a caution from the Judge, should be quashed except under some statutory power such as exists in England. But it may well be considered whether such a conviction may not in Victoria, in a proper case, be made the subject of an entry on the record that "the party convicted ought not to have been convicted"—see the Crimes Act 1890, sec. 482.

I think the present case comes within the third and fourth of these propositions. There existed, in my view, substantial evidence in corroboration of Poke's testimony as to the cause of death. He had deposed to the conversation with the accused already quoted, and if this testimony were believed, the accused had admitted facts from which his guilt might be inferred. Two portions of these conversations were most material in relation to the present point. On Tuesday, the 22nd August, the accused, according to Poke, after telling him that the girl was dead, said that he could have her body secretly buried, and that he would take her clothes in a bag to his farm at Carrum and burn them. On Tuesday, the 29th, as Poke swears, the accused told him that he had buried the body—where, he did not say—that he had burned the clothes, and that the jewellery had "gone with them." Now, independent evidence was given that on Sunday the 27th the accused was seen near the Carrum railway station going towards his farm, and carrying a gladstone bag such as he often carried on his visits thither; that he was at the farm that day; that he told a witness there that he was going to light a fire, and the witness afterwards saw a fire alight, and the accused standing by it; that the fire nearly burned down, and the accused "brightened it up again;" that the fire was about three feet across; that afterwards the witness pointed out to the detectives the site of the fire, and saw them rake over the ashes. This is where most of the remnants of burned clothing, buttons and other articles were found by the detectives on the 2nd September.

This independent evidence went to confirm the accomplice in a most material part of his story. Without evidence of the matters thus related, the jury might well have declined to believe Poke as to the admissions of the 22nd August or those of the 29th. The production of evidence as to acts of the accused intervening between his statement of intention to destroy and his announcement that he had done so, unquestionably tended to give credit to Poke's testimony on this part of the case, and if the jury on hearing it ceased to entertain as to this part the doubt that naturally attaches to the testimony of an accomplice, it was open to them to presume that he had told the truth as to the rest. If they believed the substance of his story, they believed facts from which the guilt of the accused might be inferred, if I am right in thinking, as I do, that the woman's reception at the house and treatment as a patient, his admission of her death in the house while still a patient, and his admission of his conduct in secretly burying the body and destroying as he thought he had destroyed, her clothes and other belongings, are facts which warrant such a conclusion.

As substantial corroboration existed, the Court will not disturb the conviction on the ground that the learned Chief Justice did not caution the jury against convicting on the uncorroborated evidence of an accomplice. But the corroborative evidence was sufficient for its purpose, it was left to the jury as evidence which they might believe or reject, and there can be no doubt that Poke's evidence, confirmed by it, warranted the jury in the conclusion to which they came, as to the wisdom of which I have no opinion to express. In these circumstances, I do not see any reason for disturbing the conviction on the ground of the want of a caution. The caution would have been nugatory in view of the existence of corroboration.

There is a second appeal, based on the refusal by the Supreme Court of an order nisi calling on the Chief Justice and the Attorney-General of Victoria to show cause why certain questions of law should not be reserved for the determination of the Supreme Court. These questions related to the reception of certain evidence given by one Elizabeth Coleman at the trial and the effect of the striking out of some of it. The first piece of evidence was as to the washing of some linen which had been stained with blood. The witness entirely failed to identify this linen as coming from the room occupied by Mary Davies or with any alleged unlawful conduct of the accused. It could not possibly prejudice him in the mind of any jury. It should have been excluded as being immaterial, but not having been excluded it cannot be said to have led or contributed to any substantial or any miscarriage of justice. The second piece of evidence related to a discovery in the surgery of the accused. The evidence was elicited by the Crown doubtless with the belief that it could prove that the thing discovered was a fœtus, and that similar matter had been found in the possession of the accused on previous occasions. In the course of the witness's examination it came out that nothing of the kind could be proved, and that she could say no more than that the substance discovered was "supposed" upon the suggestion of one of the detective officers, to be a fœtus. Subsequently, the learned Chief Justice intimated in Court that he would strike this evidence out, and later, in his charge to the jury, he warned them against acting on it. He told them to put it right out of their minds, to forget it altogether. His Honor's warning to the jury was evidently an emphatic one. A reference to the charge will show that he could not have put the matter more strongly. But it is urged that as the evidence was once admitted the subsequent striking out and warning could not entirely remove its effect, and therefore this is a sufficient ground for setting aside the conviction. I do not think it a sufficient ground. At worst, the evidence, stopping short where it did, could only be regarded as a failure by the Crown to prove something that it regarded as important. The failure was quite evident, and occurring in the way it did brought into relief the inability of the Crown to prove that the house of the accused was used as a place of abortion. The evidence was originally admitted under the miapprehension that it was a step in the proof of a system. When the mistake was discovered, all that was possible was done to prevent it from influencing the jury adversely to the accused. It is impossible to make the administration of justice proof against occasional accidents such as the original reception of the evidence, and when they occur the only course possible is to strike out the matter complained of and to warn the jury strongly to leave it entirely out of consideration. This is the practice adopted, so far as my knowledge and experience extend, on the criminal as well as the civil side, and is the only possible means of rectifying the mishap. To hold that on all such occasions the whole proceedings are rendered abortive would be to place a fatal obstacle in the path of the administration of justice. I do not think that the incident, dealt with as it was by his Honor, could possibly influence the jury against the accused, and I do not think that it is a ground for quashing the conviction. The second appeal therefore must be dismissed.

As the first appeal is, I understand, allowed by the whole Court, it only remains to consider whether the conviction should be quashed simpliciter or whether a new trial should be ordered under the power already referred to. The majority are of opinion that such an order should be made. The only ground on which the appeal has succeeded is that which questions the direction to the jury as to the statement of the accused. As the objections relating to the sufficiency of the evidence and to the failure to caution the jury against acting on the evidence of an accomplice are not sustained, this seems to us to be a case in which the interests of justice will be best consulted by sending the case to another jury.

O'Connor J.

The learned Judge who tried the prisoner stated the questions raised for the consideration of the Supreme Court of Victoria in the original special case in the following words:—

The additional ground relating to the prisoner's statement I shall deal with later.

The first two grounds are in substance one. Stated in a few words they amount to the objection that there was not before the jury sufficient evidence to justify a verdict of guilty. That, as it comes before this Court, is a matter of law and merely of law. Having regard to the order which I think the Court ought to make in this case I shall refrain from expressing my view as to the weight or value of the evidence to which I shall be obliged to refer any more than is necessary in dealing with the sufficiency of the evidence as a matter of law. In a case of this kind, which may come before another jury, it is most desirable that no comment should be made except such as may be necessary for the elucidation of the questions of law. I agree with my learned brother Barton that the statement alleged to have been made by the prisoner to Poke on the 19th August, however it should be interpreted, did not amount to a confession.

Taking all the facts together the case must be dealt with on the basis that there was no confession of guilt, nor any direct evidence of guilt, and that the case for the Crown was made out entirely by circumstantial evidence. A large portion of the evidence consisted of admissions by the prisoner to Poke establishing material facts. Poke is on the evidence undoubtedly an accomplice, and it will be important later on to consider whether his testimony is in fact corroborated, what direction to the jury the learned Judge should have given in regard to the necessity for his evidence being corroborated, and the consequences of failure to give that direction. For the present I shall take the circumstances proved by Poke as standing upon the same footing as any other circumstances in the case. The prisoner's statement, to which I shall refer later in more detail, is also relied upon by the Crown. Taken generally, it is in direct denial of the charge, his allegation being that the girl left his hospital on 25th August completely cured, that she took the dress basket containing clothing with her, and left him as a gift, because she was not able to pay him otherwise, the articles of jewellery afterwards found in his possession.

Such being the general nature of the evidence, I shall now address myself to the question whether the circumstances relied upon by the Crown were sufficient in law to entitle the jury to draw the inference that the prisoner was guilty. Some observations have been made by my colleagues as to the general nature of circumstantial evidence and the care that should be exercised by Judges and juries in acting upon it. In what my learned brother the Chief Justice has said in that connection I entirely concur. But the expression of my view on the subject would be incomplete if I did not add something to what he has said.

The duty of a jury in regard to circumstantial evidence is often in practice stated briefly, and, I think, accurately, in these words:—"The circumstances must be such that the jury may reasonably draw from them an inference of the prisoner's guilt, and can reasonably draw no other inference." It is, I think, necessary for the purposes of this case to add that an inference to be reasonable must rest upon something more than mere conjecture. The bare possibility of innocence should not prevent a jury from finding the prisoner guilty, if the inference of guilt is the only inference open to reasonable men upon a consideration of all the facts in evidence. There are some observations in Starkie on Evidence, 3rd ed., on this aspect that are worthy of attention. At page 577 the learned author says:—

What circumstances will amount to proof can never be matter of general definition; the legal test is the sufficiency of the evidence to satisfy the understanding and conscience of the jury. On the one hand, absolute, metaphysical and demonstrative certainty, is not essential to proof by circumstances. It is sufficient if they produce moral certainty to the exclusion of every reasonable doubt; even direct and positive testimony does not afford grounds of belief of a higher and superior nature. To acquit upon light, trivial and fanciful suppositions and remote conjectures, is a virtual violation of the juror's oath, and an offence of great magnitude against the interests of society, directly tending to the disregard of the obligation of a judicial oath, the hindrance and disparagement of justice, and the encouragment of malefactors. On the other hand, a juror ought not to condemn unless the evidence excludes from his mind all reasonable doubt as to the guilt of the accused, and, as has been well observed, unless he be so convinced by the evidence that he would venture to act upon that conviction in matters of the highest concern and importance to his own interest; and in no case, as it seems, ought the force of circumstantial evidence, sufficient to warrant conviction, to be inferior to that which is derived from the testimony of a single witness, the lowest degree of direct evidence.

In drawing an inference of guilt, or in declining to draw it, the jury must act upon the facts established in evidence, and if the only inference that can reasonably be drawn from those facts is that of the prisoner's guilt, it is their duty to draw it. They cannot evade the discharge of that duty because of the existence of some fanciful supposition or possibility not reasonably to be inferred from the facts proved. I agree with my brother the Chief Justice that it is for the Judge to determine, in regard to circumstantial evidence, as in regard to any other kind of evidence, what amounts to that degree of proof which will in law justify the jury in drawing an inference of guilt, and, if he is of opinion that the facts are insufficient in law to justify a jury in drawing that inference, he is bound to withdraw the case from them. Otherwise the jury might go through the form of bringing in a verdict of guilty which the Court on appeal would immediately set aside as being founded on facts insufficient in law. If it were not for Poke's evidence I think the learned Judge at the trial would have been bound on that ground to have withdrawn the case from the jury. But with Poke's evidence, if the jury chose to believe it, there was in my opinion abundant evidence to justify the verdict. It is not necessary for me, after the full manner in which the evidence has been dealt with by my colleagues, to refer to it in any detail. I shall therefore restrict my observations to those important portions which in my opinion justified the learned trial Judge in leaving the case to the jury.

The prisoner was a medical man 72 years of age, with 40 years experience in his profession. He kept a hospital for women only, and apparently had kept it for some years, though for how many does not exactly appear. We must take it as the case went to the jury that there was no evidence that the hospital had ever before been used as a place for procuring abortion. It was proved that the girl herself arranged with the prisoner for being treated in the hospital, and that, under that arrangement, she entered on the 9th August. With that evidence the prisoner's statement is in accord. There seems to have been some discussion as to the terms of payment, and a witness named Lack, through whom a girl friend of Mary Davies had supplied £25 for the purpose, went to the hospital, saw the prisoner there, and endeavoured to get him to take a bill, which he refused to do. No inference is to be drawn against the prisoner from his refusal to deal on credit, for, apparently, Poke and Mary Davies were at that time unknown to him. Lack's evidence, however, corroborates the circumstance of Mary Davies having entered the hospital under the name of Mrs. Nelson under some arrangement made by her with the prisoner. Whether the arrangement was as suggested by the Crown, the payment of £25, or as stated by the prisoner, £5 per week payable in advance, is left uncertain. She took with her to the hospital a dress basket containing some clothes. The clothes she was wearing when she entered and and those she took with her are described by her mother and sister who knew with some exactness, apparently from personal observation, the clothes she had in her possession. She had a serge dress on which were large metal buttons covered with the same material and blouses with mother-o'pearl buttons. She seems to have worn stocking suspenders with the usual attachments kept up by a waist belt with buckle, also women's boots of the ordinary light kind. She entered the hospital as I have said on the 9th August but nothing eventful happened until the 17th August.

Up to that time Poke had been visiting her from time to time and she appeared to him to be going on well. On the 17th August Poke went to the hospital and had a very important interview with the prisoner. The prisoner told him that the girl was suffering from puerperal fever, which, he explained, arose from absorption of poison into the blood. In answer to Poke's question he said the disease was dangerous. Poke went to her room and found the girl looking ill and drowsy. The prisoner followed him and explained that on the night before—the 16th—she had fallen down going to the lavatory and cut her face. That was said in explanation, apparently, of some cuts appearing on her hands and face. It is important to note that he did not then, nor has he at any other time attributed any ill consequences to the fall. On the next night, Friday, 18th August, Poke went to the hospital again. The girl apparently was very ill—in a critical condition. Then took place the first of a series of conversations which the jury would clearly be entitled to regard as incriminating the prisoner. Before referring to these in detail, it will be well to consider the relation in which the prisoner stood to Poke at that time and the right of the latter to be informed of her condition. It is plain that Poke was aware of the terms on which she had been received into the prisoner's hospital but we have no evidence of the details of the arrangement beyond those given by the prisoner himself. He explains in his statement to the jury that she complained of feeling ill and placed herself under his treatment, and that she was suffering from abdominal pains and a discharge of blood. He found there was a threatening of a miscarriage and advised her to take a rest and a course of treatment in his hospital for which she would have to pay £5 per week in advance. He further states that from her symptoms he thought it might be necessary, if a miscarriage did occur, to perform the operation known as curetting. That was the reason he gave for obtaining the written consent of her husband, as he then believed Poke to be, for her treatment. Notwithstanding the difference between the prisoner's case and that of the Crown in other respects, his statement corroborates the Crown case in the all important fact that the girl was in the hospital under an arrangement that he should treat her for a threatened miscarriage, which at the time was more or less imminent. In considering the relations of the prisoner and Poke it is important to note that no one but the prisoner had anything to do with the patient in the way of surgical or medical treatment. At the time of the conversation of which I am speaking the prisoner believed Poke to be the girl's husband. Under these circumstances the latter at the outset of the interview was therefore asking the former as her medical attendant as to her condition. It cannot be doubted that it was the prisoner's duty to give the information asked.

The prisoner informed him that she was in a dangerous condition and suffering from puerperal fever. The medical evidence is that the occurrence of puerperal fever under the circumstances established beyond doubt that there had been a miscarriage. It is clear therefore that the dangerous condition which Mary Davies was in on the 18th August arose from a miscarriage. On the evidence there is no possible way of accounting for her death except that it was from puerperal fever brought on by a miscarriage. The miscarriage happened in the prisoner's hospital while the girl was under his sole care. As to whether it arose in some way for which he was not to blame or whether it arose from his criminal act the prisoner himself must have known. He alone therefore was able to inform Poke, and afterwards the Court, if he thought fit, how the miscarriage had come about, why the girl died of it and what had become of her dead body. The prisoner's statement to Poke as to the girl's condition when it first became evident that she was likely to die is therefore of vital importance. The prisoner said "Mrs. Nelson is very ill. In case anything happens, what are you going to do?" To which Poke replied "I do not know." Some vague expressions made use of by the prisoner, which immediately followed, have been the subject of much discussion. Probably their real meaning was substantially this: "The best thing to say is that she came here to bring about a miscarriage, I had to attend to her, and it was necessary, in order to save worse consequences to bring about the miscarriage and treat her after." The particular form of expression matters very little. The significant fact for the jury was that the prisoner who, on the assumption of his being innocent, was treating a woman lawfully in the ordinary way in his own hospital should ask the man whom he supposed to be her husband what he was going to do if she should die. In the next sentence Poke tells him they are not married. "Then," said the doctor, "leave it to me. In that case don't come to the house again, it might breed suspicion. Meet me to-morrow night at the Melbourne Cricket Ground at 9 p.m." It may be reasonably asked whether the latter sentences of the conversation could have left the jury in any doubt as to what was in the prisoner's mind; they were certainly entitled to ask themselves whether such a conversation could be reasonably explained on any other ground than that the prisoner knew that the girl's death had occurred under such circumstances as would make its discovery a very serious matter for him and for Poke. The next interview was on the following Sunday night, the appointment for the Melbourne Cricket Ground not having been kept. It was for the jury to consider why such secrecy was observed, why the man so deeply interested in her life, was warned not to go to the hospital to see her at a time when her life was despaired of. Poke, however, on the 20th August called at the Hospital, found that she was slightly better, and arranged to meet the prisoner on Monday night following. He did so, and was then told by the prisoner that things were as bad as they could be. A meeting was arranged for the next night at the Fitzroy Gardens. They met accordingly, and an important conversation took place. That was the night of Tuesday, 22nd August. Poke asked how it was with the girl. The prisoner said "She is dead." I need not repeat the remainder of his statement on that occasion in detail. It was to the effect that the girl was dead, and that he had given her chloroform before she died to quieten her screaming. The prisoner next asked how long it would be before her relatives would inquire, and went on to say: "If they do not inquire for a fortnight I can have her secretly buried." He further said: "I'll have to let my housekeeper know; she is my right hand." In other words, he promised to have the body buried secretly, to dispose of the clothing, and undertook that no one should know anything about it except the housekeeper and himself. The death must have taken place on the 21st or 22nd August. On the 22nd the only servants in the house, two girls, were sent away. They were told by the prisoner that they might have a few days holiday. Apparently one of them asked to be allowed to stay until the next day, but he said: "No; cannot you stay with your aunt? You had better go to-day." She went accordingly, and the prisoner and his housekeeper were then the only occupants of the house. The interview of the 22nd August was the last which Poke and the prisoner had for some time. It is important to observe that at its close the latter said: "You will have to keep your mouth closed or it will be bad for both of us." Here again the jury were entitled to ask themselves whether such a warning could possibly be explained on any other supposition than that the miscarriage, of which the girl died in the prisoner's hospital and under his treatment, had been brought about by some unlawful act on his part. The next meeting was arranged for the 29th August. In the meantime, on the 27th, the prisoner was at his farm at Carrum. On that day he made a fire in one of the paddocks, and there was evidence that in that fire was burnt woman's clothing which, judging by the remains in the ashes, corresponded generally with the clothing which had been in the girl's possession in the hospital. Those were circumstances from which the jury might well conclude that the prisoner had carried out his promise to Poke to burn the girl's clothes at his farm. On Tuesday, the 29th August, Poke saw the prisoner at the hospital. The latter expressed his fear that Poke was being watched, told him that the girl's clothes had been burnt, that he had got rid of the last of them the day before, and that her jewellery had been burnt with them. Incidentally it may be mentioned that the latter statement was not true as the jewellery was afterwards found in his possession. It was also inconsistent with the statement he made in Court that the girl had handed the jewellery over to him in payment for accommodation at his hospital. On that night, the 29th August, the police entered the prisoner's premises. An interview took place which has been detailed already. The prisoner denied at first that the girl had ever been at the hospital. Finally, when the document purporting to have been signed by her husband, authorizing her to be treated by the prisoner was found by the detectives, he admitted that the girl had been there, but declined to answer any other questions. Poke was then confronted with him and went through in his presence a statement of what the prisoner had said to him on that evening. The prisoner declined to answer or explain any of Poke's statements. Such were the material facts from which the jury were asked to draw the conclusion that the prisoner was guilty of the offence with which he was charged. It was their duty, applying to the evidence the principles to which I have referred, to draw the conclusion to which all the circumstances taken together reasonably led them, to the exclusion of any other reasonable conclusion. As to some issues of fact the reasonable conclusions were not difficult. One inference inevitably followed from the evidence as a whole, namely that the girl died at the prisoner's hospital from the effect of a miscarriage. The miscarriage might have occurred without any fault on his part. On the other hand it might have been brought about by his criminal act. The crucial matter therefore to be determined was whether the inference that the miscarriage had been brought about by the prisoner's criminal act was the only inference that could reasonably be drawn from all the circumstances taken together. There was one other inference which it has been suggested might have been drawn, namely, that the prisoner had arranged the secret burial of the body and the destruction of all evidence of the girl having been at the hospital, because he feared the consequences to himself of the death through miscarriage of an unmarried woman brought to his hospital by her paramour. It is suggested as possible that, for that reason, he followed the course he took in order to conceal the evidence which an examination of the body might have disclosed. In my opinion it would be practically impossible for a jury on the evidence before them to reasonably come to the conclusion that any sane man in the prisoner's position, innocent of wrongdoing, would, for such a motive as that suggested, take upon himself the awful responsibility of concealing a death in his own hospital in the midst of a city where the law and the usages of a civilized community render it essential that every death shall be made public. The obtaining of a death certificate before burial, the registration of death, the publicity of ordinary burial in a cemetery are part of the law and the usages of most civilized societies. Yet it was argued that, for the very inadequate reason suggested, the prisoner took upon himself the risk of secretly disposing of his patient's body and concealing the fact of her death. I find it impossible to say that the existence of that possibility ought, as a matter of law, to have prevented the jury from drawing the conclusion, reasonably open to them on the evidence, that the prisoner took upon himself the responsibility of concealment because he feared the consequences to himself of the real facts as to the girl's death coming to light.

In connection with the inference which may be drawn from the conduct of persons accused I shall refer to two passages in Starkie on Evidence, 3rd ed., which contain a very clear statement of the principles applicable.

The first passage at p. 575 is as follows:—

It frequently happens, as has been seen, that where the evidence of the circumstances attending the transaction itself would be imperfect and inconclusive, it derives a conclusive nature and tendency from a consideration of the conduct of the accused. The ordinary motives of self-preservation and self-interest, common to all mankind, furnish the strongest presumption that a party would explain, by statement at all events, and by proof where it was practicable, such evidence as tended to his prejudice. Hence it is that circumstances, which abstractedly considered, would be inconclusive, acquire a conclusive character and tendency from the silence of the adversary, or his failure in attempting to explain them.

In the following passage at p. 563 of the same volume the principle is further illustrated:—

The presumption that a man will do that which tends to his obvious advantage, if he possesses the means, supplies a most important test for judging of the comparative weight of evidence. It is to be weighed according to the proof which it was in the power of one party to have produced, and in the power of the other to have contradicted.

If, on the supposition that a charge or claim is unfounded, the party against whom it is made has evidence within his reach by which he may repel that which is offered to his prejudice, his omission to do so supplies a strong presumption that the charge or claim is well founded; it would be contrary to every principle of reason, and to all experience of human conduct, to form any other conclusion.

In R. v. Burdett[16], Lord Tenterden, then Abbott C.J., makes some pertinent observations on the same subject. He says:—

In a great portion of trials, as they occur in practice, no direct proof that the party accused actually committed the crime, is or can be given; the man who is charged with theft, is rarely seen to break the house or take the goods; and, in cases of murder, it rarely happens that the eye of any witness sees the fatal blow struck or the poisonous ingredients poured into the cup. In drawing an inference or conclusion from facts proved, regard must always be had to the nature of the particular case, and the facility that appears to be afforded, either of explanation or contradiction. No person is to be required to explain or contradict, until enough has been proved to warrant a reasonable and just conclusion against him, in the absence of explanation or contradiction; but when such proof has been given, and the nature of the case is such as to admit of explanation or contradiction, if the conclusion to which the proof tends be untrue, and the accused offers no explanation or contradiction; can human reason do otherwise than adopt the conclusion to which the proof tends? The premises may lead more or less strongly to the conclusion, and care must be taken not to draw the conclusion hastily; but in matters that regard the conduct of men, the certainty of mathematical demonstration cannot be required or expected; and it is one of the peculiar advantages of our jurisprudence, that the conclusion is to be drawn by the unanimous judgment and conscience of twelve men, conversant with the affairs and business of life, and who know, that, where reasonable doubt is entertained, it is their duty to acquit; and not of one or more lawyers, whose habits might be suspected of leading them to the indulgence of too much subtilty and refinement.

Applying the principles expounded in these passages to the evidence before us, can it be said reasonably that the jury were not entitled to draw the inference of guilt against a doctor, charged as the prisoner has been, who will not explain the death of a patient in his hospital, whom he alone was attending, his secret disposal of her body, his burning of her clothes, and his attempt to remove all evidence of her ever having been in his hospital and under his care? They were not bound to draw that inference; another jury might arrive at the opposite conclusion. But I have no hesitation in affirming as a matter of law that the verdict which the jury did find was open to them on the evidence. As I have pointed out Poke's evidence forms a vital part of the Crown's case, and, before considering the next question, I shall proceed to consider whether that evidence was in fact corroborated, because it is conceded that, if the evidence was in fact corroborated, the omission of the Judge to warn the jury as to the necessity of corroboration was of no moment. And, first, it may be asked, what is corroboration? It does not mean that all the material facts have to be proved by independent evidence. What it does mean is well expounded in the following passage from the last edition of Russell on Crimes (7th ed.), at p. 2287:—

The confirmation need not extend to every part of the accomplice's evidence, for there would be no occasion to use him at all as a witness, if his narrative could be completely proved by other evidence, free from suspicion. But the question is, whether he is to be believed upon points which the confirmation does not reach. And if the jury find some part of his evidence satisfactorily corroborated, this is a good ground for them to believe him in other parts as to which there is no confirmation. So far all the authorities agree; the only point on which any difference of opinion has been supposed to exist relates to the particular part or parts of the accomplice's testimony which ought to be confirmed.

It is not sufficient to corroborate an accomplice as to the facts of the case generally. He should be corroborated as to some material fact or facts which go to prove that the prisoner was connected with the crime charged.

I agree with my learned brother Barton that there was abundance of corroboration. It is to be found in the burning of the girl's clothes by the prisoner at his farm, in his possession of her jewellery, in his false and contradictory statements about her death and in everything that he did in his endeavour to hide the fact that she had been his patient and had died in his hospital As Lord Alverstone has pointed out in R. v. Tate[17] where there is in fact corroboration of an accomplice's testimony the Court will not interfere with a verdict of guilty, even though the Judge has not followed the usual practice in warning the jury against giving credit to the accomplice's uncorroborated evidence. Holding that view, it is not really necessary for me to consider what directions ought to be given to a jury with reference to an accomplice's evidence, and the consequences of a failure to give any direction. But as the matter has been gone into, I propose to express my opinion. The first question that arises is whether the giving of the usual direction is a rule of law or a rule of practice. In my opinion the common law of Victoria on the subject is the same as the English common law. The common law of England in this respect was definitely laid down as well established in R. v. Stubbs[18], and I agree with Cussen J. in the Court below that the common law of Victoria on this matter is as stated by the learned Judges in that case. Nor have I any doubt that it is still the common law of England, notwithstanding the practice which the Court of Criminal Appeal, constituted under a late Statute, has followed in several cases. In the latest edition of Russell on Crimes (7th ed., p. 2286) the law is thus stated:—

It has long been adopted as a general rule of practice that the testimony of an accomplice ought to receive confirmation, and that, unless it be corroborated in some material part by unimpeachable evidence, the presiding Judge ought to advise the jury to acquit the prisoner. This practice of requiring some confirmation of an accomplice's evidence in strictness rests only upon the discretion of the Judge. And this indeed appears to be the only mode in which it can be made reconcilable with the doctrine already stated, that a legal conviction may take place upon the unsupported evidence of an accomplice. But the practice in question has obtained so much sanction from legal authority that it "deserves all the reverence of law," and a deviation from it in any particular case would be justly considered of questionable propriety.

The rule is there treated as still a rule of practice and not a rule of law, notwithstanding the view of the Court of Criminal Appeal in the cases cited to us. No doubt the Court of Criminal Appeal in England has virtually established the rule as invariable in that Court, not, in my opinion, as modifying in any way the old common law rule, but in the exercise of the extended powers which it possesses under the Criminal Appeal Act of 1907. The power of the Court of Criminal Appeal before the enactment of that Statute was limited to giving effect to objections of law. In R. v. Stubbs[19], Willes J., with reference to an objection that the practice now under consideration had not been followed, says, "We sit here under a Statute to decide questions of law, and questions of law can only be reserved for our opinion. This is not a question of law, but of practice."

The new Court of Criminal Appeal in England, constituted by the Act of 1907, is empowered by sec. 4 to set aside a conviction in any case in which it is satisfied there has been a miscarriage of justice, whether the objection is a legal objection or not. The Court in R. v. Tate[20] treated the failure to comply with the rule of practice as to warning juries not to convict on the uncorroborated evidence of an accomplice as having brought about a miscarriage of justice. That decision has been followed in several cases, and it seems to be now the recognized rule of that Court to so regard the failure of the trial Judge to direct the jury in accordance with that practice. But the Victorian Court has no such power. Its power is still limited to giving effect to objections of law as the power of the English Courts was limited at the time when R. v. Stubbs[21] was decided. I have therefore come to the conclusion that, even if Poke's evidence had not been corroborated, though it would have been the duty of the Judge to follow the practice of warning the jury against convicting on the uncorroborated evidence of the accomplice, his failure to give that direction would not have amounted to an error in law entitling the prisoner to have the conviction quashed.

I come now to the ground which relates to the misdirection of the presiding Judge with respect to the prisoner's statement. My learned colleagues have dealt with that question very fully and I agree with their view of the matter. I will only add that it would be impossible in my opinion to give effect to the right conferred upon the prisoner by sec. 142 of the Victorian Evidence Act 1890 if the prisoner's statement were not put before the jury for their full consideration. Any statement of a prisoner not on oath carries on the face of it certain infirmities. It has not the sanction of an oath and the prisoner who makes it is deeply interested in asserting his innocence. On the other hand, it may explain matters in evidence which otherwise would tell against him. Sec. 52 of the Victorian Evidence Act 1890 permits the prisoner "to make a statement of facts (without oath) in lieu of or in addition to any evidence in his behalf," clearly implying that the prisoner is entitled to have his statement as fully considered as his evidence would be, if he gave evidence. Madden C.J. seemed to think he had bettered his misdirection by the second summing up on this point made at the request of the prisoner's counsel. I agree with my colleagues that the correction made the position no better. I therefore concur in the view of the three learned Judges of the Supreme Court of Victoria that the summing up was in this respect a serious misdirection. The prisoner's statement was in direct contradiction of the Crown case and purported to furnish an explanation of the circumstances proved against him. But, as it was left to the jury, it is quite clear that they were prevented from considering it on its merits. On that ground I agree that the conviction must be set aside.

Then arises the important question—what consequences are to follow? Is this Court to quash the conviction and so end the prosecution, or should this Court direct a new trial? The proper course to be taken will depend largely upon whether the case is one in which the Supreme Court of Victoria ought to have exercised the power of granting a new trial, for this Court is in a position now to make any order which the Supreme Court of Victoria ought to have made. There have been many criminal cases in Victoria in which statutory authority new trials have been granted—one of them was a case of murder, R. v. Whelan[22]. There have been six or seven other criminal cases in which new trials have been granted. After carefully reading the decisions in those cases, I have been unable to gather that the Victorian Supreme Court has laid down any general principles on which it will exercise its discretion to grant a new trial in a criminal case. Perhaps it is as well that no exhaustive or rigid definition of principles should be attempted. The Court must, however, exercise a legal discretion, that is to say, must act upon some legal principle. It appears to me that one principle at least may be laid down. Where the facts proved a first trial would have been sufficient to support the conviction, if the jury had been properly directed, it seems to me that in general a new trial may be granted to enable the faulty direction to be remedied. In exercising the discretion given by the Statute the interests, not only of the prisoner, but of the efficient administration of justice ought to be considered, always providing that no injustice is done to the accused. In this case there was, as I have pointed out, ample evidence to justify a verdict of guilty, if the jury thought fit to come to that finding on the evidence. If it were not for the misdirection as to the prisoner's statement, the verdict of the jury could not in my opinion have been disturbed. I think it is now in the interests of the administration of justice, and not unjust to the prisoner, that a new trial should be granted to enable the evidence to be again submitted to another jury with a proper direction as to the prisoner's statement. I agree therefore with my learned brother Barton that the conviction should be quashed and a new trial ordered.

Appeal allowed. New trial ordered.

Solicitors for the appellant, Strongman & Crouch.

Solicitor for the respondent, Guinness, Crown Solicitor for the State of Victoria.