"Both those matters were essentially questions for

the jury. They depended on an assessment of all

the evidence, and in particular an assessment or

evaluation of the credibility of the witnesses who

gave it. Included amongst them was the accused

himself who gave evidence from the witness-box.

In all the circumstances, it may be said that the

Crown case against the accused was not a

particularly strong one. It did depend, however,

very much upon the impression of witnesses formed

by the jury in the course of the trial. That being

so, it is quite clear to my mind that, there being

evidence in (the) case on which the jury might

reasonably find a verdict of guilty, it cannot

possibly be said that they must have entertained a

reasonable doubt about the guilt of the accused.

On that footing, and having regard to what was said

by the High Court in the case of Chamberlain v.

The Queen ... it seems to me that it is simply not

open to this Court to set aside the jury verdict on

the ground that it was one to which the jury could

not reasonably have come."

The Court of Criminal Appeal accordingly dismissed the appeal against conviction as well as the application for leave to appeal against sentence.

15. The principal ground of appeal which the applicant seeks to raise in this Court is that the Court of Criminal Appeal was in error and should have concluded that the verdict was unreasonable or could not be supported having regard to the evidence. The scope of this ground of appeal was discussed in the joint judgment of Gibbs C.J. and myself in Chamberlain v. The Queen (No. 2) [1984] HCA 7; (1984) 153 CLR 521, at pp 530-534. It is unnecessary to recount the long line of cases in which the statutory provisions have been interpreted. It is sufficient to say that it is now well settled that a verdict may be set aside as unsafe and unsatisfactory notwithstanding that there was, as a matter of law, evidence upon which the accused could have been convicted: Whitehorn v. The Queen [1983] HCA 42; (1983) 152 CLR 657, at pp 660, 686; Chamberlain, at pp 532, 604 et seq., 618 et seq. In Chamberlain Gibbs C.J. and I, after noting that it was unnecessary to consider whether the jurisdiction exercised by the Court of Criminal Appeal in Australia is precisely the same as that exercised by the Court of Criminal Appeal in England, said (at p.534):

"... the proper test to be applied in Australia is,

as Dawson J. said, to ask whether the jury, acting

reasonably, must have entertained a sufficient

doubt to have entitled the accused to an acquittal,

i.e. must have entertained a reasonable doubt as to

the guilt of the accused. To say that the Court of

Criminal Appeal thinks that it was unsafe or

dangerous to convict, is another way of saying that

the Court of Criminal Appeal thinks that a

reasonable jury should have entertained such a

doubt. The function which the Court of Appeal

performs in making an independent assessment of the

evidence is performed for the purpose of deciding

that question. The responsibility of deciding upon

the verdict, whether of conviction or acquittal,

lies with the jury and we can see no justification,

in the absence of express statutory provisions

leading to a different result, for an appellate

tribunal to usurp the function of the jury and

disturb a verdict of conviction simply because it

disagrees with the jury's conclusion. We do agree

that in many cases the distinction will be of no

practical consequence; it will be merely a matter

of words. That will not generally be the case

where questions of credibility are decisive.

However, whether it matters from a practical point

of view or not in a particular case, it is not

unimportant to observe the distinction - the trial

is by jury, and (absent other sources of error)

the jury's verdict should not be interfered with

unless the Court of Criminal Appeal concludes that

a reasonable jury ought to have had a reasonable

doubt."

16. In performing the function which is discussed in the passage just quoted the Court of Criminal Appeal is deciding a question of fact. So much clearly appears from the judgment of the Court (Dixon C.J., Fullagar and Taylor JJ.) in Raspor v. The Queen [1958] HCA 30; (1958) 99 CLR 346, at p 350; Hocking v. Bell [1945] HCA 16; (1945) 71 CLR 430, at p 497. When "the court performs this duty, it is not deciding a question of law; it is supervising or reviewing the findings of a tribunal of fact", to use the words of Dixon J. in Darling Island Stevedoring & Lighterage Co. Ltd. v. Jacobsen [1945] HCA 22; (1945) 70 CLR 635, at p 643.

17. Accordingly, if all that the applicant asks us to do in the present case is to re-exercise the function of the Court of Criminal Appeal in reviewing the verdict of guilty, I would not grant special leave to appeal on this ground simply because it involves no more than a question of fact. However, it seems to me that there is another and important question at stake. That is the question whether the Court of Criminal Appeal performed its duty to make an independent assessment of the evidence as required by Chamberlain.

18. It will be recalled that McPherson J. specifically referred to the matter of the reliability of the admission made by the applicant to Mrs Bargo and concluded that it was a matter essentially for the jury, observing that it depended on an assessment of all the evidence, including the credibility of witnesses, the applicant being one of them. Although his Honour thought that the Crown case was not particularly strong, his opinion was that there was evidence on which the jury might reasonably find a verdict of guilty.

19. Putting to one side the admission made to Mrs Bargo, there was no direct evidence that the applicant started the fire. And apart from the evidence of Doniger there was no direct evidence that the applicant was the man who poured methylated spirits over the deceased.

20. The trial judge in his directions to the jury pointed out that counsel for the Crown had said "Well, we are not really relying upon his evidence to any extent." The trial judge again drew attention to this statement later in his summing-up. His Honour warned the jury that Doniger was a witness with "the obvious problems of credibility". However, his Honour told the jury that they must examine the evidence and decide whether it was so unreliable that it should be disregarded or whether it might provide some assistance in reaching a verdict. At the same time the jury was told that they should not convict unless they accepted the evidence of the admission to Mrs Bargo and that it was reliable. Counsel for the applicant challenged the sufficiency of the directions given in relation to Doniger's evidence. For the reasons given by Deane, Toohey and Gaudron JJ. there is no substance in this challenge.

21. The consequence is that the conviction depends upon the admission made by the applicant to Mrs Bargo. The jury could without any difficulty accept Mrs Bargo's evidence that the admission was made to her. The reliability of the admission was then the critical matter. The Court of Criminal Appeal did not examine this matter in detail, contenting itself with McPherson J.'s comment that it was essentially a matter for the jury, involving as it did the question of credibility.

22. In failing to direct its attention specifically to an examination of the reliability of the admission, the Court of Criminal Appeal did not carry out an independent assessment of the evidence which it was required to do in dealing with the ground of appeal advanced by the applicant. The making of a careful independent assessment was essential to the making of an informed judgment on the question whether the jury could reasonably convict on the materials before them. The Court's duty was to satisfy itself that there was "a sufficiency of legal evidence to satisfy reasonable men to the exclusion of any reasonable doubt", in the words of Dixon J. in McKay v. The King [1935] HCA 70; (1935) 54 CLR 1, at p 9. The performance of that duty extends to an examination of the probative value of a confession, notwithstanding that there is sufficient confessional evidence to be submitted to the jury. In McKay Dixon J. went on to say (at p.10):

"Even if confessional evidence might appear

sufficient to submit to a jury, yet a conviction

would doubtless be quashed if it appeared that the

jury had been allowed or encouraged to act upon

views of it which are unsafe. It is conceivable

that a direction to a jury that they might convict,

although they were unable to find confirmatory

evidence, or to accept it, might in some

circumstances have this result."

Evatt J. agreed with the reasons of Dixon J. (p.10). See also Wright v. The Queen (1977) 15 ALR 305 and note the discussion: (1977) 51 ALJR 779.

23. Merely to say that it was a matter for the jury is to assert without demonstration that the question must be answered in the affirmative. And the glancing reference to the applicant's credibility as a witness seems to assume that the jury could have regard to the applicant's demeanour as a witness as an additional reason for accepting Mrs Bargo's opinion that the applicant's admission would be reliable in preference to the opinion of the psychiatrist. It is very likely that the jury was much impressed by the clarity of the applicant's evidence and his ease in understanding and dealing with the questions put to him as a witness. And the jury may have thought that his responses in the intelligence and memory test were not entirely genuine. But the applicant had been deprived of alcohol for eight months preceding the trial and, according to Dr Quinn, his mental condition was improved at the trial, a matter to which the trial judge had directed the jury's attention. In this situation the applicant's apparent mental state at the time of the trial was not of significant value in determining whether he was or might be confabulating when the fatal incident took place.

24. The discussion by Dixon J. of the reliability of an admission made by a schizophrenic in Sinclair v. The King [1946] HCA 55; (1946) 73 CLR 316 is instructive. Sinclair was convicted of murder on the basis of confessions made by him. The medical evidence at the trial was that he was a schizophrenic and that there was a real risk that on a particular occasion such a man as Sinclair might fail to distinguish fact from fantasy and that he might construct and relate an imaginative account of something that had never really happened. The evidence did not show that he had a mind so disordered and irresponsible that it would be dangerous to pay any attention whatever to what he said. His counsel submitted that his confessions should have been excluded on the ground that they were inadmissible. It was held that they were admissible and that once admitted it was for the jury to determine the weight, if any, to be given to them. It was not a case in which the jury verdict was challenged on the ground that the verdict was unreasonable and could not be supported having regard to the evidence.

"Boyd Sinclair's mental state did not disable him

from observing, appreciating, recollecting and

recounting real occurrences, events or experiences.

The fact that his mind, in its schizophrenic state,

may have been stored with imaginary episodes ...

would ... make it impossible to place reliance upon

his confessional statements as intrinsically likely

to be true ... But it is to be noticed that his

condition did no more than make it possible that

the source of any confessional statement made, lay

in these tendencies. His was not a case in which

it could be said that the higher probability was in

favour of his confession of such a crime being the

product of imagination. Reason suggests that in

such circumstances it is for the tribunal of fact

to ascertain or verify the factual basis of the

statements of a man in such a mental condition ..."

His Honour went on to point out that the external facts in the case independently proved supplied many reasons for supposing that the confessional statements were substantially correct.

26. In the present case the admission made to Mrs Bargo was rightly admitted in evidence. Its weight was a matter for the jury. In deciding what weight should be given to the admission, they had to choose between the opinion of Mrs Bargo that in the condition in which he appeared to her to be his statement could be reliable and the opinion of Dr Quinn that there was a distinct possibility that he might be confabulating. Unlike the jury in Sinclair, the jury did not have the advantage of measuring the reliability of the admission by reference to independently established facts indicative of the truth of the matter admitted.

27. In these circumstances I have reached the conclusion that a reasonable jury could not have been satisfied beyond reasonable doubt that the admission was reliable and, accordingly, of the applicant's guilt. The applicant's demeanour as a witness would in other circumstances provide a sound basis for preferring Mrs Bargo's opinion to that of Dr Quinn on the applicant's propensity to confabulate. But here the probability of an improvement in his condition whilst in custody undermines the value of his demeanour as a witness.

28. I am not persuaded that the Court of Criminal Appeal recognized that the function which it was performing required it to make a careful and independent assessment of the critical evidence, namely the weight to be given to the admission made to Mrs Bargo and that the performance of this function entailed examining the reliability of the admission made by a person whose reliability of perception, recollection and recounting was very much in question. It seems to me that the Court of Criminal Appeal was under some misapprehension as to the task which it had to perform in assessing the qualitative worth of the admission or, if not, that it failed to perform that task.

29. Accordingly, although in the ultimate analysis the application is one which calls on this Court to decide a question of fact, the failure of the Court of Criminal Appeal to carry out its function in this respect warrants the grant of special leave to appeal.

30. I would grant the application for special leave to appeal, allow the appeal and set aside the conviction.

DEANE, TOOHEY AND GAUDRON JJ. Mervyn Joseph Morris ("the applicant") seeks special leave to appeal from a decision of the Supreme Court of Queensland, sitting as a Court of Criminal Appeal, dismissing his appeal from convictions for the manslaughter of Lionel Blades and Douglas Campbell ("the deceased persons").

2. The applicant was charged with having murdered the deceased persons, the prosecution case being that on 30 September 1985 at approximately 9.40 a.m., in the vicinity of 31 Merivale Street, South Brisbane, the applicant threw methylated spirits on the deceased persons and set fire to them, causing injuries from which they later died. At the request of the trial judge the jury indicated that verdicts of manslaughter were returned on the ground of diminished responsibility on the part of the accused.

3. The grounds upon which the applicant relies raise, in essence, two issues, viz. whether the directions given by the learned trial judge in relation to the evidence of a witness, Joseph Doniger, were adequate in the circumstances of the case and, whether the whole of the evidence was such that the jury must or should have entertained a reasonable doubt as to the applicant's guilt.

4. The applicant is a chronic alcoholic. He gave evidence that for many years his preferred drink has been methylated spirits. For a number of years he has been a resident of the St. Vincent de Paul Hostel in South Brisbane, as were the deceased persons and the witness Joseph Doniger.

5. The central issue in the trial was whether it was the applicant who did the acts which caused the injuries sustained by the deceased persons on 30 September 1985. The evidence adduced by the prosecution relevant to this issue comprised the testimony of four persons who deposed to certain events in the vicinity of 31 Merivale Street before and after the clothing of the deceased persons was ignited, the testimony of Joseph Doniger, and evidence of an admission made by the applicant to Mrs Florence Bargo, a welfare officer employed at the St. Vincent de Paul Hostel. The Crown also led evidence of the applicant's interview with police after he was taken into custody at approximately twelve noon on the day of the incident. In the record of interview taken between 2.26 p.m. and 4.14 p.m. that same day, the applicant denied being present at the place where the deceased were burnt and stated he "heard about this ... when Flo (Mrs Florence Bargo) told me".

6. There is much common ground in the evidence of the four persons who deposed as to events in the vicinity of 31 Merivale Street. Mr John Carson, who at the time worked at 27 Merivale Street, gave evidence that at 9.10 a.m. he observed four men drinking near 31 Merivale Street. Shortly after 9.30 a.m. he left the building in which he worked, and again observed four men. The clothing of two of those men was in flames. He rendered assistance, and whilst so doing observed the other two men to remain in the vicinity. He identified Joseph Doniger as one of those men. He failed to identify the applicant as the fourth person.

7. Mrs Kerri Guthrie, Mr Geoffrey Fergus and Mr Henry St. Ledger also gave evidence as to four men being present in the laneway at about the time of the events described by Mr Carson. None identified the applicant as being present, but Mrs Guthrie deposed to having seen a man pouring clear liquid over another man whilst she was engaged in parking her car in the area. She described the person pouring liquid as "slight to medium build, he appeared to be short and he had dark hair", a description apparently consistent with the appearance of the applicant. Some little time later, when she had parked the car, she observed the deceased persons to be on fire. Mr Fergus gave evidence as to a man, who was not the accused, "sloshing" fluid from a bottle across the legs of two other men sitting in the laneway.

8. The evidence of Joseph Doniger was both confused and confusing. His evidence in chief was that he and a companion, Morton Howell, walked past the applicant and the deceased persons in Merivale Street and, after he had gone 100 yards, looked back and saw a burst of flame. Mr Howell had died before the trial, and the deposition of his evidence at the committal proceedings was read to the jury. That evidence went to events which occurred prior to 9.00 a.m. On the prosecutor's application, Mr Doniger was declared a hostile witness and under cross-examination by the prosecutor gave evidence of former inconsistent statements. He admitted that he had previously stated to the police that he had been with the applicant and the deceased persons in Merivale Street, when the applicant "started to toss the metho around over the other blokes who were sitting down", whereupon he had walked some 5 yards away, noticed that Douglas Campbell was on fire, and had returned to render assistance. He also admitted that he had previously given evidence in the Magistrate's Court in which he deposed to having seen the applicant in the company of the deceased persons and having observed the applicant "toss metho around", whereupon he walked off (apparently in company with someone) and, after proceeding some distance, turned around and saw the applicant run away and throw a bottle. During the course of this cross-examination he also deposed to the truth of the various matters contained in his prior statements which were orally put to him. Cross-examined by defence counsel, Joseph Doniger gave several versions of his movements on the day including that, although he had not seen what happened, he had seen the men lying on the ground injured and had gone over to see what was happening.

9. None of the evidence above related touched the question central to the prosecution case whether the applicant set fire to the deceased persons. The evidence was to the effect that the pouring of the liquid did not result in an instantaneous conflagration. Any one of the four men present might have sparked the subsequent flames. The only evidence to connect the applicant with the flames was that of Mrs Florence Bargo. She testified that she spoke to the applicant at approximately 10.00 a.m., and that he said "I think I have killed my two mates. I threw metho over them, struck a match and burnt them."

10. It was submitted that specific directions should have been given by the trial judge in relation to the evidence of Mr Doniger by reference to the matters specified in s.102 of the Evidence Act 1977 (Q.). By s.101 of that Act a previous inconsistent statement of a witness is rendered admissible as evidence of any fact stated therein of which direct oral evidence by the witness would be admissible. By s.102, in estimating the weight, if any, to be given to any statement rendered admissible by, inter alia, s.101, regard is to be had to the circumstances from which an inference can reasonably be drawn as to its accuracy, including the contemporaneity of the statement with the occurrence of facts to which it relates and any incentive the maker might have had to conceal or misrepresent the facts.

11. The receipt into evidence of a prior inconsistent statement as evidence of the facts therein contained poses particular problems in a criminal trial, especially where the prior inconsistent statement is more damaging to an accused person than is the testimony of the witness. Where the prosecution seeks to adduce such evidence from a prosecution witness, an issue may well arise as to whether the prejudicial nature of the statement does not outweigh its probative value, such that as a matter of judicial discretion it should be excluded: see Harris v. Director of Public Prosecutions (1952) AC 694, at p 707; Kuruma v. The Queen (1955) AC 197, at p 204; Driscoll v. The Queen [1977] HCA 43; (1977) 137 CLR 517, at p 541 and Cleland v. The Queen [1982] HCA 67; (1982) 151 CLR 1. If however, such a statement is admitted, it will usually be necessary for the trial judge to give very careful and very precise instructions to a jury as to the weight the evidence should be given. The nature of the instructions will necessarily depend on the particular case. It is difficult to conceive that in a case where the prior inconsistent statement is more damaging to the accused person than the evidence given by the witness, a mere invitation to the jury to consider the matters referred to in s.102 of the Evidence Act would be a sufficient instruction. In many cases such an invitation may be to the disadvantage of the defence case. However, in our view, these issues do not arise in the present case. The prior inconsistent statements were not received into evidence pursuant to s.101 of the Evidence Act. As noted above, their contents actually became part of the oral evidence of the witness and as such the evidence of Mr Doniger fell for consideration in accordance with ordinary principles, which, it was further submitted, should have resulted in a direction that the evidence be disregarded.

12. The question of appropriate directions to be given by the trial judge to a jury in relation to the evidence of a witness who has made a prior inconsistent statement was considered in Driscoll. In that case Gibbs J., speaking on this issue for the whole Court, stated at p.536:

"In some cases the circumstances might be such that

it would be highly desirable, if not necessary, for

the judge to warn the jury against accepting the

evidence of the witness. From the point of view of

the accused this warning would be particularly

necessary when the testimony of the witness was

more damaging to the accused than the previous

statement. In some cases the unreliability of the

witness might be so obvious as to make a warning on

the subject almost superfluous. It is possible to

conceive other cases in which the evidence given by

a witness might be regarded as reliable

notwithstanding that he had made an earlier

statement inconsistent with his testimony."

His Honour went on (at p.537) to reject a suggestion in the judgment of the Court of Criminal Appeal in Reg. v. Golder, Jones and Porritt (1960) 45 Cr App R 5, at p 11, that "there (was) an inflexible rule of law or practice that the jury should be directed that the evidence should be regarded as unreliable". In the present case the evidence of Mr Doniger was unsatisfactory, not only by reason of his prior inconsistent statements, but by reason of the inconsistent versions to which he deposed.

13. It must have been obvious to the jury that the evidence of the witness Doniger was inherently unreliable. It was warned as to "the obvious problems of credibility" and twice reminded that the Crown was not relying upon his evidence. The jury was told that it was not free to ignore the evidence, but was required to look at it and thereafter decide whether it was so unreliable that it should be put out of mind or whether some part of it might provide assistance in reaching a correct verdict. These directions were given in a context in which it was made clear that a conviction should ensue only if the jury accepted the evidence of Mrs Bargo as to the applicant's admission, and further accepted that such admission was reliable. Given these matters, we are not persuaded that there is substance in the submissions directed to establishing the inadequacy of the warnings given by the trial judge in relation to the evidence of Mr Doniger.

14. Although confessional material is rightly regarded as evidence of a most persuasive nature, it has long been recognized that there are occasions when such evidence cannot be relied upon. The exclusion from evidence of confessional statements not shown to have been voluntarily made was, as noted by Dawson J. in Cleland (at p.27), for the reason "at least in its origins, because such statements were unreliable as evidence". Also, as was conceded by Dixon J. in Sinclair v. The King [1946] HCA 55; (1946) 73 CLR 316, at p 338, confessional material might properly be rejected if "made by a person whose unsoundness of mind is such that no account ought to be taken of his self-incriminating statements for any evidentiary purpose as proof of the criminal acts alleged against him". It is neither clear on existing authority nor relevant to the present case to decide whether the rejection of confessional statements in such circumstances is through the operation of an exclusionary rule or depends upon the exercise of the trial judge's undoubted discretion to exclude confessional statements when it would be unfair to the accused to admit them. However, even if the confessional material is properly to be admitted, it may be that when it is viewed in the light of all the evidence before the court it would be unsafe to rely upon this material to ground a verdict. So much is evident in the judgment of Dixon J. in Sinclair (see p.338) and his judgment, with which Evatt J. agreed, in McKay v. The King [1935] HCA 70; (1935) 54 CLR 1, at pp 9-10. In Wright v. The Queen (1977) 15 ALR 305 this Court held that the jury's verdict, based on a confession by the applicant, was unsafe and the conviction was quashed. Whitehorn v. The Queen [1983] HCA 42; (1983) 152 CLR 657 provides another, and more recent example, of a verdict which was unsafe by reason that the only evidence to support it was a confessional statement made in circumstances which brought the reliability of the statement into question.

15. In the present case no objection was made at the trial to the admission of Mrs Bargo's evidence of the accused's confessional statement. However, the reliability of the confessional material was put in issue by evidence of the applicant's intelligence, which was close to borderline mental defective, his defective memory and brain damage, including frontal lobe damage, as a result of his severe alcoholism. There was evidence from Mrs Bargo that the applicant had been drinking prior to her conversation with him. The applicant also testified as to his consumption of methylated spirits on the morning in question. Evidence was called from Dr Quinn, a psychiatrist with extensive experience of the effects of alcoholism. Dr Quinn assessed the alcohol content of the quantity of methylated spirits which the applicant deposed to drinking on the morning in question as equivalent to that contained in a bottle of scotch. He gave evidence that that intake would result in a state of intoxication such that it was likely that the applicant would be prone to confabulation. Confabulation is a process whereby defective recall, a condition which may be experienced by severe alcoholics, is hidden by the imparting of modified information, including information based in delusion. Information so imparted may be based entirely on fantasy, although it may on occasions have a factual basis, even though not accurate in the circumstances. It is in the context of this evidence, and evidence that when interviewed by the investigating police on the afternoon of 30 September the applicant denied any involvement in the events causing the injuries to the deceased persons, that it is necessary to consider the decision of the Court of Criminal Appeal that it could not be said that the jury must have entertained a reasonable doubt about the guilt of the applicant.

16. By s.668E of the Criminal Code (Q.) the Court of Criminal Appeal is required to allow an appeal against conviction if it is of the opinion that the verdict of the jury is unreasonable, or cannot be supported by the evidence, or that the judgment of the court of trial should be set aside on the ground of the wrong decision of any question of law, or that on any ground whatsoever there was a miscarriage of justice, subject to the proviso that it may dismiss an appeal if it considers that no substantial miscarriage of justice has actually occurred.

17. Under a provision such as s.668E of the Criminal Code, a Court of Criminal Appeal may set aside a verdict that is unsafe or unsatisfactory notwithstanding that there is, as a matter of law, sufficient evidence to support it: Chamberlain v. The Queen (No.2) [1984] HCA 7; (1984) 153 CLR 521, at pp 532-534 and pp 605-606. See also Raspor v. The Queen [1958] HCA 30; (1958) 99 CLR 346, at pp 350-352; Plomp v. The Queen [1963] HCA 44; (1963) 110 CLR 234, at pp 244 and 250 and Whitehorn.

18. In Whitehorn Dawson J. (at p.686), with whom Gibbs C.J. and Brennan J. agreed, expressed the relevant function of a Court of Criminal Appeal when an issue is raised as to whether a verdict is unsafe or unsatisfactory as deciding "whether the appellate court thinks that upon the whole of the evidence it was open to the jury to be satisfied beyond reasonable doubt that the accused was guilty". In Chamberlain (No.2) that test was accepted as correct by Gibbs C.J. and Mason J. (at p.534) and by Brennan J. (at p.607). Gibbs C.J. and Mason J. said at p.534: "To say that the Court of Criminal Appeal thinks that it was unsafe or dangerous to convict, is another way of saying that the Court of Criminal Appeal thinks that a reasonable jury should have entertained such a doubt". Although of no practical effect in the present case, the Court of Criminal Appeal, in adopting the formulation of Gibbs C.J. and Mason J., incorrectly substituted the word "would" for the word "should" in the cited passage and, perhaps, thereby suggested a somewhat stricter test.

19. The test as expressed in Chamberlain (No.2) by Gibbs C.J. and Mason J., suggests that a verdict which in the view of a Court of Criminal Appeal is unsafe or unsatisfactory is necessarily a verdict that is unreasonable, or is not supported, at least upon the requisite standard of proof, by the evidence. For our part, we would think that there might be verdicts falling within the concept of miscarriage of justice, as that expression is used in the common criminal appeal provisions, by reason of some defect or weakness of the evidence even though on the evidence it was open to the jury to be satisfied of guilt beyond reasonable doubt, as, for example, where there is some feature of the evidence which raises a substantial possibility that the jury may have been mistaken or misled: see Davies and Cody v. The King [1937] HCA 27; (1937) 57 CLR 170, at p 180. Whether or not this be so, it is clear that the question whether a verdict is unsafe or unsatisfactory involves a Court of Criminal Appeal undertaking an independent examination of the relevant evidence to determine whether it was open to the jury to be satisfied beyond reasonable doubt as to the guilt of the accused. That function is not discharged merely by a consideration of whether there was a sufficiency of evidence to sustain a conviction, for it is clear that a verdict may be unsafe or unsatisfactory notwithstanding that there was evidence sufficient to entitle a reasonable jury to convict. So much is clear from the following passage in the joint judgment of Gibbs C.J. and Mason J. in Chamberlain (No.2) at p.531:

"In Raspor v. The Queen and Plomp v. The Queen,

it was recognized that a court of criminal appeal

may interfere with a verdict which is unsafe or

unsatisfactory even if there is sufficient evidence

to support it as a matter of law, and even though

there has been no misdirection, erroneous reception

or rejection of evidence, and no other complaint as

to the course of the trial. In other words, even

if there is some evidence on which a reasonable

jury might be entitled to convict, a Court of

Criminal Appeal has the responsibility to consider

whether 'none the less it would be dangerous in all

the circumstances to allow the verdict of guilty to

stand': Hayes v. The Queen."

20. A Court of Criminal Appeal must make an independent assessment of the evidence, both as to its sufficiency and its quality. In the present case, the Court of Criminal Appeal held, correctly, that the evidence of Mrs Bargo "was plainly sufficient to justify a conclusion on the part of the jury that the accused was responsible for the act of throwing the methylated spirits on the deceased and setting them alight." However, there was a further obligation upon the Court of Criminal Appeal to assess the reliability of the admission to which Mrs Bargo deposed. On that issue the Court of Criminal Appeal noted that the evidence of Dr Quinn, to which reference has already been made, was relevant to the intention of the accused and the reliability of his admission to Mrs Bargo, but stated that "those matters were essentially questions for the jury. They depended on an assessment of all the evidence, and in particular an assessment or evaluation of the credibility of the witnesses who gave it."

21. In treating the question of the reliability of the admission made as one of credibility to be determined by the jury, the Court of Criminal Appeal failed to perform an independent assessment of the only evidence which linked the applicant with setting fire to the deceased persons. That evidence fell for assessment in the light of a single body of evidence, relevant not only to the issue of the reliability of any admission made to Mrs Bargo but also to the question of diminished responsibility on the part of the applicant, which the jury found was established.

22. Although it was open to the jury to accept so much of the evidence as went to the question of diminished responsibility and to reject the evidence going to the likelihood of the applicant having engaged in the process of confabulation when speaking to Mrs Bargo, the question was not simply a question of the credibility of witnesses, or even a question of the credibility of Dr Quinn. It was a question of the reliability of an admission, uncorroborated by any other evidence as to the applicant's setting fire to the deceased persons. That reliability had to be assessed in the light of the applicant having consumed a quantity of alcohol prior to the making of the admission, in the light of his mental capacities, and in the light of his impaired brain function. It also had to be assessed in the context of his later denial of any involvement in the incident when interviewed by police that afternoon. Even apart from Dr Quinn's evidence as to the likelihood of the applicant having engaged in the process of confabulation, these were matters which rendered it doubtful that any confessional statement made by the applicant was, in the words of Dixon J. in Sinclair (at p.338), "intrinsically likely to be true". In the absence of any other evidence to connect the applicant with setting fire to the deceased persons, these matters, when considered in the context of the prosecution's obligation to prove beyond reasonable doubt that the applicant set fire to the deceased persons, should have caused the jury to entertain a reasonable doubt as to the reliability of any admission made to Mrs Bargo and accordingly as to the applicant's guilt.

23. Because the Court of Criminal Appeal did not make the independent assessment of the evidence required of it, we are of the opinion that this is a proper case for the grant of special leave to appeal. The case is thus not one which calls for a consideration of the extent of the authority of the decision in Liberato v. The Queen [1985] HCA 66; (1985) 159 CLR 507. Because we are of the further opinion that the jury should have entertained a reasonable doubt as to the applicant's guilt, we would allow the appeal and set aside the convictions.

DAWSON J. All too often in applications before this Court insufficient attention is given to the distinction between leave to appeal and special leave to appeal. Leave to appeal may be given where an applicant makes out a prima facie case of error. That is not enough to attract special leave to appeal, which should only be given where, in addition, there is some special feature of the case which warrants the attention of this Court. It would be wrong to attempt to formulate rules which would confine the exercise of the jurisdiction to grant special leave. It remains a discretionary jurisdiction. Sufficient has, however, been established in the decided cases to provide a clear guide to those occasions when it is appropriate to grant special leave and those when it is not.

2. Section 35A of the Judiciary Act 1903 (Cth) allows the Court to have regard to any matters that it considers relevant, which confirms the discretionary nature of the jurisdiction to grant special leave. The section goes on to provide that, in exercising that jurisdiction, the Court shall have regard to the public importance of any question of law and the need to resolve judicial differences of opinion concerning the state of the law. In addition the Court shall, under the section, consider whether the administration of justice, either generally or in the particular case, requires an application for special leave to be granted. Those criteria are to be found in previous decisions of the Court and the section, as far as it goes, is declaratory.

3. In exercising its wide discretion to grant special leave to appeal, it is proper that the Court should be influenced by the function which it performs as the final appellate court in the judicial hierarchy. The number of applications for special leave increases year by year, particularly in the criminal law. Since the number of cases with which the Court can properly deal in any one year is limited, it is inevitable that a careful choice must be made having regard to the duty which the Court has to develop and clarify the law and to maintain procedural regularity in the courts below. The Court must necessarily place greater emphasis upon its public role in the evolution of the law than upon the private rights of the litigants before it. Whilst procedurally and otherwise this Court performs in many ways a truly appellate function, more significantly it operates as a court of review and this must ultimately be the most important factor in the selection of those cases in which special leave to appeal is to be granted. See generally Jolowicz, "Appeal and Review in Comparative Law: Similarities, Differences and Purposes", [1986] MelbULawRw 17; (1986) 15 Melbourne University Law Review 618.

4. From the beginning the Court has refused to grant special leave to appeal in criminal cases upon questions of fact. See Bataillard v. The King [1907] HCA 17; (1907) 4 CLR 1282, at p 1289; Collis v. Smith [1909] HCA 53; (1909) 9 CLR 490, at p 495. The most recent example of this approach is to be found in Liberato v. The Queen [1985] HCA 66; (1985) 159 CLR 507. The majority in that case said, at p 509:

"It has been repeatedly affirmed by this Court

that it is not a court of criminal appeal and that

it will not grant special leave to appeal in

criminal cases unless some point of general

importance is involved, which, if wrongly decided,

might seriously interfere with the administration

of criminal justice: see now Judiciary Act 1903

(Cth), s.35A. It would not be in accordance with

that practice to grant special leave to appeal in

this case where no question of law is involved and

where this Court is merely being asked to

substitute for the view taken by the Court of

Criminal Appeal a different view of the evidence

and of the effect of the summing up: cf. Reg. v.

Howe [1958] HCA 38; (1958) 100 CLR 448, at p.457."

But perhaps the most striking example of this invariable attitude is Raspor v. The Queen [1958] HCA 30; (1958) 99 CLR 346. In that case the trial judge had advised the jury to stop the trial and acquit the accused because of the inadequacy of the evidence identifying him as the culprit. The jury rejected this advice and proceeded to convict. There was no error in the directions given by the trial judge to the jury. The Court of Criminal Appeal of Victoria rejected an application for leave to appeal and the accused sought special leave to appeal to this Court.

5. This Court (Dixon C.J., Fullagar and Taylor JJ.) stated the accepted practice as follows, at pp.349-350:

"In the circumstances it was clearly a matter for

the Court of Criminal Appeal to decide whether they

considered that upon the evidence the conviction

should be allowed to stand. If there were no more

in the case, no ground for hesitation on our part

would exist; there would be no foundation for an

application to the special discretion of this

Court."

The Court then examined whether the Court of Criminal Appeal had taken an unduly narrow view of its function and, having concluded that it had not, refused special leave to appeal because the case turned on nothing but a question of fact.

6. This case is no different and turns on nothing but a question of fact.

7. The applicant was charged with the murder of two persons who died from burns which they received when a quantity of methylated spirits was poured on them and set alight. The applicant was convicted of manslaughter upon the basis of diminished responsibility.

8. The applicant, the two deceased and one of the prosecution witnesses, Doniger, were residents at the St. Vincent de Paul Hostel in South Brisbane. All of them suffered to some degree from alcoholism and, indeed, the deceased were on the morning on which they died drinking methylated spirits with at least two other men in or near a laneway not far from the hostel. The manager of a nearby hardware store saw two men, as well as the two deceased, at the scene of the crime at the time of the fatal incident. One of the men he identified as Doniger. He could not say whether the other man was the applicant, although he was "vaguely familiar". A woman, a bank employee, who was at the scene of the crime, saw one man pouring a clear liquid over another man and subsequently saw the two deceased alight. She was unable to identify either the applicant or Doniger.

9. Doniger was called as a witness by the prosecution and said that the applicant was present when the deceased were set on fire although he himself was some 100 yards away. He said that he did not see the applicant do anything more with methylated spirits than drink it. However, after being declared a hostile witness, Doniger agreed that his earlier statements to the police were correct. Those statements were to the effect that he was present when the deceased were set alight and that he had seen the applicant dousing them with methylated spirits. Under cross-examination he contradicted this evidence.

10. A Mrs Bargo, a welfare officer employed at the hostel, gave evidence that she had known the applicant for about ten years. She said that at about 9.30 a.m. on the morning in question she saw the applicant and Doniger walking out through the front gate of the hostel. About 10.00 a.m. on the same morning she said that she saw the applicant at the hostel looking very upset. She asked him what the matter was and, according to her, he said "I am waiting for them to pick me up." She asked who "they" were and he replied "I am waiting for the boys in blue." She asked why and he replied "I think I have killed my two mates. I threw metho over them, struck a match and burnt them." When asked had she noticed whether the applicant had been drinking when he spoke to her, Mrs Bargo said "He wasn't - he had a few." In explanation of that answer she said "Well, I have seen Shorty (a nickname of the applicant) very, very drunk, but he wasn't as drunk as I have seen other times. This was a bit early in the morning." She also said that the applicant was making sense when he spoke to her. When asked in cross-examination whether the applicant was a reliable person when he was drunk, Mrs Bargo replied "Not when he is real drunk; when he is paralytic."

11. The applicant gave evidence on oath that he made no admission to Mrs Bargo, that he did not start the fire, that he was not at the scene of the crime at the relevant time, that he did not see anyone else set the deceased on fire and that he was back in the hostel before 9.00 a.m.

12. A psychologist called on behalf of the applicant swore that the results of tests which he performed on the applicant were consistent with frontal lobe deterioration of the brain due to the excessive consumption of alcohol. A psychiatrist was also called on behalf of the applicant. He said that he had spoken to the applicant for half an hour in the cells of the Supreme Court during the trial. The trial was some seven months after the events in question and the applicant was in custody during that time. The psychiatrist said that he had read the psychologist's report. He accepted the view of the latter that there was frontal lobe deterioration of the applicant's brain. He gave evidence that brain damage of that kind can give rise to confabulation "where an individual has defective recall and fills in events, almost at random". He said that such a condition could be chronic, but added "I am not suggesting Mr Morris is in that chronic state, but that state can occur in a damaged brain that is intoxicated, quite readily, so that memories are haphazard and accounts are haphazard." The psychiatrist expressed the opinion that it was safe to assume that there was no deterioration in the applicant's brain functioning during the period he was in custody and that there had been a degree of recovery during that time.

13. In his application to the Court of Criminal Appeal for leave to appeal against conviction, the applicant sought to rely upon that part of s.668E of the Criminal Code 1899 (Q.) which requires the Court to allow an appeal if it is of the opinion that the verdict of the jury should be set aside on the ground that it is unreasonable, or cannot be supported having regard to the evidence. That ground of appeal requires the Court to discern whether the evidence is capable upon the required standard of proof of supporting a verdict of guilty. The required standard of proof is, of course, proof beyond reasonable doubt. The proper test to be applied by the Court is whether the jury, acting reasonably, must have entertained a sufficient doubt to have entitled the accused to an acquittal, that is, must have entertained a reasonable doubt as to the guilt of the accused. For the purpose of deciding that question the Court of Criminal Appeal must make an independent assessment of the evidence. It is, however, a limited function. As Gibbs C.J. and Mason J. pointed out in Chamberlain v. The Queen (No.2) [1984] HCA 7; (1984) 153 CLR 521, at p 534:

"The responsibility of deciding upon the verdict,

whether of conviction or acquittal, lies with the

jury and we can see no justification, in the

absence of express statutory provisions leading to

a different result, for an appellate tribunal to

usurp the function of the jury and disturb a

verdict of conviction simply because it disagrees

with the jury's conclusion. We do agree that in

many cases the distinction will be of no practical

consequence; it will be merely a matter of words.

That will not generally be the case where questions

of credibility are decisive. However, whether it

matters from a practical point of view or not in a

particular case, it is not unimportant to observe

the distinction - the trial is by jury, and (absent

other sources of error) the jury's verdict should

not be interfered with unless the Court of Criminal

Appeal concludes that a reasonable jury ought to

have had a reasonable doubt."

See also Whitehorn v. The Queen [1983] HCA 42; (1983) 152 CLR 657, at p 687.

14. In this application for special leave to appeal, counsel for the applicant sought to raise the same matters as he had raised in the Court of Criminal Appeal. At one stage he also submitted that pursuant to s.102 of the Evidence Act 1977 (Q.) the trial judge ought to have instructed the jury to have regard to the various matters to which the section refers in determining the inference which could reasonably be drawn from the previous inconsistent statement of the witness Doniger upon the basis of which he was declared a hostile witness. However, that submission could not be sustained because the witness adopted the statement in his evidence and the point became academic. It was not, in any event, a point which had been raised at the trial or before the Court of Criminal Appeal.

15. The Court of Criminal Appeal in no way misconceived its function in considering whether on the evidence the verdict ought to be allowed to stand. In his reasons for judgment, with which the other two members of the Court (Andrews C.J. and Demack J.) agreed, McPherson J. referred to both Chamberlain (No.2) and Whitehorn and there can be no doubt that in reaching his conclusion he correctly applied the principle contained in those cases. His Honour summarized the evidence which, he said, apart from the evidence of Mrs Bargo, "might very well not have justified a conviction". He referred to the direction of the trial judge to the jury that unless they accepted the evidence of Mrs Bargo they should be loath to convict. Nevertheless, McPherson J. expressed the view, plainly correct, that Mrs Bargo's evidence, if accepted, was sufficient to justify the jury concluding that the applicant was the individual responsible for setting fire to the deceased. He went on to deal with the evidence of the psychologist and the psychiatrist, the combined effect of which was that there was the possibility of confabulation on the part of the applicant. That evidence, he said, had a double relevance in the case of the applicant. It was relevant to the question whether the proper inference was that he would have foreseen and intended the results of the act of throwing methylated spirits at the deceased and then setting them on fire; it also had a relevance to the matter of the reliability of the admission or confession that he was alleged to have made to Mrs Bargo. He concluded:

"In all the circumstances, it may be said that

the Crown case against the accused was not a

particularly strong one. It did depend, however,

very much upon the impression of witnesses formed

by the jury in the course of the trial. That being

so, it is quite clear to my mind that, there being

evidence in (the) case on which the jury might

reasonably find a verdict of guilty, it cannot

possibly be said that they must have entertained a

reasonable doubt about the guilt of the accused.

On that footing, and having regard to what was said

by the High Court in the case of Chamberlain v. The

Queen to which I have referred, it seems to me that

it is simply not open to this Court to set aside

the jury verdict on the ground that it was one to

which the jury could not reasonably have come."

16. The assessment made of the evidence by McPherson J. was as complete as was possible at an appellate level but the case was, as he pointed out, one in which credibility was a significant issue. That was essentially a matter for the jury and not for an appellate court. The jury had an opportunity to observe the witnesses and, in particular, the applicant, and that circumstance must have been of crucial importance to them in reaching a conclusion. The evidence by the psychiatrist that the applicant may have been confabulating was somewhat speculative, being based on a considerably lesser observation of the applicant than that which the jury themselves had an opportunity to make and being necessarily inconclusive concerning the actual mental condition of the applicant either at the time of the alleged offences or at the time of the trial. Indeed, the trial judge expressed surprise, as he was entitled to do, at the breadth of the opinion expressed by the psychiatrist in view of the limited time which he had spent with the applicant. It is hardly remarkable that the jury, exercising due caution, apparently accepted the evidence of Mrs Bargo, who over a period of ten years had known the applicant in her capacity as a welfare officer and had observed him and other alcoholics in varying states of intoxication.

17. True it is that the psychiatrist gave evidence that it might be assumed that the mental condition of the applicant had improved during the time he was held in custody. The jury, of course, heard that evidence. Moreover, the trial judge brought it specifically to their attention in his charge, saying "Dr Quinn said that you would expect he would be better now, better when he was examined by the psychologist than he was then, when he was drinking regularly, because he has been deprived of his alcohol for a few months." There is no reason to suppose that the jury did not bear this in mind when assessing the evidence of the applicant and the reliability, and hence the probative force, of the confession which Mrs Bargo said the applicant had made to her.

18. It has been suggested that the Court of Criminal Appeal did not make an independent assessment of the evidence. The assessment which it made was necessarily independent of that made by the jury. It is apparent that it considered all of the evidence in making its assessment. It cannot be said that the probative value of the applicant's confession to Mrs Bargo did not claim the attention of the Court. However, the probative value of that confession depended largely upon whether Mrs Bargo's evidence was accepted or not by the jury. Her evidence was to the effect that the applicant was reliable when he was not intoxicated to a high degree and that on the morning of the alleged offence the applicant was not very drunk and was making sense. If that evidence was accepted, the largely theoretical evidence of the psychiatrist, which does not appear to have been based upon any clinical examination of the applicant, may not have left the jury in any real doubt. But it was a matter for the jury and beyond observing, as it did, that this was the case, there was little else the Court of Criminal Appeal could have said about it. If the jury were bound to reject the confession which the applicant made to Mrs Bargo, then it ought not to have been admitted in evidence and, if necessary, a voir dire should have been held for the purpose of determining that issue. No suggestion was made that such a course should have been taken. Once the confession was admitted in evidence, it was for the jury to place upon it what reliance they thought proper in all the circumstances, after receiving appropriate directions from the trial judge. That is the effect of Sinclair v. The King [1946] HCA 55; (1946) 73 CLR 316 where the question examined was whether a confession was inadmissible in evidence. It was determined that the confession in that case was admissible and special leave to appeal was refused.

19. I agree with the conclusion reached by the Court of Criminal Appeal, but even if I did not, this case would not possess any of the features necessary to attract special leave to appeal. The questions which it raises are entirely questions of fact. In so far as they do not depend upon credibility involving the observation of the witnesses, the Court of Criminal Appeal examined those questions and concluded that the verdict of the jury should stand. It would be plainly wrong for this Court to re-examine those same questions merely because it might come to a different conclusion.

20. It is true that in Whitehorn this Court allowed an appeal where it considered that a confession, which was the only evidence against the applicant, ought to have been rejected by the jury. But Whitehorn was a case which raised more than one point of general concern; it was applied by the Court of Criminal Appeal in this case. It was, therefore, a case which warranted the grant of special leave to appeal and, special leave having been granted, it was proper for this Court to deal with the matter as an appeal and substitute its view for that of the court below. In this case there is nothing which would warrant special leave being granted.

Order of the Court of Criminal appeal of Queensland dated 23 September 1986 set aside and in lieu thereof allow the appeal and quash the convictions.