R v B D X [2009] VSCA 28

2 In May 2008, BDX (hereafter referred to as 'the appellant', leave to appeal having been granted) was presented for trial in the County Court on three counts of incest. The complainant was his stepdaughter.

3 Count 1 alleged an act of digital penetration between 1 January 1997 and 31 December 1998. Count 2 alleged an act of sexual intercourse between 1 March 1998 and 30 April 1998. Count 3 alleged a further act of sexual intercourse between 1 January 2001 and 31 December 2002.

4 The appellant was convicted on counts 1 and 3, but acquitted on count 2. He was sentenced to three years and six months' imprisonment on each of the two counts on which he was convicted, with a year of the sentence imposed on count 1 ordered to be served cumulatively upon the sentence imposed on count 3. That made a total effective sentence of four years and six months' imprisonment. A non-parole period of two years and three months' imprisonment was fixed.

5 The background facts may be briefly stated. The complainant was born in August 1990. She was therefore about seven or eight years of age at the time of the offence that gave rise to count 1, and about 10 or 11 years of age at the time of the offence that gave rise to count 3. By the time of the trial, she was aged 17.

6 The Crown case depended essentially upon the complainant's evidence. In classical terms, it was a case of oath against oath. There was no corroboration of the complainant's testimony and no circumstantial or other evidence to support her account. Plainly, the issue for the jury to determine was whether they accepted the complainant as a credible witness. Her credit was therefore critical.

7 The appellant relied upon six grounds in support of his application for leave to appeal. Relevantly, they were as follows:

1. The trial miscarried as a result of the trial judge excluding evidence in the defence case sought to be called from a witness, Yvonne Zubrycki, impeaching the complainant's credibility; and in particular -

(a) as to the complainant's reputation for veracity;

(b) whether, in the opinion of the witness, the complainant was to be believed upon her oath.

...

1. The trial miscarried as a result of the trial judge excluding evidence in the defence case sought to be called from Dr Andrew Gibbs; and in particular, the evidence was admissible -

(a) as bearing upon the improbability of the complainant's evidence;

(b) as bearing upon motive.

...

1. A miscarriage of justice flowed from the directions to the jury concerning delay in complaint; and in particular, the trial judge failed to direct sufficiently or at all that -

(a) the jury could use delay in complaint as bearing upon the credibility of the complainant;

(b) in the circumstances of the case, it would be dangerous to convict on the evidence of the complainant alone.

...

4A. A miscarriage of justice was occasioned by the judge giving the jury the following direction or making the following 'comment'-

'[I]n this case where you have got a, obviously, a sharp conflict of evidence, it seems to me - and this is a comment which I make which you can accept or you can reject, it does not bind you in any way, that unless the Crown can satisfy you to reject the accused's answers in the record of interview and his sworn evidence, and to accept the version given by [the complainant], then you could not convict.'

...

4B. A miscarriage of justice flowed from the judge's failure to give the jury directions that -

(a) even if they preferred the evidence for the prosecution, they should not convict unless satisfied beyond reasonable doubt of the truth of that evidence;

(b) even if they did not positively believe the evidence for the defence, they could not find an issue against the applicant contrary to that evidence if that evidence gave rise to a reasonable doubt as to that issue.

1. The trial miscarried from the judge's failure to direct the jury sufficiently or at all that even if they rejected the motive for the complainant to lie put forward by the applicant -

(a) that did not mean the complainant was telling the truth;

(b) the prosecution still needed to prove that the witness was telling the truth.

...

1. An aggregate of the errors contained in grounds 1, 2, 3, 4A, 4B and 5 caused a miscarriage of justice.

8 Before dealing with these grounds, it is perhaps useful to explain why the Court was constituted by five judges, rather than the usual three, for this appeal. That decision was taken when it became clear that the appellant wished to challenge the correctness of a recent decision of this Court in R v Taylor (No 2).[1] It was conceded on his behalf that ground 3 could not be maintained unless that decision was overruled. In those circumstances, the President elected, pursuant to s 11(1) of the Supreme Court Act 1986, to convene a court of five.

9 It will be convenient to deal with the grounds of appeal in the order in which they were set out in the draft notice.

10 Early in the trial, and in the absence of the jury, counsel who appeared for the appellant below raised with the judge his intention to cross-examine the complainant with regard to what he described as her 'bad character'. Just why he felt compelled to do so is unclear.

11 Section 37A of the Evidence Act 1958 imposes limits upon the cross-examination of complainants in trials for sexual offences. However, that section does not prevent cross-examination with a view to demonstrating bad character as such. It only restricts questions as to prior sexual activity. It is apparent that trial counsel had no intention of cross-examining the complainant as to such matters. Accordingly, he did not require leave of the judge to attack the complainant's credibility.

12 In any event, and for whatever reason, trial counsel went on to inform his Honour that he proposed to cross-examine the complainant about a reputation that he claimed that she had acquired for making 'false allegations' at her former place of employment.

13 Not content with flagging his intention in that regard, trial counsel next informed the judge that irrespective of what the complainant might say when cross-examined along these lines, he proposed to call her former employer, Ms Yvonne Zubrycki, to testify that she would not believe the complainant on her oath. He referred to a number of English authorities which held that, at common law, a witness could be called to say that he or she would not believe another witness on their oath.[2]

14 The prosecutor seemed taken aback by this suggestion. She expressed doubts as to whether any such evidence could be led. Her reply was curious given that, shortly thereafter, she informed the judge that she had herself on other occasions, as defence counsel, called precisely such evidence.

15 Plainly, the judge had reservations as to the course proposed on behalf of the accused. He expressed 'disquiet' at the notion that any such evidence should be led. Indeed, his provisional view was that the complainant's former employer should not be permitted to give this evidence. However, the point was left in abeyance at that stage.

16 Much later in the trial, after the complainant had been cross-examined and it had been squarely put to her that she had, in the past, made false allegations against her former employer (which she denied), the issue of calling the former employer, Ms Zubrycki, again arose. His Honour determined that there should be a voir dire regarding that evidence.

17 On the voir dire, Ms Zubrycki gave evidence that the complainant had worked for her for about six months in the latter half of 2007. She said that she was aware of the complainant's general reputation and, in particular, her reputation for truthfulness. She said that by the time the complainant left her employ, she would not have believed anything the complainant said, even if said on oath.

18 Under cross-examination, Ms Zubrycki clarified the false allegations which she claimed the complainant had made against her. In substance, Ms Zubrycki said that she had falsely been accused of assaulting the complainant and interfering with her mobile phone.

19 The prosecutor then put the following questions in cross-examination:

You say [the complainant] is not to be believed on her oath, right?

---Yes

Are you saying that she would commit perjury?

---Yes, if that's worthwhile, I wouldn't ...

Perjury is lying under oath, wilful and corrupt perjury?

--- Yes, I would.

Are you familiar with her reputation for honesty within her community?

--- I wouldn't say I'm familiar with her for honesty. I would be familiar with the fact that she's lied.

That she's lied?

--- Mm.

Are you familiar with her community, that is, her circle of friends, the people she associates with, the people she is known to?

--- I am only familiar with the people that she worked with.

Their knowledge of their reputation for honesty, that is, telling the truth?

--- Well, all I can say is that they told me after the incident that they didn't believe a word that she said, and that applied to everyone in my employer [sic].

Sorry?

--- That applied to all the staff that worked for myself, and its ...

Your entire staff has the view that she's a liar?

--- Well, they know that I would not have done what she said. Yes, they said that she was lying about it.

20 His Honour's response to this evidence evinced little enthusiasm. He characterised it as nothing more than 'a dispute between employer and employee'.

21 Undeterred, trial counsel maintained that the evidence went further and demonstrated 'a history of lies' and a 'reputation for lies'. His Honour then warned that if counsel persisted in calling Ms Zubrycki, he might well have to comment upon her evidence. Counsel responded, entirely reasonably, that he could not prevent his Honour from doing so. The judge then said: 'No, of course you can't. It just seems absolute nonsense.'

22 That was by no means the end of the debate. Trial counsel submitted that it was perfectly clear, on the authorities, that he was entitled to call Ms Zubrycki. He added, perhaps somewhat petulantly, that his Honour's insistence that there be a voir dire had given the Crown an advantage to which it was not entitled.

23 At that stage, the judge appeared to still be vacillating as to whether to permit Ms Zubrycki to give evidence. The prosecutor maintained her position that the evidence was inadmissible. She submitted that the course that was proposed was exceedingly rare. She invited his Honour to take into account the devastating impact that it might have upon the complainant if evidence were to be given that she could not be believed on her oath. Finally, she submitted that evidence of that kind could not be led without an adequate foundation for its reception. She submitted that the dispute between Ms Zubrycki and the complainant did not provide any such foundation.

24 It may have been better if the prosecutor had reflected somewhat before making these submissions. The judge clearly regarded them as persuasive. Thereafter, trial counsel tried repeatedly, but in vain, to persuade his Honour that they were without substance.

25 In any event, the judge finally determined that the evidence of Ms Zubrycki should not be received. What his Honour said was succinct but to the point:

I don't propose to admit it. You can call her but not ask the question that you wouldn't believe her on oath.

26 Since the question whether Ms Zubrycki was prepared to believe the complainant on her oath was the very question that counsel wanted answered, and the answer to which the only reason that she was being called, his Honour's ruling that she could give evidence but not be asked that question, provided no comfort to the defence. Counsel pressed further but his Honour would not be moved. The judge regarded the matter as involving a 'head-on conflict between employer and employee', which in his mind was incapable of being resolved. He added, somewhat prophetically: 'Well, I'll give you an appeal point ...'

27 In our opinion, the judge erred in preventing the defence from leading this evidence. The rule that allows such evidence to be led is well established.[3] According to Wigmore, it dates back to 1664.[4]

28 The modern formulation of the rule is stated in Archbold, Criminal Pleading, Evidence and Practice 2009, in the following terms:

Whether a witness has or has not been convicted, witnesses may be called to speak as to his general character, although not as to any particular offence of which he may be guilty: 2 Hawk_._ c. 46 s.207_; R v Rookwood_ (1696) 13 St Tr. 139 at 211; and R v Watson (1817) 32 St Tr.1. In order to impeach the credit of a witness for veracity, witnesses may be called by the other side to prove that his general reputation is such that they would not believe him upon his oath: R v Brown and Hedley (1867) LR 1 CCR 70.

...

In practice the question usually put is 'From your knowledge of the witness would you believe him on his oath?'[5]

29 After referring to the decision of the Court of Appeal in R v Richardson; R v Longman,[6] (hereafter 'Richardson') to which we shall shortly return, Archbold continues:

The impeaching witness may not, in examination-in-chief, give reasons for his belief, but he may be asked for his reasons in cross-examination, and his answers in cross-examination cannot be contradicted: R v Gunewardene [1951] 2 KB 600, 35 Cr App R 80, CCA; R v Richardson, ante; cf. Toohey v Metropolitan Police Commr, post.[7]

30 In R v Gunewardene,[8] upon which the Court of Appeal in Richardson most heavily relied, the appellant, a medical practitioner, was convicted of manslaughter. The case against him was that he had counselled or procured an illegal abortion, which had been performed on a woman who had died as a consequence thereof. It was alleged, therefore, that he was a principal in the second degree to her death.

31 At the trial, the prosecution called a witness who gave evidence that was highly damaging to the appellant. After the appellant had himself given evidence, his counsel wished to call a doctor to testify as to the reliability of the prosecution witness and, particularly, as to the state of that witness's mind. When the prosecutor objected, counsel for the appellant submitted that the evidence was admissible on the ground of reputation for truthfulness and honesty.

32 Lord Goddard CJ delivered the judgment of the Court of Criminal Appeal (Lynskey and Devlin JJ agreeing). His Lordship said:

That witnesses can be called to say that they would not believe a particular witness called by the other side, whether for the prosecution in a criminal case or for a party in a civil case, is, in the opinion of the court, undoubted; but the nature of the discrediting evidence and how far the witness can go in stating the grounds for his belief are the matters which the court has to determine. [9]

33 After citing authorities going back as far as the seventeenth century, Lord Goddard noted that originally a witness called to impeach the credibility of another was confined to giving evidence as to general reputation, and could not state his individual opinion.[10] His Lordship continued:

But later cases show that the rule has at any rate been relaxed, though Lord Ellenborough seems to have taken much the same view in Mawson v. Hartsink, where, after some discussion, he allowed the question to be put in this way ... 'Have you the means of knowing what the general character of this witness was? and from such knowledge of his general character, would you believe him on oath?'

34 Mawson v Hartsink,[11] to which his Lordship referred, was but one of a number of cases decided in the early part of the nineteenth century regarding the scope of this rule. In Carlos v Brook,[12] Lord Eldon observed that a witness may be asked whether he would 'believe a man on his oath', but that it was not competent to ask him the ground of his opinion. In R v Bispham,[13] it was held that a person called to give evidence need not have personal knowledge of false testimony by the impugned witness in order to be permitted to testify in accordance with this rule. Several years later, in R v Hemp,[14] it was held that even if the person called had such personal knowledge, he or she would not be allowed to refer to it under the ban on collateral matters. Other early nineteenth century cases dealing with this rule include R v Rudge[15]and Harris v Tippett.[16]

35 In Gunewardene, having discussed Mawson v Hartsink, Lord Goddard next referred to R v Watson,[17] where it was held that while a witness could be called to say that he would not believe a previous witness on his oath, he could not give evidence in-chief of the particular facts that led him to that conclusion.

36 His Lordship then referred to Stephen's Digest of the Law of Evidence:[18]

The credit of any witness may be impeached by the opposite party, by the evidence of persons who swear that they, from their knowledge of the witness, believe him to be unworthy of credit upon his oath. Such persons may not upon their examination in chief give reasons for their belief, but they may be asked their reasons in cross-examination, and their answers cannot be contradicted.

37 It should be noted that Gunewardene was considered, and cited with approval, by the House of Lords in Toohey v Metropolitan Police Commissioner.[19]

38 There is a particularly helpful discussion of the rule in Richardson.[20] The defendants were charged with having conspired to pervert the course of justice by attempting to influence a jury and with having suborned witnesses at a trial in which the brother of one of the defendants was among those tried.

39 At the trial, the main prosecution witness gave evidence. In order to discredit her, the defence called a witness who was asked whether he would believe her on her oath. He replied that 'in certain particulars' he would.[21] However, the trial judge intervened and refused to allow the witness to be asked the further question whether, from his personal knowledge, he would believe her on her oath. The trial judge also refused to allow him to qualify his previous answer.[22]

40 The judgment of the Court of Criminal Appeal was delivered by Edmund Davies LJ, with whom Widgery and Lyell JJ agreed. His Lordship described the method, which counsel for the defence had sought to use to discredit the main prosecution witness, as having had 'considerable antiquity' in the law.[23] He then summarised the position at common law as follows:

1. A witness may be asked whether he has knowledge of the impugned witness's general reputation for veracity and whether (from such knowledge) he would believe the impugned witness's sworn testimony.
2. The witness called to impeach the credibility of a previous witness may also express his individual opinion (based upon his personal knowledge) as to whether the latter is to be believed upon his oath and is not confined to giving evidence merely of general reputation.
3. But whether his opinion as to the impugned witness's credibility be based simply upon the latter's general reputation for veracity or upon his personal knowledge, the witness cannot be permitted to indicate during his examination-in-chief the particular facts, circumstances or incidents which formed the basis of his opinion, although he may be cross-examined as to them.

This method of attacking a witness's veracity, though ancient, is used with exceeding rarity. Nevertheless it was sought to be made use of in the present case ...[24]

41 It is clear from this discussion of the authorities that the common law rule that a witness can be called to attack the veracity of another witness, though rarely invoked in practice, is alive and well in England.

42 Before this Court, the Crown submitted that irrespective of the position in England, the rule is no longer to be regarded as part of the common law of Australia. This was a somewhat bold submission given that, in Bell v The Queen,[25] the Full Court of the Federal Court, as recently as 1985, expressly affirmed the continued existence of the rule.

43 In a joint judgment, Wilcox and Miles JJ (with whom Davies J relevantly agreed) referred specifically, and in some detail, to Richardson. Their Honours cited with approval the passage from the judgment of Edmund Davies LJ to which we have previously referred.[26] They clearly endorsed his Lordship's statement of the position at common law.

44 The position in Victoria is that the rule has been considered and applied.[27] The same is true for New South Wales, where the rule has been applied comparatively recently. In R v Hanrahan,[28] Manning J observed that it was 'beyond doubt that evidence of the general reputation for veracity of a witness may still be given'. However, his Honour acknowledged that such evidence was rarely led in modern times. It is perhaps worth noting that Richardson is referred to in the joint judgment of McHugh, Gummow and Hayne JJ in Bull v R,[29] although in a somewhat different context.

45 The Crown sought to deal with these authorities by referring to the criticisms that had been levelled at the rule by various commentators. It noted that the Australian Law Reform Commission, in one of its early reports on codification of the law of evidence, appeared to favour its abolition.[30] In addition, the rule has been described as 'cumbersome, anomalous, and unconvincing' by no less an authority than J D Heydon in Cross on Evidence.[31]

46 With great respect, so far as this Court is concerned, none of this can possibly matter. If the rule continues to exist as part of the common law, it is irrelevant that it has its critics. The appellant was entitled to be tried according to law. That is the law as it stands, not as some think it ought to be. Whether Ms Zubrycki's evidence, had it been led, would have influenced the jury to reject the complainant's account can never be known. However, in our view, the trial judge erred in excluding it.

47 The Crown submitted that, even if his Honour did so err, this was a case which warranted the application of the proviso. The Crown's invocation of the proviso requires this Court to review the whole of the record of the trial, make its own independent assessment of the evidence and determine whether, making due allowance for the natural limitations that exist in the case of an appellate court proceeding wholly or substantially on the record, the accused was proved beyond reasonable doubt to be guilty of the offence of which he was convicted.[32]

48 It was submitted on behalf of the Crown that whatever opinion Ms Zubrycki might have held as to the complainant's reputation for veracity, it was hardly likely to have carried any great weight. That may be so, but there are obvious difficulties with invoking the proviso where the complainant's credibility was of such critical importance. For all we know, Ms Zubrycki might have been an extraordinarily impressive witness whose evidence cast real doubt upon the complainant's truthfulness. The transcript of the voir dire, being necessarily limited in scope, casts little light upon that issue.

49 The Crown, of course, had the burden of persuading this Court that the proviso should be invoked. That burden might have been difficult to discharge. However, for reasons that follow, it is unnecessary for us to express any final conclusion regarding that matter.

50 As with ground 1, this ground, as formulated, arose out of an application foreshadowed on behalf of the defence at the commencement of the trial. Counsel indicated that he proposed to call as a witness for the defence a clinical neuropsychologist, Dr Andrew Gibbs. It was the trial judge's rejection of that evidence that gave rise to this ground of appeal.

51 Prior to the commencement of the trial, the Crown proposed to lead evidence from the complainant to the effect that the appellant had indecently assaulted her in Queensland when she was only three years of age. Of course, that matter could not have been the subject of a charge in Victoria. It therefore involved what is usually described as an 'uncharged act'.

52 Ultimately, the Crown resiled from its intent to lead that evidence. However, trial counsel was not content to let the matter rest there. He indicated that he proposed to cross-examine the complainant regarding a statement that she had made to the police about that matter. He further indicated that, once he had cross-examined the uncharged act into evidence, he would seek to discredit the complainant by calling Dr Gibbs to testify that what she had said about that incident could not possibly be correct.

53 It is likely that counsel believed that he needed leave, pursuant to s 37A of the Evidence Act, to question the complainant about this matter. Of course, his intention was to go much further than merely challenging her in cross-examination.

54 There was no dispute at trial about Dr Gibbs' expertise in the field of neuropsychology. Among his other qualifications, he holds a PhD from the University of Melbourne. He was, at the relevant time, a consultant at the Mercy Private Hospital in East Melbourne.

55 Dr Gibbs prepared a report dated 29 April 2008. In that report, he said that he had read the various accounts given by the complainant of the appellant's conduct towards her. Those accounts included what she had said about the indecent assault allegedly committed in Queensland in 1993, when she was three years old. He described her statement regarding that matter as 'very detailed'.

56 The opinion that Dr Gibbs expressed regarding the statement was in the following terms:

1. Infantile amnesia would not make possible such detailed memories claimed for Queensland (Nerang) at age three years where these are highly detailed by the time of the police statement. A scientific explanation for this period is that the brain and verbal skills are not sufficiently developed to allow encoding of such 'memories'. Infantile amnesia occurs up to 4 years of age. The nature of the complainant's 'memory' changes from being 'vague' in her handwritten statements to being extremely detailed in the police statement.
2. This Queensland event revolves around discussions of 'first impressions' between the complainant and friend. It appears that first 'Disclosure' to friend revolves around claims that relate to a period of infantile amnesia at age three years. Interestingly, at committal friend retracts knowing of prior events to Philip Island (1997) - and in particular being informed of an indecent touch in Queensland during such period. (She states in her police statement that the complainant had a bad first impression of the defendant because of something he did. Italics added. Complainant states in police statement she told friend she was touched, and at committal p 8/9 states she told the complainant she was touched in QLD as well as events claimed at Phillip Island and Bonnie Doon.) This raises concern about the first statements to [the friend] that include initially vague 'memories' given these occur at age 3 years, well in the period known as infantile amnesia where such detailed recollection would not be possible. I note these initially vague 'memories' are elaborated into great detail by the time of the complainant's eventual police statement.
3. By the time of police statements, the complainant has an highly detailed account of claims relating to Queensland and coincide with period of infantile amnesia. These change from being a 'vague' memory in her handwritten statements to being highly detailed accounts of an indecent touch on the vagina at a time when the complainant was no older than 3 years and 3 months if the family moved from the location (Nerang) in November 1993. This raises some questions as to the process through which such 'recollections' occurred. Such processes include exposure to post-hoc information (ie discussion or material) that becomes incorporated into one's narrative, exposure to direct suggestion, or use of specific techniques (eg: exploration of symptoms, 'journaling/writing' exercises, imagery or hypnosis like methods, dream analysis ...) to encourage such recollection where these might impact on the reliability and validity of such recall.
4. By the time of the police statements, there is reference to matters of a spatial nature including the positioning of the complainant in relation to the defendant on left or right, as well as whether the left or right hand of the complainant or defendant occurred. At age 3 or 4 years, children are not able to reliably know left from right, and the ability to know the spatial or 'crossed perspective' of the other is more advanced and only reliable to later in childhood. With respect to memory, such detail does not necessarily imply the historical accuracy of the event in the absence of independent reliable corroboration.[33]

57 As previously indicated, counsel proposed to call Dr Gibbs with a view to discrediting the complainant generally. He indicated that his objective was to establish that she had been untruthful in her account of the Queensland matter, and that she should therefore be regarded as biased against the appellant.

58 His Honour, plainly troubled by this explanation, called on the prosecutor to respond. She submitted that the question of any indecent assault in 1993 went only to credit, and not to any issue in the trial. She further submitted that any opinion that Dr Gibbs expressed regarding infantile amnesia should be excluded. In her terms:

... One does not have to be a neuropsychologist to be aware of the fact that memory varies enormously from person to person. It's more than possible for a person to have a fleeting memory of something that happened at the age of three.

59 The prosecutor then proffered the information that her own earliest memory went back to the age of three. She even supplied details. She submitted that whether or not the complainant should be believed in relation to the uncharged act was a matter for the jury alone to determine, and that expert evidence of the kind sought to be led from Dr Gibbs should therefore not be admitted.

60 The prosecutor next submitted that a distinction should be drawn between children below the age of, say, 12 months, and those of about three years. She acknowledged, however, that she had little familiarity with the literature on this subject, and was basing this submission solely upon her own impressions.

61 The prosecutor then submitted that the only purpose for which Dr Gibbs was to be called was to demonstrate to the jury that the complainant had lied about the Queensland matter.

62 In fact, that was not strictly correct. Dr Gibbs' point was that the complainant's evidence regarding what had occurred in 1993 was necessarily unreliable, and not that it had to be viewed as being untruthful. Whether the jury took the next step, and concluded that the complainant was lying when she claimed to have a detailed recollection of that event, was, ultimately, a matter for them.

63 Having submitted that Dr Gibbs' evidence went solely to credit, the prosecutor drew attention to R v Umanski,[34] a case that dealt with what is ordinarily described as 'the collateral evidence rule'. She then referred to the decision of the House of Lords in Toohey v Metropolitan Police Commissioner, which held that medical evidence might be admissible as an exception to that rule.[35] She sought, however, to distinguish Toohey from the present case.

64 Counsel for the accused replied by referring to s 37E of the Evidence Act 1958. That section provides as follows:

Despite any rule of law to the contrary, in any legal proceeding that relates (wholly or partly) to a charge for a sexual offence, the court may receive evidence of a person's opinion that is based on that person's specialised knowledge (acquired through training, study or experience) of-

(a) the nature of sexual offences; and

(b) the social, psychological and cultural factors that may affect the behaviour of a person who has been the victim, or who alleges that he or she has been the victim, of a sexual offence, including the reasons that may contribute to a delay on the part of the victim to report the offence.

65 Presumably, counsel was relying upon s 37E(b) as the basis upon which Dr Gibbs' evidence should be received.

... I propose to rule that the rules as to expert evidence remain and they don't go out the window and that an expert cannot express an opinion which is an opinion to be determined by the jury. What I'm trying to say in general terms, and I'll give more detailed reasons later, is that he would be usurping the function of the jury.

I rule that the defence or Dr Gibbs cannot express an opinion that what the complainant alleged in the Queensland incident was a lie, and I rule that it goes to credit only, that incident.

67 His Honour indicated that he would publish more detailed reasons for the exclusion of Dr Gibbs' evidence at a later time. However, he did not return to that subject.

68 It should be emphasised that the Crown did not, at any stage, challenge Dr Gibbs' qualifications as an expert in the field of infantile amnesia. That was hardly surprising. The only issue raised below, and again on appeal to this Court, was whether the expert evidence as to that subject ought to have been admitted.

69 The Crown submitted that questions regarding the fallibility of human memory were entirely a matter for submission and solely within the province of the jury. They were not a matter for expert evidence. Accordingly, no such evidence could be led. The appellant, on the other hand, submitted that Dr Gibb's evidence was admissible and should have been received.

70 It is, of course, understood that witnesses are generally prohibited from giving evidence as to their opinions. They are usually only permitted to give evidence of facts that they have observed.[36] However, the position is different in relation to experts. A properly qualified expert may be permitted to give evidence of his or her opinion about a matter, at least where the jury would be unlikely to form a correct judgment of that matter without such assistance.[37]

71 In general, in order for expert evidence to be admissible, the subject matter must be beyond the scope of common knowledge, it must form part of a recognised body of knowledge, and that body of knowledge must be capable of assisting the jury in their fact-ascertainment task. Where expert evidence is admitted, the witness should state the factual basis for any conclusions drawn, and how those conclusions have been reached.

72 The first matter to be determined is whether the field in relation to which the opinion is expressed is one on which expert evidence can be called.

73 At common law, even those who are expert in a particular area are generally precluded from expressing opinions in circumstances where an ordinary person is as capable of forming a correct view on the question as anyone else.[38] The behaviour of ordinary people, and the process of assessing a witness's credibility, are matters typically within a jury's common knowledge. As such, expert evidence may not normally be led as to whether a witness's testimony is accurate or otherwise credible.[39]

74 The primary objection to Dr Gibbs' evidence, both at trial and before this Court, was that his views regarding infantile amnesia could be of no assistance to the jury because they would know, from their own experience, how human memory operates. They would also fully appreciate the constraints upon the capacity to recall of those who were very young when the relevant events occurred.

75 In Cross on Evidence, it is suggested that no useful purpose can be served by enumerating the matters which have been treated by the courts as fit subjects for expert opinion.[40] However, some examples taken from the cases may be instructive.

76 It has been held that, in appropriate cases, expert evidence can be led regarding a medical or psychiatric condition that affects, or may affect, a witness's ability to give reliable evidence. That is, as long as that evidence goes beyond the ordinary knowledge and experience of jurors.[41] In R v Runjanjic,[42] it was held that expert evidence could be led as to both the existence and characteristics of battered wife syndrome. In R v Bartlett,[43] the same conclusion was reached in relation to repressed memory syndrome.

77 The basic principles are those set out by the High Court in Clark v Ryan.[44] In that case, it was held that a consulting engineer, who over many years had been engaged in investigating road accidents for insurance companies, could not give evidence as to the movements and tendencies of a semi-trailer in an articulated vehicle. He was thereby precluded from expressing an opinion as to how a particular collision between a panel van and semi-trailer had occurred.

78 Dixon CJ put forward two reasons for this conclusion. The first related to the expert's lack of qualifications, whether formal or through experience, to express an opinion on this matter. The second was that:

... it was an attempt to guide the jury upon matters which it was within the ordinary capacity of jurors to determine for themselves.[45]

79 In R v Parker,[46] an early authority establishing the evidentiary use of fingerprints to prove identity, Cussen J said that expert witnesses may give in evidence statements based on their own experience or study, but cannot be permitted to attempt to point out to the jury matters which they could determine for themselves.

80 However, it is clear that there are limits even upon the admissibility of scientific evidence. For example, in R v Chard,[47] the English Court of Appeal held that a trial judge had rightly excluded medical evidence concerning the intention, at the critical time, of someone charged with murder. In that particular case, there was no question of any mental illness or other disability.

81 A useful illustration of the operation of these principles can be seen in Murphy v R.[48] That was an appeal from a conviction for murder following a trial in which a record of interview containing confessional statements had been tendered. The interview had been reduced to writing and read aloud to the accused, who then signed each page. In an unsworn statement, he denied having made the admissions. The defence sought unsuccessfully to call evidence from a psychologist to say that the accused had limited intellectual capacity by reason of a disturbed childhood and inadequate educational opportunities. The witness proposed to say that the accused would have had difficulty reading the record of interview and that certain expressions attributed to him were not those that he would have been capable of using. The trial judge rejected the admission of this evidence. By majority, the High Court held that it ought to have been admitted.

82 In summary, the test for admissibility of evidence of this nature seems to be whether the behaviour which the jury is called upon to assess is outside their experience and knowledge, and whether scientific evidence may be held to provide them with an understanding of it. If it is outside their ordinary experience then, provided there is a recognised body of science which deals with this aspect of behaviour, a suitably qualified expert may be called to express an opinion about it.

The justification for receiving such evidence, then, is that the conduct of mentally defective persons, persons suffering from mental illness or severe personality disorder, or persons under the influence of drugs, or children who have been the victims of trauma or sexual assault or persons suffering from a psychological state caused by stress is beyond the experience of the ordinary juror.[49]

84 In R v H(JR) (Childhood Amnesia),[50] a decision of the English Court of Appeal which is directly in point in relation to this ground of appeal, the appellant had been convicted of having indecently assaulted his daughter when she was aged four or five. Her witness statement, prepared when she was aged 19, gave a detailed account of the incident, including a description of her own emotional reaction to what had occurred. Following an unsuccessful appeal against conviction, the Criminal Cases Review Commission referred the case to the Court of Appeal.

85 That Court agreed to hear evidence from a psychologist who was an expert in the field of memory formation and development. In substance, he said that memories of early childhood were qualitatively different from memories of later events. Adults could not usually remember events of early childhood so as to be able to give a coherent narrative account, the only recall being one which was fragmentary, disjointed and idiosyncratic. The psychologist's evidence was that this 'period of childhood amnesia' lasted until the age of about seven. Evidence of an event said to have occurred at an early age containing detail and extraneous facts might well be unreliable yet, nonetheless, appear credible.

86 The Court of Appeal received the evidence but sounded a note of warning in doing so. Their Lordships said:

We would not wish to leave this case without sounding a note of caution about the introduction of evidence of the kind given by Professor Conway in this case. It will only be in the most unusual of circumstances that such evidence will be relevant and admissible at the trial of allegations of child abuse. The evidence would be relevant only in those rare cases in which the complainant provides a description of very early events which appears to contain an unrealistic amount of detail. That, in the experience of this court, does not happen often.[51]

87 The passage highlighted above seems to us to be apposite to this case. The whole point of Dr Gibbs' evidence would have been to establish that the complainant in the present case had given a description of early events which contained an unrealistic amount of detail. That meets the note of caution sounded by the Court of Appeal, and renders this evidence both relevant and admissible. On that basis, it ought to have been received.

88 The trial judge, in holding to the contrary, seems to have had in mind the general rule, based on the desirability of avoiding a multiplicity of issues, that answers given by a witness to questions put in cross-examination concerning collateral facts must be treated as final. They may or may not be accepted by the jury, but the cross-examiner must take them as they come, and cannot contradict them by other evidence.[52]

89 Regrettably, his Honour failed to appreciate that there are a number of important exceptions to this rule, at least one of which, namely, an allegation of bias, seems to us to have been applicable.

90 Toohey provides a useful example of the scope of the exceptions to the collateral evidence rule. In that case, the House of Lords held that evidence could be given of a witness' lack of opportunity, or capacity, to perceive the events about which he testified.[53]

91 The collateral evidence rule is no longer universally regarded as inflexible and unyielding. The categories of exception to that rule are not closed. In Palmer v R, McHugh J cited with approval the description of the rule as one of 'convenience and not of principle'.[54] His Honour said that to elevate such a rule into a fixed rule of law would be a mistake.[55]

92 In addition, as Cross suggests, 'it is a rule of practice related to the proper management of litigation',[56] and should be applied as such. A trial judge should not be prevented from determining, in an appropriate case, that the matter on which a witness' credit is being tested is sufficiently relevant to that credit as it bears upon the issues in the case that such evidence should be admitted.[57]

93 That is not to say that the collateral evidence rule no longer exists at common law. Nothing said by the High Court has had that effect. The rule must be applied, but with a measure of commonsense and flexibility.

94 The Crown submitted that Dr Gibbs' evidence with regard to infantile amnesia went no further than stating what any jury would regard as obvious. We reject that contention. It is true that most people can readily appreciate the difficulties that others have in remembering events that occurred long ago. However, in our opinion, it is not within the ordinary scope of knowledge that childhood amnesia exists as a recognised scientific phenomenon.

95 Based on his training and experience, Dr Gibbs was able to say that the complainant's detailed account of the events of 1993 could not possibly be accurate. The plain implication of that evidence was that she had not been truthful to the police, and that she bore malice towards the appellant. That went to the issue of bias. The appellant was entitled to have that evidence put before the jury as part of his overall defence to the charges brought against him.

96 The Crown very properly did not suggest that the proviso could be invoked in relation to this ground of appeal. It was essentially for that reason, and based on this ground, that the Court indicated at the conclusion of the hearing that the appeal would be allowed, and a new trial ordered.

97 This ground complains of inadequate directions on the part of the trial judge regarding the complainant's delay in telling anyone about these offences. In particular, it contends that his Honour failed to direct the jury adequately or at all that:

(a) the jury could use delay in complaint as bearing upon the credibility of the complainant;

(b) in the circumstances of this case, it would be dangerous to convict on the evidence of the complainant alone.

98 Speaking broadly, ground 3 complains of a failure by the trial judge to direct the jury in accordance with the requirements of two High Court decisions: Kilby v R[58] in relation to warning (a) and Longman v R[59] in relation to warning (b).

99 The difficulty with this ground, as formulated, lies in the fact that last year this Court held in R v Taylor (No 2)[60] that s 61 of the Crimes Act 1958 no longer applied to trials conducted after the commencement of amendments made by the Crimes (Sexual Offences Further Amendment) Act 2006 ('the amending Act'). This means that the less favourable position for accused persons with respect to delays in making complaint, which now applies by operation of s 61(1)(b)(ii) and (iii), also applies to any trial in which a presentment is filed after 1 December 2006, even if the charges were laid prior to that date.

100 Part 2 of the amending Act amended s 61 of the Crimes Act in relation to jury warnings. That Part had its own transitional provision, now s 607 of the Crimes Act. That section provides that the amendments made by the relevant provisions of the amending Act apply to:

any proceeding that commences on or after the commencement of that section, irrespective of when the offence to which the proceeding relates is alleged to have been committed.

• what was the proceeding to which s 607 referred? and

102 In Taylor (No 2), it was submitted on behalf of the applicant that the reference to 'any proceeding that commences' in s 607 had to be a reference to s 26(1) of the Magistrates' Court Act 1989. That sub-section provides that 'a criminal proceeding must be commenced by filing a charge'. For reasons that are not readily apparent, the Crown raised no contrary argument in that case.

103 The Court of Appeal in Taylor (No 2) comprised Ashley and Kellam JJA and Hansen AJA. In three separate, but concurring, judgments, their Honours held that s 61 should have been applied in its amended form because the proceeding commenced, for the purpose of s 607, at the time when the new presentments in that case were filed over.

104 Ashley JA delivered the leading judgment. His Honour explored the notion of the commencement of a proceeding through analogous concepts, including the commencement of a prosecution. He noted that, in England, it had been held that, in the context of statutory time limits pertaining to the prosecution of particular offences, the commencement of the prosecution occurred at any of a number of stages. These included the preferring of the indictment (when it was sent up without a preliminary inquiry), or the laying of the information or complaint, or the arrest of the accused person, or the application for a summons or warrant in respect of the offence.[61]

105 His Honour discussed various Australian authorities which were to the same effect.[62] He observed that the principle underlying these cases was that the laying of charges was, for some purposes, to be regarded as the commencement of the prosecution, which ended in conviction or acquittal in the County or Supreme Court.

106 However, as his Honour noted, context was of critical importance when deciding what was meant in s 607 by both the terms 'proceeding' and 'commencement'. His Honour said:

It does not follow, because for the purposes of addressing a statutory time bar a criminal prosecution commences when a charge is filed that something is thereafter identifiable at all times and for all purposes as 'the proceeding' which was commenced.[63]

107 Ashley JA then explained why, in his view, the more general approach taken to the commencement of a prosecution should not be applied when construing s 607. In broad terms, his Honour observed:

• the amendments to s 61 were relevant only to trials by jury of particular categories of criminal matters. Section 607 was directed to the commencement of those provisions. Such trials could not be held in the Magistrates' Court. It followed that s 26(1) of the Magistrates' Court Act could have no application to the construction of s 607;

• for a trial to be held in the County or Supreme Court, that Court must be seised of the matter. It would only be so seised, most commonly, when presentment was made in that Court: see Crimes Act, s 353(1), and note the analysis of that section in R v Parker;[64]

• if a trial is conducted and a verdict returned on a defective presentment, the trial is a nullity. The logical corollary of this is that if a presentment is regularly filed, a proceeding is thereby commenced in the court in which it is filed;

• section 4(4) of the Crimes (Criminal Trials) Act 1999 provides that 'on the making and filing of a fresh presentment, proceedings on any presentment previously made and filed in relation to the same offence or a related offence are permanently stayed'. That provision suggests that a new proceeding commences when a fresh presentment is filed over. This interpretation is supported by authority;[65]

• such a conclusion is consistent with those circumstances, rare though they might be, where a person is presented directly to a superior court for trial; and

• although a person may be committed on a number of charges by the Magistrates' Court, the presentment ultimately filed may differ to an extent from the charges upon which the person was committed.[66]

108 These were the main, but by no means only, factors that persuaded Ashley JA to the conclusion that he reached. His Honour arrived at that conclusion only after a thorough, and indeed painstaking, analysis. He recognised that whichever interpretation one adopted of the language used in s 607, there would be some odd consequences. He considered every relevant canon of construction before concluding that the 'proceeding' was that upon the filed-over presentment on which the applicant stood his trial.[67]

109 Kellam JA confined his analysis to the adequacy of the direction given pursuant to Kilby v R, and the sufficiency of the Longman warning. His Honour said, in relation to s 607, that he agreed with Ashley JA that the time of commencement of a criminal proceeding in the County or Supreme Court was when the presentment was filed. He referred specifically to s 4(4) of the Crimes (Criminal Trials) Act 1999 in support of that conclusion.[68]

110 Hansen AJA observed that Ashley JA had called into question the correctness of the assumption of both the applicant and the Crown that the proceeding in Taylor (No 2) had commenced prior to 1 December 2006, when the amendments came into force. He said that he agreed with Ashley JA that s 61(1)(b) applied in its amended form because the proceeding commenced, for the purpose of s 607, when the new presentments were filed over.[69]

111 Despite that conclusion, the Court of Appeal in Taylor (No 2) proceeded to determine the application before it upon the same basis as both the trial, and the appeal, had been conducted. In other words, their Honours assumed, contrary to their actual finding, that the more generous provisions that had applied before the enactment of the amending Act continued to apply to the instant case. That meant that both a Kilby direction and a Longman warning ought to have been given.

112 In relation to Kilby, it was held that the trial judge should have instructed the jury that in evaluating the evidence of the complainant, they could take into account the fact that there had been no complaint made at the earliest opportunity. In relation to Longman, it was held that the jury should have been warned, in undiluted terms, that it was dangerous to convict on the uncorroborated evidence of the complainant, and that they should not do so unless satisfied of the truth and accuracy of that evidence.

113 In the present case, the trial judge was clearly alive to the dictates of Taylor (No 2). His Honour observed that the presentment had been filed after 1 December 2006 which, in light of Taylor (No 2), meant that the amending Act applied.

114 Counsel who appeared for the appellant at trial did not press for a full Longman warning. He recognised that, at face value, the decision in Taylor (No 2) meant that the applicant was not entitled to any such direction. He did, however, press for a direction that would accommodate the rationale for the Longman warning, notwithstanding his concession that the 'dangerous to convict' component of that warning could no longer be used.

115 In summary, the position is as follows. The appellant was charged in November 2006, though the presentment was not filed until after 1 December 2006. It was submitted before this Court that, but for the decision in Taylor (No 2), the appellant would have been entitled to both a Kilby direction, and what was described as a full Longman warning.

116 Before this Court, senior counsel for the appellant submitted that Taylor (No 2) was wrongly decided, and should be overruled. It was in anticipation of that submission that a five-judge court was convened to hear this appeal.

117 Senior counsel's argument that Taylor (No 2) was wrongly decided can be summarised briefly. It was submitted that the reasoning failed to recognise that, generally speaking, a criminal proceeding can be commenced in several ways. In most cases, such proceedings are begun by filing a charge in the Magistrates' Court. However, on rare occasions, they can be commenced by filing a direct presentment.

118 It was submitted that Taylor (No 2) did not allow for the fact that a criminal proceeding is an integrated process, which involves a number of different, but connected, steps. Taken to its logical conclusion, Taylor (No 2) suggests that a gap exists between committal for trial and the filing of a presentment. Thus, Taylor (No 2) posits what was said to be the wholly untenable position that no court will be seised of the matter during the intermediate period since, by definition, there is no underlying proceeding on foot.

119 It was submitted on behalf of the appellant that the reasoning in Taylor (No 2) was misconceived in other ways as well. As previously noted, Ashley JA referred to s 4(4) of the Crimes (Criminal Trials) Act in support of the conclusion that the filing over of a fresh presentment involved the commencement of a criminal proceeding.

120 It was submitted, however, that this reasoning was fallacious. Section 4(4) was directed to 'proceedings on any presentment'. Ashley JA concluded that the provision suggested that 'there is a proceeding on a presentment' and that 'the proceeding on an earlier presentment is effectively brought to an end, and a new proceeding commences, when a fresh presentment is filed over'.[70] It was submitted that his Honour's analysis failed to take into account that the sub-section related solely to any proceeding 'on a presentment', and not to any proceeding generally. The latter typically involves a number of steps before a presentment is filed; for example, committal proceedings, which are clearly not proceedings 'on a presentment'.

121 As previously indicated, if the appellant's attack upon Taylor (No 2) were to succeed, and that case were to be overruled, the next step in his argument would be that he was entitled to both a Kilby direction and what has been termed a full Longman warning.

122 In our view, the appellant's argument breaks down at a number of points. First, it assumes that this Court, having been specially constituted as a court of five, will more readily hold that Taylor (No 2) was wrongly decided than would a court of three. That assumption may be unwarranted. Taylor (No 2) is a recent, carefully considered, and unanimous, judgment of the Court of Appeal. While there is some support for the view that a Court of Appeal comprising five judges should be more readily prepared to depart from an earlier decision of the Court composed in the normal way of three judges, this has not been definitively established. There is also a tenable view in support of the proposition that the ordinary rules of stare decisis continue to apply even to a five judge Court of Appeal.[71]

123 Of course, the Court of Appeal is not bound by its own previous decisions. Nonetheless, it will not lightly depart from its own carefully considered decision, delivered only relatively recently, in which the very issue now pressed was fully argued, and finally resolved.

124 Senior counsel for the appellant did not submit that the reasoning in Taylor (No 2) was per incuriam. Nor could he. Nor did he submit that some compelling argument presented in that case had been overlooked. All that was suggested was that the matter should be reconsidered by this Court because there was a preferable interpretation of s 607 available.

125 The principles which govern stare decisis at the level of intermediate appellate courts in this country have been laid down by the High Court. In Nguyen v Nguyen,[72] it was said:

The extent to which the Full Court of the Supreme Court of a State regards itself as free to depart from its own previous decisions must be a matter of practice for the court to determine for itself. An example of such a determination is the Practice Statement of the House of Lords in 1966: Practice Statement (Judicial Precedent).[73]

Where a court of appeal holds itself free to depart from an earlier decision it should do so cautiously and only when compelled to the conclusion that the earlier decision is wrong. The occasions upon which the departure from previous authority is warranted are infrequent and exceptional and pose no real threat to the doctrine of precedent and the predictability of the law.

...

In these circumstances, it would seem inappropriate that the appeal courts of the Supreme Courts and of the Federal Court should regard themselves as strictly bound by their own previous decisions. In cases where an appeal is not available or is not taken to this Court, rigid adherence to precedent is likely on occasions to perpetuate error without, as experience has shown, significantly increasing the corresponding advantage of certainty.[74]

127 The difficulty lies in determining whether the principles laid down by the High Court apply in precisely the same way to the deliberations of a specially constituted court of five.

128 Section 11(1) of the Supreme Court Act 1986 provides that any three or more Judges of Appeal constitute, and may exercise, all the jurisdiction and powers of the Court of Appeal. However, whether constituted by three or five, there is still only one Court of Appeal.

129 In Nguyen, reference was made to the long-standing practice whereby, in this State, a Court of five or more judges was convened if a decision of a Full Court of three judges was to be reviewed.[75] It may assist to consider some of the cases in which that practice was adopted.

130 In Forster v Forster,[76] reported in 1907, the Full Court of the Supreme Court considered a question that had been referred to it by a single judge. Part of the proceeding concerned the correctness of Firkins v Firkins, an earlier decision of the Full Court reported in 1898.[77] The Full Court in Firkins v Firkins had been constituted by three judges. The report of the case in Forster v Forster, though not part of the judgment itself, states:

As the first question involved the correctness of the decision of the Full Court in Firkins v Firkins ... it was heard and determined by the Full Court, consisting of five Judges, the second question being heard and determined by an ordinary Full Court, consisting of three Judges.[78]

131 The Full Court (a'Beckett ACJ, Hodges, Hood, Cussen and Chomley JJ) in Forster v Forster said:

So far as the first question involved the correctness of the decision in Firkins v Firkins, it was thought desirable, following several precedents, that the matter should be considered by the Full Court, consisting of five Judges.[79]

132 Their Honours did not identify the 'several precedents' to which they referred.

133 In McKinnon v Gange,[80] the issue to be determined was whether a decision of the Full Court in Clutterbuck v Curry[81] should be followed. Madden CJ, delivering the judgment of the Court (Madden CJ, Hood and Cussen JJ), said:

Of course, this Court always follows its own decisions, except in those cases where a subsequent Court ordinarily constituted is convinced by new authorities that its former decision is wrong, or where arguments of a new kind are presented to it, which carry conviction and seem to make the former decision not tenable. Then this Court brings in the whole strength of the Full Court to say whether the previous decision is still good, and to overrule it and get it out of the way if it is found to be bad.[82]

134 In context, the Chief Justice's allusion to 'the whole strength of the Full Court' was probably intended to connote the entire membership of the Supreme Court bench at that time. Forster v Forster had been decided by five judges in 1906, two of whom, Hood and Cussen JJ, were on the Court in McKinnon v Gange.

135 One possible explanation for constituting a bench of five in Forster v Forster was simply that the whole Court would thereby consider the matter in banc. Prior to the Judicature Acts 1873-1875, sittings of the Queens Bench, Common Pleas, and Exchequer at Westminster for determination of questions of law were described as 'in banc'.[83]

136 On that view, there was no longer any basis, in Victoria, for constituting a court of five after 1919, when a sixth justice was appointed to the Supreme Court. Certainly, there is not now, and never has been, any magic in the number five.

137 In more recent years, what has been described as a 'Full Bench' of five judges was occasionally constituted to consider the correctness of an earlier decision of a Full Court of three. In R v Yates,[84] the Full Bench said:

It appeared to that Court that the argument might raise for consideration the question whether the Court is confined to looking at the sentence actually imposed or whether it is entitled to consider the length of time which the applicant will probably have to serve in gaol. It was accordingly decided that the application should be reargued before a Court of five judges so that the Court could, if necessary, reconsider some earlier decisions of this Court which might not be thought to be any longer applicable.[85]

138 Over the past decade or so, since the creation of the Court of Appeal, the Court has been constituted as a bench of five far more frequently than was previously the case.[86]

139 In R v Tait,[87] Callaway JA explained the reasons for this:

The Court of Appeal would usually follow a decision of the Full Court and convene a Full Bench of five or more judges if the decision were to be reconsidered. Compare Nguyen v Nguyen ... There were, nevertheless, exceptional circumstances in which the Full Court constituted by three judges was at liberty not to follow a prior decision. A recent example is Avco Financial Services Ltd v Abschinski ... It may be that in future we would extend those exceptional circumstances to enable a greater number of Full Court, and in due course some of our own, previous decisions to be reviewed by a court of three. There are some old precedents, and some extempore judgments, where that course might be more in the interests of efficiency and justice than convening a Full Bench. But, where an important point of principle is involved, I would not propose a departure from the existing practice, which is long established and beneficial if properly applied. Compare McKinnon v Gange ... [88]

140 In Fernandez v Director of Public Prosecutions,[89] Winneke P said:

At the outset of these proceedings, a question of the competence of the appeal arose because of the decision of the Appeal Division of this court in Beljajev v Director of Public Prosecutions to the effect that no appeal lay to the Full Court from a decision of a single judge made pursuant to s 18A of the Bail Act. Counsel for the appellant indicated that he wished to challenge the correctness of that decision, particularly in the light of the recent decision of the High Court in Roy Morgan Research Centre Pty Ltd v Commissioner of State Revenue (Vic). Accordingly, five judges have constituted the court for the hearing and determination of this appeal.[90]

141 In Dodoro v Knighting,[91] Winneke P said:

Because the practices which I have described in the preceding paragraph were becoming entrenched, I have regarded it as desirable to convene a court of five judges for the purpose of finally deciding the status of 'appeals' to this court from refusals by the County Court to grant leave to applicants pursuant to s 93(4)(d) of the Act.[92]

142 Intermediate appellate courts of other states also, on occasion, constitute themselves as a bench of five.[93] So too does the Federal Court. The Full Court of that Court regards itself as free to depart from its earlier decisions, though only when they are regarded as being 'plainly wrong'.[94] The Chief Justice of the Federal Court sometimes constitutes a court of five in response to an application to do so, usually where a challenge is mounted to the correctness of one of its earlier decisions.[95]

143 It is instructive, in this regard, to consider the position of the English Court of Appeal. On occasion, that Court has been constituted by more than three judges.

144 In Kelly & Co v Kellond,[96] Lord Esher MR, sitting as a member of the Court of Appeal, said of that Court's powers:

This Court is one composed of six members, and if at any time a decision of a lesser number is called in question, and a difficulty arises about the accuracy of it, I think this Court is entitled, sitting as a full Court, to decide whether we will follow or not the decision arrived at by the smaller number.[97]

145 However, in Young v Bristol Aeroplane Co Ltd,[98] the Court of Appeal came to a different conclusion, with Lord Greene MR saying:

It is true that in this and similar cases the court which held itself to be bound by previous decisions consisted of three members only, but we can find no warrant for the argument that what is conveniently but inaccurately called the full court has any greater power in this respect than a division of the court consisting of three members only.[99]

146 In Davis v Johnson,[100] a specially constituted five judge Court of Appeal considered the correctness of Young v Bristol Aeroplane Co. On behalf of the majority, Lord Denning said:

Whenever it appears to this court that a previous decision was wrong, we should be at liberty to depart from it if we think it right to do so. Normally -- in nearly every case of course -- we would adhere to it. But in an exceptional case we are at liberty to depart from it.

Alternatively, in my opinion, we should extend the exceptions in Young v. Bristol Aeroplane Co. Ltd. [1944] K.B. 718 when it appears to be a proper case to do so. I realise that this comes virtually to the same thing, but such new exceptions have been created since Young v. Bristol Aeroplane Co. Ltd ... we extended the exceptions by holding that we could depart from a previous decision where there were conflicting principles -- as distinct from conflicting decisions -- of this court. Likewise we extended the notion of per incuriam in Industrial Properties (Barton Hill) Ltd. v. Associated Electrical Industries Ltd. [1977] Q.B. 614.[101]

147 Goff LJ, in dissent, took a different view. His Lordship said:

Alternatively, it has been suggested that the present court, being five in number and not three, has power not possessed by three to disregard previous decisions at least when in those cases the court consisted of not more than three members. For reasons which I will develop later, I cannot accept any of these submissions.[102]

148 Cumming-Bruce LJ agreed with Goff LJ. His Lordship said:

.. as was pointed out in Young v. Bristol Aeroplane, a 'full court' of five judges of the Court of Appeal has no greater jurisdiction or higher authority than a normal division of the court consisting of three judges, we cannot help thinking that, if the Attorney General's argument were accepted, there would be a strong tendency in cases of public interest and importance, to invite a 'full court' in effect to usurp the function of the House of Lords and to reverse a previous decision of the Court of Appeal. Such a result would plainly be inconsistent with the maintenance of the principle of stare decisis in our courts.[103]

149 On appeal to the House of Lords, the principles stated in Young v Bristol Aeroplane Co Ltd were affirmed.[104] It is interesting to note that Lord Diplock described Lord Denning's efforts as:

... a one-man crusade with the object of freeing the Court of Appeal from the shackles, which the doctrine of stare decisis imposed upon its liberty of decision by the application of the rule laid down in the Bristol Aeroplane case to its own previous decisions ...[105]

150 As can be seen from the above, in England the Court of Appeal is bound by its own previous decisions. It makes no difference how many members of that Court sit on any particular case, the rules as to stare decisis remain the same.

151 The position in Victoria is, in certain respects, quite different. The Court of Appeal in this State is not bound by its own previous decisions. Nor does it regard itself as bound by previous decisions of the Full Court.

152 Nonetheless, the doctrine of stare decisis remains important, and the Court will be slow to depart from one of its own earlier decisions. If there is a difference when the Court is constituted by five judges, as has been suggested,[106] that difference is likely to be marginal.

153 That takes us back to the present case, and ground 3. Regrettably, modern statutes are often drafted in a style that is highly convoluted, and presents difficulties of interpretation. Section 607 is no exception.

154 The starting point in construing any statutory provision must always be the language of the section itself. However, the task of construction also involves consideration of context and purpose. In that task, as in so many other areas of the law, reasonable minds may differ as to what ought to be the correct outcome.

155 The conclusion reached by the Court of Appeal in Taylor (No 2) provides a good example of this difficulty. Perhaps not everyone would arrive at the same interpretation of s 607 as did the Full Court in that case. At the same time, the reasoning, and particularly that of Ashley JA who wrote the lead judgment, cannot sensibly be said to be 'plainly wrong'. His Honour's judgment sets out in great detail, and with considerable care, a powerful case for concluding that s 607 should be construed in the manner therein outlined.

156 In our view, this Court cannot say that the construction for which the appellant now contends is preferable to that arrived at in Taylor (No 2). All that can be said in that regard is that whatever view one takes of s 607, the legislation may lead to peculiar, and even irrational, outcomes.

157 Of course, if the Parliament is dissatisfied with the interpretation given to s 607 in Taylor (No 2), it can easily accommodate that by amending the section. In our view, Taylor (No 2) should be followed. Accordingly, ground 3 fails in limine.

158 As it happens, in the particular circumstances of this case, none of this matters in the slightest. Even if Taylor (No 2) were wrongly decided, ground 3 would still necessarily fail.

159 In order to explain why we have come to that conclusion, it is useful to set out, in summary, some parts of what the trial judge told the jury regarding the matter of delay.

160 His Honour reminded the jury that although the offences were alleged to have occurred between 1997 and 2002, the complainant had not raised any complaint about them until February 2006. Accordingly, there was nothing that could be described as a 'recent complaint'. Nonetheless, as a matter of law, there might be good reasons why a victim of a sexual assault might delay or hesitate in complaining about it. In that regard, the complainant had explained that she had been frightened to tell anyone of what her stepfather had done to her. Her failure to complain could be damaging to her credit, though delay of itself would not indicate that her allegations were false. Complaints were not always made immediately, or even shortly after, sexual assaults. Ultimately, the weight to be given to this factor was a matter for the jury.

I now inform you of a significant consequence of the fact of delay and this is the impact of the delay on [BDX's] duty to defend himself against these charges. In assessing the evidence in this case, you will need to have regard to the following significant considerations because of the delay, especially since the offences are alleged to have been committed in about 1997 up to 2001 or 2002, and we are now of course in almost the middle of 2008.

You must regard these significant considerations that [BDX] lost the opportunity to make enquiries at or close to the time of the alleged incidents. That is number one. Number two, that [BDX] has lost the ability to explore the alleged circumstances in detail soon after the offences were said to have occurred, and such an exploration may have uncovered evidence which would have thrown doubt upon the complainant's allegations or confirmed [BDX's] denial of the charges, although you must bear in mind he does not have to prove or disprove anything. [BDX], thirdly, lost the means of testing the complainant's allegations which would have been available had there been no delay in prosecution. The advantage he lost was testing them at an earlier point in time, and other witnesses, apart from those who did give evidence, who may have been able to give evidence contradicting the complainant's allegation or allegations, can no longer remember relevant details, and those witnesses who did give evidence, the [Ls] and the other man who gave evidence, they also have difficulty in remembering relevant details.

I instruct you as a direction of law that you have got to take those disadvantages into consideration when determining whether the prosecution has proved [BDX's] guilt beyond reasonable doubt and I think it is probably worthwhile if I repeat what I have just said that in assessing the evidence you need to have regard to these significant considerations because of the delay, that [BDX] had lost the opportunity to make enquiries at or close to the time of the alleged incident. Secondly, that [BDX] had lost the ability to explore the alleged circumstances in detail soon after the offences were said to have occurred, and such an exploration may have uncovered evidence which would have thrown doubt on the complainant's allegations, or confirmed his denial of the charges, although as I said, there is no onus on him to prove or disprove anything.

[BDX], thirdly, lost the means of testing the complainant's allegations at an earlier point in time, and that would have been available had there been no delay in prosecution, and that is not a criticism of the prosecution, of course, because there was no complaint made until 20 February 2006, and other witnesses who may have been able to give evidence contradicting the complainant's allegations, can no longer remember relevant details, and those who did give evidence may no longer be able to remember relevant and pertinent details about the times that they spent with the accused man, and I repeat my instruction to you as a direction of law that you have got to take these disadvantages into consideration when considering whether the prosecution has proved [BDX's] guilt beyond reasonable doubt.

Another consequence of the delay is its possible effect, especially on the reliability of the complainant's memory, as well as indeed, of course, [BDX's] memory, because the accused's memory may be faulty on some matters because after all he is part of the human race too, and his memory may be non-existent on other matters which could perhaps exculpate him.

It is a matter of common human experience that honest people are subject to faulty recollection or inaccuracy, and you will easily understand that the passage of time may affect any witness' memory and people may convince themselves that some incident occurred in a particular way in the past, and one cannot ignore the effective imagination in motion, prejudice or suggestion, on how past events are remembered after many years. You will understand also that some people honestly and accurately recall incidents in the past with clarity and can swear to them sufficient for a jury to accept and believe what the witness is saying after the jury has given that evidence careful scrutiny.

You have got to consider the question of delay and its effect on memory and possible effect, especially on the reliability of the complainant's memory. As I just said, you will easily understand that the passage of time can affect any witness' memory, while in some cases people simply forget things. In other cases, their memory can become distorted, that is, they may come to remember things that did not really happen.

Experience has shown that human recollection is frequently erroneous and liable to distortion in this way and the likelihood of this error increases with the length of delay, and the risk is enhanced if the complainant was young when the alleged offence took place, and one's recollection of events occurring in childhood is often distorted or mistaken. It is therefore important that you carefully consider, not only whether the complainant's evidence is honest, in the sense that she believes it to be true, but also whether it is in fact true. Whilst you should use your common sense and experience in life in assessing the effect of the delay upon the complainant's memory, you must keep in mind the possibility that she honestly believes what she is saying, but is mistaken due to the distortion of her memory because of the delay that has occurred.

162 The judge went on to point out that there was no evidence whatsoever to support the complainant's account of what had occurred. His Honour reminded the jury that it was not necessary, as a matter of law, to have any such supporting evidence. He then said:

You have got to look, and must look at all the circumstances in which the complainant alleges that each of the offences was committed, and you would no doubt bear in mind that a sexual offence is not likely to be committed in a blaze of publicity. But you, the jury, may consider that the potential for error to be greater in the absence of any supporting evidence.

...

The fact remains that [the complainant's] evidence provides the sole evidence of the crimes alleged in the three counts on the presentment, and her evidence is crucial to this case, concerning these charges ... For this reason alone I direct you that you must scrutinise [the complainant's] evidence with great care before convicting on the basis of that evidence alone. You must be satisfied beyond reasonable doubt that [the complainant's] evidence is honest, accurate, and reliable.

163 In our opinion, in directing the jury in these terms, his Honour brought home to them the very reasons why it has been held that, where appropriate, both a Kilby direction and a Longman warning should be given. The only thing that his Honour did not do was to recite verbatim the contents of those directions, as endorsed by the High Court.

164 Before this Court, the only criticism levelled by counsel for the appellant in relation to the supposed failure to give a Longman warning was his Honour's failure to use the word 'dangerous' when telling the jury about the consequences of delay.

165 In evaluating that criticism, it must be remembered that the trial judge reminded the jury of the disadvantages that the appellant suffered by reason of the long delay that had ensued before the complainant raised these matters with the police. His Honour spelt out in clear terms the need to scrutinise the complainant's evidence 'with great care'. He warned the jury repeatedly that there was no corroboration, or other support, for the complainant's testimony. The fact that he did not use the word 'dangerous', as some of the cases (including Longman) suggest he ought to have done, does not mean that the jury were thereby misdirected. The risks of acting on the unsupported evidence of the complainant, and the need for the jury to scrutinise her evidence carefully, were spelt out in considerable detail. The purpose for which both a Kilby direction, and a Longman warning, are normally given was thereby adequately achieved.

166 A trial judge's charge need not follow any particular formula. It should not be regarded as a mantra. It must be tailored to the needs of the individual case. In our view, this charge was sufficient in spelling out to the jury precisely how they should approach their task of evaluating the complainant's evidence.

167 It follows that irrespective of whether Taylor (No 2) was correctly decided, there is no substance to ground 3.

168 This ground complains of a failure on the part of the trial judge to give what has come to be known as a Liberato direction.[107] In that case, Brennan and Deane JJ, in separate judgments, observed that when a case turns on a conflict between the evidence of a prosecution witness and that of a defence witness, a judge should not leave the jury with the belief that the answer to that question concludes the issue of innocence or guilt. The jury should be told that, even if they prefer the evidence for the prosecution, they should not convict unless satisfied beyond reasonable doubt of the truth of that evidence. They should also be told that, even if they do not positively believe the evidence for the defence, they cannot find an issue against the accused contrary to that evidence if that evidence gives rise to a reasonable doubt as to that issue.[108] It should be noted that their Honours were in dissent on the question whether special leave to appeal should be granted.[109]

169 Ground 4 has two components. The first complains of a gloss that the trial judge put upon the conflict between the account given by the complainant and that of the appellant. His Honour, having referred to that conflict, added:

... and this is a comment which I make which you can accept or you can reject, it does not bind you in any way.

... unless the Crown can satisfy you to reject the accused's answers in the record of interview and his sworn evidence, and to accept the version given by [the complainant], then you could not convict.

171 The second component of this ground complains that there was a miscarriage of justice because the trial judge failed to direct the jury that even if they preferred the evidence for the prosecution, they should not convict unless satisfied beyond reasonable doubt of the truth of that evidence. Added to that component was said to be the failure of the trial judge to direct the jury that even if they did not positively believe the evidence led on behalf of the appellant, they could not convict if that evidence gave rise to a reasonable doubt as to an element of the offence.

172 There is plainly substance to the first limb of this ground. His Honour should not have couched his direction to the jury as a 'comment' which they were free to disregard. It was, in fact, a direction of law which they were bound to follow. Nonetheless, when viewed in context, and having regard to the charge as a whole, we are not persuaded that this one sentence, uttered incautiously, gave rise to any miscarriage of justice.

173 In that regard, it should be noted that the passage to which this ground relates was preceded by the following directions:

Let me, before I go on ... turn to a trap that juries sometimes fall into, which may appear to be an attractive way of going about it, saying, 'right, we will sort this out by working out who is lying'. The attraction is real, but it is superficial and the problem is that you may not be able to resolve the conflict of evidence ... However, you can only convict [BDX] ... if you are satisfied, first of all, of the truth of the allegations made on behalf of the Crown, and the problem with the 'who's lying?' approach, is that it sometimes ignores that the onus of proof is on the Crown and the standard of that proof is proof beyond reasonable doubt.

174 We are fortified in our conclusion that the jury were not left under any misapprehension as to the onus of proof by his Honour's use, shortly after this passage, of the word 'comment' by the fact that no exception was taken to this aspect of his Honour's charge. That suggests that if there were a misdirection, it did not give rise to a fundamental irregularity in the trial. Certainly, the error even if it could be described as material, did not lead to a miscarriage of justice. It did not deprive the appellant of a chance fairly open to him of being acquitted.

175 As regards the second component of ground 4, the jury were told repeatedly of the need to be satisfied beyond reasonable doubt of the appellant's guilt before they could convict. The trial judge was not asked to give a direction of the kind now said to have been indispensable to a fair trial. In our view, there is no substance to this second component of this ground.

176 This ground contends that the trial miscarried by reason of the trial judge's failure to direct the jury sufficiently, or at all, that even if they rejected the motive put forward by the appellant for the complainant to lie, that did not mean that she was telling the truth. The ground contends that the jury should have been told that the prosecution still needed to establish that her evidence was truthful.

177 The ground is based upon a well-known passage in Palmer v R,[110] where, in a joint judgment, Brennan CJ, Gaudron and Gummow JJ said:

It is necessary to distinguish between cross-examination of a witness as to the motive of that witness to lie and cross-examination of another witness designed to show that that witness does not know of any fact from which to infer that the first witness had a motive to lie. The distinction was referred to in the context of a criminal trial by Hunt CJ at CL in another passage from his judgment in Uhrig:

'What this Court said in R v F and in R v E should not be interpreted as excluding arguments being put to the jury, by either counsel or the judge, relating to the validity of the motive to lie which has been asserted in relation to a witness in the particular case. That is so notwithstanding that there is no requirement for the accused to prove such a motive, although in many such cases where the evidence of that witness is vital to the Crown case it would be appropriate for the judge to direct the jury that, even if they reject the motive to lie put forward by the accused, that does not mean that the witness is necessarily telling the truth, and to emphasize that the Crown must still satisfy them that the witness is telling the truth. I believe that it is necessary for such a distinction to be stated expressly, in order to avoid skilful advocates attempting to persuade trial judges that a necessary consequence of this Court's decisions in those two cases is that arguments relating to a motive to lie are excluded in every case. That is not a necessary consequence at all.'

In our opinion, the view adopted in Queensland and New South Wales is correct in principle.[111]

178 The Uhrig direction, as it was described, to which this passage was addressed, was, of course, prefaced by the qualification that it should be given 'in many such cases'.[112] It was not stated as a mandatory requirement.

179 In any event, the trial judge's charge to the jury needs to be considered in this context. His Honour said:

One further matter I should direct you on. In a question which was put in the record of interview by the informant, [BDX] was asked this question. 'OK, you've obviously denied these allegations?' [BDX] replied, 'Yes'. The informant said, 'All right, why would [the complainant] make these allegations?' [BDX], in fact, provided explanations, but I direct you that [BDX], at no stage in this trial, bears any onus of proof. He does not have to prove or disprove anything at any time. The onus of proof is on the prosecution, on the Crown, from beginning to end and the law does not require [BDX] to give any explanation, nor provide or suggest any motive for ... the complainant, making what [BDX] maintained in answers to questions put to him in the record of interview were false allegations as to any form of sexual penetration with [the complainant].

An accused person, any accused person, is not required to give any explanation or provide any motive and the effect of a question such as this, 'Why would [the complainant] make these allegations,' may have the effect, or may lead to reversing the onus of proof. You must not draw any inference adverse to [BDX] because, for example, on the explanation or explanations which he gave, or the lack of some other explanation that perhaps he could have given, and now gives. No accused person can be expected to see into a complainant's mind, and be held accountable for failing to discern whatever motive there may be for what he says and maintains are false allegations. To ask a question, such as was asked here, is to invite you, the jury, to speculate as to what may be a possible motive for lying, and that would be - would not be trying the case on the evidence, but to speculate about unproven facts, so I emphasise and stress and direct you, that the law does not require [BDX] to give any explanation, or suggest any motive at all, for [the complainant] making what he claimed were false allegations.

The onus of proof is on the prosecution and an accused person does not have to prove or disprove anything.

180 In our view this ground was without substance. The directions given were substantially in accord with both Palmer and Uhrig. There was no risk that the jury would have been misled as to the onus of proof with regard to any motive on the part of the complainant to lie.

181 The complaint here was based on what was described as 'an aggregate of errors'. The principle under which such a ground can be entertained is that laid down in R v Kotzmann.[113]

182 We have already explained that ground 2, standing alone, was sufficient in our view to require a new trial. There is no purpose to be served by considering whether ground 6 adds anything to the remaining grounds so as to justify further the decision reached by the Court.

183 Ground 2, having been made out, the Court determined, at the conclusion of argument on the hearing of the appeal, that a new trial should be ordered.

184 We have had the advantage of reading in draft the joint reasons for judgment which substantially reflect our reasons for agreeing in the orders that the appeal should be allowed and that it should be ordered that a new trial be had.

185 We agree that the judge erred in excluding the evidence proposed to be given by Mrs Zubrycki that she would not believe the complainant on oath. On appeal the respondent conceded that the rule was part of the common law of the United Kingdom but contended that it did not form part of the common law of Australia. That contention must be rejected for the reasons advanced by Vincent and Weinberg JJA.

186 If that were the only error in the trial, we consider that the proviso could be applied; just as it was applied by the English Court of Appeal in Reg v Richardson[114] in relation to a similar error. Though the error relates to the credibility of a principal Crown witness, and errors of that kind will commonly preclude the application of the proviso,[115] that is not always so, and in our view it would not be so in this case. As the trial judge rightly said, Mrs Zubrycki's evidence was about a dispute between employer and employee which had nothing at all to do with the complainant's credibility or reliability as to matters in issue. Consequently, in our view a jury properly instructed as to the significance of her evidence could not have failed to conclude that it had no bearing on the issues which they needed to decide. In the result, we consider that the irregularity could not reasonably be supposed to be one that would have influenced the verdict.[116]

Ground 2 - Evidence of infantile amnesia and the exception to the finality rule

187 We agree that the evidence of Dr Gibbs, a neuro-psychologist, as to the improbability of the complainant being able to recall events which she said had occurred when she was only three years of age, was both relevant and admissible. In our judgment, the learned trial judge was in error in excluding this evidence and it was for that reason that we joined in the orders pronounced at the conclusion of the appeal that the appeal be allowed, the conviction quashed and that there be a new trial.

188 At the trial, the appellant sought to rely upon the evidence of Dr Gibbs to support a contention that the complainant had given false evidence about events which were said to have occurred when she was three years' old, and thereby to demonstrate that the complainant had a bias against the appellant. Counsel at trial also relied upon s 37E of the Evidence Act 1958.

189 The prosecutor submitted in opposition to that contention that Dr Gibbs' evidence was inadmissible in that it went only to the complainant's credit and involved the expression of an opinion which usurped the jury's function of determining for itself whether the complainant was reliable. But the prosecutor, made no submission with respect to the defence contention that it should be admitted as proof of the complainant's bias or that it was admissible under s 37E.

190 The trial judge ruled the evidence inadmissible on the basis that it went only to the complainant's credit and involved the expression of an opinion which was the sole province of the jury. His Honour, however, also failed to address the question of bias and he too made no mention of the operation of s 37E.

191 Under the finality rule, independent evidence is inadmissible to contradict answers given by a witness in cross-examination as to credit. But the rule is subject to the well established qualification that evidence in contradiction of answers to questions going to credit is admissible if it tends to show that the witness is biased or partial in relation to the accused. In this case the evidence of Dr Gibbs would have had that effect - by demonstrating that the complainant was disposed to make incredible allegations against the appellant of having committed sexual offences against her.

192 Senior counsel for the respondent on appeal, rightly in our view, abandoned the submission advanced at trial that the evidence was inadmissible on the ground that it breached the finality rule. He conceded that the evidence fell within the exception which permits evidence which tends to show a disposition to make or support false complaints and which may thus bear upon whether the offence has been committed.[117] But he sought to maintain an argument that the opinion of Dr Gibbs was inadmissible as expert evidence, upon the basis that the reliability of the complainant to relate events when she was three years was a matter solely for the jury. In our view that argument should be rejected.

193 We agree for the reasons given by Vincent and Weinberg JJA that the law governing the admission of expert evidence as to matters which may affect a witness's ability to give reliable evidence entitled the appellant to adduce the evidence of Dr Gibbs. Such evidence will ordinarily be admitted to show a lack of capacity or opportunity by the witness to perceive an event which is in issue. The decision of the House of Lords in Toohey v Metropolitan Police Commissioner[118] is an example. Here, though the evidence did not relate to an issue in the case but to credit, it was admissible under the exception to the finality rule to which we have referred.

194 We wish, however, to add that we take a different view to Vincent and Weinberg JJA concerning the scope of the finality rule. Despite the observations of McHugh J in Nicholls & Coates v The Queen, it does not appear to us that the majority of the High Court accepted that the finality rule should be seen as a mere guide to discretionary case management. Nor was it argued on this appeal that the admission of Dr Gibbs' evidence required any relaxation of the finality rule or that its admission could be rested upon the exercise of a discretion by the trial judge based upon case management considerations. With great respect to the observations made by Vincent and Weinberg JJA in their joint judgment, we should not be disposed to adopt such a view until and unless the High Court has done so. In our opinion, the exceptions to the finality rule continue to define the ambit of evidence which may be relevant and admissible as to the credit of a witness.

195 For the sake of completeness, we should also say that that we do not consider that s 37E added anything to the admissibility of Dr Gibbs' evidence.

196 Section 37E makes admissible expert opinion evidence of:

(a) the nature of sexual offences; and

(b) the social, psychological and cultural factors that may affect the behaviour of a person who has been the victim, or who alleges that he or she has been the victim, of a sexual offence, including the reasons that may contribute to a delay on the part of the victim to report the offence.

197 The scope of the opinion evidence is confined to 'social psychological and cultural factors' that may affect behaviour which occurred prior to, at the time of or subsequent to the commission of the offence charged. Dr Gibbs' opinion did not relate to any relevant behaviour of the victim or to the particular factors specified in s 37E(b) which bear upon such behaviour. In our view, his opinion was not admissible pursuant to s 37E.

198 In Taylor (No 2)[119] this Court decided that the amendments made to s 6L of the Crimes Act 1958 by the Crimes (Sexual Offences Further Amendments) Act 2006 applies to trials initiated by presentment filed or further presentment filed over on or after the commencement of that Act. The appellant contends that the decision in Taylor (No 2) was wrong and, consequently, that the trial judge was wrong to direct the jury as if the amendments did apply.

199 In our view, Taylor (No 2) was correctly decided and nothing advanced in argument causes us any doubt about the reasoning which supports it. One case not mentioned in argument but which might perhaps be thought to bear upon the problem is Cornwell v The Queen.[120] In that case the High Court held that a second trial following a mistrial or successful appeal was not a new proceeding but rather that the first trial and the second trial were parts of the one proceeding for the purposes of s 128(7) of the Evidence Act 2008. Significantly, the Court considered that to be the case regardless of whether an accused were put up for a second trial on the original indictment or a fresh indictment, providing the charges in each case were the same. It seems to us, however, that the decision turned on the terms and purpose of s 128(7) of the Evidence Act 2008, which are of course very different to the terms and purpose of s 607 of the Crimes Act 1958. We do not consider that it throws any doubt on the decision in Taylor (No 2).

200 The joint judgment of Vincent and Weinberg JJA raises the question of the extent to which the full bench of five judges is free to depart from an earlier decision of this Court comprising three judges. That question was also pursued by Weinberg JA in the appeal of R v Roussety[121] which was heard by a Court comprising five judges on the same day as the present appeal.

Weinberg JA has raised the question of the extent to which a full bench of five judges is free to depart from an earlier decision of a Court of three. He has pointed out that it has generally been assumed that a Court of five has greater freedom to depart from an earlier appellate decision of a Court of three. No party advanced argument on the appeal that this was not an appropriate case in which a Court comprising five judges could reconsider its earlier decision by a Court of three concerning the question of statutory construction. As the matter has not been the subject of argument, this case is not the appropriate vehicle for consideration of when a Court of five may depart from an earlier majority decision of a Court of three.[122]

Similarly, in this case, neither party made any submission as to when a court of comprised of five judges may reconsider an earlier decision of a court comprised of three judges.

202 We do not regard the present case as an appropriate vehicle for consideration of when a Court of five may depart from an earlier decision of a Court of three. We would, therefore, not wish it to be understood that we necessarily agree with the view expressed by Vincent and Weinberg JJA that any difference between the approach which may be adopted by a Court of three or five judges would only be marginal.

203 For the reasons given by Vincent and Weinberg JJA, we agree that even if a Kilby v The Queen[123] or Longman v The Queen[124] direction were required, the instructions given to the jury by the learned trial judge were sufficient in this case.

204 As this Court said in R v KDY,[125] the dissenting judgments of Brennan and Dean JJ in Liberato explain that the need for a Liberato direction arises where the jury has been left with the impression that the evidence on which the accused relies can give rise to a reasonable doubt only if they believe that evidence to be true. That was the suggestion that had been made in Liberato. It therefore required a clear and unequivocal direction about the criminal onus and standard of proof so that the jury would not mistake the 'choice' between witnesses as the real question or as concluding the issue whether the prosecution had proved its case beyond reasonable doubt. But the present case was not one in which the trial judge had 'overlaid'

directions on the burden of proof or raised questions about a choice between conflicting witnesses which may have misled the jury in its process of reasoning.[126] We agree that for the reasons given by Vincent and Weinberg JJA that the jury were not left under any misapprehension as to the onus of proof or that any Liberato direction was called for.

Grounds 5 and 6 - Palmer/Uhrig directions and Kotzman

205 There is nothing we can usefully add to what Vincent and Weinberg JJA have written on these questions. We agree with their Honours for the reasons they have given that ground 5 fail and that the validity of ground 6 need not be decided.

206 I joined in the orders that the appeal be allowed and that a new trial be had because, in my opinion, ground 2 had been made out. The joint reasons of Vincent and Weinberg JJA and of Nettle and Redlich JJA which I have had the advantage of reading in draft, explain my conclusion. In short, Dr Gibbs' proposed evidence, contradicting the complainant's answers to questions going to credit, would arguably have tended to show that the complainant was biased against the appellant. It would have done so by calling in aid a matter of expertise, that is, the phenomenon of infantile amnesia. That was enough to establish ground 2.

207 I also agree with the other members of the Court, for the reasons which they give, that error was established in respect of ground 1, that (even if R v Taylor (No 2)[127] was wrongly decided) no part of ground 3 was made out, and that grounds 4 and 5 were not established. Like Vincent and Weinberg JJA, I refrain from expressing a conclusion whether ground 6 was made out, it being unnecessary to do so.

208 I add three matters. First, although it is unnecessary to decide the point, I doubt that the proviso could have been applied if ground 1 had been the sole ground of appeal which was established. I would be reluctant to speculate what impact the proposed evidence of Mrs Zubrycki might have had upon the jury in a case in which the complainant's credit was under attack, and in which the proposed evidence of Dr Gibbs, regardless of the precise basis of its admissibility, would have tended to reinforce that attack.

209 Second, despite the submissions advanced by senior counsel for the appellant, I remain of the opinion that the construction which I placed upon s 607 of the Crimes Act 1958 in Taylor (No 2) is the correct one. I should add that I respectfully agree with what is said by Nettle and Redlich JJA concerning Cornwell v The Queen.[128]

210 Third, on the view which I take of Taylor (No 2), it is unnecessary for me to add to what to I said in R v Roussety[129] concerning the hearing and determination of an appeal by a five person bench.

[1] [2008] VSCA 57; (2008) 18 VR 613.

[2] R v Gunewardene [1951] 2 KB 600; Toohey v Metropolitan Police Commissioner [1965] AC 595 and R v Richardson; R & Longman [1969] 1 QB 299.

[3] See generally J D Heydon, Cross on Evidence (7th Australian ed, 2004) [19045]; P K Waight and C R Williams, Evidence - Commentary and Materials (7th ed, 2006) 336-7; A Ligertwood, Australian Evidence (4th ed, 2004) 555-6; D Ross, Ross on Crime (3rd ed, 2007) [3.1210]; G Roberts, Evidence - Proof and Practice (1998) 276-7 and J Hunter, C Cameron and T Henning, Litigation II - Evidence and Criminal Process (7th ed, 2005) [22.83]-[22.84].

[4] J D Heydon, Cross on Evidence (7th Australian ed, 2004) [19045], n 67.

[5] Archbold, Criminal Pleading, Evidence and Practice 2009, [8-153] (emphasis added).

[6] [1969] 1 QB 299.

[7] Archbold, Criminal Pleading, Evidence and Practice 2009, [8-153].

[8] [1951] 2 KB 600.

[11] (1802) 4 Esp 102; 170 ER 656. Cross On Evidence (7th Australian ed, 2004) at [19045] also refers to R v O'Coigly (1798) 26 St Tr 1191 as authority for the proposition that the impugning witness must speak from personal knowledge.

[12] (1804) 10 Ves 49.

[13] [1830] EngR 795; (1830) 4 C & P 392; 172 ER 754.

[14] [1833] EngR 419; (1835) 5 C & P 468; 172 ER 1057.

[15] [1805] EngR 10; (1805) 170 ER 255.

[16] [1811] EngR 198; (1811) 170 ER 1277.

[17] [1817] EngR 133; (1817) 2 Stark 116, 149.

[19] [1965] AC 595, 606.

[20] [1969] 1 QB 299.

[25] Bell v The Queen [1985] FCA 454; (1985) 7 FCR 555.

[26] [1985] FCA 454; (1985) 7 FCR 555, 563.

[27] See, for example, R v Moloney [1888] VicLawRp 163; (1888) 14 VLR 934.

[28] [1967] 2 NSWR 717, 718-9.

[29] [2000] HCA 24; (2000) 201 CLR 443, 455.

[30] Australian Law Reform Commission, Evidence, Report No 26 (Interim) (1985) Vol 2 [183].

[31] Cross on Evidence (7th Australian ed, 2004), [19045].

[32] Weiss v R [2005] HCA 81; (2005) 224 CLR 300.

[34] [1961] VicRp 38; [1961] VR 242.

[35] [1965] AC 595.

[36] R v Holmes [2008] VSCA 128.

[37] Clark v Ryan [1960] HCA 42; (1960) 103 CLR 486; R v Darrington & McGauley [1980] VicRp 36; [1980] VR 353 and R v Smith [1987] VicRp 74; [1987] VR 907.

[38] Farrell v R (1998) 194 CLR 286.

[39] R v Turner [1975] QB 834 and R v O'Callaghan [1976] VicRp 43; [1976] VR 441.

[40] J D Heydon, Cross on Evidence (7th Australian ed, 2004), [29050].

[41] Farrell v R (1998) 194 CLR 286 and R v Hickey & Komljenovic (1995) 89 A Crim R 554.

[42] [1991] SASC 2951; (1991) 56 SASR 114.

[43] [1996] VICSC 47; [1996] 2 VR 687.

[44] [1960] HCA 42; (1960) 103 CLR 486.

[46] [1912] VicLawRp 32; (1912) VLR 152, 159-60.

[47] (1971) 56 Cr App R 268.

[48] [1989] HCA 28; (1989) 167 CLR 94.

[49] J D Heydon, Cross on Evidence (7th Australian ed, 2004), [29050] (citations omitted).

[50] [2006] 1 Cr App R 195.

[52] See generally Attorney-General v Hitchcock [1847] EngR 616; (1847) 154 ER 38 and Piddington v Bennett & Wood Pty Ltd [1940] HCA 2; (1940) 63 CLR 533.

[53] Toohey v Metropolitan Police Commissioner [1965] AC 595, 608.

[54] [1998] HCA 2; (1998) 193 CLR 1, 23.

[56] J D Heydon, Cross on Evidence (7th Australian ed, 2004), [17590].

[57] Natta v Canham [1991] FCA 470; (1991) 32 FCR 282, 298 and 300.

[58] [1973] HCA 30; (1973) 129 CLR 460.

[59] [1989] HCA 60; (1989) 168 CLR 79.

[60] [2008] VSCA 57; (2008) 18 VR 613.

[62] Ibid [15]-[18]. See, for example, R v Kelly [1921] VicLawRp 86; [1921] VLR 489, 491; R v Hackett [1995] SASC 5210; (1995) 64 SASR 471; and Sankey v Whitlam (1978) 142 CLR 1, 26.

[64] [1977] VicRp 3; [1977] VR 22.

[65] R v Harris and Others (No. 2) [1990] VicRp 27; [1990] VR 305.

[66] Taylor (No 2) [2008] VSCA 57, [20]-[27].

[71] R v Roussety [2008] VSCA 259, [80]-[81] (Weinberg JA, with whom Vincent JA relevantly agreed). Cf Ashley JA, [57], where his Honour expressed the provisional view that there was a 'different and lesser inhibition' for a court of five to depart from a decision of a court of three.

[72] [1990] HCA 9; (1990) 169 CLR 245.

[76] [1907] VicLawRp 31; [1907] VLR 159.

[77] (1898) 4 ALR 74.

[78] [1907] VicLawRp 31; [1907] VLR 159, 160.

[80] [1910] VicLawRp 7; [1910] VLR 32.

[81] [1885] VicLawRp 150; [1885] 11 VLR 810.

[82] [1910] VicLawRp 7; [1910] VLR 32, 35.

[83] See In re Hastings (No 2) [1959] 1 QB 358, 367 (Lord Parker CJ). There is a helpful discussion by Bell J of the meaning of the term 'in banc' in Engebretson v Bartlett [2007] VSC 163; (2007) 16 VR 417, 424-30. His Honour notes that it was not until 1883, when the Judicature Act was passed in Victoria, that provision was made for a Full Court consisting of all of the judges of the Court, or not less than three of them.

[84] [1985] VicRp 3; [1985] VR 41.

[86] See, for example: Fernandez v Director of Public Prosecutions [2002] VSCA 115; (2002) 5 VR 374; Dodoro v Knighting [2004] VSCA 217; (2004) 10 VR 277; R v Pidoto and O'Dea [2006] VSCA 185; (2006) 14 VR 269; Re Zoudi [2006] VSCA 298; (2006) 14 VR 580; R v Raad [2006] VSCA 67; (2006) 15 VR 338; Kelso v Tatiara Meat Co Pty Ltd [2007] VSCA 267; (2007) 17 VR 592; R v Piacentino [2007] VSCA 49; (2007) 15 VR 501; and R v MacNeil-Brown [2008] VSCA 190.

[87] [1996] VicRp 48; [1996] 1 VR 662.

[89] [2002] VSCA 115; (2002) 5 VR 374.

[91] [2004] VSCA 217; (2004) 10 VR 277.

[93] See, for example, R v Simpson (2001) 53 NSWLR 704, in which the New South Wales Court of Criminal Appeal was constituted as a court of five in order to consider the interpretation of a new sentencing provision. See also Yates v State of Western Australia [2008] WASCA 144, where a court of five sat to determine the effect upon sentencing of transitional provisions regarding the abolition of remissions.

[94] See, for example, SZEEU v Minister for Immigration and Multicultural and Indigenous Affairs [2006] FCAFC 2; (2006) 150 FCR 214, [146]-[148] (Weinberg J).

[95] See, for example, New Zealand v Moloney [2006] FCAFC 143; (2006) 154 FCR 250.

[96] (1888) 20 QBD 569.

[98] [1944] KB 718.

[100] [1978] UKHL 1; [1979] AC 264 (CA).

[104] [1978] UKHL 1; [1979] AC 264 (HL(E)).

[106] R v Roussety [2008] VSCA 259, [57] (Ashley JA).

[107] Liberato v R [1985] HCA 66; (1985) 159 CLR 507.

[108] Ibid [515]-[517], [519]-[521].

[110] [1998] HCA 2; (1998) 193 CLR 1.

[112] R v Uhrig (Unreported, New South Wales Court of Criminal Appeal, Hunt CJ, Newman and Ireland JJ, 24 October 1996).

[113] [1999] VSCA 27; [1999] 2 VR 123.

[114] (1969) 1 QB 299, 307.

[115] R v Rajakaruna [2006] VSCA 277; (2006) 15 VR 592, 612-3 [71]-[72]; The Queen v KDY (2008) VSCA 104 [38]-[41].

[116] Stokes v The Queen [1960] HCA 95; (1960) 105 CLR 279 [284]-[285] (Dixon CJ, Fullagher and Kitto JJ; McIvor v The Queen [2006] HCA 34; (2006) 227 CLR 373 [405] (Gleeson CJ, Gummow, Haydon and Crennan JJ); Bounds v The Queen [2006] HCA 39; (2006) 228 ALR 190.

[117] Nicholls v R [2005] HCA 1; (2005) 219 CLR 196 [52] (McHugh J) [170] (Gummow and Callinan JJ) [205]-[206] (Kirby J); R v Harrington (1998) 3 VR 531 [539]; R v Lawrence [2001] QCA 441; [2002] 2 Qd R 400, 410.

[118] [1965] AC 595.

[119] R v Taylor (No 2) [2008] VSCA 57; (2008) 18 VR 613.

[120] [2007] HCA 12; (2007) 231 CLR 260.

[121] (2008) VSCA 259.

[123] [1973] HCA 30; (1973) 129 CLR 460.

[124] [1989] HCA 60; (1989) 168 CLR 79.

[125] (2008) VSCA 104 [26]-[27].

[127] [2008] VSCA 57; (2008) 18 VR 613.

[128] [2007] HCA 12; (2007) 231 CLR 260.

[129] [2008] VSCA 259, [57].