Appeal - Criminal law and procedure - Evidence - Warnings - Longman warning - Children - Inconsistency - Mental state - Drug use - Sleeping - Delay

Evidence Act 1906 (WA), s 36BD

1 This is yet another appeal against conviction which is based, to a significant degree, upon the alleged failure of the trial judge to give, in a sexual assault case, warnings which it is asserted were necessary in order to avoid a perceptible risk of a miscarriage of justice. The warnings which it is alleged that the trial judge in this case should have given, but did not give, were referred to by counsel for the appellant by the shorthand term of "Longman" warning (Longman v The Queen [1989] HCA 60; (1989) 168 CLR 79). That term is often apt to lead to confusion. I should therefore begin by noting that, in the present case, the one thing which it is not suggested that the judge failed to do, is to warn about the forensic disadvantage faced by an accused person where there is a significant period of time which has elapsed between the date of the alleged offences and the accused first becoming aware of a complaint about them. There was a significant delay in this case, and it is accepted that her Honour explained to the jury that forensic disadvantage with which Longman was concerned.

2 However, the expression "Longman warning" was used by the appellant in the present case as a shorthand for the warning(s) which should have been given, because Longman made it clear that, despite the prohibition of any practice of warning that sexual assault complainants as a class are unreliable, there remains in every criminal trial a requirement that a trial judge give a warning wherever necessary to avoid a perceptible risk of a miscarriage of justice arising from the circumstances of the case. It seems that it is often in cases of a sexual nature that courts have perceived particular circumstances which require such a warning, but the principle extends beyond that category of cases.

3 Given that the task of assessment of the facts is one for the jury, a question obviously arises as to how, as a matter of principle, a "warning" to the jury concerning factors which they must take into account, and directing them that they must scrutinise the evidence of a witness with care in the light of those factors, can be justified. In the cases of White v The Queen [2006] WASCA 62 [70], [78] and Winmar v The State of Western Australia [2007] WASCA 244 [21] - [22], this court examined authority and concluded that, as a matter of principle, it appeared that the need to give a warning arises where two factors are present. First, there must be some circumstance which makes it likely that evidence will or may be unreliable. Second, the risk inherent in the evidence must arise from a factor of which the courts have special knowledge, experience or awareness, so that the jury would not appreciate the risk without a judicial warning. As to the second factor, alternatively, it may be sufficient if, notwithstanding that the jury has the ability to understand and assess the risk, the particular circumstances of the case are such that the jury may be led to overlook it, or to place insufficient weight upon the potential risk. A similar view of the effect of relevant authority seems to have been taken in Victoria: see R v Mazzolini [1999] VSCA 150; [1999] 3 VR 113 [55] (Ormiston JA), R v Miletic [1997] 1 VR 593, 606 (Winneke P, Charles and Callaway JJA).

4 While the principle can be readily stated, it cannot be so readily applied. It is a principle which gives rise to difficulty and confusion, not only for trial judges, but also for courts of appeal. A search of a database of Australian decisions in the 18 years since Longman was decided throws up hundreds of cases grappling with the question of whether a warning was or was not required in the light of particular circumstances. Trial judges are, of course, generally alive to the need to give a warning where it is necessary to avoid a perceptible risk of a miscarriage of justice, and are anxious to ensure that no risk of a miscarriage of justice arises in the cases over which they preside, so that the cases do not as a rule stem from any overlooking of relevant principle. The difficulty is not one of principle, but of fact (see, for example, the differing views expressed in Tully v The Queen [2006] HCA 56; (2006) 81 ALJR 391 [57], [87] and [91], [132], [151], [186]).

5 In my experience, it is rare for counsel who argue that a warning should have been given to explain the reasoning process which would lead to the conclusion that a warning was necessary. Rather, it is usually simply suggested, without further analysis, that some feature of the case under discussion bears some factual similarity to a decided authority. The High Court has noted that this approach is unsatisfactory: Tully [53] (Kirby J), [131] (Callinan J).

6 In the case of Winmar, this court considered, in relation to identification evidence, the problem of identifying aspects of evidence about which courts have special experience or expertise. The reasons in that case note that there is a danger, when judges attempt to assist juries by warning them about particular aspects of the evidence, that judges themselves may be basing their views upon their own misapprehensions of "general" experience, or of human psychology, or of the state of scientific understanding. That danger is increased by the tendency, in appeals of this kind, for counsel to put forward a "grab bag" of "factors" which may bear upon the reliability of the evidence, without any coherent explanation of their significance, in an attempt to persuade either the trial judge, or an appellate court, that these are factors which common sense, or universal judicial experience, demonstrate must always affect reliability. That has been the approach adopted in this case. It is an approach which, if successful, creates unnecessary and undesirable uncertainty in the conduct of criminal trials. It is in an attempt to reduce uncertainty of that kind, that I make the following observations.

7 It appears to me that, as a general rule, it will not be possible to demonstrate that a trial judge was in error in failing to warn a jury about a particular factor or combination of factors unless counsel can demonstrate one or more of the following:

(1) that binding authority requires a warning to be given in respect of that, or of those, factor(s), either in every case, or in cases which relevantly are similar to that in issue;

(2) that logic, or repeated experience demonstrates that some aspect of the evidence must give rise to a real risk of unreliability (that is, it will not be enough simply to make the assertion) and that, for reasons which can be clearly articulated in the circumstances of the particular case, a jury would not have been able to appreciate, or would have been likely to have overlooked, the relevant risk;

(3) that scientific works of authority demonstrate a clear consensus of view to the effect that evidence of a particular kind carries inherent risks, and that there is some reason for supposing that a jury may not appreciate those risks.

8 Against that background, I turn to describe the course of evidence at this trial, and then deal with the grounds of appeal and the appellant's submissions in support of them (which traversed a number of matters not encompassed within the grounds, but did so without objection by the respondent).

9 The appellant, FGC, was, it is common ground, a respected member of the Noongar community. The complainant was his granddaughter, who was 22 years of age and a full-time university student at the time of trial. She said that, as a child, together with her older sister, she would stay with her grandmother and FGC on a regular basis, between pre-primary school and about grade seven. It appears that she stayed there overnight from time to time.

10 The first count the jury had to consider was count 2 in the indictment. That was because, at the completion of the evidence, the learned trial judge drew to counsel's attention the fact that it appeared that count 1 charged an offence which was not known to the law at the time at which it was alleged to have occurred. That count was withdrawn from the jury.

11 In relation to count 2, the complainant said that, when she was about 5 years of age, she stayed at the appellant's home, sleeping in a spare bedroom. At around 9.30 one night, the appellant fetched her from her bed and they walked to his bedroom, where her grandmother was asleep in the bed. Her grandmother was ill at the time. It appears that her grandmother's illness lasted for some time. The appellant told the complainant to remove her clothes, as she would not require them in bed. She got into the bed with the appellant. After some cuddling and open-mouthed kissing of her, it was said that count 2 occurred when the appellant placed his finger in her vagina. She said that this hurt and that he removed his finger and returned to cuddling her. She remained in the bed for a time after this incident and the appellant then took her back to her room. The following morning, she felt a distinctive pain in the vaginal area when she went to the toilet.

12 Counts 3 and 4 occurred after an occasion which the complainant particularly recalled because she had been driving around the Brookton area with an auntie who had to be taken to hospital for treatment because she had a "bad fit". In due course, the auntie took her back to the appellant's house some time after 6 pm. The complainant's grandmother had recently died. The complainant slept in the appellant's room. They had some conversation about whether the complainant loved him, and she told him that she did because she did not want to hurt or upset him. Again, the appellant commenced by cuddling and touching her, and there was a touching of the breast which constituted count 3, and a rubbing of the clitoris which constituted count 4. The complainant also gave evidence that there was on that occasion penile penetration, but that was not charged in the indictment and was dealt with by her Honour as relationship evidence.

13 Counts 5, 6 and 7 were alleged to have occurred when the complainant was about eight or nine years of age. She was staying in the spare room at the appellant's house. She recalled that the appellant appeared to be upset and depressed during that evening. He came into the spare room where she was staying at about 9 pm and kissed her on the lips for a few moments (count 5). He told her that he would return for her later and did so between half an hour and an hour later. In the meantime, she had fallen asleep, but he woke her and led her to his bedroom, where she got into bed with him. Counts 6 and 7 relate to allegations of touching of the vaginal area and penile penetration on that occasion.

14 It was the complainant's evidence that there had been other sexual touching, so that she was "sort of ... trained" to comply with the appellant's wishes. She said that she felt sorry for him on the occasion of counts 3 and 4, because he had recently lost his wife, and that, as a small child, she had been taught to respect her elders, which she said was particularly important for a Noongar woman. She did not mention to anyone what had occurred at first because, at a young age, she thought that was the way a grandfather "shared his love" for his granddaughter. However, in about year six, she saw at school a video entitled something like "Kids Have the Right to Say No". That, in effect, confirmed her suspicions that her grandfather's behaviour was inappropriate.

15 The complainant was cross-examined about a number of issues. In relation to the unlikelihood of her grandfather behaving in that way with her grandmother asleep in the bed, she said that she had no doubt that her grandmother was awake for "many of the counts that may have happened", but she said that she knew why her grandmother never said anything, or reacted in any way. She was not asked what that reason may have been; rather, it was put to her that she was merely speculating, which she denied. It was put to her that she was "wanting to make a statement about this on behalf of Noongar women". She denied that. Concerning certain internal inconsistencies in her evidence, she said that, because there had been many attempts to make her drop the case and because of family pressures, she found the trial a "very scary experience". She was cross-examined at some length about discrepancies between an earlier and later police statement, particularly about the question of whether she had told her mother about the abuse when she was in about grade seven at primary school. Her explanation was that she did tell her mother, but that, as it was her view that her mother would never admit to being told, she removed that information from the later statement.

16 It was the complainant's evidence that she had told a female school psychologist, whose name she did not recall, about the abuse, and that she told a male school chaplain things about the appellant, but did not specifically inform him of the abuse. There was some cross-examination, to which I will turn in more detail later, about the complainant having been a voluntary patient at the Alma Street Clinic, having had an addiction to Valium at one point, having attempted self-harm by cutting her arms on many occasions, and about some drug use. There was cross-examination about why the complainant had not complained of the abuse at the time at which it occurred, or shortly thereafter. The answer to one of those questions is the subject of two grounds of appeal.

17 The complainant's mother confirmed that the complainant and her older sister would frequently stay with the appellant, and with the appellant's wife until the latter died. She said that, for eight to 10 years, she was comfortable with the arrangement, which involved her dropping the children off from time to time so that she and her husband could attend to other matters. It was put to her in cross-examination that at no time did she have concerns, from what she saw or heard, about the safety of the children when they were with their grandparents. Her answer was, "Not at that time, no". When she was asked in cross-examination whether she had noticed anything about the demeanour of either girl, she said that the complainant had behavioural problems "right through from primary school". She confirmed that the complainant had seen a school psychologist. It was her evidence that she knew at the time that the complainant was unhappy and that the complainant would say things like, "I hate myself, I'm ugly".

18 She described an occasion when the complainant was about 18 years of age, by which time the complainant had been living independently of her parents for some four years. The complainant asked for money to pay for books and her parents offered to buy the books rather than to provide her with the money. About half an hour to three-quarters of an hour later, the complainant became angry and said to the complainant's father, words to the effect of, "Your father raped me and he's a paedophile". Perhaps importantly for two of the grounds of appeal, it was put to the complainant's mother that this was the first time that she had heard anything from the complainant about any allegation of improper behaviour by the appellant. Her response was:

There was things said earlier without any substance. That's why we didn't take it into consideration, but when the word 'paedophile' came out and that she'd been raped, that was the - I mean, that was - to us that was the - it was the words that we needed to hear that that's what actually did happen ... We'd heard things before in the past but we didn't - we wanted to be sure. You just don't ... (ts 92)

19 There was a somewhat inconclusive cross-examination about whether the complainant's mother had said anything along those lines in her police statement, and the flavour of her evidence seems to have been that, whether the statement said so expressly or not, she had been meaning to allude to concerns which she had earlier harboured, when she referred in her statement to certain discussions with her nephews and nieces. Cross-examination on this point concluded with the proposition being put to the complainant's mother that the complainant had not said anything to her about the complainant's grandfather at any time prior to the incident described. Her response was, "Not directly, no".

20 The complainant's father gave evidence. He was the appellant's son. He recalled a time when the complainant first told him that the appellant had raped her. He recalled where it was, but did not otherwise recall the surrounding circumstances.

21 A school chaplain from the school the complainant had attended gave evidence. He recalled the complainant seeing him formally a couple of times and having some informal conversations with her. It was his evidence that there was no disclosure by the complainant of "inappropriate conduct" by the appellant, but he said that there was "just a comment about feeling uncomfortable about certain behaviours and things that were said".

22 One of the investigating police officers was called. It was his evidence in cross-examination (given without objection) that the appellant had, when interviewed, said he was innocent, but had declined to answer further questions. He said he had located the chaplain who had given evidence, but that he had not attempted to locate the school psychologist.

23 The appellant gave evidence. He had been prominent in indigenous issues for a considerable time. He had 10 children and about 30 or so grandchildren. He denied that the complainant had ever shared a bed with him and his wife. She had certainly never shared a bed with him. It was his evidence that none of the grandchildren stayed with him after his wife had died, because they were frightened of the grandmother's spirit, that being an aspect of Aboriginal culture. The complainant and her sister did stay with him after he remarried and lived in a different house on one occasion. The complainant and her sister did not come around to see him at all after she was in about grade seven, and he had only brief contact with the complainant's mother, his daughter-in-law, after that time. The only time he had physical contact with the complainant at all was when she was about seven or eight years of age and he rubbed her head and shoulders in an effort to extract the "evil spirits" from her, since her mother had said that the complainant "had the devil in her".

24 At the conclusion of the evidence, her Honour specifically raised with counsel her view that it was an appropriate case for a Longman warning, and advised counsel of her intention to give a warning about the need to scrutinise the complainant's evidence with care because of the passage of time. She also canvassed briefly with counsel the question of delay in complaint, and the use of relationship evidence. She invited counsel to raise any issues which were of concern to them, and as a result of certain submissions by counsel for the appellant, her Honour noted that she would be directing the jury that they could not infer, from the abrupt cessation of the complainant's contact with her grandfather, that the reason for that cessation had been the complainant's sexual abuse, and also that they could not infer that the personal difficulties (involving self-harm attempts and so on) from which the complainant suffered were the result of sexual abuse. She did, in due course, direct the jury in those terms.

25 Neither before nor after her Honour's direction to the jury did counsel suggest to her that any of the matters which now form part of the grounds of appeal should be the subject of a specific direction.

26 The second of the grounds of appeal was abandoned at the hearing. The third and fourth are in the alternative, and concern the same issue. There are therefore, in substance, two grounds of appeal. However, the first has a number of particulars.

The learned trial judge erred in that the Longman direction was inadequate in that it did not point out the particular features of the complainant's evidence that might affect her reliability.

28 The particulars to this ground suggest that her Honour should have, in the course of a warning, directed the jury in relation to inconsistencies within the complainant's evidence and as against the evidence of other State witnesses, and in relation to the complainant's use of illicit drugs, addiction to Valium, and psychiatric illness. However, although those are the only particulars, the submissions in support of this ground (written and oral) went on to canvass, and to suggest that her Honour should have warned the jury about, a number of other matters. Those other matters are that the complainant had been woken from sleep relatively shortly before some of the counts, the complainant's young age at the time of the alleged offences, the film which the complainant said she had seen at school, and the complainant's delay in complaining of the offences. I deal with each of these issues separately.

29 The first observation which should be made about the proposition that a judge is required to warn a jury about "inconsistencies" in the evidence of a complainant, is that the proposition as formulated is far too broad, and is not compatible with the respective roles of judge and jury. Cross-examination of a complainant and other State witnesses usually has, as its central aim, that of exposing and exploring inconsistencies between the evidence of those witnesses and internal inconsistencies in the evidence of those witnesses. The assessment of the importance of any inconsistencies which may exist, and the evaluation of their significance for the credibility of the witness and for the overall strength of the State case, lies at the core of a jury's task. It is well within the range of ordinary human experience to consider the extent to which inconsistencies between what a witness says in court, and what a witness may have said to others on another occasion, stems from inaccurate recollection, untruthfulness, a simple lapse of memory about a relatively insignificant matter, confusion occasioned by the manner of questioning, reticence to talk to strangers about intimate matters, or the myriad other possible causes of inconsistencies in evidence.

30 It may be, of course, that inconsistencies are of such significance to the State case that a trial judge is required to refer specifically to them and to explain their significance. For example, in Robinson v The Queen [1999] HCA 42; (1999) 197 CLR 162, it appears that there was an inconsistency in some aspects of the complainant's evidence as to whether penetration had occurred [25]. As penetration was an element of the offence charged, it is not surprising that the view was taken that the trial judge in that case was required to refer specifically to that aspect of the evidence. That is far from the present case.

31 The inconsistencies pointed to in the present case, although described rather generally, are said to be inconsistencies between the complainant's evidence of complaint on a variety of occasions, and the evidence of those to whom complaint was made. There are three aspects of the alleged inconsistencies.

32 First, she said that she had told the school psychologist that her grandfather had sexually abused her. There was no evidence from the school psychologist. However, there was evidence from the investigating officer that he had not endeavoured to locate the psychologist. There was evidence from the complainant's mother that the complainant had seen a psychologist during her school years. There was, therefore, no relevant inconsistency in that respect.

33 So far as the school chaplain was concerned, the complainant's evidence in cross-examination was to the effect that she had not said "anything specific" to the school chaplain, although she agreed that she had, in her statement to police, told them that she had complained to the chaplain that her grandfather had sexually abused her. The chaplain's evidence was that there had not been a specific disclosure of abuse, but there had been comments about being "uncomfortable about certain behaviours and things that were said".

34 So far as the complainant's mother was concerned, the position was more complex. It appeared that the complainant had, in an earlier police statement, said that she had told her mother of the abuse, but that she had later changed the statement. Her explanation was that she believed that her mother would not admit that she had been told. In cross-examination, she explained what she saw as the difficulties in telling her mother and she said that her parents had "trouble to this day trying to deal with the fact that they never, ever saw what was going on". As I have noted, the mother's evidence was that the complainant had not "directly" complained of abuse, although she said that the complainant's parents had "heard things" in the past (whether from the complainant or from other family members was not clear).

35 There was, in my view, no reason for the trial judge to take the view that there were such inconsistencies in the complainant's evidence, concerning the matter of complaint, as to require any kind of warning. It would have been open to the jury to have formed the view that the inconsistencies in the complainant's evidence, in her mother's evidence, and between the evidence of the complainant and that of her mother, were such that neither the complainant nor her mother could be regarded as an accurate historian. However, it was equally open to the jury to form the view that the evidence demonstrated that the complainant, while she had not directly and explicitly complained of abuse, had demonstrated from an early age that she was not comfortable with the way that her grandfather behaved towards her. A detailed direction concerning this evidence would hardly have been likely to have assisted the appellant. A warning was not required.

36 The relevant particular refers to "[t]he complainant's use of illicit drugs, addiction to Valium and psychiatric illness". All of these were explored in cross-examination.

37 The evidence in relation to the complainant's use of illicit drugs was very scanty. There was hearsay evidence from the complainant's mother, without objection, to the effect that the complainant had admitted to her at one time that she was using illicit drugs. The complainant was asked about admissions to her mother and said that she had admitted to her mother that she had used speed or ecstasy. The admission was made "[a]round that time, possibly when I was young, dumb and experimental". The complainant said that she certainly did not "do that crap any more". It was put to her that she was a "substantial" user of speed and ecstasy and she denied that. Evidence of some unspecified use of illicit drugs at some time in the past, for what the complainant's evidence suggests was a relatively brief period, would not require a warning.

38 So far as the complainant's Valium addiction was concerned, she said in cross-examination that she had voluntarily admitted herself to the Alma Street Clinic because she had an addiction to Valium. The effect of that drug upon her was the subject of cross-examination. Her evidence was:

You said that under Valium the concept of time is a little hard?---At the time when I was addicted to Valium and I walked into Alma Street Clinic it wasn't on my mind to record the date.

No?---That's not an overall misconception of time or memory.

So time and memory?---I said it's not, meaning that my conception of time and memory is still like that, especially when I haven't been on those tablets for a very long time.

So your recollection of time and of events is affected by your previous addiction to Valium. Is that what you're saying?---No, that's not what I'm saying. You're twisting my words around.

No, I don't mean to do that. I don't understand what it is you're conveying?---I think I have made it quite clear. At the time that I had the addiction to Valium, which was now almost two to three years ago, it was a weird - the conception of time and stuff was a bit obviously muddled. Everything was muddled. That's why I admitted myself to Alma Street, because I wanted to kill myself. I couldn't deal with it any more. No-one would help me. No one would support me.

You said your conception of time and stuff was muddled. What are you referring to when you use the word 'stuff'?---Dates, times, at that particular time.

And events?---At that particular time.

And events?---It doesn't mean that I fabricated any of this in my own head. (ts 75)

39 This was not a case in which "dates and times", her memory for which she said had been affected, were of critical importance. Although the complainant's evidence was also that Valium had affected her memory for events at the time of her addiction, that well post-dated the events the subject of the indictment. There was no medical evidence concerning the use of Valium, either as prescribed or to excess. That is not a subject about which trial judges have particular experience or knowledge. Had her Honour warned the jury that the Valium addiction might have made the complainant's recollection or evidence unreliable, in respects other than those indicated by the complainant, she would have been presenting to the jury as fact a proposition which was completely unsubstantiated and which may or may not be correct.

40 So far as self-harm was concerned, it was the complainant's evidence that she had attempted to harm herself by cutting her arms, that she had tried that on many occasions, and that she had probably done that more often since the trial had begun. It was also plain from other evidence, that the complainant had been troubled in her early teenage years. She had left home at the age of 14. She had lived a rather nomadic existence for a time, although she always kept in touch with her parents. Her mother's evidence had been that she had had behavioural problems through primary school and was plainly very unhappy for a long time.

41 Implicit, however, in the appellant's proposition that the jury should have been "warned" about these matters, is the proposition that the complainant's deep unhappiness, which may have amounted to a psychiatric illness (although there was no evidence of any formal diagnosis) would or could have made her untruthful or inaccurate in her recollection of, or evidence concerning, sexual abuse. There is no evidence supporting this proposition. No logical argument is advanced in support of it. To the extent that general experience may be a guide, it is a matter of general knowledge that depression in one form or another is a very common form of illness in the general population. If conditions of this kind caused inaccuracy or untruthfulness, one would expect that to be the subject of general knowledge. I have found only one reference concerned directly with the relationship, if any, between psychological trauma and sexual abuse; it suggests that the severity of post-traumatic stress disorder is positively, rather than negatively, correlated with accuracy of memory: Alexander et al "Traumatic Impact Predicts Long-Term Memory for Documented Child Sexual Abuse" (2005) 16(1), Psychological Science 33.

42 To the extent that it may be relevant to consider matters of which judges have particular experience, it is to be noted that her Honour and both counsel had, apparently, only one concern with the aspect of the complainant's evidence which dealt with her troubled adolescence and attempts at self-harm. That concern (which resulted in an appropriate direction) was that the jury might infer from those facts that they had been caused by sexual abuse, thereby rendering the complainant's account of what had occurred more probable (ts 139-141). That is no doubt because illicit drug use, running away from home, suicidal thoughts, and attempts at self-harm are precisely the sorts of symptoms which are often seen resulting from sexual abuse in childhood: see VIM v The State of Western Australia [2005] WASCA 233; (2005) 31 WAR 1 [291]. See also Browne and Finkelhor, "Impact of Child Sexual Abuse: A Review of the Research" (1986) 99(1) Psychological Bulletin 66; Mullen et al, "Childhood Sexual Abuse and Mental Health in Adult Life" (1993) 163 British Journal of Psychiatry 721; Pettigrew and Burcham, "Effects of Childhood Sexual Abuse in Adult Female Psychiatric Patients" (1997) 31(2) Australian and New Zealand Journal of Psychiatry 208.

43 It would be surprising, therefore, if a warning were required in every case in which a complainant had some signs of emotional instability since sexual abuse in childhood very frequently results in significant psychological consequences. The complainant who says, for example, that she was sexually abused by her grandfather from a very early age, but that it had had no effect upon her whatever, would be, experience suggests, an unusually resilient person.

44 So far as authority is concerned, there are only two cases which may be relevant. The first is Bromley v The Queen [1986] HCA 49; (1986) 161 CLR 315. Bromley concerned, relevantly, the evidence of a witness who had schizophrenia and who had said during the course of cross-examination things which plainly could not have been correct; for example, he said that he thought he had seen the devil (319). In that case, Gibbs CJ, with whom Mason, Wilson and Dawson JJ agreed, said:

If it appears that a witness whose evidence is important has some mental disability which may affect his or her capacity to give reliable evidence, common sense clearly dictates that the jury should be given a warning, appropriate to the circumstances of the case, of the possible danger of basing a conviction on the testimony of that witness unless it is confirmed by other evidence. The warning should be clear and, in a case in which a lay juror might not understand why the evidence of the witness was potentially unreliable, it should be explained to the jury why that is so. (319)

45 Those observations are inapplicable to the present case. The complainant did not have a mental disability. Her drug use was in the past. While she had apparently recently attempted self-harm, that was not, as I have already explained, a matter which it appears to me the judge should have considered might affect her capacity to give reliable evidence. It is significant that in the passage I have quoted, Gibbs CJ discussed the need for the judge to explain to the jury why the evidence of a witness may be potentially unreliable. In the present case, I note that, although invited to do so, counsel for the appellant did not explain to us why it was that the complainant's attempts at self-harm indicated that her evidence might be unreliable.

46 This is not, in my view, a case in which the reasons of the majority in Bromley have application. Rather, it appears to me that it is one calling for the application of the remarks of Brennan J. His Honour dissented in the result, but his views appear to be referable to a different view of the facts, rather than any different application of principle. His Honour observed:

... the courts have no scientific knowledge about the danger of acting on evidence given under the influence of mental disorders of various kinds and degrees, nor have they acquired any experience of the danger of acting on evidence given by persons suffering from a mental disorder wider than the experience of the general public. (324)

47 Fleming v The Queen [1998] HCA 68; (1998) 197 CLR 250 was concerned as a matter of principle with the question of whether, at a trial by judge alone pursuant to the Criminal Procedure Act 1986 (NSW), a judge was required in his or her reasons to refer to any warnings about the reliability of the complainant's evidence which it would have been necessary to have given to a jury. The court held that it was necessary for a judge to do so. Most of the decision is concerned with the discussion of that question of principle, and the report of the circumstances of the case which were said to give rise to the warning is very brief.

48 All that the report discloses of the relevant facts, is that the complainant was a 15-year-old female pupil at the school where the appellant was deputy principal. She had some emotional problems arising from an eating disorder and a stressful family situation, which she discussed with him relatively frequently. It appears that the prosecution case was that she had been infatuated with him and had been a willing participant in each of the relevant alleged offences.

49 It is not clear from the report whether or not the Crown accepted, for the purposes of the appeal, that a warning of some kind should have been given; the respondent's submission, recorded at [36], suggests that the Crown conducted the appeal on the basis that a warning had been given, rather than that none was required. While, therefore, it appears, reading [36] and [38] together, that the High Court considered that a warning was required in that case by reason of the complainant's "age, emotional instability and infatuation with the appellant", there does not appear to be any statement of more widely applicable principle suggesting that a warning is required in every case in which a complainant's evidence demonstrates some emotional instability. One fact that does emerge, which distinguishes Fleming from the present case, is that in Fleming it appears that the emotional instability pre-dated the alleged abuse. That does not appear to be so in this case. I would not regard Fleming as establishing any general principle which is applicable wherever a witness may disclose emotional difficulties, irrespective of their onset or particular characteristics.

50 So far as the relevance of waking from sleep is concerned, consideration of the authorities begins with Longman.

51 In that case, the complainant's evidence was that she had been asleep on each occasion and was awakened by the appellant's touching of her genitalia, but that she pretended to remain asleep. The majority (Brennan, Dawson and Toohey JJ) noted:

There were several significant circumstances in the case: the delay in prosecution, the nature of the allegations, the age of the complainant at the time of the events alleged in the two counts in the indictment, the alleged awakening of a sleeping child by indecent acts and the absence of complaint either to the applicant or to the complainant's mother. It would not have been surprising if these circumstances had elicited some comment from the trial judge, for it would have been proper to remind the jury of considerations relevant to the evaluation of the evidence. (90) (Emphasis added)

52 Deane J stated that "[t]he possibility of child fantasy about sexual matters, particularly in relation to occurrences when the child is half-asleep or between periods of sleep, cannot be ignored" (101) (emphasis added). McHugh J, having expressed his opinion that recollection of events which occurred in childhood is particularly susceptible to error, went on to note:

The likelihood of error was increased by the circumstances in which the complainant said the incidents occurred. The opportunity for error in recalling, twenty years later, two incidents of childhood which are alleged to have occurred as the complainant awoke, and then pretended to be asleep, are obvious. (108)

53 The majority view, in Longman, was that while a comment on the complainant's awakening from sleep would not have been surprising neither comment nor warning was required. Longman, then, is contrary to the appellant's submission that some warning is required where a complainant is, or has been, sleeping. The views of Deane and McHugh JJ appear to be influenced by factors peculiar to the recollections of children, with which I deal shortly.

54 Robinson was another case in which a complainant had been sleeping immediately prior to the commission of the alleged offence. The complainant was aged eight at the time of the offences. As I have noted earlier, there were in that case inconsistencies in the complainant's account of whether penetration had actually occurred. Some of these inconsistencies, which were summarised in the reasons of Lee J in the Court of Appeal (Robinson (1998) 102 A Crim R 89, 126) may have been attributable to the complainant's confused state on waking from sleep. In that case, then, the deficiency in the evidence concerning an element of the offence was potentially linked to the complainant having been sleeping.

55 In Doggett v The Queen [2001] HCA 46; (2001) 208 CLR 343, the complainant said that, after waking from a bad dream, she went to the appellant's room and informed him that she was scared. The appellant told her to get into bed with him, which she did. She awoke later to find the appellant fondling her. The complainant went back to sleep and was awakened once more by the appellant doing a similar thing. These events comprised counts 1 and 2. There were five other counts the subject of the appeal. The appeal was allowed by a majority on the basis that a Longman direction ought to have been given (Gaudron, Kirby and Callinan JJ; Gleeson CJ and McHugh J dissenting).

56 One of the reasons which Gaudron and Callinan JJ considered gave rise to the need for a Longman direction was that "the complainant's evidence with respect to the circumstances surrounding the first three counts made the point that the complainant's recollection of some matters was, to say the least, questionable" [47]. It does not appear that the complainant having been awoken from sleep was one of the "circumstances" to which their Honours alluded, since count 3 was alleged to have occurred in quite different circumstances, some three months after the first two counts and seemingly while the complainant was fully awake. It is more likely that the "circumstances" to which their Honours were referring related to confusion in the complainant's evidence as to the timing of the first three counts: see R v Doggett [1999] QCA 441 [4] - [7]. Kirby J, in the context of a discussion about honest but erroneous memory, quoted Deane J's comments about "[t]he possibility of fantasy about sexual matters, particularly in relation to occurrences when the child is half-asleep or between periods of sleep" [124], although in the course of that discussion, his Honour did not expressly refer to the fact that the complainant had been asleep at the time counts 1 and 2 were alleged to have occurred. The complainant having been asleep was not, it would appear, a factor of particular significance in that case.

57 The only other relevant High Court case touching this issue appears to be Tully. One of the counts the subject of that appeal was alleged to have arisen when the appellant woke the complainant and persuaded her to go to his bedroom where he pushed her onto the bed and asked her to have sex with him. The majority (Callinan, Heydon and Crennan JJ) dismissed the appeal on the basis that a Longman warning was not necessary. Crennan J noted that the appellant had not contended that the warning was imperative "because events occurred when the complainant was in the twilight state between waking and sleeping" [185], but her Honour expressed no view about the relevance of that factor, had it existed. In their respective dissenting judgments, Kirby and Hayne JJ considered that a Longman warning ought to have been given on a number of bases, none of which appear to have related to the fact that the complainant had been asleep immediately before the sexual penetration was alleged to have occurred [57], [87], [91].

58 It appears from these authorities that a warning detailing the dangers of relying on a complainant's testimony as to incidents that occurred while the complainant was "half-asleep or between periods of sleep" may only be required if there is some inconsistency in the complainant's evidence as to what actually happened, or some other factor of significance which might detract from the credibility of the testimony and to which the complainant having been asleep may be linked. Even where there are such factors, of course, much will depend upon their significance in the context of the whole of the evidence, the way in which the trial was fought, and the possibility that a jury may overlook them.

59 This case is factually much more similar to Tully than to any of the other cases. It was the complainant's evidence in this case that on the relevant occasions she had been woken, and had then been taken to the appellant's bedroom, where she had undressed and got into bed. The acts alleged had then occurred. This was not, on any view, a case of a child who was asleep or half-asleep at the time at which the alleged abuse occurred.

60 In Longman Brennan, Dawson and Toohey JJ considered that the age of the complainant at the time of the events alleged in the two counts in the indictment (namely, six and 10 years of age) was, in combination with other factors such as the length of the delay, a "significant circumstance" which might have elicited some comment (not warning) from the trial judge to remind the jury of considerations relevant to the evaluation of the evidence (90). However, their Honours did not explain why the complainant's age was a factor which might have elicited a comment. For Deane J, who considered a warning necessary, the significance of the complainant's age appears to have been linked to the view that "[t]he possibility of child fantasy about sexual matters ... cannot be ignored" (101). I discuss this notion of child fantasy below. By contrast, McHugh J appeared to suggest that age was significant in the context of assessing the risks associated with the passage of time:

The longer the period between an 'event' and its recall, the greater the margin for error ... Recollection of events which occurred in childhood is particularly susceptible to error and is also subject to the possibility that it may not even be genuine: Hunter, Memory, rev. ed. (1964), pp 269-270.

... The jury should have been warned that, in evaluating her evidence, they had to bear in mind that it was uncorroborated, that over twenty years had elapsed since the last of the alleged offences occurred, that experience has shown that human recollection, and particularly the recollection of events occurring in childhood, is frequently erroneous and liable to distortion by reason of various factors. (107 - 108) (Emphasis added)

61 In Doggett, Kirby J agreed with the views of Deane and McHugh JJ concerning the fallibility of memory [126] - [127].

62 In B v The Queen [1992] HCA 68; (1992) 175 CLR 599, the court was invited to consider whether a trial judge was obliged to give a warning pointing out the dangers of acting on a child's evidence. In their joint judgment Dawson and Gaudron JJ highlighted some of the considerations said to be relevant to warning jurors about the uncorroborated evidence of children:

The fact that young children may be under the influence of others and are apt to allow their imaginations to run away with them and to invent untrue stories is one justification which has been advanced [See R v Dossi (1918) 13 Cr App R 158 at 161]. Again, it has been said that the warning is required in relation to 'children who, though old enough to understand the nature of an oath and so competent to give sworn evidence, are yet so young that their comprehension of events and of questions put to them or their own powers of expression may be imperfect' [DPP v Hester [1973] AC 296 at 325]. (616)

63 It should be noted that their Honours did not expressly endorse any of the justifications advanced for requiring such a warning, and that they did not consider that a warning would have been mandatory in that case by reason of the complainant's age (11 at the date of the earliest offences). It should also be noted that the reasons referred to by their Honours would appear to be inapplicable to an adult who gives evidence, even if the evidence concerns events which occurred when he or she was a child.

64 In Robinson, the High Court observed that the reasons for the warning discussed in both Longman and B "included what are now rejected as 'stereotypical assumptions'" [19]. In support of that proposition, their Honours cited R v Ewanchuk [1999] 1 SCR 330, 336 in which the Supreme Court of Canada said that "[c]omplainants should be able to rely on a system free from such myths and stereotypes, and on a judiciary whose impartiality is not compromised by these biased assumptions." Nevertheless, the High Court in Robinson held that there were a number of particular features of the case which demanded a suitable warning; among these was the age of the complainant, who was eight at the time of alleged offences and 11 at the time of trial [25]. While their Honours did not explain why age was relevant in that case, it may be that the court considered that some of the material inconsistencies in the complainant's evidence in relation to whether there was penetration (which I have noted earlier) were attributable to his youthfulness, either alone or in combination with other factors.

65 In Doggett, Gaudron and Callinan JJ appeared to suggest that a complainant's age at the time of the offences may be relevant to the extent that it, coupled with the passage of time, might detract from a complainant's ability to recall details of the offences:

[W]e will identify and discuss the features of the case which, the appellant argues, required that a Longman direction be given.

First, there is the fact that the complainant was only eight years old when the first and second offences are alleged to have been committed. Whilst it may be accepted that, of course, in adulthood people may have vivid and clear recollections of events which occurred when they were children, the evidence in this case itself demonstrates the truism that accounts of events remote in time need to be carefully scrutinised. [42] - [43]

66 It appears, then, that their Honours considered that the age of the complainant at the time of the offending was relevant, not because of any unreliability peculiar to children, but because of the difficulty of recalling, on occasion, the detail of events occurring a considerable time before. In that case, their Honours noted that the complainant's recollection of some matters appeared "questionable" [46] - [47].

67 By contrast in Tully, the majority (Kirby and Hayne JJ dissenting) held that a warning was not required in a case in which the complainant was aged nine and 10 at the time of the relevant offences.

68 In my respectful view, the principal thrust of the authorities discussed above is that a warning may be required only if there is something in the facts of the case (in addition to the complainant's age) which suggests that a complainant's age at the time of the offences has impaired his or her ability to recall relevant matters accurately. So far as the observations of Deane and McHugh JJ in Longman are concerned, I would respectfully agree with the conclusions of Spigelman CJ in JJB v The Queen [2006] NSWCCA 126; (2006) 161 A Crim R 187:

... The observations by Deane J and McHugh J are just that - observations. They have never been given authoritative force either from the High Court or from any intermediate court of criminal appeal.

Their Honours' observations are based on assumptions about child psychology which are widely held but which are not necessarily well founded. Many judges share a conventional wisdom about human behaviour, which may represent the limitations of their background. This has been shown to be so in sexual assault cases. (See R v Johnston (1998) 45 NSWLR 362 at 367-368.)

Legislative intervention was required to overcome the tendency of male judges to treat sexual assault complainants as prone to be unreliable. The observations of Deane J and McHugh J in Longman reflect a similar legal tradition that treated children as unreliable witnesses. In the past both categories of witnesses required corroboration.

Jurors may also reflect these widely held assumptions about children, as they may also do about sexual assault complainants. Such prejudices may be reinforced by the profession and the bench in the conduct of a criminal case. (See Quas JA et al 'Do Jurors "Know" What Isn't So about Child Witnesses?' (2005) 29 Law and Human Behaviour 425.)

There is a significant debate as to whether expert evidence should be admissible about the ability of children to give accurate evidence, especially in child sexual assault proceedings. See, most recently, Uniform Evidence Laws Report ALRC Report 102, NSWLRC Report 112, VLRC Final Report, December 2005 at 9.138-9.158; Criminal Justice Sexual Offences Task Force, Responding to Sexual Assault, Final Report, Sydney, December 2005, pp 165-176. These two recent reports refer to a range of earlier studies and reports. They also outline the legislation that already exists in some jurisdictions to permit such evidence and make recommendations for further legislative intervention.

There is a substantial body of psychological research indicating that children, even very young children, give reliable evidence. (See eg the references in Ligertwood Australian Evidence (4th ed) LexisNexis Butterworths, Australia, 2004 para 7.31 fn 10 and 11.) These are complex issues, as reflected in reviews of the research on the ability of young children to distinguish fantasy from reality (see Woolley JD 'Thinking About Fantasy: Are Children Fundamentally Different Thinkers and Believers from Adults' (1997) 68 Child Development 991; Taylor M 'The Role of Creative Control and Culture in Children's Fantasy/Reality Judgments' (1997) 68 Child Development 1015; Sharon T and Woolley JD 'Do Monsters Dream? Young Children's Understanding of the Fantasy/Reality Distinction' (2004) 24 British Journal of Developmental Psychology 293 at 294-296). The same is true of research about a child's ability to accurately recall stressful events (see McNally RJ, Remembering Trauma, Harvard Uni P, Cambridge Mass 2003 at pp 58-62).

The complexity of these issues is not reflected in the observations of Deane J and McHugh J in Longman, which should, accordingly, be treated with caution. [2] - [8]

69 To the extent that the need to warn about the evidence of child witnesses stems in part from a view that children may fantasise about "sexual matters" and may then confuse that fantasy with reality, I would add only the following brief observation, which appears to be relevant to the present case. The expression "sexual matters" is apt to interfere with logical analysis, covering as it does a very wide variety of types of matters. In Longman itself, for example, the acts which Deane J considered could potentially have had some source in childhood fantasy, involved no more than the touching of the young girl's external genitalia. Even if one accepted that a female child might fantasise about matters of that kind, it is a very different proposition to consider that a child would fantasise about the type of penetration which was the subject of count 7 in this case. That is for obvious physiological reasons. A young girl's external genitalia are readily accessible and it is probable that most young girls at some stage will discover that touching of that area is pleasant. By contrast, a young girl's vagina is not readily accessible, and it is unlikely that she will have any reason, based in experience, to consider that inserting anything into it is likely to be a pleasant experience. Even if she has had some education in sexual matters, so that she appreciates intellectually how intercourse takes place, her basis for fantasising about those matters, grounded in any understanding of how she is likely to experience such an event, is plainly limited. To extrapolate from Deane J's observations in Longman, so as to require a warning in a case such as the present, would be illogical. I would add that it would, as a matter of authority in this State, be an error to direct a jury in terms such as "children do fantasise": Reference of a Question of Law (No 1 of 1999) [1999] WASCA 53; (1999) 106 A Crim R 408 [9].

70 Although this factor was not formally abandoned, it was clear during the course of argument that the appellant's counsel was not able to advance any reason why the film which the complainant had seen at school might have been a matter which affected the reliability of her evidence. He expressly disclaimed any reliance on the proposition that seeing a film dealing with sexual abuse could lead a girl to believe mistakenly that she had been a victim of sexual abuse herself. Indeed, put so bluntly, the proposition appears to be ridiculous. However, no alternative basis for a warning about the film was advanced.

71 In any event, I note that the fact of the production and showing of such a film in a school demonstrates that those experienced in dealing with young children consider that children may have difficulty either in understanding the unacceptability of sexual abuse, or in knowing how to complain about it. Having regard to the complainant's evidence in the present case that it took her some time to appreciate the wrongfulness of the appellant's conduct, and to complain about it, the film is hardly a matter upon which the appellant would have wished her Honour to have placed any emphasis.

72 As the issue of delay is contained in submissions in the appellant's case which are concerned with her Honour's alleged failure to give an adequate warning, I assume that it is (impliedly) submitted that it is necessary to warn a jury (rather than comment) that, in some circumstances (not specified in the submissions), a failure to complain at an early opportunity is a factor making it dangerous for a jury to act upon the evidence of the complainant. That view is far from being supported by the ordinary experience of the courts. It is a view which may be open upon one reading of Crofts v The Queen [1996] HCA 22; (1996) 186 CLR 427, but such a reading would be, in my respectful opinion, erroneous.

73 Turning first to the experience of the courts, it was noted in R v Mazzolini (Ormiston JA):

Many unfortunate cases of long delayed complaints have come before the courts in recent years and, more significantly, a very large number have led to guilty pleas, so that no questions there were raised as to the veracity of the complaints though many years delayed. [24]

74 This observation is borne out by the Victorian Law Reform Commission's Sexual Offences Interim Report of 2003, which noted that between 1994 and 2002, for offences such as incest and other offences against children involving penetration, almost one-third of reports related to assaults that occurred more than five years prior [2.35].

75 In similar vein, Kirby J noted in Doggett:

In recent years, there has been a significant increase in the number of complaints about sexual offences alleged to have been committed at a time when the complainant was a child. Many cases coming before this Court illustrate this trend. [99]

76 The experience in Western Australia is similar. I have examined the cases which are collected in sch A of VIM, being cases in which there was a plea of guilty. In many of those, it is not possible to tell when the offences occurred. However, in a significant number of them a report to any relevant authority must have been first made many years after the offending commenced, and in a number of cases many years after the period of offending had ceased. The time between the first offence and the complaint in the cases contained in that table appears to have ranged from one to over 20 years, with delays of the order of three to five years most common.

77 The greater judicial experience with long delayed complaints which recent decades affords, has in turn led to a better understanding, gleaned from evidence at trial and from victim impact statements, of why it is that children, in particular, may not disclose sexual abuse at the time at which it occurs. The many reasons commonly encountered include the following. Obviously, there may be threats (express or implied) of physical harm, of rejection, or of being sent away. There is fear of being disbelieved, especially if complaint might pit the child's word against that of an adult, with the resulting stigma of being thought to have lied about a matter of significance. There is a belief that adults may already know of the abuse (particularly where the offences take place in circumstances where discovery is likely), in which case it would be pointless to complain. There is a belief - or knowledge - that other children are being abused or have been abused, leading to a belief either that the behaviour is normal or that the offender is so powerful that there is no point in complaining. There is, very often, the desire to protect a parent from the upsetting reality, particularly if the abuser is a step-parent or grandparent.

78 Finally, and most importantly, the courts now have a better understanding of the way in which three important influences can interact. One is the sense of privacy and, to a degree, shame, which often accompanies discussion of sexual matters, so that a child is often reluctant to broach such issues; another is a genuine confusion on the part of most young children about what behaviour is or is not normal or acceptable; and finally, there is the element of "grooming" so often present in cases of this kind, so that the abuse co-exists with innocent and pleasurable experiences and expressions of love or affection. It is that grooming process which makes it difficult for even the most articulate of children (and often even when grown up) to explain how they feel or felt about the offender and why it was that they made no disclosure at the time.

79 Although the experience of the courts in recent decades has led to a greater understanding of the likelihood of, and reasons for, delay in making a complaint of sexual abuse in childhood, it is not clear to what extent that knowledge forms part of the knowledge and understanding of the wider community. As Anderson J observed in Cook v The Queen [2000] WASCA 78; (2000) 22 WAR 67:

The problem is not that [the jury] will fail to pay regard to [delay] on the question of whether the complainants' story is false or not, but that they may make too much of it on that question. The tendency to be suspicious of the truth of an allegation of sexual misconduct, about which nothing was said at the time, and the tendency to make an assumption against the complainant as to the truth of the allegation in those circumstances, would appear to be the very reason why s 36BD [of the Evidence Act] was enacted. [108]

80 For a trial judge, in the general run of cases involving child sexual abuse, to warn a jury that delay in complaint may mean that the complainant's account is less credible would, perversely, be likely to have an effect precisely the opposite of that sought to be achieved by the principle which underlies judicial warnings. That is, such a warning may tend to reinforce assumptions which are likely to be made by a jury, but which run counter to the experience of the court.

81 So far as authority is concerned, the most relevant High Court decisions appear to be Kilby v The Queen [1973] HCA 30; (1973) 129 CLR 460, Longman, Crofts and Papakosmas v The Queen [1999] HCA 37; (1999) 196 CLR 297.

82 Kilby is authority for the proposition that a trial judge is not entitled to instruct a jury, on a complaint of rape, that the failure of the complainant to complain at the earliest possible opportunity is evidence of consent. However, there was in that case a survey of the legal relevance of consent, and some discussion of the types of directions which ought to be given to juries in relation to delay. In relation to the appropriate direction, Barwick CJ (McTiernan, Stephen and Mason JJ agreeing) said:

It would no doubt be proper for a trial judge to instruct a jury that in evaluating the evidence of a woman who claims to have been the victim of a rape and in determining whether to believe her, they could take into account that she had made no complaint at the earliest reasonable opportunity. Indeed, in my opinion, such a direction would not only be proper but, depending of course on the particular circumstances of the case, ought as a general rule to be given. (465)

83 Even in the case of an adult woman, Barwick CJ noted that "... there may be many reasons why a complaint is not made ..." (472). Similar observations were made by Menzies J (with whom Mason J agreed):

It is, of course, true that in a particular case the explanation of an absence of complaint might be that there was consent, but the correct way to deal with that possibility, in an appropriate case, is for the judge to tell the jury that in considering the evidence that the intercourse was had without consent, it is a circumstance that is to be taken into account that there was no complaint by the woman at the first reasonable opportunity for making it. (475)

84 There is a slight difference of emphasis between the observations of Barwick CJ and of Menzies J, the former expressing the view that a direction of the kind indicated ought to be given as a general rule and the latter apparently taking the view that the direction should be given only "in an appropriate case". The difference would not appear to be of significance, since both their Honours were plainly of the view that there would be cases in which it was either inappropriate or unnecessary to direct the jury in that way. It should be noted that in Kilby, complaint seems to have been seen as of relevance primarily to the complainant's evidence relating to consent, and not as relevant to her evidence concerning whether the acts occurred at all. And, of course, the principle was enunciated in relation to an adult woman, not a child. Importantly, delay was said by Barwick CJ to be something a jury could be told they "could" (not must) take into account.

85 The significance of delay in complaint as affecting the complainant's credibility, as well as its significance in creating a forensic disadvantage for the accused, was considered in Longman. In that case, the majority judgment noted that the absence of complaint to the complainant's mother was one of the circumstances in respect of which it would not have been surprising if it had elicited "some comment" from the trial judge (90). Their Honours went on to note that any comment of that kind would of course require it to be balanced by the statutory direction that there may be "good reasons" why the complainant may hesitate in making or refrain from making a complaint. However, the court was at that point plainly concerned to distinguish between circumstances which might elicit a comment and those which require a warning, and placed the absence of complaint in the former category.

86 Although Deane and McHugh JJ in Longman took a somewhat broader view than the majority of the circumstances which would have required a warning, the absence of complaint at the first opportunity was not a circumstance which either of their Honours appeared to consider required a warning. At 94, Deane J, referring to the time-worn suggestion that complaints of sexual assault may be made with ease, noted the distress and humiliation to which a complainant in a sexual case is commonly subjected, and suggested that those factors would make that suggestion inappropriate, particularly when sexual assault within a family unit is involved. At 99 - 100, his Honour dealt shortly with a suggestion that the absence of contemporaneous complaint required a warning. He noted the complainant's explanation for the delay and went on to observe, in the context of the express statutory recognition, that there may be good reasons for the absence of contemporaneous complaint, "the absence of such contemporaneous complaint could not properly be seen as of itself requiring that the judge be satisfied that a warning was justified" (100). McHugh J did refer (108) to the fact that no complaint was made to the complainant's mother as a circumstance requiring a warning, but that appears to have been in the context of explaining the difficulties created for the accused in testing the complainant's account, the absence of complaint being one of the circumstances which meant that the accused had been denied the opportunity adequately to test the complainant's memory.

87 Crofts is the case upon which the appellant relies. It was a case in which the court was concerned to correct a mistaken view of the Victorian equivalent of s 36BD of the Evidence Act 1906 (WA). The trial judge had directed the jury that they should not, as a matter of law, draw an inference that "because the complainant did not complain about the offences immediately after they happened, that they did not happen or that she was not bothered by them or that she consented". The majority (Toohey, Gaudron, Gummow and Kirby JJ) considered the quoted passage in the sense in which the jury would undoubtedly have understood it; that is, as a direction that the jury was not entitled to rely upon evidence of the lack of complaint to ground an inference that the complainant was not telling the truth (444). That direction, the majority held, was clearly erroneous. Provisions such as s 36BD were not intended to effect a fundamental alteration in the balance of a direction. That was plain from the use of the adverb "necessarily". The court explained that:

Delay in complaining may not necessarily indicate that an allegation is false. But in the particular circumstances of a case, the delay may be so long, so inexplicable, or so unexplained, that the jury could properly take it into account in concluding that, in the particular case, the allegation was false. (448)

88 Pausing there, Crofts appears to be authority for the proposition that delay in complaining may properly found an inference adverse to the complainant's credit only where it is "so long, so inexplicable, or so unexplained". That would tend to suggest that a direction positively suggesting to a jury that it would be proper to reason from a delay in complaint adversely to a complainant's credit, would be relatively rare. That understanding is reinforced by the discussion of the circumstances which in that particular case cast an affirmative duty upon the trial judge to give such a direction. At 445 (and also, it appears, at 450), the majority considered that an affirmative duty to that effect was imposed upon the judge "by the terms in which he had given the earlier instruction" (that is, it was necessary in order to correct the earlier erroneous direction of law). In the same passage, the majority expressed the view that in many cases the failure of a judge to "redress this balance" by directing a jury that delay in complaint could found an inference adverse to the complainant's credit might be protected from appellate interference by the clear intention of the equivalent of s 36BD.

89 However, in the concluding paragraphs of Crofts, the majority said:

Two qualifications to the duty to provide the warning suggested by Kilby may be accepted. The first is where the peculiar facts of the case and the conduct of the trial do not suggest the need for a warning to restore a balance of fairness ... The second is that the warning should not be expressed in such terms as to undermine the purpose of the amending Act by suggesting ... that delay in making a complaint about an alleged sexual offence is invariably a sign that the complainant's evidence is false. (451) (Emphasis added)

90 The two aspects of this passage which are not entirely easy to reconcile with earlier parts of the joint reasons are the reference to a "warning", and the suggestion that there is a general duty to warn, subject only to exceptions in particular cases. The second matter is more one of emphasis, since the underlying proposition would appear to be the general one that a direction is required where the facts of the case call for it, and is not required where the facts do not call for it. The emphasis is of some interest to appellate courts however, since this later passage suggests a somewhat closer scrutiny of a direction which omits a passage of the type suggested by Kilby, than is suggested by the earlier parts of the joint reasons.

91 So far as the reference to a "warning" is concerned, the expression is used again a number of times at 452, which is the concluding page of the joint reasons. Notwithstanding its repetition, I do not think that the court was intending to suggest that there would generally be a duty cast upon a trial judge to "warn" a jury about a complainant's delay in complaining or failing to complain. There are a number of reasons for that view. One is that the expression used a number of times is a "Kilby warning"; however, Kilby speaks in terms of a direction concerning the way in which a jury is permitted to approach evidence of delay in complaint, rather than a warning to the jury that this is an aspect of the complainant's evidence which gives rise to a need for careful scrutiny of it. More importantly, a requirement to give a "warning" about delay in complaint, as the term "warning" is now understood, would appear to be inconsistent not only with Kilby, but also with Longman and with the earlier passages in the joint reasons in Crofts itself, to which I have referred.

92 I would therefore conclude that the court in Kilby was not intending, in the concluding pages of the joint reasons, to suggest that there was a duty to warn, as that term is now understood, concerning delay in complaint. Although in recent years the High Court has insisted upon a sharp distinction between warnings and comments or observations (for example, Azzopardi v The Queen [2001] HCA 25; (2001) 205 CLR 50 [49] - [52]; Mahmood v The State of Western Australia [2008] HCA 1 [16] (Gleeson CJ, Gummow, Kirby and Kiefel JJ)), the distinction does not always seem to have been as clear, or as significant as it now is. For example, the discussion of the duty of the trial judge in Peacock v The King [1911] HCA 66; (1911) 13 CLR 619 appears to use interchangeably the terms "caution", "advise", "direct" and "warn" (eg Griffith CJ at 639, 641 - 643; Barton J at 655 - 656, 657 - 658). In the course of his reasons in that case, Barton J at 655 cited with apparent approval R v Mullins (1848) 3 Cox CC 526, in which, in directing a jury, Maule J said:

I have been in the habit of telling juries here and elsewhere, that they would probably think the evidence of an accomplice required corroboration, but directions of judges so given are not directions in point of law which juries are bound to adopt, but observations respecting facts which judges are very properly in the habit of making, because it is their duty as well as that of counsel, to assist the jury in coming to a just conclusion.

93 In the United Kingdom, as recently as 1986, the House of Lords appeared to distinguish between what had by then become the "accomplice warning" and the obligation "to advise" a jury to proceed with caution where there was material to suggest that witnesses' evidence may be tainted (R v Spencer [1987] UKHL 2; [1987] AC 128, 140 citing R v Beck [1982] 1 WLR 461, 469). It may be that the concluding paragraphs of Crofts can be readily reconciled with the earlier portions of the joint reasons, if the court is understood as having used the expression "warning" in its older, and looser, sense.

94 For the reasons given above, I would understand Crofts as saying no more than that a trial judge's direction, where a case involves a delay in complaining in respect of offences of a sexual nature, must make it clear to the jury that the jury is permitted to consider whether that failure casts doubt upon the complainant's credibility. Of course, the statutory direction required by s 36BD must always be given.

95 Finally, I mention Papakosmas v The Queen in which the High Court discussed, but no member of it directly agreed with, the historical view that late complaint was necessarily relevant to credibility (Gleeson CJ and Hayne J at [12] - [20], Gaudron and Kirby JJ at [49], McHugh J at [76]).

96 Turning to the direction concerning delay in the present case, it fell into two parts. First, her Honour, as a matter of law, explained the jury's task in some detail, directing them that before they found the appellant guilty they must of course be satisfied beyond reasonable doubt that he committed the act complained of . She directed the jury that it was not her Honour's task to form a view as to the complainant's credibility or reliability, but the jury's. She reminded them of the earlier Longman direction which she had given concerning the relevance of the passage of a number of years between the alleged events and the trial. She then said the following:

The acceptance or rejection of the complainant's evidence is entirely within your province, but I am alerting you to matters which you must and which you have to take into account in making that assessment.

It is the case that absence or delay in complaining by a complainant does not necessarily indicate that allegations she makes are false. You will appreciate that there might be good reason why a victim of an offence or offences of the type alleged in this particular case might hesitate in or even refrain from making a complaint about the behaviour. Fear, embarrassment or perhaps a desire not to cause problems within the family, or not to get into trouble oneself, might explain the absence of complaint or delay in complaint. (ts 162 - 163)

97 She then reminded the jury that the complainant agreed that although she said, in her first statement to police, that she had told her mother about the abuse when she was in grade seven, she conceded that in her second statement she had said that she did not tell her mother. She reminded the jury that the reason the complainant gave for the failure to complain and the delay in complaining was "essentially that it was to keep the peace as a result of family pressure". It involved a desire not to lose family support by being banished from the family and she said there were also cultural pressures on her not to complain.

98 In that part of the direction which put to the jury the defence case, her Honour noted that the complainant's credibility was central, and that it was suggested on behalf of the appellant that she was not a reliable witness. In that context, she reminded the jury of the submission made on behalf of the appellant that it would be unusual, if a child had pain following the events the subject of the allegation in count 2, that the child would not mention that to her mother. She reminded the jury of the defence submission that even after the complainant saw a film in grade seven, which clearly suggested that the activity was wrong, she nevertheless did not make allegations in order, she said, to preserve her relationship with her family.

99 Although her Honour did not in terms tell the jury that it was open to them to consider that the complainant's delay in complaining affected her credibility, that must have been blindingly obvious from the way in which she did direct them. As has been noted in a number of cases, including Crofts, the word "necessarily" itself, in the statutory direction, indicates that there may be circumstances in which absence of complaint or failure to complain may tend to indicate that a complainant's story is not true. Further, the jury can only have understood, from the fact that her Honour summarised the evidence concerning delay, and reminded them of the defence case concerning delay, that the delay and the reasons given for the delay were matters which they could consider when evaluating the complainant's evidence.

100 In a case of alleged sexual abuse of an adult complainant, who was a child at the time of the abuse, neither binding authority nor logic nor judicial experience require that a trial judge must necessarily warn a jury about any of the following factors:

(a) inconsistencies internal to the complainant's evidence;

(b) inconsistencies between the evidence of the complainant and that of other witnesses;

(c) depression, self-harm or similar mental states on the part of the complainant;

(d) evidence that the complainant was asleep prior to the commission of the offences;

(e) evidence that the complainant delayed in complaining of the abuse.

101 It may be necessary to warn a jury in relation to any or all of these matters only where there is reason to believe that, in the particular circumstances of the case, the factor gives rise to a danger of which jurors would otherwise be unaware or would not appreciate. Such cases are likely to be rare. The present is not such a case. A judge may, of course, comment in a fair and balanced way upon the evidence, and may refer to the factors listed above, either in the course of such comment or in summarising the respective cases of the accused and the State. The summary in the present case was fair and balanced.

103 Almost the last questions and answers given in cross-examination were as follows:

None of what you have described today in any of the three events that you have described today occurred at all?---Incorrect.

Why do you say you didn't tell anyone about this for so long?---This has been a secret in our family for so many years. There's aunties, nieces, nephews, whole lots of people that have been abused by this one man and for some reason due to what a lot of people have said is cultural right, old men have the right to do this. You may have seen on the news. This man has gone through and acted pervertedly for, you know, years and years and years. As far as I was concerned, I was told that he had that right and then I was told that if I went and told anybody I would shame the family name, black fellows had a hard enough time trying to get respect in the community as it was, so why would you want to make it any harder. So, yeah, I mean, as far as I was concerned, I was pretty much warned and then pretty much banished from my family for going through this. Now, do you want to do that for so many years because I wanted a family. I didn't want to be banished. I didn't want to be on my own in this world, like I have for the last two years. If this really didn't happen, I wouldn't have put myself through this much hell. (ts 79)

104 Counsel did not attempt to interrupt the witness. No objection was taken at the time by counsel for the appellant. Despite the fact that her Honour gave both counsel ample opportunity, both before her direction to the jury and afterwards, to raise issues with her, there was no application that she give the jury a specific direction about this long answer.

105 Despite the apparent absence of any concern about the matter at trial, it is now argued on behalf of the appellant that her Honour had a duty either to discharge the jury of her own motion or, alternatively, to direct the jury not to take into account the complainant's statement that the appellant had abused others. It is alleged that the complainant's "outburst" was unexpected, prejudicial, and non-responsive to the question.

106 Dealing with each of these submissions, there is no evidence before us that the outburst was unexpected, in the sense that it was something which cross-examining counsel for the appellant could not have anticipated from either the depositions which he had or from the complainant's cross-examination to that point. So far as the cultural issues and the complainant's concern about ostracism were concerned, she had already made reference to those concerns during the course of her evidence-in-chief and the earlier cross-examination. Indeed, it had earlier been put to her that she was attempting to make a statement - it seems some sort of political statement - "on behalf of Noongar women". So far as the reference to the abuse of others is concerned, I have already noted that there seems to have been, at least in the statement of the complainant's mother, a reference to nieces and nephews which might have alerted counsel to at least the possibility for such allegations to be raised. Whether counsel did understand, or could have understood, that there might have been other concerns about the appellant's behaviour, we simply do not know.

107 So far as responsiveness to the question is concerned, it seems to me that the whole of the answer was responsive. The complainant was saying, in effect, a number of things about why she had not complained earlier. One was that it was not surprising or unusual that she had not complained, as others also had not complained. Her reason for not complaining was, in part, a mistaken view of Aboriginal culture, but a mistake shared by others. Further, she was saying that she had been told directly that she would bring shame on the family and she was concerned about being "banished".

108 The only irrelevant portion of the answer appears to be the reference to something "seen on the news". It was argued before us that this must have been a reference to some specific incident, and was prejudicial by linking the appellant with it. However, there is no evidence before us of what was "on the news" at the time. There have been reports from many parts of Australia in recent years about the misuse of Aboriginal culture to attempt to excuse criminal conduct. There have been reports of law reform bodies and other inquiries concerning the extent to which culture should be taken into account by way of defence or mitigation. There is nothing before us to indicate which of these very many possible reports the complainant may have been either referring to, or been understood by the jury as referring to.

109 The complainant's response was prejudicial, to the extent that it suggested uncharged acts relating to people other than the complainant. There is, of course, some circularity of reasoning involved in reaching that view. Before the jury could conclude that the appellant had abused others, and therefore before they could, impermissibly, leap to the conclusion that it could therefore be accepted that he had abused the complainant, the jury would have to accept, as a witness of truth, the person making the allegation that others had been abused. That person was the complainant. To find that her answer was unduly prejudicial, one would have to assume that the jury might not, without it, have accepted her evidence as honest and accurate, but that they had nevertheless both accepted her evidence about that particular matter, and been improperly persuaded by it.

110 In any event, it appears to me sufficient, to deal with these grounds, to note that the answer was given in response to a question deliberately asked on behalf of the appellant, and was very relevant to the subject matter of that question. There was no reason for her Honour to discharge the jury. As to a warning about possible impermissible reasoning from that answer, in the context of her evidence as a whole, it formed one of the less significant parts of the complainant's explanation of why she had not complained at an earlier time. Her Honour may well have formed the view that to remind the jury of it, in the course of warning about impermissible reasoning, would have increased rather than reduced the possibility of the jury placing excessive weight upon that portion of the evidence.

111 In respect of the two counts of aggravated sexual penetration without consent and the three counts of indecent dealing, the appellant was sentenced to a total effective term of 6 years' imprisonment with eligibility for parole. It is not necessary to be concerned with the individual sentences imposed, since the appellant's submissions concede, properly, that they were within an appropriate range for offences of the kind committed.

112 The sole ground of appeal is that the sentence was manifestly excessive, having regard to the appellant's advanced age, poor health and limited life expectancy, his previous good character, the length of time since the commission of the offences and the nature of the offences which the appellant characterises as "at the lower end of the scale for offences of their type".

113 Dealing with each of those matters in turn, the appellant will be 81 years of age at his earliest date of eligibility for release. It is accepted that that is an advanced age. However, there was nothing before her Honour, and there is nothing before us, to suggest that he suffers any particular ill-health. He suffers from a variety of common medical conditions, such as hypertension, but there is no medical report indicating that he suffers from any of them to a significant degree, or that any of them is such that imprisonment is likely to bear more harshly upon him than it would upon other offenders. Having regard to the principles enunciated by Steytler P in Gulyas v The State of Western Australia [2007] WASCA 263 [54] (McLure and Miller JJA agreeing), there is no reason to consider that her Honour erred in not giving greater weight to the appellant's age.

114 So far as his previous good character is concerned, it is properly conceded that that is not a factor which can be given a great deal of weight in cases such as the present. That is because in cases of sexual abuse of children, it is often the fact of an offender's apparent good character, reputation, and position within the community, which assists the offender in carrying out the offences undetected. The length of time since the commission of the offences may be of weight in cases where, for one reason or another, a threatened prosecution has taken some time to occur, so that the offences have been hanging over the offender's head as it were, or where there has been clear evidence of rehabilitation during the period in question. However, mere length of time, too, is a factor of limited weight.

115 The contention that the offences were "at the lower end of the scale" for offences of their type, requires closer examination. It is submitted that the offences do not display aggravating features over and above those inherent in this type of offending. It is submitted that breach of trust and disparity of age are features in almost every case of this type, but that the offending did not take place over an unusually long period and there was no force or threat used against the complainant. Reference is made to the complainant's evidence that the appellant was "very gentle". These submissions display the misunderstanding about the primary reasons why offending of this kind is so serious, which the court sought to correct in VIM: see [288] - [293]. It is perfectly true that these offences could have been much worse. They could have been violent, or sadistic, or more numerous. However, it is plain that the learned sentencing judge recognised that fact, since the sentences which she imposed were not in the upper range of sentences for such offending. Nor was the total effective sentence in the upper range of effective sentences imposed in relation to repeated sexual offending upon a small child by a trusted family member (see VIM [301] - [312]).

116 The appellant's submissions in relation to sentence, in the end, appear to turn on a misconception of the role of an appellate court. The written submissions suggest that "... some measure of leniency or mercy should have been afforded". While it is open to sentencing judges to impose a merciful sentence in cases where, for particular reasons, they consider that course desirable (see R v Osenkowski (1982) 30 SASR 212),

a disposition which is "merciful" is, by definition, one which is more lenient than a strict application of sentencing principle would suggest. The function of an appellate court is to ensure that sentencing principle is applied, not to exercise that discretion which can appropriately only be exercised by a judge at first instance who has had the advantage either of presiding over a trial, or at least of gauging the reaction of the offender as he or she comes to be sentenced, in the context of the submissions made to the sentencing judge. I would therefore dismiss the appeal against sentence.

117 The appellant was sentenced on 30 June 2006. The applications for leave to appeal were not made until 29 January 2007. There has been sworn an affidavit by the appellant's solicitor explaining the delay between December 2006, when he was asked to consider the merits of an appeal against conviction and/or sentence, and the date of the filing of the appeal notices. The delay between the time of sentence and the time at which the solicitor was approached by the Legal Aid Commission, is completely unexplained.

118 The respondent did not oppose an extension of time however, and we did not require from the appellant's counsel any submissions in relation to it. That being so, I would extend time, but would dismiss both appeals.

119 BUSS JA: The material facts, the grounds of appeal and the issues in the appeal are set out in the reasons of Wheeler JA. The appellant abandoned ground 2 at the hearing of the appeal, and now relies on grounds 1, 3 and 4. I would dismiss the appeal. My reasons are as follows.

120 In Bropho v The State of Western Australia [2006] WASCA 109, I reviewed several decisions of the High Court in connection with the circumstances in which a trial judge is required to give a warning to the jury, in the context of the trial of a person for a sexual offence, in relation to the evidence of the complainant. The authorities I reviewed included Longman v The Queen [1989] HCA 60; (1989) 168 CLR 79, Crampton v The Queen [2000] HCA 60; (2000) 206 CLR 261, Fleming v The Queen [1998] HCA 68; (1998) 197 CLR 250 and Carr v The Queen [1988] HCA 47; (1988) 165 CLR 314. Also see, in a different context, my review in Cecez v The State of Western Australia [2007] WASCA 260. It is unnecessary, in these reasons, again to review the decisions in question.

121 In my opinion, the authorities require that a trial judge must warn the jury as to the need for careful scrutiny of the complainant's evidence whenever a warning is necessary to avoid a perceptible risk of miscarriage of justice in relation to the assessment of his or her evidence which arises from the circumstances of the case. The necessity for such a warning will depend, in each case, on the particular circumstances, including whether there is a risk that the jury may be unable to appreciate without assistance, or may overlook, some deficiency or danger associated with the complainant or his or her evidence.

122 In the present case, the appellant relied on Crofts v The Queen [1996] HCA 22; (1996) 186 CLR 427 in relation to the complainant's delay in making complaint, and contended that the learned trial judge erred in failing to direct the jury that they could take the delay into account in assessing the complainant's credit. Crofts concerned an accused charged with thirteen counts of committing sexual offences against a child aged between 10 and 16 years. The complainant had delayed in making complaints about the alleged offending for periods of between six months and six years from the dates on which the acts were alleged to have occurred. Section 61 of the Crimes Act 1958 (Vic) provided, relevantly, that on the trial of a person for certain sexual offences, the judge must not warn or suggest in any way to the jury, that the law regards complainants in sexual cases as an unreliable class of witness. Section 61 also contained provisions substantially similar to s 36BD of the Western Australian Evidence Act 1906 (WA). Toohey, Gaudron, Gummow and Kirby JJ referred to the statement of Barwick CJ (with whom McTiernan, Stephen and Mason JJ concurred) in Kilby v The Queen [1973] HCA 30; (1973) 129 CLR 460, 472, in relation to the significance of a want of timely complaint, where his Honour said:

In my opinion, quite apart from the fact that there may be many reasons why a complaint is not made, the want of a complaint does not found an inference of consent. It does tell against the consistency of the woman's account and accordingly is clearly relevant to her credibility in that respect (448).

Their Honours recorded that after the decision in Kilby:

In appropriate cases, juries were told that the absence of timely complaint (if that be their conclusion) was relevant to the credibility of the complainant and a fact to be considered in evaluating the consistency of the complainant's evidence. There might be 'many reasons' to explain why the complaint was not made promptly. It was for the jury to weigh up the significance of the delay in complaining (448).

123 In Crofts, Toohey, Gaudron, Gummow and Kirby JJ approved the following observations of Hunt J (with whom Slattery CJ at CL and Carruthers J agreed) in R v Davies (1985) 3 NSWLR 276, 278, in relation to s 405B of the Crimes Act 1900 (NSW), which is comparable to s 36BD of the Western Australian Evidence Act:

I am unable to see from the statute generally any legislative intention ... to preclude the trial judge - as a matter of common fairness - giving direction on the other side of the coin to those required by that section. Making obligatory the giving of directions that there may be good reasons for the absence of a complaint or for the delay in making it, whether or not such reasons were suggested in the evidence, is certainly to be seen as tilting the balance in favour of the complainant - no doubt because of the difficulty she may often have in articulating those reasons herself - but it should not be seen as standing the law on its head to exclude what in common fairness and common experience should also be taken into account in favour of the accused (449).

The purpose [of s 61] was not to convert complainants in sexual misconduct cases into an especially trustworthy class of witnesses (Longman v The Queen [1989] HCA 60; (1989) 168 CLR 79 at 86 - 87). It was simply to correct what had previously been standard practice by which, based on supposed 'human experience' and the 'experience of courts', judges were required to instruct juries that complainants of sexual misconduct were specially suspect, those complained against specially vulnerable and delay in complaining invariably critical. In restoring the balance, the intention of the legislature was not to 'sterilise' complainants from critical comment where the particular facts of the case, and the justice of the circumstances, suggested that the judge should put such comments before the jury for their consideration (Longman at 86, citing R v Pahuja (1987) 49 SASR 191 at 199, per King CJ; R v Miletic (unreported; Supreme Court of Victoria (Court of Appeal); 9 August 1996) at 20). The overriding duty of the trial judge remains to ensure that the accused secures a fair trial (cf M v The Queen [1994] HCA 63; (1994) 181 CLR 487 at 515, per Gaudron J). It would require much clearer language than appears in s 61 of the Act to oblige a judge, in a case otherwise calling for comment, to refrain from drawing to the notice of the jury aspects of the facts of the case which, on ordinary human experience, would be material to the evaluation of those facts (Longman at 86).

...

Two qualifications to the duty to provide the warning suggested by Kilby may be accepted. The first is where the peculiar facts of the case and the conduct of the trial do not suggest the need for a warning to restore a balance of fairness (R v Murray (1987) 11 NSWLR 12 at 18; M v The Queen [1994] HCA 63; (1994) 181 CLR 487 at 514 - 515). The second is that the warning should not be expressed in such terms as to undermine the purpose of the amending Act by suggesting a stereotyped view that complainants in sexual assault cases are unreliable or that delay in making a complaint about an alleged sexual offence is invariably a sign that the complainant's evidence is false (451).

124 In M v The Queen [1994] HCA 63; (1994) 181 CLR 487, Gaudron J referred to the comments of Hunt J in Davies as to the existence of a 'general rule' to give a direction in accordance with Kilby, and noted:

However, that is not a hard and fast rule and a conviction will not be set aside simply because there has not been a direction of that kind (See Reg v Preval [1984] 3 NSWLR 647; Reg v Murray (1987) 11 NSWLR 12) (514).

Her Honour identified a class of case where the 'general rule' might not apply:

In cases [of sexual assault on a child by a person who has the child's trust and confidence], the victim may be reluctant to resist the offender or to protest and, on that account, reluctant also to complain. As well, a child in that situation may be reluctant to complain from fear that he or she will not be believed, from fear of punishment or, even, fear of rejection by the offender. In my view, the present matter should be approached with those considerations in mind. And if approached on that basis, the argument with respect to the absence of complaint and with respect to the complainant's apparently normal conduct on the weekends in question loses all force (515).

125 In RBK v The Queen [2004] WASCA 216, a child sexual offence case, the trial judge directed the jury in accordance with s 36BD, set out the complainant's evidence as to why there was a delay of two to three years before her complaint, and then said to the jury:

It's a matter for you to look at all the evidence to consider her explanation and this is a matter for you to consider when you're judging whether you should believe her or not when you're assessing her credit [32].

The Court of Criminal Appeal dismissed the appellant's ground of appeal which alleged that this direction did not make clear to the jury the use to which they could put the failure to make a timely complaint. See Malcolm CJ [57] - [58] and Murray J (with whom Wheeler J agreed) [102] - [104].

126 In Christophers v The Queen [2003] WASCA 214, also a child sexual offence case, the trial judge said, in summing up:

The complaint in this case was made a considerable time after the events the subject of the charges. That evidence may be taken into account by you in considering the consistency and therefore the credibility of the complainant....[The fact that she has said something to her sister is] simply a matter which you can and you should take into account in considering the consistency and therefore the credibility of what the complainant has said [57].

The Court of Criminal Appeal rejected the appellant's argument that this direction 'was not balanced by a direction that the jury could nonetheless take the delay into account when assessing the complainant's credibility'. See Hasluck J (with whom Murray J agreed) [68] - [70]. Also see Tanner v The Queen [2001] WASCA 60 [31] - [38].

127 In Cook v The Queen [2000] WASCA 78; (2000) 22 WAR 67, Anderson J (with whom Pidgeon and Wallwork JJ agreed) held that a 'balancing' direction was not necessary, as 'in the context of the trial as a whole, it must have been crystal clear to the jury that they were permitted to use the evidence of delay in complaining as relevant to the complaints' credibility' [113].

128 In Kailis v The Queen [1999] WASCA 29; (1999) 21 WAR 100, Malcolm CJ said in relation to the type of direction in question:

[T]here is much to be said for the proposition that the interests of justice are better served if each case is approached in the light of its own facts, with directions to the jury being fashioned to take account of the assumption, if any, to which those facts might give rise, rather than by reference to an assumption expressed in general terms, but which, clearly, cannot hold good in all cases [125].

Ground 1: inconsistencies in complainant's evidence

129 I agree with Wheeler JA, for the reasons she gives, that a warning was not required, in the particular circumstances of this case, as a result of the alleged inconsistencies in the complainant's evidence.

130 I agree with Wheeler JA, for the reasons she gives, that a warning was not required, in the particular circumstances of this case, as a result of the evidence concerning the complainant's drug use and self-harm.

131 As Wheeler JA has noted, it was the complainant's evidence in this case that on the relevant occasions she had been woken and had then been taken to the appellant's bedroom, where she undressed and got into his bed. The alleged sexual penetration or unlawful and indecent dealing then occurred. I agree with Wheeler JA that this case did not, on any view, involve a child who was asleep or half-asleep at the time when the alleged abuse occurred. A warning was not required, in the particular circumstances of this case, as a result of the possible effect of sleepiness.

132 In my opinion, there was nothing in the facts of this case to suggest that the complainant's age, at the time of the alleged offences, had impaired her ability to give a reliable account of matters in issue. Accordingly, it was unnecessary, in the particular circumstances of this case, for a warning to be given.

Ground 1: the film seen by the complainant at school

133 I agree with Wheeler JA, for the reasons she gives, that it was unnecessary, in the particular circumstances of this case, for a warning to be given about the film seen by the complainant at school.

134 The relevant directions which the learned trial judge gave to the jury in this case are recounted in Wheeler JA's reasons.

135 In my opinion, the learned trial judge directed the jury in terms that would have left them in no doubt that the complainant's delay in making complaint was a factor to be taken into account in evaluating her evidence and considering her credibility. No further direction or warning was required to ensure the appellant received a fair trial.

137 I agree with Wheeler JA, for the reasons she gives, that grounds 3 and 4 are without merit.

138 I agree with Wheeler JA, for the reasons she gives, that the learned trial judge did not make any material error of fact or law in sentencing the appellant. The appellant did not challenge the individual sentences. In my opinion, for the reasons given by Wheeler JA, the total effective sentence imposed was not manifestly excessive (strictly, did not breach the totality principle) having regard to the matters raised by the appellant.

139 I agree with Wheeler JA, for the reasons she gives, that the applications for an extension of time should be granted.

140 I would dismiss the appeal against conviction, and also the appeal against sentence.

141 MURRAY AJA: In relation to the appeal against conviction in this case, I have had the considerable advantage of reading in draft the reasons for decision of Wheeler JA. I agree with her Honour's view that the appeal against conviction should be dismissed. If an extension of time is still required in relation to the appeal against sentence, I would grant it, but I agree with her Honour that the appeal against sentence should also be dismissed.

142 I have nothing to add to the views expressed by Wheeler JA in relation to sentence. As to conviction, I note the valuable review of authority leading to her Honour's identification, with which I respectfully agree, of the overarching principle by reference to which the adequacy of the comments and instructions of the trial judge, generally described in ground 1 as a 'Longman direction', are to be judged. Her Honour's concluding summary is one with which I also respectfully agree.

143 To this analysis, with which I respectfully agree, can be added the discussion by the Court of Appeal (a bench of five judges) in Winmar v The State of Western Australia [2007] WASCA 244, particularly at [21] - [22], and the discussion by Buss JA in Cecez v The State of Western Australia [2007] WASCA 260. In the latter case (which was heard after Winmar was argued, but before judgment in Winmar was delivered), I also attempted an analysis of authorities concerned with diverse areas of evidence and the directions to be given by a trial judge before identifying an applicable general principle, which I expressed at [220] - [221] as follows:

It amounts to this, I think. As part of the trial judge's function to direct the jury as to the law and make such observations about the facts as are required to secure a fair trial, in accordance with the general statement of the duty by the High Court in Alford v Magee [1952] HCA 3; (1952) 85 CLR 437 at 466, when the real issues in the case are identified, the trial judge will be obliged to consider, in relation to matters of fact and evidence, whether any, and if so what, instruction should be given to the jury if a perceptible risk of a miscarriage of justice is to be avoided. The instruction to be

given should be one related specifically to the issue identified and the circumstances of the case. It should be an instruction designed to assist the jury to understand how to approach a particular body of evidence, any limitations there may be as to how it may be used and as to how they may reason, having regard to the evidence, upon the issue of the guilt of the accused.

An instruction or warning will be required where the experience of trial judges reveals a danger to the process of achieving a fair trial which the jury may not appreciate. If that is not the case, no special instruction or warning is required to avoid a miscarriage of justice and the trial judge will not err if no special instruction is given, but the jury is left to their fact-finding task guided by their own wisdom and experience. In my view, it may be added that unless the circumstances positively call for a special instruction or a warning, the judge should do no more than remind the jury of the issues truly raised by the case which will require their decision.

144 Wheeler JA discusses the particular criticisms made by the appellant of the directions given by the trial judge. Her Honour discusses cases which bear on those particular areas of fact and which have provided guidance as to whether, and if so in what terms, directions are necessary if the risk of a miscarriage of justice is to be avoided. In my opinion, all these cases can be seen to exemplify the application of the central principle, and are related to the particular circumstances of the cases in question. For me, however, it is sufficient that I add to the reasons of Wheeler JA only the comment that I too could detect nothing in the matters covered in the trial judge's direction relative to ground 1, or omitted from it, which would cause this court to hold that the trial judge in any way fell into error.

145 I would only add one further brief comment on this aspect of the case. In introducing the discussion of decided authority upon which I embarked in Cecez, I said, at [206], that in the context of that discussion I would draw no distinction between a comment on the evidence and a warning, citing the terms of the Criminal Procedure Act 2004 (WA), s 112, which requires a trial judge to instruct the jury on the law applicable to the case and make any observations about the evidence that the judge thinks necessary in the interests of justice.

146 It is, I think, a debate about semantics. There will be occasions when the judge's observations about the evidence involve directions of law. The important thing will be, having regard to the point at issue, whether the jury are to be given a direction binding on them as a matter of law, or whether they are to be warned of the danger of acting upon the evidence in a particular way, a matter of comment in respect of which they may accept or reject the guidance offered by the trial judge. In Mahmood v The State of Western Australia [2008] HCA 1 at [16], a majority of the High Court said:

The distinction between a direction and a comment by a trial judge is referred to in Azzopardi v The Queen [2001] 205 CLR 50 at 69 - 70 [49] - [52]. It reflects the fundamental division of functions in a criminal trial between the judge and the jury. The distinction is important. Telling a jury that they may attach particular significance to a fact, or in this case suggesting that other evidence may be considered of greater weight, is comment. Because it is comment it may be ignored by the jury, a matter about which the jury should be told. A direction, on the other hand, may contain warnings about the care needed in assessing some evidence or the use to which it may be put. A direction is something which the law requires the trial judge to give to the jury and which they must heed.

147 The point is, of course, that whether comment in the sense of the word as used in Mahmood, or a direction is required will depend upon the particular subject matter to be dealt with and the question how the perceived risk of a miscarriage of justice is to be alleviated.