APPLICATION for special leave to appeal, and APPEAL, from the Court of Criminal Appeal of Victoria.

DIXON, MCTIERNAN, WEBB AND KITTO JJ. This is an application by a prisoner Court of Victoria refusing him leave to appeal against his conviction. (at p371)

2. The questions upon which the application to this court depends cannot be understood without a brief account of some of the circumstances of the crime. (at p371)

3. The prisoner was convicted of murdering a young woman, twenty-two years of age, with whom he had been living. Her name was Sylvia Holmes. Her death was caused by manual throttling, but before her death she had received a blow of some kind on the left side of the chin where bruising had developed. She and the prisoner lived in a very small semi-detached dwelling in Fitzroy. She worked as an office cleaner in the city but she was also a prostitute. The prisoner worked at a factory. Their association had gone on for about six years, for the last four years of which they had been living together, apparently in this dwelling. Sylvia Holmes was killed on the evening of Friday 5th May 1950. On the afternoon and evening of that day the prisoner had been drinking beer, first in a neighbouring hotel and afterwards in the house of another prostitute a couple of doors down the street. This woman's name was Ada Adkins. It appears from the evidence that Sylvia Holmes, who presumably had returned from cleaning, came and called the prisoner to come to his tea. After some delay he went. He encountered two companions in the street with whom he drank more beer. One then accompanied him to his dwelling and sat with Sylvia Holmes and the prisoner in their kitchen while they ate their meal. According to him their behaviour was amicable. After that the prisoner returned to Ada Adkins' house. The time would then have been about half past seven. Ada Adkins deposed that after his return the prisoner made use of strong expressions of resentment at the behaviour of Sylvia Holmes, the grounds being in effect that she was complaining of his visiting Ada Adkins and that she would not allow him to have sexual intercourse with her without his using a contraceptive. According to the witness he said that he would knock her out if she did not shut up. He sat with a party of people drinking in Ada Adkins' house. The beer ran out and she set forth to procure more. She says that the prisoner left the house with her and that when they reached his house, where the light was on, he said: "You go and get the beer and I will slip in and see how Sylvia is". He entered the house and she went on. The prisoner's version is that he did not accompany Ada Adkins from her house when she set out to obtain more beer, but remained for some time, he going back, however, to his house before she had returned. Ada Adkins said that she went for the beer, obtained it and carried it back, her route taking her again past the prisoner's dwelling. On her account twenty minutes or more must have elapsed. Just as she had passed the prisoner's house he called to her from his front door. On entering she found the body of Sylvia Holmes lying halfway on the bed in the front room, the room upon which the door opened. The prisoner's story was that he had just gone in the front door on his return from Ada Adkins' house and had found the body in this position. He had seen Ada Adkins approaching and had also seen near by another woman named Phyllis Gunn. Having seen Sylvia Holmes' body and having cried out to her without response, he rushed out of the door and called first to Ada Adkins and then to Phyllis Gunn. The latter, according to him, was the first to come. Phyllis Gunn, who fixed the time as about five minutes past nine, agrees with the prisoner that she entered first and as to this disagrees with Ada Adkins. Phyllis Gunn felt Sylvia Holmes' heart and concluded that she was dead. The prisoner told her to "get out" and set off for the police. Ada Adkins' evidence puts a definite interval between her own entry and the arrival of Phyllis Gunn. She swore that she said to the prisoner: "What happened? Is she dead or alive?" He answered that she was dead and added: "She has had it coming to her for a long time but I did not mean to kill her". He said he hit her but he did not mean to do it. She then took the beer to her house, returning at once with two bottles. It was then, as her evidence appears to state, that Phyllis Gunn came in. The latter, however, went to her own house as the prisoner ran off for the police, while Ada Adkins remained in the prisoner's dwelling with the body of Sylvia Holmes. When the police took charge they kept the prisoner out of the front room. He behaved in an excited, not to say frantic, manner. A doctor came who pronounced the girl dead but he did not look for the cause of death. At a late hour police photographers came and then the government pathologist. Except that Ada Adkins had straightened a mat and that the dead girl's leg which hung over the bedside had been placed on the bed, it is said that nothing had been moved. The photographs show the body lying face uppermost, with the right buttock over the edge of the bed and the trunk and head extending diagnolly across the bed, a double bed. The coverlet had been thrown back to the right-hand side of the bed, the pillows were displaced and the undersheet was also pushed back, so that except where a pillow was partly under it the body lay on a blanket. The girl's legs were bare, her dress being pushed up to her thighs. Her panties were at the head of the bed, her watch under her body and her head scarf under her elbow. On the floor was a clean contraceptive. Phyllis Gunn said that the bedclothes were not in this condition when she came in, but were undisturbed with the body lying on the coverlet. Ada Adkins in effect said that they were in the same condition. The bruise on the chin of the dead girl was visible, but it is doubtful if any marks on her neck and throat were distinctly noticeable. Until the government pathologist came, no-one, according to the case for the prosecution, knew what the cause of death was. But one officer of police said that in the meantime the prisoner, after making a violent attempt to escape from them to the room where the body lay, told a constable who said that he had been at more than beer that that had nothing to do with him and continued: "Find the bastard who throttled Sylvia and don't stay around here". The Crown prosecutor, in opening to the jury, used this against the prisoner because at that stage of the evening no-one but the man who killed her could know that Sylvia Holmes had been throttled. The other police officer, however, gave a different version. According to his version the prisoner said: "Stop sitting around here and get the bastard who did Sylvia" or words to that effect. The prisoner was not then arrested. It appears that Ada Adkins did not at that time tell the police of the statement he made to her when she first entered the room where the dead girl lay, namely, that he had hit her but did not mean to kill her. Indeed it was not until 6th July 19518 over a year later, that she told them. A brother of Sylvia Holmes said that on the day following her death the prisoner remarked to him that if he had been at work it would not have happened, a remark the prisoner explained in his evidence as meaning that he would not have had the shock of finding her body. But naturally it was used against him as really meaning that he would not have killed her had he been at work. The medical evidence showed that Sylvia Holmes had been strangled or throttled by someone's bare hands. It showed too that she had not long been dead. But it proved that she must have received the blow upon her chin at some interval before death not less than ten minutes. (at p374)

4. The foregoing outline is limited to evidence of which the admissibility or the reception at the trial was not in question upon this application. It is better to state the nature of the evidence, the reception of which was in question in considering whether it ought to have been received. It will be seen that if Ada Adkins' account of the prisoner's statements to her and of his movements was accepted, the case to show that it was at the hands of the prisoner that Sylvia Holmes died was by no means weak. There was ample time while Ada Adkins was getting the further supply of beer for him to do it. Throttling would not take a minute, and it is not necessary to suppose that the throttling followed the blow on the chin in rapid succession. He might have given her a blow on the chin early in a scene of violence and mounting passion. Because the medical evidence was inconsistent with death having occurred immediately after the blow, an hypothesis was put forward that the prisoner delivered the blow which had rendered her unconscious and that he had gone for the police leaving her in this condition. The hypothesis was that she was then not dead and that she was strangled during his absence. This it was suggested might account for the discrepancy in the evidence as to the state of the bed. It is enough to state this hypothesis to show that it may safely be neglected, as, in fact, it was at the trial. (at p374)

5. The learned judge who presided at the trial put the case to the jury as one in which it was open to them to find a verdict of murder in one or other of three aspects. They might find that the prisoner brought about the young woman's death by a voluntary act done intending to kill her. They might find that, although an intention to kill was absent, he did intend to inflict grievous bodily harm and did an act likely to lead to death. The third alternative was that without an intention to kill her he caused her death in the course of attempting to have sexual connection with her against her will and by force, caused it by force of a kind likely to bring about her death. If they were not satisfied of any of these alternatives but were satisfied that he had killed her laying his hands on her against her will the jury might find the prisoner guilty of manslaughter. But his Honour warned them against finding a verdict of manslaughter by way of compromise. (at p374)

6. Upon this direction it may be remarked that, assuming always that the jury were satisfied beyond reasonable doubt that it was the prisoner who caused Sylvia Holme's death, there was nothing in the facts to suggest that she did anything that would afford provocation reducing the homicide to manslaughter. On the other hand, the circumstances did, on the same assumption, make it rather more likely than not that the attack upon the woman in which she was throttled was made in an outburst of passionate violence on the part of the prisoner, who was not, in making the particular attack, animated by a specific intent to effect a further purpose. It may be that it all arose from demands for sexual connection without using a contraceptive. That might account for what the photographs show. But it does not follow that when he throttled her he was actually attempting to commit rape upon her body or doing something in furtherance of that object. Indeed there is an absence of some of the evidence of this that might be expected. These considerations do not directly affect the questions upon which our decision depends, but they show that much turned upon a precise application of the distinction between murder and manslaughter and they bear upon the importance to be attached to the reception of evidence the admissibility of which is denied. (at p375)

7. At the trial the Crown prosecutor, without objection from counsel for the prisoner, led a great deal of evidence of doubtful or at all events of disputed relevance. The admissibility of the evidence is put by the Crown on the ground that it tends to show either motive or possible motive, or grounds of resentment on the part of Sylvia Holmes and consequent anger and violence on the prisoner's part, or the prisoner's hostility to the use of a contraceptive in sexual intercourse or threats to Sylvia Holmes betokening hostility or a desire, presumably an intermittently recurrent desire, to kill her or else that it tends to rebut a possible defence that the prisoner bore such affection to her that he would be unlikely to kill her. (at p375)

8. A young woman named Kathleen Adams was called to say that fifteen months before Sylvia Holmes' death she had begun to have sexual relations with the prisoner, that he did not use a contraceptive, that at the time of the crime she was pregnant to him of a child born on 3rd August 1950, that she had not seen the prisoner from the time of the crime until after the birth of her child, but that at the time of his arrest she was living with him, and that about eight months before her death Sylvia Holmes had shown her hostility to her in an insulting manner. This evidence appears to be entirely inadmissible. It is of course theoretically possible that because another woman was pregnant to him he wanted to get rid of Sylvia Holmes. But that is a speculation with no relation to the realities of the situation which the foregoing summary depicts. He was not married to Sylvia Holmes, who was a prostitute with whom he chose to live. Apparently he was married to some other woman. To the police the prisoner did remark that he loved Sylvia and would not kill her, but that does not justify the introduction into the case of his relations with other women. The evidence of this witness is not made relevant in any manner to the crime charged. In the evidence of Ada Adkins, who, it will be remembered, was another prostitute, were descriptions of how the prisoner had sexual connection with her and did not use a contraceptive and how Sylvia Holmes came looking for him and how she exhibited her hostility to the witness. Strangely enough, this was not made the subject of distinct objection upon the hearing of the application for leave to appeal in the Supreme Court. Only in a very exceptional case would this Court entertain an objection not taken before the Court of Criminal Appeal as a ground for granting special leave. But, in the view we take of the whole case, it is perhaps proper to observe that, except that the prisoner's statement on 5th May 1950 to Ada Adkins to the effect that Sylvia resented his coming to her house and the manner in which she behaved that evening when coming near or into contact with the witness may make the dead woman's attitude to Ada Adkins relevant, the passages summarized above from her evidence do not seem admissible. (at p376)

9. A male witness was called who deposed among other things that two or three months before her death he had intercourse several times with Sylvia Holmes and that she requested him to use a contraceptive which she produced from a drawer. This evidence was irrelevant and inadmissible, but, except for its contribution to the distraction of the minds of the jurors from the vital issues, it could hardly matter much. (at p376)

10. A complaint is made that evidence was led of threats of violence or exhibitions of violence towards Sylvia Holmes by the prisoner. In the main this evidence was admissible because it might reasonably be considered to tend to show the prisoner's violent behaviour towards her as the woman with whom he was living. But there are two or three observations to be made in reference to such evidence. First some of it consisted in descriptions of marks of injury borne by Sylvia Holmes, no direct evidence being tendered to show that the prisoner inflicted them. This evidence could only be made admissible by circumstantial evidence raising a presumptive inference that he was the author of the injuries. In most of these instances there is not enough for this purpose, as the transcript stands. In the second place, some of the evidence of violence and threats of violence relates to a time somewhat remote from the date of the crime, remote enough to make it likely that, had it been objected to, it would have been excluded. That, however, leads to the third observation, namely, that in the particular circumstances of this case the frequent use of violence by the prisoner towards the woman with whom he was living and threats that one day he would kill her, especially when repeated over a long period, do not necessarily tend to support the inference that he deliberately killed her and do anything but tend to show that he killed her in the course of attempting to commit rape. (at p377)

11. It is now necessary to pass to what is perhaps the most important question in the case, a question about which the Supreme Court was divided in opinion, Lowe and Martin JJ. taking a view against the prisoner and Smith J. a view in his favour. When the two police officers gave in evidence respectively the differing versions of what the prisoner had said about pursuing the man who had killed Sylvia, it must have seemed antecedently probable that if the prisoner gave evidence on his own behalf, as in the event he did, he would deny the version which attributed to him the word "throttle" and adopt the version which made him use the expression "who did Sylvia". Clearly the latter could not incriminate him whilst the word "throttle" could only be used by him, if he had not committed the crime, because either from something he had heard or something he had seen, he thought that was the manner of Sylvia Holmes' death. But at the stage when he made this statement nothing, so far as the evidence went, had occurred or appeared which would indicate that she had been throttled. The prisoner's counsel cross-examined the officer, who deposed to the word "throttle" to show that the other constable had an equally good opportunity of hearing the word actually used, that is to say, the constable who deposed to the words "who did Sylvia". The transcript of the evidence discloses other grounds for supposing that the prisoner was expected by his counsel to say in evidence that the word he employed was not "throttle", and that the other police officer was right in saying that he spoke of the man "who did Sylvia" or words to that effect. But when the prisoner was being examined in chief he stated that he did say: "Stop sitting round here and go and find the bastard that throttled Sylvia". He was then asked: "Well, now, why did you use the word 'throttled'?" and answered: "Well I heard the police talking about it and I took it she had been throttled. I heard the police talking about it, I heard them say she had been throttled or strangled". He went on to say in effect that he took it for granted that it was right, that the police ought to know and that otherwise he had no knowledge as to what caused her death. (at p378)

12. The Crown Prosecutor lost no time in seeking to turn to account this doubtless unexpected admission. Before beginning the cross-examination of the prisoner he applied to recall the police witnesses. The prisoner's counsel objected but the presiding Judge decided to permit it. In the result after the close of the case for the defence seven officers of police were recalled and examined and cross-examined as to the possibility of someone having said that the dead girl was throttled or strangled and of the prisoner having heard it. When these witnesses had been called during the case for the prosecution they had in some instances given times when they came to the dwelling in a way which made it seem, erroneously as it turned out, as if that was their first arrival. It was necessary to clear these misapprehensions away. Although the evidence of none of the seven witnesses given when recalled was long, the cumulative effect was to bring into strong relief the admission made by the prisoner that he had used the word "throttled", to give great emphasis to the point made by the Crown Prosecutor upon it, as well as to the contradiction which was involved of the prisoner's evidence by so many witnesses and also to detract from any advantage the prisoner might obtain in placing before the jury by his personal evidence the answer he made to the charge as the final thing before the addresses of counsel and the charge to the jury by the judge. (at p378)

13. Now an extremely strict rule has been adopted in the English Court of Criminal Appeal with respect to the discretionary power of a presiding judge to allow the prosecutor to call evidence after the close of the case for the defence. The rule is adopted from the language attributed to Tindal C.J. in the report of R. v. Frost (1839) 4 St Tr (NS) 86, at p 386. His Lordship is reported to have said: - "There is no doubt that the general rule is that where the Crown begins its case, like a plaintiff in a civil suit, they cannot afterwards support their case by calling fresh witnesses because they are met by certain evidence that contradicts it. They stand or fall by the evidence they have given. They must close their case before the defence begins; but if any matter arises ex improviso, which no human ingenuity can foresee, on the part of a defendant in a civil suit, or a prisoner in a criminal case, there seems to me no reason why that matter which so arose ex improviso may not be answered by contrary evidence on the part of the Crown". The great limitation contained in this language lies, not in the expression "ex improviso", which means no more than "unexpectedly" or "suddenly", but in the words "which no human ingenuity can foresee". In the report in Carrington & Payne (1839) 9 Car & P, at p 159 (173 ER, at p 784) the words ascribed to Tindal C.J. are: "But if any matter arises ex improviso, which the Crown could not foresee, supposing it to be entirely new matter, which they may be able to answer only by contradictory evidence, they may give evidence in reply". The decisions in England allow the presiding judge at a criminal trial to call a witness if he thinks the imperative demands of justice require it. This view was acted on in Victoria (R. v. Collins [1907] VicLawRp 55; (1907) VLR 292). But in Titheradge v. The King [1917] HCA 76; (1917) 24 CLR 107 this Court denied the power. In England the rule expressed in the language of Tindal C.J. is applied alike to the power of the judge to call a witness and to his power to allow the prosecution to recall a witness or call a new witness once the case for the prosecution has closed and the prisoner has gone into his case. In R. v. Frost (1839) 4 St Tr (NS) 86 it was probably not intended to state an exhaustive rule. But how rigidly the principle against reopening the case for the prosecution was applied may be seen from the striking decision by Garrow B. in R. v. Stimpson (1826) 2 Car & P 415 (172 ER 188). In the end the language of Tindal C.J. has come to be used in England as a criterion (R. v. Harris (1927) 2 KB 587; R. v. Liddle (1928) 21 Cr App R 3; R. v. McMahon (1933) 24 Cr App R 95; R. v. Day (1940) 27 Cr App R 168). The earlier cases of R. v. Crippen (1911) 1 KB 149 and R. v. Sullivan (1923) 1 KB 47 seem to have allowed more flexibility. The formula adopted from Tindal C.J. has little to commend it. The words "which no human ingenuity can foresee" hardly express a legal principle. They are rhetorical, but if literally understood they lay down a test which could almost never be satisfied. Clearly the principle is that the prosecution must present its case completely before the prisoner's answer is made. There are issues the proof of which do not lie upon the prosecution and in such cases it may have a rebutting case, as when the defence is insanity. When the prisoner seeks to prove good character evidence may be allowed in reply. But the prosecution may not split its case on any issue. The Court possesses a power to allow further evidence to be called, but it must be exercised according to rule and the rule is against reopening the Crown case unless the circumstances are most exceptional. We are not disposed to lay down the rule in the terms adopted from Tindal C.J. in R. v. Frost (1839) 4 St Tr (NS), at p 386. It is a matter of practice and procedure, and in such matters, even where the procedure is criminal and is directed to safeguarding the position of the accused, there is less reason for closely following English authority than where the development of the substantive law is involved. It is, for example, difficult to apply the rule where the jury ask for the recall of a witness or further proof or disproof of a fact. It seems to us unsafe to adopt a rigid formula in view of the almost infinite variety of difficulties that may arise at a criminal trial. It is probably enough to say that the occasion must be very special or exceptional to warrant a departure from the principle that the prosecution must offer all its proofs during the progress of the Crown case and before the prisoner is called upon for his defence. The argument that ss. 430, 431 and 448 of the Crimes Act 1928 (Vict.) combine to make it impossible for the prosecution to call further evidence after the prisoner has entered upon his defence, or at all events concluded it, is, we think, unsound. But the policy of these provisions strengthens the principle which makes a departure from the rule allowable only in exceptional circumstances. Further, although we have not thought it proper to adopt the formula of Sir Nicholas Tindal, the English cases make it plain enough that generally speaking an occasion will not suffice for allowing an exceptional course if it ought reasonably to have been foreseen. Again, it may be pointed out that even an unexpected occasion may be of such a nature that it would have been covered, had the Crown case been fully and strictly proved. (at p380)

14. The application of these principles ought, we think, to have resulted in a refusal of the Crown Prosecutor's application to recall the seven officers of police. If the prosecution relied upon the fact, as an incriminating circumstance, that the prisoner used the word "throttle" at a time when only the guilty man could be aware how the dead girl was killed it was necessary as part of the Crown case first to prove that he used the word and then to exclude the reasonable possibility of his having become aware or of his supposing that she was throttled because of anything he saw or heard. Whether he admitted or denied the use of the word it had no incriminatory effect unless the possibility of his deriving knowledge or belief that she was throttled from any other source was reasonably disproved. No doubt it seemed hardly worth while embarking upon an attempt to make full proof of the fact that the prisoner could not have thought Sylvia Holmes was so killed unless he killed her, until the prisoner admitted that he used the word. It may be conceded that the cross-examination of the witness deposing to the word "throttle" confirmed this view. But that did not alter the necessity, if reliance was to be placed upon the use of the word, of proving the whole of the subsidiary issue. The situation was one in which the prosecutor very naturally sought to exploit an unexpected advantage arising from the prisoner's admission. We think that consistently with principle he ought not to have been allowed to do so. (at p381)

15. No doubt a discretionary judgment must be exercised by a presiding judge and he has many advantages in its exercise denied to a court of appeal. But we think that the present case was not within the principle which was applicable. (at p381)

16. There remains the question whether this is a proper case for the grant of special leave to appeal. After some consideration we have arrived at the conclusion that an affirmative answer should be given to this question. It possesses a "special" character within the rule stated in In re Eather v. The King [1915] HCA 86; (1915) 20 CLR 147 , because we think in a case where the issues were, not simply whether the prisoner was the man at whose hands the young woman died, but, in addition, if so, whether his crime was murder or manslaughter, the recalling of the seven officers of police at the end of the case, which we think was wrong, and the presence in the case of so much inadmissible evidence was very likely to operate unfavourably to the prisoner. The prosecution must take the responsibility for leading the inadmissible evidence. The failure of the prisoner's counsel to object could not make it admissible, but it was calculated to place the presiding judge in a difficult and embarrassing position. For he could not be expected to see in advance that the evidence would never be made relevant. Moreover, the failure of counsel to object to inadmissible evidence may provide ground for an inference that the reception of the evidence can give rise to no substantial miscarriage of justice. In the present case, however, we think that the elements to which we have referred combine to make it proper to order a new trial. (at p381)

17. We grant special leave to appeal, treat the appeal as instituted and heard instanter, allow the appeal, set aside the refusal of leave to appeal by the Supreme Court and in lieu thereof grant such leave, set aside the conviction and sentence and order a new trial. (at p382)

FULLAGAR J. I have felt, in the course of considering this case, some doubt as to whether this Court ought to interfere, but I have come to the conclusion that special leave to appeal ought to be granted. The Court has not adhered to the very strict rule laid down in Eather v. The King [1914] HCA 77; (1914) 19 CLR 409, and the dissenting judgment of Isaacs J. in that case may be said to have prevailed almost from the outset. The discretion is "unfettered", though it is necessary for an applicant to make "a prima-facie case showing special circumstances": see In re Eather v. The King [1915] HCA 86; (1915) 20 CLR 147. In the present case the crime of which the applicant was convicted is a capital crime. I am clearly of opinion that the Crown ought not to have been permitted, after the close of the case for the defence, to adduce the further evidence which it did adduce. Further, I think it highly probable that the learned trial Judge, in admitting that evidence, not merely exercised his discretion wrongly but exercised it without having present to his mind certain considerations which (whether one accepts recent English authority with or without qualification) undoubtedly demanded attention. The applicant was entitled to have those considerations weighed, and, in a case of this kind, it can make no difference that he was represented by a barrister who ought to have drawn his Honour's attention to them. And, while I do not myself think it very likely that the evidence in question made any difference to the result of the trial, I cannot feel satisfied that it did not carry substantial weight with the jury. When this position is reached, it seems to me that, to adopt the words of Isaacs J. in Eather v. The King (1914) 19 CLR, at p 428 , "the cause of justice is better served by granting than by refusing leave to appeal". And, in the circumstances of this case, it almost necessarily follows that the appeal must be allowed. (at p382)

2. I should not have thought it right to grant special leave on the ground that evidence was wrongly admitted, although I am clearly of opinion that the evidence of Kathleen Adams was inadmissible, and although I think that other evidence was admitted which went to show no more than that the applicant was a man of violent disposition. If objection had been taken to the evidence of Adams and to this other evidence, I feel sure that his Honour would have rejected it. On the other hand, I do not think that any of the evidence given by Ada Adkins was inadmissible, and in any case I am unable to regard any of the evidence in question as having had any important effect on the issue of the trial. On what I regard as the real question in the case, the question which was the subject of the dissenting judgment of Smith J., I have little to add to what has already been said. (at p383)

3. I entirely agree that the language which Tindal C.J. is reported in the State Trials as having used in R. v. Frost (1839) 4 St Tr (NS), at p 386 is altogether too strict and rigid, and ought not to be accepted as a formula. In the first place, I would regard the authority of the statement so often quoted as extremely doubtful. The words "ex improviso" and the words "which no human ingenuity can foresee" convey radically different meanings, the latter being in no sense a translation of the former. The latter words do not occur in the report in Carrington & Payne (1839) 9 Car & P 129, at p 159 [1839] EngR 1190; (173 ER 771, at p 784), and they have, to my mind, the almost unmistakable character of a "gloss" in the correct sense of that much abused word - a note by a misguided commentator, which has by some mischance become incorporated in the text. Moreover, the rule is stated as applying equally to criminal and to civil cases, and I should not imagine that anybody would suppose that the question should be approached from the same point of view in both classes of case. In the second place, the rule so stated is on its face fundamentally unsound and calculated not to aid but to impede the administration of justice. After all, the aim of legal proceedings, including criminal proceedings, is supposed to be to elicit the truth so far as human imperfection permits. I cannot feel the slightest doubt that the course taken by Cussen J. in R. v. Collins [1907] VicLawRp 55; (1907) VLR 292 was entirely correct and proper. (at p383)

4. These things having been said, however, it remains true and important that the Crown should be permitted to adduce evidence after the close of the case for the defence only in exceptional circumstances and when it is reasonably clear that the accused will not be unfairly prejudiced by the admission of the evidence. A wide discretion must be conceded to the judge presiding at the trial, but it should be regarded as limited in that way, and it is to be remembered that the practical effect of evidence on the minds of a jury may differ according as the evidence is adduced in chief or by way of replication. But the discretion ought not to be regarded as further limited or as governed by any rigid rule or formula. I would add that it is by no means clear to me that the same considerations are appropriate when the Crown proposes to call further evidence as are appropriate when the judge is thinking of calling a witness himself. It is impossible to foresee and provide a priori for the infinite variety of circumstances in which either question may arise. (at p384)

5. To the extent indicated above I would differ, with respect, from Smith J. But with his analysis of the situation in this particular case I am in complete agreement, and, to my mind, it follows from that analysis - even if one refuses to accept the State Trials formula - that the evidence in question ought not to have been allowed to be given. It is tempting, of course, to say that it could not make much difference whether the evidence were given in chief or by way of replication. But it might have made a great difference. There were, as Smith J. points out, reasons why the Crown should refrain from leading the evidence in question in the normal course of its case. The point which the Crown was seeking to make by it was a subordinate point, and - in view of the difference between the evidence of Detective Witham and that of Constable Nailon - it rested on a very weak foundation. To seek by witness after witness to exclude all possibility of an innocent explanation of words, the uttering of which would probably be denied, may well have seemed to involve the tactical disadvantage which often results from the magnification of a comparatively unimportant issue. When the accused (probably "ex improviso" of everybody concerned in the case) admitted that he had used the word "throttled", the position changed, to my mind, more in appearance than in reality. In view of the conflicting recollections of Constable Nailon and Detective Witham, it could have been put with great force to the jury that it would be very unsafe to find that the word "throttled" had been used. When the accused admitted that he had used the word, the conflict between Nailon and Witham may have seemed to be resolved in favour of Witham. But actually it might well have been thought still very unsafe to rely even on the accused's recollection of the use of a particular word nearly eighteen months before in circumstances of high excitement and tension, although the Crown could have quite fairly and properly relied on the "admission" and put to the jury the general improbability of his explanation of his use of the word, commenting on the cross-examination of Witham. When, however, a whole series of witnesses was called with a view to excluding or minimising the possibility of the truth of the accused's explanation, there was danger that, in the eyes of the jury, a comparatively unimportant issue might be magnified out of all proportion and that altogether undue weight might be attached to evidence which, on any view, ought to have been scanned with caution. Nor was this danger mitigated when the learned judge, towards the end of his charge to the jury, referred to the difference between Nailon's version and Witham's version as a "slight" difference. The verbal difference was indeed slight, but the difference in significance was the difference between the innocuous and the possibly damning. (at p385)

Order that special leave to appeal be granted and that the appeal be treated as instituted and heard instanter. Allow the appeal. Set aside the order of the Supreme Court refusing leave to appeal and in lieu thereof grant such leave to appeal and set aside the conviction and sentence and order a new trial.